Assessment of nutrition intervention for patients with...

Assessment of nutrition intervention for patients with unresectable pancreatic cancer in a fish oil

supplement trial – does it make a difference?

Wendy Louise Davidson

BSc Qld, GradDipNutDiet QUT

Centre for Health Research School of Public Health

Queensland University of Technology

submitted for Master in Applied Science (Research)

2003

ii

Keywords

Nutrition, carcinoma of the pancreas, pancreatic neoplasms, weight loss, cachexia,

quality of life, dietetics, prognosis, survival

iii

Abstract

Severe and progressive weight loss is a feature of pancreatic cancer but it has been

unclear whether dietetic intervention can improve patient outcomes. This study is a

post hoc analysis of the extensive data available from the multicentre BH80 Cancer

Cachexia trial to examine how patients with unresectable pancreatic cancer respond

to intensive dietetic intervention. The BH80 study compared an n-3 fatty acid

enriched oral supplement with an isonitrogenous, isocaloric supplement given over

an eight week period. Additional qualitative data was also collected and examined

for the patients recruited by the Australian sites.

The aims were to determine whether achieving weight stabilisation is an appropriate

goal of nutrition intervention for people with unresectable pancreatic cancer; to

identify determinants of weight stabilisation; and to describe nutrition-related

features of patients recruited by the Australian sites, prior to and during intensive

nutrition intervention.

Data from 107 patients for whom weight change data over an eight week nutrition

intervention period was available, was divided into weight losing (> 1kg lost) and

weight stable (≤ 1 kg lost) patients. Group survival duration (Kaplan Meier log rank

test) and global quality of life (EORTC QLQ-C30 global health status/quality of life)

were compared. Variables including energy intake, BMI, presence of pain, nausea

and vomiting, and appetite loss, C reactive protein and stage of disease were also

compared to determine predictors of weight stability using logistic regression

analysis.

This study demonstrates that weight stabilisation is not only achievable for some

patients in the short term, but it is also associated with improved outcomes. Those

patients who were able to stabilise their weight after eight weeks of oral nutrition

support lived longer from baseline and reported better quality of life than those who

continued to lose weight. Weight stabilisation is therefore a reasonable and

worthwhile goal for this patient group.

iv

Patients who halted their weight loss had greater energy and protein intakes than

those who continued to lose weight. They also had lower BMI’s and were less likely

to have pain, appetite loss or an acute phase protein response at baseline. Female

gender and the absence of nausea or vomiting at baseline, were independently

associated with an increased likelihood of weight stabilisation.

This study also confirms reports that patients with unresectable pancreatic cancer

experience a wide range of symptoms that impact negatively on their nutritional

status. Common issues include loss of appetite, pain, nausea and vomiting, taste

alterations, pancreatic insufficiency, fatigue and depression. Nutrition intervention

needs to be individualised and intensive to adjust to disease progression and to

address the patient specific barriers to intake. Further research is recommended such

as randomised studies of oral nutrition support versus standard care with the goal of

weight stabilisation, methods for predicting survival time and whether optimal

management of nausea and vomiting improves outcomes in patients with advanced

cancer.

Table of Contents

Keywords ii

Abstract iii

Table of Contents v

List of Tables ix

List of Figures x

List of Abbreviations xii

Acknowledgments xv

1 INTRODUCTION 1

1.1 Overall aim 1

1.2 Specific aims 3 1.2.1. Aim 1 3 1.2.2. Aim 2 3 1.2.3. Aim 3 4

2 LITERATURE REVIEW 5

2.1 Pancreatic cancer 5 2.1.1 Incidence 5 2.1.2. Risk factors 6 2.1.3. Symptoms and features 7 2.1.4. Prognostic features 10 2.1.5. Treatment 11

2.2 Cancer associated weight loss 12 2.2.1. Prevalence of weight loss in cancer 12 2.2.2. Effects of weight loss on patient outcomes 14 2.2.3. Metabolic factors affecting weight loss 15 2.2.4. Resting energy expenditure 17 2.2.5. Acute phase protein response 19 2.2.6. Proinflammatory cytokines 20 2.2.7. Catabolic factors 21 2.2.8. Pancreatic insufficiency 21

2.3 Causes of reduced dietary intake 22 2.3.1. Appetite and taste changes 26 2.3.2. Gastrointestinal symptoms 29 2.3.3. Pain and depression 31

2.4 Effect of nutrition support 32 2.4.1. Studies of nutritional support in cancer 32 2.4.2. Research on oral nutrition support 35

2.5 Evidence-based goals for nutrition support 37 2.5.1. Current dietetic practice in oncology 37 2.5.2. Increased survival time 40 2.5.3. Improved quality of life 40

vi

2.5.4. Nutritional requirements 43 2.5.5. Body weight & lean body mass 44 2.5.6. Functional improvements 45

2.6 Summary 46

3 METHODS – AIMS 1 &2 47

3.1 BH80 trial 47 3.1.1. BH80 study design 47 3.1.2. Subjects 49 3.1.3. Ethical issues 50 3.1.4. Anthropometry 50 3.1.5. Nutrient intake 51 3.1.6 Biochemical analysis 52 3.1.7. Quality of Life 52 3.1.8. Appetite & Satiety 53 3.1.9. Functional status 54 3.1.10. Survival duration 54

3.2 Preparation of data for secondary analysis 54 3.2.1 Rationale for pooling subjects 54 3.2.2 Creation of weight losing or weight stable groups 56 3.2.3 Inclusion criteria 56 3.2.4 Confirming normal distribution 57

3.3 Data analysis 57 3.3.1 Outcomes of dietetic intervention 57 3.3.2 Determinants of weight stabilisation 59

4 RESULTS – AIMS 1 & 2 67

4.1 Description of patients excluded from the study 67 4.1.1 Reasons for excluding patients from the study 67 4.1.2 Comparison of included & excluded patients 68 4.1.3 Description of included patients 70

4.2 Outcomes of nutrition intervention 71 4.2.1. Survival Duration 71 4.2.2. Quality of Life 73

4.3 Determinants of weight stabilisation 73 4.3.1. Comparison of the WL and WS groups 75 4.3.2. Determinants of weight stability - multivariate analysis 77

4.4 Summary 79

5 METHODS - AIM 3 81

5.1 Comparison of qualitative and quantitative methods 81

5.2 Subjects 84

5.3 Data from the BH80 study 84 5.3.1. Accuracy of dietary recording methods 85 5.3.2. Nutrient intake 86 5.3.3. Anthropometry 87

vii

5.3.4. Quality of life questionnaire 88 5.3.5. Appetite 88 5.3.6. Functional status 88 5.3.7. Statistical methods 88

5.4 Sub-protocol 89 5.4.1. Ethical issues 89 5.4.2. Structured Interview 90 5.4.3. Recent Food Intake Report 90 5.4.4. Semi-structured Interview (Narrative) 90

6 RESULTS – AIM 3 93

6.1 Description of Australian participants 93 6.1.1. Comparison with BH80 participants. 94

6.2 Body weight changes 95

6.3 Changes in dietary intake 98 6.3.1. Energy intake. 98 6.3.2 Macronutrients 103 6.3.3. Protein intake 103

6.4 Nutrition supplements 105 6.4.1. Supplement intake 105 6.4.2. The role of supplements 106

6.5 Changes in appetite and taste perceptions 107 6.5.1. Appetite changes 107 6.5.2. Early satiety 108 6.5.3. Taste changes 109

6.6 Symptoms and quality of life 110 6.6.1. Pain 111 6.6.2. Nausea & vomiting 111 6.6.3. Asthenia 111 6.6.4. Constipation/diarrhoea 112 6.6.5. Xerostomia 112

6.7 Summary 113

6.8 Case Studies 113 6.8.1. Case 1 - Mrs A 113 6.8.2. Case 2 - Mr B 115

7 DISCUSSION 119

7.1 Study design issues 119

7.2 Outcomes for weight stabilisation 121 7.2.1. Survival duration 122 7.2.2. Quality of life 123 7.2.3. Summary - Aim 1 123

7.3 Determinants of weight stabilisation 124 7.3.1. Independent determinants 125 7.3.2. Determinants significant by bivariate analysis 128 7.3.3. Characteristics which were not predictive 134

viii

7.3.4. Summary - Aim 2 137

7.4 Additional nutrition related factors 139 7.4.1. Bodyweight changes 139 7.4.2. Nutrient intake 140 7.4.3. Taste & appetite changes 141 7.4.4. Symptoms and quality of life 142 7.4.5. Summary - Aim 3 143

7.5 Implications for nutrition intervention 143 7.5.1. Appropriate level of nutrition intervention 144 7.5.2. Individualised and intensive intervention 146

8 CONCLUSION 151

APPENDICES 155

Appendix 1: BH80 study sites 155

Appendix 2: BH80 Inclusion/Exclusion Criteria 156

Appendix 3: BH80 patient information and consent 158

Appendix 4: EORTC QLQ-C30 (version 3) 162

Appendix 5: Karnofsky Performance Status Scale 164

Appendix 6: Staging for Pancreatic Cancer 165

Appendix 7: Sub-protocol patient information and consent (ver 2) 166

Appendix 8: Identification coding of patients from the Australian sites 169

Appendix 9: T=0 structured interview (ver 1) 170

Appendix 10: Recent food intake report (ver 1) 172

Appendix 11: T=9/10 semi-structured interview/narrative 173

Appendix 12: Notation used in transcription of interviews 175

Appendix 13: Published paper (in press) - Clinical Nutrition 176

Appendix 14: Oral presentation – DAA 21st National Conference 185

Appendix 15: Poster presentation – DAA 20th National Conference 186

Appendix 16: Descriptive statistics for study participants including data to test for normality187

Appendix 17: Splett’s cascade model of outcome measures 188

REFERENCES 189

ix

List of Tables Table 2.1 Prevalence of symptoms reported by patients with pancreatic cancer referred to a

palliative care service 8 Table 2.2 Categories of degree of weight loss 14 Table 2.3 Summary of studies describing metabolic changes in pancreatic and other gastrointestinal

cancers. 18 Table 2.4 Summary of cross-sectional studies describing the dietary protein and energy intake of

patients with cancer, as well as baseline intakes from intervention studies. 24 Table 2.5 Factors proposed as contributing to cancer anorexia 26 Table 2.6 Summary of studies of oral nutrition support in cancer 33 Table 2.7 Summary of nutrition-intervention studies in pancreatic cancer 36 Table 2.8 Recommendations regarding energy requirements for cancer patients. 42 Table 2.9 Recommendations regarding protein requirements for cancer patients. 44 Table 3.1 Outcome variables of the BH80 study. 47 Table 3.2 Major components of BH80 study supplements 48 Table 3.3 Schedule of data collection in BH80 study 49 Table 3.4 Description of variables used for assessing outcomes for weight stabilisation in patients

with unresectable pancreatic cancer receiving nutrition intervention over eight weeks (Aim 1) 59

Table 3.5 Description of baseline variables used to identify determinants of weight stability in

patients with unresectable pancreatic cancer receiving nutrition intervention over eight weeks (Aim 2) 61

Table 3.6 Description of Week 8 variables used to identify determinants of weight stability in

patients with unresectable pancreatic cancer receiving nutrition intervention over eight weeks (Aim 2) 62

Table 3.7 Plasma EPA as a percentage of total phospholipids, with and without fish oil or EPA

supplementation. 65 Table 4.1 Comparison of baseline characteristics of patients with unresectable pancreatic cancer who

were included in analysis with those who were excluded (continuous variables). 68 Table 4.2 Comparison of baseline characteristics of patients with unresectable pancreatic cancer who

were included in analysis with those who were excluded (categorical variables). 69 Table 4.3 Comparison of baseline characteristics of 107 weight losing (WL) and weight stable (WS)

patients with unresectable pancreatic cancer receiving eight weeks of nutrition intervention. 74

Table 4.4 Comparison of self-reported pre-illness body mass index (BMI) for weight losing (WL)

and weight stable (WS) pancreatic cancer patients 75 Table 4.5 Comparison of plasma eicosapentanoic acid (EPA) distributions for weight losing and

weight stable pancreatic cancer patients following eight weeks of nutrition intervention 76

x

Table 4.6 Comparison of characteristics of weight losing (WL) and weight stable (WS) pancreatic

cancer patients at baseline and week 8 76 Table 4.7 Comparison of energy intake distributions at week 8 for weight losing and weight stable

pancreatic cancer patients 77 Table 4.8 The effect of different variables on the likelihood of patients with pancreatic cancer being

weight stable following eight weeks of nutrition intervention (n=83) – logistic regression analysis. 78

Table 5.1 Comparison of quantitative and qualitative research methods 82 Table 6.1 Comparison of baseline characteristics of patients with unresectable pancreatic cancer

from the Australian sites versus the combined sites. 95 Table 6.2 Changes in body weight (kg) over time, of patients with unresectable pancreatic cancer

recruited into the BH80 study from Australian sites 97 Table 6.3. Macronutrient contributions to total energy for patients with unresectable pancreatic cancer

recruited into the BH80 study from Australian sites. 102 Table 6.4 Total protein intake for patients with unresectable pancreatic cancer recruited into the

BH80 study from Australian sites 104 Table 6.5 Self-reported lack of appetite scores for patients with unresectable pancreatic cancer at the

Australian sites immediately prior to baseline, week 4 and week 8 assessments over eight weeks of nutrition intervention 108

List of Figures Figure 1.1 Model representing the three aims of this study 2 Figure 2.1 Frequency and degree of weight loss in patients with advanced cancer 13 Figure 2.2 Otterey’s algorithm of optimal nutritional intervention 38 Figure 3.1 Model representing the first two aims of this study of nutrition intervention in

patients with unresectable pancreatic cancer 55 Figure 4.1 Reasons for exclusion of some BH80 participants from secondary analysis 67 Figure 4.2 Model representing the first aim of this study of nutrition intervention in patients

with unresectable pancreatic cancer 71 Figure 4.3 Comparison of survival time from baseline for weight losing (n=44) and weight

stable pancreatic cancer patients – Kaplan Meier survival plot 72 Figure 4.4 Model representing the second aim of this study of nutrition intervention in patients

with unresectable pancreatic cancer 73 Figure 5.1 Model representing Aim 3 81 Figure 6.1 Percentage of pre-illness body weight lost by each Australian patient with

unresectable pancreatic cancer until final weight measurement 96

xi

Figure 6.2 Change in energy intake between baseline and week 4 for patients with unresectable pancreatic cancer recruited into the BH80 study from Australian sites. 99

Figure 6.3 Energy intake at baseline (T0) and week 4 (T4) for patients with unresectable

pancreatic cancer recruited into the BH80 study from Australian sites who survived less than ten weeks from baseline (Group 1) 100

Figure 6.4 Energy intake at baseline, week 4 and week 8 for patients with unresectable

pancreatic cancer recruited into the BH80 study from Australian sites who survived more than ten weeks from baseline (Group 2). 101

Figure 7.1 Model representing Aims 1 & 2 of this study of nutrition intervention in

unresectable pancreatic cancer 138 Figure 7.2 Clinical pathway for medical nutrition therapy in unresectable pancreatic cancer

148

xii

List of Abbreviations

APPR Acute phase-protein response

BIA Bioelectrical impedance analysis

BMI Body Mass Index

BW Body weight

CI Confidence interval

CRP C reactive protein

CT Chemotherapy

DV Dependent variable

ECOG Eastern Cooperative Oncology Group

EI/BMR Ratio of energy intake to basal metabolic rate

EORTC European Organisation for Research and Treatment of Cancer

EORTC QLQ-C30 EORTC quality of life questionnaire

EPA Eicosapentaenoic acid

EQ5D EuroQol quality of life questionnaire

Global QoL EORTC QLQ-C30 Global health status/QoL scale

IL-1 Interleukin-1

IL-6 Interleukin-6

INFγ Interferon gamma

IV Independent variable

KPS Karnofsky Performance Status

LAS Linear analogue scale

LBM Lean body mass

LMF Lipid mobilising factor

ONS Oral nutrition support

OR Odds ratio

PAH Princess Alexandra Hospital

PC Pancreatic cancer

PG-SGA Patient Generated – Subjective Global Assessment

PIF Proteolysis inducing factor

xiii

Pt Patient

QoL Quality of life

RCT Randomised control trial

REE Resting energy expenditure

RT Radiotherapy

T0 Baseline data collection time point

T4 Week 4 data collection time point

T8 Week 8 data collection time point

TBW Total body water

TNF Tumour necrosis factor

TWH The Wesley Hospital

WL Weight losing (loss of >1kg over eight weeks)

WS Weight stable (loss of ≤ 1kg over eight weeks)

xiv

The work contained in this thesis has not been previously submitted for a degree or diploma at any other higher education institution. To the best of my knowledge and belief, the thesis contains no material previously published or written by another person except where due reference is made. Signed: _______________________________ Dated: ________________________________

xv

Acknowledgments

To the people with pancreatic cancer and their carers who gave so generously of their time and to the Cancer Cachexia Study Group which made access to this data possible. To my supervisors, Dr Susan Ash and Prof Sandra Capra, whose enthusiasm, insight and dedication to the advancement of nutrition and dietetics are inspirational. To Dr Diana Battistutta for her valuable statistical advice and to the members of the Nutrition Practice Research Group. To my husband, Gary, my family, Jacob, Elliott, Dylan, Tegan and Paavi, and to my parents, Barbara and Collin Davidson, for their love and support throughout the process.

1 Introduction

Severe and progressive weight loss is a feature of adenocarcinoma of the pancreas

and this wasting process is distressing to the patient and their family. It is unclear

whether dietetic intervention can improve patient outcomes. There are no known

cures currently available for patients with unresectable pancreatic cancer and

expected survival time is short.

Medical management focuses on palliative measures, such as the reduction of pain,

to improve the patient’s quality of life. Nutrition therapy for advanced cachexia has

been considered by some doctors to be futile – an ineffective and possibly unethical

use of resources. Is dietary intervention appropriate in the palliative care setting for

patients who are losing weight, especially if we have no reliable method for helping

patients to restore that lost weight?

A rare opportunity was available through the multicentre BH80 Cancer Cachexia

trial, to examine how patients with unresectable pancreatic cancer respond to

intensive dietetic intervention. The trial was controlled to determine the effect of

adding fish oil to a protein and energy-dense oral supplement, therefore both arms of

the study received a form of dietary intervention. All patients were initially losing

weight, as a criteria for entrance to the trial, and were expected to experience

ongoing weight loss - the usual disease process.

This study is a post hoc analysis of the extensive data available from the BH80 trial,

pooling both arms of the trial to analyse outcomes for this large, relatively

homogeneous group of patients with an advanced cancer. More detailed data has also

been collected and examined for the patients recruited by the Australian sites.

1.1. Overall aim

To determine appropriate goals for effective nutrition intervention for weight losing

patients with unresectable pancreatic cancer, which would assist dietitian

nutritionists in providing appropriate clinical services (Fig 1.1).

Introduction 2

Figure 1.1 Model representing the three aims of this study

Aim 1 QoL

Weight change

Survival

Aim 2

Weight

change Energy

intake T8BMI T0

PS T0

Stage of disease

Symptoms T0

Energy intake T0

ONS

Aim 3

Weight

change Intake T8

ONS

pain, n&v, app,

taste, depression

Symptoms T0

Intake T0

BMI T0

PS T0

Stage of disease Psychosocial factors

Survival

Patient’s

perceptions

QoL

beyond T8 E, Pr, F,

CHOfood texture

E, Pr, F,

CHO

Key to symbols:

intervention independent variable dependent variable

ONS – oral nutrition support, QoL – quality of life, BMI – body mass index, PS – performance status, n&v –

nausea & vomiting, app – appetite, E – energy, Pr – protein, F – fat, CHO – carbohydrate, T0 – baseline, T8 –

week 8.

Introduction 3

1.2. Specific aims

1.1.1 Aim 1

To determine whether achieving weight stabilisation is an appropriate goal of

nutrition intervention for people with unresectable pancreatic cancer.

Hypotheses

a) Patients who lose no more than 1kg over eight weeks of nutrition intervention

will have increased survival time compared to those who lose more than 1kg.

b) Patients who lose no more than 1kg over eight weeks of nutrition intervention

will have better quality of life than those who lose more than 1kg.

If these hypotheses are supported then weight stabilisation could be considered a

measure of “effective nutrition intervention” and therefore an appropriate goal for

medical nutrition therapy for this patient group.

1.1.2 Aim 2

To identify determinants of weight stabilisation in patients with unresectable

pancreatic cancer.

Hypotheses

a) Patients who would be likely to benefit from nutrition intervention, (ie.

weight stabilisation is feasible), can be identified using a variety of features

on presentation – higher BMI, greater energy intake, better performance

status, less pain, less loss of appetite and less nausea and vomiting.

b) Patients who would be likely to benefit from nutrition intervention, (ie.

weight stabilisation is feasible), will be those who can achieve greater energy

intake by week eight.

Introduction 4

1.1.3 Aim 3

To describe nutrition-related features of patients with unresectable pancreatic cancer

recruited by the Australian sites, prior to and during intensive nutrition intervention.

a) To describe changes to body weight, dietary intake, taste perceptions and

appetite and the role of dietary supplements.

b) To explore patients’ perceptions regarding changes in weight, appetite and

quality of life through case studies.

This section uses both quantitative and qualitative data in order to provide a clearer

picture of the illness as experienced by the patient, to assist in providing patient-

focussed dietetic care.

The methods and results for Aims 1 and 2 (Chapters 3 & 4) will be considered

separately from those for Aim 3 (Chapters 5 & 6).

2 Literature Review

2.1 Pancreatic cancer

Pancreatic cancer is a particularly challenging disease for dietitian nutritionists

because of its severe impact on nutritional status. The pancreas is the site of

production of digestive enzymes and disease often spreads to adjacent parts of the

gastrointestinal tract. Weight loss is severe and patients have a range of nutrition

related symptoms.

Most cases of pancreatic cancer are not suitable for potentially curative surgery

(Lillemoe, Yeo, & Cameron, 2000), therefore the focus of care is palliation – to

minimise symptoms and maximise quality of life.

2.1.1 Incidence

Pancreatic cancer is the fourth leading cause of cancer mortality in the USA with

more than 28,000 deaths/year (Greenlee, Hill-Harmon, Murray, & Thun, 2001). In

Australia it is the fourth most common cause of cancer death for women and fifth

most common cause for men. 1788 Australians (916 men and 872 women) were

diagnosed as having pancreatic cancer in 1999 (Australian Institute of Health and

Welfare, 2002).

Exocrine tumours account for 95% pancreatic cancers, the majority of which are

adenocarcinomas. The remaining 5% are islet cell tumours, most of which are

benign.

Crude incidence rates have been increasing for both men and women in Australia.

For both sexes the crude rates were 7.3 cases diagnosed per 100,000 population in

1983 and 9.4 per 100,000 in 1999 (Australian Institute of Health and Welfare, 2002).

This increase is likely to be due largely to the ageing population because the age

adjusted rates for Australians have been relatively stable.

Literature Review 6

2.1.2 Risk factors

The incidence of pancreatic cancer rises steadily with age, with 84% of Australian

cases in 1999 occurring in people over 60 years of age. The disease is uncommon in

people younger than 40 years (Australian Institute of Health and Welfare, 2002;

Greenlee et al., 2001; Lillemoe et al., 2000). Men are more likely to develop cancer

of the pancreas than women. Lillemoe has reported a narrowing of gender

differences in incidence rates in the USA over recent years (Lillemoe et al., 2000).

Australian age standardised data (Australian Institute of Health and Welfare, 2002)

show a slight increase in incidence for females in the 1980’s but this appears to have

steadied in the 1990’s.

Inherited genetic alterations may be a factor in 5 – 10% cases of pancreatic cancer

(Howe & Conlon, 1997). Occupational exposure to certain chemicals may also

increase the risk (Hoppin et al., 2000).

Chronic pancreatitis is associated with an increased risk of pancreatic cancer,

however this may be due to a failure to adjust for common risk factors such as

smoking. Evidence linking diabetes mellitus and pancreatic cancer is also

inconsistent (Lillemoe et al., 2000; Silverman et al., 1999). Diabetes may develop as

an early symptom of pancreatic cancer rather than being a causal factor.

The best established risk factor for pancreatic cancer is cigarette smoking with about

30% cases estimated to be due to smoking. This increases the risk of developing

pancreatic cancer with odds ratios for current smokers between 1.3 and 5.5 (Howe &

Burch, 1996; Silverman et al., 1994). Between 1989 and 1995, the male incidence

rate for smoking-related cancers in Australia fell by an average of 0.9% per year,

while the rate for females rose by 1.3% per year [Australian Institute of Health and

Welfare, 1998 #289].

The evidence for other dietary or lifestyle risk factors is less clear. Although

increased risk of pancreatic cancer has been associated with consumption of alcohol

and coffee, neither were associated with an increased risk based on food frequency

questionnaires from the Health Professionals Follow-Up Study and the Nurses

Literature Review 7

Health Study (Michaud, Giovannucci, Willett, Colditz, & Fuchs, 2001a). No

association with coffee consumption was found in a case-control study of 436 cases

(Silverman et al., 1998). Folate appears to be protective (Kim, 1999; Stolzenberg-

Solomon et al., 1999).

Positive relationships have been shown between body mass index and pancreatic

cancer (Michaud et al., 2001b) (Silverman et al., 1998). In the study by Michaud,

obese individuals were 72% more likely to be diagnosed with pancreatic cancer than

those who were neither obese nor overweight, however physical activity had a

protective effect for overweight and obese people. These findings have led to

suggestions that chronic hyperinsulinaemia may be involved (Kaaks & Lukanova,

2001).

2.1.3 Symptoms and features

Most patients with pancreatic cancer experience multiple symptoms (Krech &

Walsh, 1991; Lillemoe, 1998; Ottery, 1996a). Krech et al reported an average of 11

symptoms per patient in their study of 39 people with pancreatic cancer referred to a

palliative care service at the Cleveland Clinic (Krech & Walsh, 1991). The

percentage of patients reporting various symptoms is listed in Table 2.1. The

following physical features were also present; 44% patients were cachectic on

presentation, 33% had serum albumin less than 3.5mg/dL, palpable abdominal mass

in 36%, ascites 23%, jaundice 21% and metastasis to at least one major organ was

documented in 64% of patients.

Pain, being easily fatigued, and anorexia are commonly found in advanced cancer.

They were consistently among the ten most prevalent and clinically important

symptoms in a study of the prevalence and severity of symptoms in 1000 patients

regardless of tumour site (Donnelly, Walsh, & Rybicki, 1995).

Signs and symptoms that accompany pancreatic cancer are particularly likely to

affect nutrient intake or utilisation due to the location of the tumour. Many symptoms

relate to compression or invasion of related physical structures, eg. approximately

50% of people with pancreatic cancer will experience jaundice during the course of

Literature Review 8

the disease. This is usually resolved by biliary bypass surgery or endoscopic insertion

of a biliary stent. Duodenal obstruction with nausea and vomiting, is usually a late

manifestation of a tumour in the head of the pancreas. Liver metastases and

retroperitoneal implants are the most common sites of distant spread, therefore

ascites and oedema frequently occur in the final weeks of life (Sauter & Coleman,

1999).

Table 2.1 Prevalence of symptoms reported by patients with pancreatic cancer

referred to a palliative care service (n=39). Symptom Percentage of patients

pain 82%

anorexia 64%

early satiety 62%

xerostomia 54%

sleep problems 54%

weight loss 51%

easily fatigued 46%

weakness 41%

nausea 41%

constipation 41%

depression 39%

dyspepsia 36%

vomiting 31%

hoarse 26%

taste changes 23%

bloating 23%

belching 23%

dyspnoea 21%

dizzy 21%

oedema 21%

cough 18%

diarrhoea 15%

hiccoughs 15%

itch 15%

dysphagia 3%

Source: Krech 1991(Krech & Walsh, 1991)

Literature Review 9

Pancreatic cancer may also cause early satiety - a sense of feeling too full to continue

eating that occurs much sooner than the individual would normally expect

(Gallagher-Allred, 1989). Delayed gastric emptying, with no mechanical obstruction,

is also common in advanced cancer (Ottery, 1996a) and could contribute to early

satiety.

Pancreatic insufficiency is reported to be a common problem (Ottery, 1996a),

however there is little objective evidence in the literature. Reports on the prevalence

of diabetes conflict, ranging from 15-20% (Sauter & Coleman, 1999) to 70-80% of

patients (Permert et al., 1993).

Anecdotal reports of taste alterations such as a metallic taste to food as well as an

aversion to the smell of food during preparation (Ottery, 1996a), and aversions to

“proteinaceous foods such as beef and pork” (Gallagher-Allred, 1989) can be found.

There are no reported studies however investigating in detail the taste changes that

occur in pancreatic cancer.

Loss of weight has been reported in more detail than the other features of

unresectable pancreatic cancer with general agreement that weight loss is common,

severe and progressive. Wigmore et al observed changes in the nutritional status of

20 patients with unresectable pancreatic cancer from diagnosis to death (Wigmore,

Plester, Richardson, & Fearon, 1997b). Eighty-five percent of patients had lost

weight at the time of diagnosis, and all had lost weight as death approached, with

median BMI dropping from 24.9 kg/m2 premorbidly to 20.7 kg/m2 at diagnosis and

17.7 kg/m2 near death. Patients had a median weight loss of 14.2% at diagnosis (10-

20% IQR) increasing to 24.5% near death (11.5-29.7% IQR).

Marked and progressive weight loss was also demonstrated in a study of 102 patients

with unresectable pancreatic cancer (Falconer et al., 1995). Median weight loss was

11% at diagnosis and 20% within six weeks of death.

In summary, some features of pancreatic cancer, such as the marked and progressive

weight loss, have been described in detail, but for others there is little agreement or

evidence in the literature.

Literature Review 10

2.1.4 Prognostic features

Unfortunately the nonspecific nature of early symptoms, including anorexia, weight

loss, abdominal discomfort and nausea, may contribute to a delay in diagnosis and

therefore make curative treatment options less likely (Lillemoe et al., 2000).

The prognosis for people with pancreatic cancer is known to be poor, but it is

difficult even for experienced clinicians to predict individual survival times. The

ability to better predict survival would assist in clinical decision making.

The five year relative survival rates for pancreatic cancer over the past three decades

(adjusted for normal life expectancy) have been 3-4% (Greenlee et al., 2001) with

fewer than 20% of patients surviving one year from diagnosis (Lillemoe, 1998).

Advanced age, male gender, liver metastases and large tumour diameter have been

reported as unfavourable prognostic factors (Lillemoe et al., 2000; van den Bosch et

al., 1994).

Weight loss has been shown to be a negative prognostic indicator for a range of

cancers, however differences in survival duration for pancreatic cancer patients with

and without weight loss in two large studies (Andreyev, Norman, Oates, &

Cunningham, 1998; DeWys et al., 1980) did not reach statistical significance. This

may reflect the very short expected survival times for this disease (median survival

times of 12 vs 14 weeks in DeWys’ study) and the confounding effect of oedema and

ascites on body weight measurement.

In a study of 102 patients with unresectable pancreatic cancer, percentage weight loss

at diagnosis was inversely related to survival duration (Falconer et al., 1995). Weight

loss however did not have independent prognostic value after adjusting for stage of

disease and presence of an acute phase protein response (APPR). The presence of an

APPR, defined as elevated C reactive protein (CRP) levels, was demonstrated to be a

useful independent prognostic indicator. Median survival for patients with an APPR

was 66 days compared to 222 days for those without an APPR.


Length of survival for a group of 25 pancreatic cancer patients who presented to a

palliative care service within one month of diagnosis, was analysed according to the

presence of a range of reported symptoms. Survival time was unaffected by all

symptoms except dyspnoea. Those with poorer performance status had a reduced

length of survival (Krech & Walsh, 1991). The importance of the ability to perform

activities of daily living, also sometimes referred to as functional status, is discussed

in more detail in section 2.5.6.

A prognostic score has been developed from pooled data from 1020 patients with

unresectable pancreatic cancer to assist in determining the most appropriate

treatment – endoscopic management for those who are expected to have shorter

survival and palliative surgery for those with better prognostic scores (Terwee et al.,

2000). Prognosis was found to be poorer for those with metastases (Stage IV

disease), pain or weight loss at diagnosis, while the presence of jaundice was a

relatively good prognostic sign. Older men faired worse than women and younger

men. The prognostic score however is derived from only gender, age (for men) and

the presence of metastases, along with a hospital-specific six month mortality rate.

2.1.5 Treatment

Tumour resection with curative intent may be possible for only 10-20% of patients

(Cooperman, Kini, Snady, Bruckner, & Chamberlain, 2000). In addition to the

patient being fit for surgery, there must be no vascular invasion or metastases and the

tumour needs to be small in size. Surgical intervention usually takes the form of a

Whipple procedure in which all or part of the pancreas, part of the stomach,

duodenum, gallbladder and part of the common bile duct are removed. This

complicated surgery can have significant side effects including delayed gastric

emptying, and a lengthy recovery period.

Resection is not an option for the majority of patients, and, because the prognosis for

unresectable pancreatic cancer is so poor, medical care focuses on the management

of symptoms to improve quality of life (Alter, 1996; Ellison, Chevlen, Still, &

Dubagunta, 2002).


Non-curative treatment is directed primarily at relieving obstructive jaundice,

duodenal obstruction and pain, as well as the management of individual issues such

as nausea and steatorrhoea. Obstruction of the common bile duct or duodenum

frequently occur and so palliative endoscopic or open surgical procedures are often

required. In a study of the natural history of unresectable pancreatic cancer, 19 out of

the 20 patients underwent either palliative bypass procedures or endoscopic stenting

(Wigmore et al., 1997b).

Treatment strategies for unresectable cancer vary considerably, with palliative

chemotherapy and/or radiotherapy offered at some centres and not others due to lack

of consistent evidence that benefits to the patient in terms of palliation or survival

time outweigh side effects, patient burden and cost. A number of randomised clinical

trials to determine the optimal use of drugs such as gemcitabine are underway

(Neoptolemos et al., 2003).

In some centres, patients with locally advanced tumours, previously considered

unresectable, are now undergoing preoperative, adjuvant chemoradiotherapy

(Cooperman, 2001; Wanebo et al., 2000). Other areas of research into the treatment

of pancreatic cancer include gene therapy, immunotherapy, gut hormone therapy and

the use of anti-angiogenesis factors.

2.2 Cancer associated weight loss

2.2.1 Prevalence of weight loss in cancer

Unintentional weight loss occurs commonly in cancer but more frequently with some

tumour types than others, making it inappropriate to pool such groups in nutrition

support studies.

An analysis of more than 3000 patients with cancers unsuitable for curative surgery

or radiation, who participated in chemotherapy trials by the Eastern Cooperative

Oncology Group (DeWys et al., 1980) demonstrated the differences in frequency of

pretreatment weight loss for different cancer types (Figure 2.1).


The highest frequency of weight loss occurred in patients with pancreatic or gastric

cancers, with approximately 60% of patients having lost at least 5% of their body

weight prior to commencing chemotherapy. These results are supported by a study of

1555 patients with locally advanced or metastatic gastrointestinal cancer preparing

for chemotherapy in which 70% of patients with cancer of the oesophagus, stomach

or pancreas, but only a third of the patients with colorectal cancer, had weight loss

(Andreyev et al., 1998). Figures for prevalence of weight loss at diagnosis of

unresectable pancreatic cancer have ranged from 51 – 85% (Krech & Walsh, 1991;

Wigmore et al., 1997b).

Figure 2.1 Frequency and degree of weight loss in patients with advanced cancer

Source: (DeWys et al., 1980)

Weight lost over previous six months (two months for pancreatic cancer)

NHL, non-Hodgkin’s lymphoma; Leukaemia, acute nonlymphocytic leukaemia

thompsdc

This figure is not available online. Please consult the hardcopy thesis available at the QUT Library.


Definitions for “weight loss” vary in these studies, but it is clear that gastrointestinal

cancers in general, and pancreatic cancer in particular, are associated with

considerable weight loss.

2.2.2 Effects of weight loss on patient outcomes

Weight loss is not only a common feature of cancer, it can be substantial, and for at

least 20% of cancer patients, death is attributed to the effects of malnutrition (Langer,

Hoffman, & Ottery, 2001). Studies involving weight losing cancer patients typically

deal with median weight losses of 15-20% (O'Gorman, McMillan, & McArdle, 1998;

Tchekmedyian et al., 1992). In Wigmore et al’s descriptive study of patients with

unresectable pancreatic cancer, 15% of patients had lost more than 20% of their

usual body weight at diagnosis and 60% had done so just prior to death (Wigmore et

al., 1997b).

Blackburn classified unintentional weight loss of 15-20% as severe, even if it occurs

over a six month period (Blackburn, Bistrian, Maini, & al, 1977). Blackburn’s

categories, published 25 years ago are still widely referred to in nutrition support

literature (Bloch, 1993; Langer et al., 2001; McCallum & Polisena, 2000). The

definitions for significant and severe weight loss are shown in Table 2.2. This degree

of loss, some of which will be from lean tissue stores, would be expected to impair

physiological functions such as respiratory function.

Table 2.2 Categories of degree of weight loss

Source: (Blackburn et al., 1977)

thompsdc

This table is not available online. Please consult the hardcopy thesis available at the QUT Library.


Many studies have demonstrated that malnourished patients have poorer health

outcomes;

reduced survival time (Andreyev et al., 1998; DeWys et al., 1980; Persson,

Sjoden, & Glimelius, 1999)

reduced tolerance of, and response to chemotherapy (Andreyev et al., 1998;

DeWys et al., 1980; Ottery, 2000)

poorer surgical outcomes (van Bokhorst de van der Scheuren et al., 1997;

Windsor & Hill, 1988)

impaired immune response and increased risk of infection (Chandra, 1995;

Lesourd, 1997)

delayed wound healing (Demling & De Santi, 1998)

increased health care costs (Ottery, 1996b)

reduced functional status, impaired muscle function and strength (DeWys et al.,

1980; Lopes, Russell, Whitwell, & Jeejeebhoy, 1982; O'Gorman et al., 1998;

Ollenschlager, Thomas, Konkol, Diehl, & Roth, 1992)

decreased quality of life (Larsson, Akerlind, Permerth, & Hornqvist, 1995;

O'Gorman et al., 1998; Ovesen, Hannibal, & Mortensen, 1993b)

Weight loss has been shown to be a negative prognostic indicator for cancers in

general, although for pancreatic cancer the association is less clear as was discussed

in section 2.1.4. Similarly, in a study of patients about to receive chemotherapy for

advanced cancer, decreasing weight was correlated with decreasing performance

status, except for patients with pancreatic and gastric cancer (DeWys et al., 1980).

Weight loss with resulting malnutrition is never desirable, however weight gain may

not be beneficial in all situations. For example, weight gain is thought to be an

adverse prognostic factor for women with breast cancer receiving adjuvant therapy,

possibly as the result of increases in circulating hormone levels from increased body

fat stores (Nixon, 1996). There is no evidence however that tumours of non-

reproductive organs such as the pancreas, respond in this way.

2.2.3 Metabolic factors affecting weight loss


Cancer associated weight loss appears to involve not only reduced nutrient intake but

also increased requirements and barriers to the utilisation of nutrients. Nutrition

support studies (Chlebowski, Palomares, Lillington, & Grosvenor, 1996; Evans et al.,

1987; Ovesen, Allingstrup, Hannibal, Mortensen, & Hansen, 1993a) suggest that

increasing intake alone does not reverse this process. The field of altered metabolism

in cancer is an active one with in vitro, animal and human studies continuing to

expand knowledge of the multiple processes that appear to be involved (Barber,

Ross, Voss, Tisdale, & Fearon, 1999b; Haslett, 1998; Inui, 1999; Langer et al., 2001;

Loprinzi et al., 1999; McMillan et al., 1999).

Cancer associated weight loss differs from starvation in that the normal adaptive

mechanisms to conserve body protein do not function (Brennan, 1977). Adipose

tissue is lost to a greater degree in starvation with fat stores utilised to replace

glucose as the principal energy source. This spares lean body tissue, minimising loss

of skeletal muscle. In contrast, the wasting process of cancer involves the loss of

both fat and lean body mass. Skeletal muscle in particular, is lost, with relative

conservation of visceral protein stores (Tisdale, 1997a).

Anaerobic glycolysis by the tumour results in increased lactate levels in the blood.

This is converted back to glucose in the liver but results in a net loss of ATP. This

Cori cycle accounts for 20% of glucose turnover in the healthy person but can be

50% in cancer (Tisdale, 1997b). Loss of muscle mass is increased as a result of

activation of the ATP-ubiquitin-dependent proteolytic pathway (Tisdale, 2002).

Several mediators have been proposed including cytokines (tumour necrosis factor

and interleukin-6) and proteolysis inducing factor (PIF).

Alterations in metabolic rate (Wigmore, Plester, Ross, & Fearon, 1997c),

proinflammatory cytokines (Falconer, Fearon, Plester, Ross, & Carter, 1994;

Karayiannakis et al., 2001) and catabolic factors such as PIF and lipid mobilising

factor (LMF) (Hirai, Hussey, Barber, Price, & Tisdale, 1998; Todorov et al., 1996;

Wigmore et al., 2000b) have been reported in weight losing pancreatic cancer

patients.


Cancer associated weight loss is often referred to in the literature as “cachexia”.

While there have been many studies investigating cachexia, it remains a poorly

defined syndrome with the result that the specific causes of the substantial weight

loss of the cachectic patients referred to in studies is unclear. For example, while

weight loss occurring in a group of patients with acute myelogenous leukaemia

undergoing chemotherapy was determined to be primarily due to inadequate oral

intake (Ollenschlager, Konkol, & Modder, 1988) this may not be the only reason for

weight loss in hypermetabolic patients with pancreatic cancer receiving no active

treatment. Unfortunately, in studies of weight loss in cancer, the patient groups are

often not homogeneous for cancer type or presence of cachexia, making results

difficult to interpret.

In 1977 Brennan defined cancer cachexia as “the syndrome of emaciation,

debilitation and malnutrition that accompanies cancer” (Brennan, 1977). More recent

descriptions incorporate the expanding knowledge of the cachectic process - “cancer

cachexia is a complex, multifactorial syndrome that results from a reduction in food

intake, a variety of metabolic abnormalities (including hypermetabolism) or more

often a combination of the two” (Fearon & Moses, 2002). Cachexia has also been

described in a number of chronic diseases such as HIV-AIDS, renal and liver failure,

and arthritis, as well as in relation to the aging process. There are, however, no

diagnostic criteria for cancer cachexia.

Some of the metabolic abnormalities and nutrition-related symptoms that occur in

weight losing cancer patients are discussed below. Issues that relate specifically to

chemotherapy, radiotherapy or surgical interventions have not been included.

2.2.4 Resting energy expenditure

Studies of resting energy expenditure (REE) in weight-losing cancer patients have

failed to show consistent alterations. Both increases and decreases have been

demonstrated which may reflect the many different causes of weight loss in these

patients. Neither advanced stage of disease nor tumour bulk has been consistently

associated with hypermetabolism although successful resection has been shown to

ameliorate it.


Table 2.3 Summary of studies describing metabolic changes in pancreatic and other gastrointestinal cancers.

Author, Year, Country

Patient Group Parameters Results

Falconer 1994 UK

21 PC pts (mean wt loss 18%) 16 controls

CRP (>10mg/L = APPR) REE (indirect calorimetry) BCM, FFM (BIA), weight

Higher REE* for pts than controls (25.9 vs 19.4 kcal/kg BW/d) Higher REE* for pts with APPR than without (28.7 vs 23.8 kcal/kg BW/d) * whether as a proportion of BW, BCM or FFM

Falconer 1995 UK

102 PC pts (median wt loss 11%)

CRP (>10mg/L = APPR) survival duration

Shorter median survival for pts with APPR than without (66d vs 222d)

Wigmore 1995 UK

16 PC pts (median wt loss 17%) 17 controls

CRP, REE (indirect calorimetry) FFM (BIA), weight

Higher REE for pts than controls (25.6 vs 19.2 kcal/kg BW/d). Also significant as a proportion of FFM. Mean CRP for pts was 51 mg/L – not detectable in controls

Wigmore 1997 UK

35 PC pts CRP (≥10mg/L = APPR) REE (indirect calorimetry) predicted REE (Schofield eqn)

Measured REE greater than predicted REE Higher REE for pts with APPR than without (26.6 vs 23.3 kcal/kg BW/d)

O’Gorman 1998 UK

119 GI cancer pts Grp1 = wt loss >5% (n=97) (median wt loss 17%) Grp2 = steady wt (n=22)

CRP (>10mg/L = APPR) albumin, weight

More pts with an APPR in Grp1 than Grp2 (71% vs 27%) CRP greater in Grp1 than Grp2 (29 vs <6 mg/L) Lower albumin levels in Grp1 than Grp2 (39 vs 42 g/L)

Wigmore 2000 UK

55 PC pts Grp1 = PIF(+ve) (n=44) Grp2 = PIF (-ve) (n=11)

REE (indirect calorimetry) PIF (urine), CRP, weight loss (median)

No difference in REE between grps Greater weight loss in Grp1 than Grp2 No association between PIF status & CRP level

PC, unresectable pancreatic cancer; GI, gastrointestinal; REE, resting energy expenditure; CRP, C-reactive protein; PIF, proteolysis inducing factor; BW, body weight; FFM, fat-free mass; BCM, body cell mass; BIA, bioelectrical impedance analysis; APPR, acute phase protein response


Increased REE measurements have been demonstrated in some patients with

pancreatic and other gastrointestinal cancers as summarised in Table 2.3, but they

were not raised for all patients. REE measurements ranged from 16.6 to 34.0 kcal/kg

body weight/day in one study of 55 pancreatic cancer patients (Wigmore et al.,

2000b). Mean REE’s were demonstrated to be 30% higher for patients with

pancreatic cancer than healthy controls in two studies (Falconer et al., 1994;

Wigmore et al., 1995). Most of the studies available are from the one research group

and patient population.

2.2.5 Acute phase protein response

The acute phase protein response (APPR) is an inherent part of the body’s immune

response. If, however, this inflammatory response continues for a lengthy period of

time, as in cachexia, the ongoing production of acute phase proteins can be a drain on

lean body stores. In the absence of dietary protein 10g skeletal muscle protein would

be required to produce just 1g of acute phase proteins due to the higher proportion of

aromatic amino acids in the latter (Reeds, Fjeld, & Jahoor, 1994).

C-reactive protein (CRP) is one of the positive acute phase reactants, ie. serum levels

increase during an APPR. Other positive acute phase proteins include ceruloplasmin,

alpha-1-antitrypsin, haptoglobin and fibrinogen. Proteins that decrease in

concentration during an inflammatory response (negative acute phase reactants)

include albumin, prealbumin and transferrin.

CRP is commonly used as a marker of the APPR. Raised CRP has been associated

with increased REE, reduced survival time and weight loss in pancreatic cancer

patients (Falconer et al., 1994; Falconer et al., 1995; O'Gorman et al., 1998;

Wigmore et al., 1997c) as summarised in Table 2.3. Poorer appetite and performance

status has been shown in gastrointestinal cancer patients with a marked inflammatory

response (CRP >30mg/L) compared to those with no inflammatory response (CRP

<10mg/L) (O'Gorman et al., 1998). Elevated CRP levels have also been

demonstrated to be associated with hypermetabolism and weight loss in non-small

cell lung cancer (Staal-Van, Dentener, Schols, Buurman, & Wouters, 1995).


CRP is often undetectable in healthy controls (Wigmore et al., 1995). In some studies

CRP has been used as a continuous variable while in others it has been converted

into a dichotomous variable (levels raised or not). Cutoffs for raised CRP levels have

been variously defined as 5 mg/L (McMillan et al., 1994), 10 mg/L (Falconer et al.,

1995; O'Gorman et al., 1998; O'Riordain, Falconer, Maingay, Fearon, & Ross, 1999;

Staal-Van et al., 1995; Wigmore et al., 1995) or even separated into <10 vs >30mg/L

as described above (O'Gorman et al., 1998). The limits of detection for the analytical

methods used in earlier studies appear to have determined these cut-offs.

Short-term use of ibuprofen (a cyclo-oxygenase inhibitor) (Wigmore et al., 1995) and

eicosapentaenoic acid (EPA) (Wigmore, Fearon, Maingay, & Ross, 1997a) in cancer

patients with an APPR have resulted in reductions in CRP levels and REE,

suggesting that at least some forms of hypermetabolism may be able to be attenuated.

2.2.6 Proinflammatory cytokines

Cytokines are small non-structural proteins that can be produced both by tumour and

host cells. Cytokines such as tumour necrosis factor–alpha (TNF-α), interleukin-1

(IL-1), interleukin-6 (IL-6) and interferon gamma (IFNγ) have been shown to cause

anorexia and inhibition of lipoprotein lipase and to promote the APPR, and therefore

have been suggested as mediators of cachexia.

Cytokines have been examined extensively in human, animal and in vitro studies.

The measurement of systemic levels, however, can be misleading because cytokines

act in a paracrine manner rather than endocrine manner. High levels of cytokines

acting at a local level may not be detectable in a blood sample.

In a comparison of pancreatic cancer patients with and without an APPR, there was

no difference in IL-6 levels between the two groups and TNF was not even detected

in the serum. In vitro production of TNF and IL-6 from peripheral blood monocytes

however, was greater in the group with an APPR (Falconer et al., 1994). TNF-α has

been demonstrated in the serum of pancreatic cancer patients (present in 37% of 63


patients) and levels were inversely correlated with body weight, BMI and albumin

levels (Karayiannakis et al., 2001).

2.2.7 Catabolic factors

Proteolysis inducing factor (PIF), a 24kDa sulphated glycoprotein has been reported

in the urine of weight losing pancreatic cancer patients (Todorov et al., 1996) and has

been shown to induce muscle catabolism in mice. This effect was attenuated by EPA.

The presence of PIF in pancreatic cancer patients has been shown to be associated

with both greater weight loss and faster rate of weight loss although REE was no

greater than for the patients without PIF (Wigmore et al., 2000b).

Lipid mobilizing factors produced by both animal and human tumours, are believed

to act by stimulation of cyclic AMP formation (Tisdale & Beck, 1991) and may

explain the increased levels of free fatty acids in the circulation of cachectic patients.

The role of a range of tumour or host-derived substances in cancer associated weight

loss is not yet clear. Cortisol, insulin-glucagon ratios, leptin, neuroendocrine

hormones and neurotransmitters (neuropeptide Y, serotonin, orexin) are some of the

molecules also being investigated.

2.2.8 Pancreatic insufficiency

The frequency with which pancreatic insufficiency occurs in pancreatic cancer is

unclear. A diagnosis of fat malabsorption is usually made clinically, based on the

presence of diarrhoea and oily, foul-smelling stools. Alternative signs of steatorrhoea

in patients with pancreatic cancer, however, may be crampy, abdominal pain and

bloating due to the constipation that can accompany narcotic analgesic use (Ottery,

1996a).

Perez et al (Perez, Newcomer, Moertel, Go, & Dimagno, 1983) demonstrated fat

malabsorption in nine out of twelve patients with ductal pancreatic cancer, and

protein malabsorption in six. Coefficients of malabsorption correlated with loss of


body weight (r=0.59, p<0.05) and pancreatic enzyme replacement significantly

improved moderate to severe malabsorption in this group.

A study of 21 patients with cancer of the head of pancreas and occlusion of the

pancreatic duct, randomised to an eight week period of weekly dietary counselling

and high dose pancreatic enzyme supplementation, resulted in weight stability

compared to placebo plus counselling (Bruno, Haverkort, Tijssen, Tytgat, & van

Leeuwen, 1998).

2.3 Causes of reduced dietary intake

Many of the symptoms experienced by patients with cancer would be expected to

hinder food intake. Abdominal fullness, taste changes, vomiting and mouth dryness

were the symptoms found to be associated with weight loss in a group of patients

with a variety of unresectable cancers (Grosvenor, Bulcavage, & Chlebowski, 1989).

The presence of a tumour, as well as its treatment, have both a psychological and a

physiological impact on the control of food intake (Padilla, 1986).

Symptoms (“a person’s perception of an abnormal physical, emotional or cognitive

state” (Anderson & Burckhardt, 1999)), and signs (observable conditions such as

vomiting) are both referred to within the simpler term “symptoms” for the purpose of

this work.

It is difficult to determine the relative contributions to cancer associated weight loss

of altered metabolism, impaired absorption and reduced intake. A study of 35

pancreatic cancer patients showed a small but significant increase in REE for patients

with an APPR, however, it was proposed that the greatest contribution to energy

deficit for these patients was reduction in food intake (Wigmore et al., 1997c).

Anorexia is commonly reported in people with cancer but the effect this has on their

food intake is not easy to quantify in a meaningful way. The limitations of methods

for estimating energy requirements, even in weight stable, healthy individuals, have

been described (Reeves & Capra, 2003). Energy requirements in cancer are discussed


further in section 2.5.4 and studies describing the dietary intake of cancer patients are

summarised in Table 2.4.

There does not appear to be any consensus, or even much debate in the literature

regarding the most appropriate way to express the energy intake of weight losing

individuals with cancer. Energy intake can be considered in absolute terms, as a

proportion of estimated requirements, or as a proportion of body weight, lean body

mass or body cell mass. If weight losing patients lost predominantly fat stores then

using usual body weight might be justified, but in cachexia, skeletal muscle is also

lost, presumably reducing basal requirements. The simple method of calculating

kJ/kg actual body weight can at least be applied consistently without assumptions

about proportions of lean or fat tissue lost.

A small study of inpatients with advanced cancer and anorexia who had lost at least

15% body weight, showed significantly lower energy intakes compared to patients

hospitalised for other conditions, both in absolute terms and as a proportion of their

body weight (Levine & Morgan, 1998).

Bruera et al also found that for patients with advanced solid tumours awaiting

chemotherapy, malnourished patients consumed significantly less energy and protein

than well-nourished ones (Bruera, Carraro, Roca, Cedaro, & Chacon, 1984). Weight

loss was one of the components of the method used for categorising these patients by

nutritional status, so it is reasonable to assume that body weight was less for the

malnourished group, however this data was not provided. The two groups did differ

for “mean caloric balance” (calculated from Mayo Clinic energy requirement tables)

with the malnourished patients in negative balance. It is not clear, however, on which

form of body weight the calculations were based.

Grosvenor et al, however, found no significant difference in energy intake (kcal/kg

BW/day) between patients with a range of unresectable cancers with and without

weight loss of more than 5% (Grosvenor et al., 1989). Ovesen et al similarly reported

no difference in energy intake for patients who had lost more than 5% body weight

over the previous three months compared to those who had not (Ovesen et al.,

1993a), however detailed data was not provided.


Table 2.4 Summary of cross-sectional studies describing the dietary protein and energy intake of patients with cancer, as well as baseline

intakes from intervention studies. Author, Year, Country

Study population Measure of intake Outcomes Comments

Levine 1998 UK

Grp1 = 10 pts with advanced cancer, anorexia & >15% wt loss over 1 yr (no recent CT/RT) Grp2 = 20 non-cancer pts

Pts selected 3 day menus (wide variety choices & flexible meal times) Tray items weighed pre & postmeal

Grp1 had significantly lower E & Pr intakes than Grp2 but no difference in macronutrient proportions. 6.0 vs 9.5 MJ/d (~1400 vs 2300 kcal/d) 0.1 vs 0.14 MJ/kgBW/d 0.9 vs 1.2 g Pr/kg/d

all hospitalised

Wigmore 1997 UK

35 pancreatic cancer pts Grp1 = CRP <10mg/L (n=16) Grp2 = CRP ≥10 mg/L (n=19)

4 day food records Grp2 had significantly lower median energy intake than Grp1 (1553 vs 2192 kcal/d)

Giacosa 1996 Italy

281 malnourished advanced cancer pts

3 day food record Mean energy intake 1207 kcal/d Mean protein intake 51g/d

Limited detail

Ovesen 1993 Denmark

105 cancer pts (lung, ovary, breast) pre-CT Grp1 = intervention Grp2 = controls

3 day diaries predicted BEE (Harris Benedict equation)

Baseline intake (Grp1, Grp2) 7.7 ± 1.9, 7.5 ± 2.3 MJ/d 1.4, 1.3 x pred. BEE 61 ± 18, 58 ± 19 g Pr/d 0.9, 0.9 g Pr/kg/d

Est. req EI = 1.5-1.7 x BEE Pr = 1.0-1.2 g/kg/d

Ollenschlager1992 Germany

29 leukemia pts hospitalised for CT

selected menus daily offered 1-2g Pr/kg & 30-50kcal/kgIBW (EI measure not described)

Mean EI during weeks of; - weight loss = 23.2 kcal/kgIBW/d - weight stability = 30.9 kcal/kgIBW/d - weight gain = 39.3 kcal/kgIBW/d


Table 2.4 cont.

Author, Year, Country

Study population Measure of intake Outcomes Comments

Grosvenor 1989 USA

254 pts with unresectable cancer Grp1 = ≥5% wt loss (n=170) Grp2 = steady wt (n=84)

24hr recall & food frequency

No difference in mean energy intake between grps (31.4 vs 30.5 kcal/kg CBW/d). But approached significance (p=0.053) when expressed relative to UBW (28.9 vs 33.5 kcal/kg UBW/d)

Evans 1987 USA

180 advanced lung & colorectal cancer pts & CT (intervention study)

24 hr recall & food frequency predicted BEE (Harris Benedict equation)

Baseline intake 62-70% targetted caloric intake (TCI)

if<5%wt loss TCI = 1.7 x predBEE if >5%wt loss TCI = 1.95 x predBEE

Bruera 1984 Argentina

72 advanced cancer pts pre-CT Grp1 = malnourished (n=35) Grp2 = well nourished (n= 37)

7 day diet recall nutritional status based on anthropometry, albumin & creatinine height index (% std)

Grp1 had significantly lower E & Pr intakes than Grp2. 1214 ± 360 vs 2088 ± 607 kcal/d 58 ± 15 vs 91 ± 14 g Pr/d

Grp1 = negative caloric balance, Grp2 = positive (based on Mayo Clinic tables)

DeWys 1977 USA

40 cancer pts 5 day food record 90% pts had energy intake <30 kcal/kgBW/d

EI, energy intake; Pr, protein; CT, chemotherapy; RT, radiotherapy; BW, body weight; CBW, current body weight; IBW, ideal body weight; UBW, usual body weight;

pred BEE, predicted basal energy expenditure; CRP, C-reactive protein


Overall, many studies have shown a lower energy intake for weight losing or

malnourished patients with advanced cancer compared to people with and without

cancer who are weight stable. Comparisons are made difficult, however, because

there appears to be no consensus on the most meaningful way of reporting energy

intake when body composition changes.

2.3.1 Appetite and taste changes

Loss of appetite for food is frequently reported by patients with advanced cancer

(Donnelly et al., 1995) and, according to Ottery, is “the most common symptom

contributing to poor nutrient intake in most cancer patients” (Ottery, 1995). Anorexia

can also have a detrimental effect on the patient’s quality of life because of the loss

of the pleasure of eating and sharing of food with friends and family. Appetite and

the ability to eat were found to be the most important factors in the “physical well

being” aspects of a quality of life study of 126 patients receiving chemotherapy or

radiotherapy (Padilla et al., 1983).

Table 2.5 Factors proposed as contributing to cancer anorexia

Dysgeusia Cytokine production

Dysosmia Gut obstruction

Food aversions ↑ production of serotonin

Pain ↑ lactate

Psychogenic anorexia ↓ response to insulin

Treatment induced Altered serum free fatty acid profile

Altered serum amino acid profile Ref: (Nitenberg & Raynard, 2000; Ottery, 1995; Tisdale, 1997a)

Many factors are considered to be involved in the development of cancer anorexia

(Table 2.5) but few effective strategies are available for reducing its impact

(Nitenberg & Raynard, 2000). Dexamethasone and megestrol acetate have been

shown to increase both appetite and energy intake in patients with advanced cancer

(Loprinzi et al., 1999; Tchekmedyian et al., 1992) but each of these drugs has


limitations. Improvements in appetite with dexamethasone are only short-term, and,

like all corticosteroids would actually increase protein requirements. The use of

progestational agents has not been shown to result in increased lean body mass

(MacDonald, Easson, Mazurak, Dunn, & Baracos, 2003; Ottery, 1995). Dronabinol,

a synthetic cannabinoid, has been shown to increase appetite in cancer patients

(Nelson, Walsh, Deeter, & Sheehan, 1994), but no randomised controlled trials have

been conducted (Walsh, Nelson, & Mahmoud, 2003). Cyproheptadine and hydrazine

sulphate have failed to show a benefit as orexigens (Chlebowski et al., 1996; Murphy

& Von Roenn, 2000).

One of the difficulties in studying the prevalence of anorexia or the efficacy of

treatments, is the subjective nature of the symptom. Sensations of hunger can be

rated but not measured objectively (Silverstone, 1982). Tools used to measure

appetite in cancer have included categorical appetite scales (Macqueen & Frost,

1998; Tchekmedyian et al., 1992) and visual or linear analogue scales (LAS)

(O'Gorman et al., 1998) which are also incorporated into “loss of appetite” subscales

within quality of life (QoL) questionnaires (Aaronson et al., 1993; O'Gorman et al.,

1998; Padilla et al., 1983).

A simple question on appetite “How would you compare your appetite now with

what it was before your present illness?” (increased, the same, slightly reduced,

moderately reduced or markedly reduced) contributed as strongly to the prediction of

overall prognosis as three other elements (KPS, ECOG-performance status and

physician estimate of survival) in 700 patients with advanced lung and colorectal

cancer (Sloan et al., 2001). The Good/Bad/Uncertain Index (GBU) was developed

from these as an easily applied stratification tool for oncology clinical trials to reduce

the chance of sicker patients being allocated to one study arm than another.

Some scales include separate elements of appetite as recommended by Hill (Hill &

Blundell, 1982), for example to distinguish the physical sensations of “hunger” from

“appetite” (the desire to eat). Linear analogue scales have been shown to be reliable

in the examination of the effect of appetite suppressing drugs on healthy individuals

(Silverstone, 1982). Hunger ratings and the amount of food consumed, however, are


not always well correlated (Hill & Blundell, 1982; Mattes, 1990; Raben, Tagliabue,

& Astrup, 1995) making the validity of the tools difficult to establish.

So although loss of appetite is common and is frequently reported as a concern by

patients and their carers, it is difficult to measure and we have few strategies for

overcoming its effect on food intake.

The sensation of taste can be altered (dysgeusia), increased (hypergeusia),

diminished (hypogeusia), or appear to be absent (ageusia). Taste changes occur with

aging due to thinning of the epithelium of taste cells, reduced blood supply and

decline in the sense of smell related to changes in olfactory receptors, as well as the

impact of medications and medical conditions (Schiffman, 1997; Strohl, 1992).

Taste changes are frequently reported by cancer patients and have been associated

with weight loss (DeWys, 1977; Grosvenor et al., 1989). Patients undergoing

chemotherapy or radiation therapy to the oral cavity may experience diminished taste

due to loss of taste buds as these have a high rate of cell turnover. Learnt food

aversions in response to an unpleasant experience such as nausea from chemotherapy

have been well described (Bernstein, 1999). Taste changes are also common in

patients who have not received chemotherapy and can appear quite early in the

course of the disease (Grosvenor et al., 1989; Nielsen, Theoglides, & Vickers, 1980).

Suggested mechanisms have been zinc deficiency, circulating substances (eg. some

amino acids have a bitter taste), altered volume and nature of saliva, depression, and

food aversions developed as a conditioned response to effects of the disease itself.

Many of the medications used in oncology, such as antidepressants, antibiotics and

anti-inflammatory drugs, have been associated with taste changes (Schiffman, 1997).

As with anorexia, taste changes are very difficult to measure. The threshold at which

a person detects or recognises specific tastes can be objectively measured. The same

is not true, however, for recording whether the taste is enjoyed (Hill & Blundell,

1982). Specific differences in threshold levels have been reported in patients with

cancer compared to healthy controls but there appears to be little consistency

between studies except for tastes changes that occur in response to certain drugs


(Ames, Gee, & Hawrysh, 1993; Bruera et al., 1984; Carson & Gormican, 1977;

Tchekmedyian et al., 1992; Trant, Serin, & Douglass, 1982).

Changes in taste thresholds have also not consistently been reflected by changes in

food preferences. De Wys reported an association between lowered threshold levels

for detecting bitter taste and an aversion to meat in a mixed cancer population

(DeWys, 1977). Although patients with gastrointestinal cancer in a different study,

were found to have lowered thresholds for recognition of bitter taste compared to

controls, this did not correspond to an aversion to meat (Hall, Staniland, & Giles,

1980).

The highly individualised character of hedonic responses to food was demonstrated

in a study by Trant et al in which both the intensity and pleasantness of a range of

tastes were examined in patients with upper gastrointestinal and thoracic tumours

(Trant et al., 1982). Other methods for recording taste changes have varied from

simple “have you experienced taste changes?” and LAS’s, to detailed questionnaires

asking about enjoyment of specific foods (Carson & Gormican, 1977; Nielsen et al.,

1980; Schiffman, 1997; Stubbs, 1989; Tchekmedyian et al., 1992; Wickham et al.,

1999). An attempt has also been made to classify dysgeusia by the range of foods

affected (Markley, Mattes-Kulig, & Henkin, 1983).

Over the last 30 years of study no solutions have been found. The subjective nature

of taste perceptions, the complexity of the changes and the varied individual

responses to such changes make this an area in which qualitative methods may

provide more meaningful data.

2.3.2 Gastrointestinal symptoms

Gastrointestinal symptoms are common and can occur early in cancer, independent

of tumour site and with or without chemotherapy or radiotherapy. A sense of

abdominal fullness was reported in 61% of the 254 patients with varied unresectable

cancers studied by Grosvenor et al (Grosvenor et al., 1989), and it was significantly

more common in those patients who had lost weight. The following symptoms were

also reported; constipation (42%), nausea (39%), vomiting (27%) and abdominal


pain (27%). Similar frequencies were noted in a smaller group of pancreatic cancer

patients; early satiety (62%), constipation (41%), dyspepsia (36%), nausea (41%),

vomiting (31%) and bloating (23%) (Krech & Walsh, 1991)

Early satiety is common in patients with anorexia (Trant et al., 1982) and may be due

to delayed gastric emptying or systemic causes. Gastroparesis may be due to the

anticholinergic effect of opiate analgesia, or to mechanical resistance to emptying

which can be caused by abdominal masses, hepatomegaly or ascites (Rhodes &

McDaniel, 2001).

Opioid drugs cause constipation by suppressing peristalsis and raising sphincter tone.

Combined with dietary changes and inactivity, it is not surprising that many cancer

patients requiring morphine are prone to constipation (Doyle, Hanks, & MacDonald,

1998.). Persistent constipation was found to be strongly associated with deteriorating

performance status in an observational study of 50 patients with advanced cancer.

The constipating effect of morphine was, however, found to be independent of dose

and resolves in some patients (Fallon & Hanks, 1999).

Diarrhoea can result from chemotherapy, radiotherapy to the abdomen or intestinal

resections. It can also be due to jaundice or lactose intolerance. In pancreatic cancer

it may also be due to pancreatic insufficiency as discussed in section 2.2.8

Nausea and vomiting, especially in advanced cancer, can have multiple causes;

analgesic or cytotoxic drugs, gastric stasis, hypercalcaemia, pharyngeal or

gastrointestinal irritation, hepatic or renal failure, gastric outlet or bowel obstruction,

treatment induced release of serotonin in the gut, emetogenic tumour products,

anxiety, and cerebral metastases (Doyle et al., 1998.; Rhodes & McDaniel, 2001).

Younger patients and women report more chemotherapy-induced nausea and

vomiting (Rhodes & McDaniel, 2001).

Measuring the occurrence and severity of nausea, as for other subjective experiences,

is difficult. It is further complicated by the difficulty patients often experience in

describing the symptom and the fact that terminology is not readily understood

(Rhodes & McDaniel, 2001). Vomiting and retching, while observable, can vary in


their frequency and the degree of distress experienced. Vomiting can also reduce the

validity of recording dietary intake.

Methods for recording gastrointestinal symptoms include simple symptom checklists,

linear analogue scales, and Likert scales that may incorporate aspects of duration and

severity as well as frequency. Numerous tools have been developed to examine

nausea and vomiting (Cotanch, 1992; Rhodes & McDaniel, 2001). Symptom scales

and questions have been included in quality of life questionnaires (Aaronson et al.,

1993) and the Patient-generated Subjective Global Assessment (PG-SGA) (Ottery,

1995) because of their recognised impact on quality of life.

Concerns have been raised that, in the absence of evidence that nutritional support

can increase survival time, encouraging increased oral intake might reduce quality of

life due to a potential increase in gastrointestinal discomfort (Grosvenor et al., 1989;

Padilla, 1986). Although there is no evidence for this at present, quality of life

measures should be included in nutrition intervention studies in pancreatic cancer to

address this concern.

2.3.3 Pain and depression

Pain was reported by 82% of patients in a study of patients with advanced pancreatic

cancer (Krech & Walsh, 1991). The presence of pain can reduce food intake, and

food consumption has been reported in some cases to increase pain (Held-

Warmkessel, Volpe, & Waldman, 1998).

Depression has been said to occur more frequently in patients with pancreatic cancer

and may precede diagnosis, suggesting a physiological basis (Passik & Breitbart,

1996). Levels of depression vary greatly between studies (Donnelly et al., 1995).

This may relate partly to the wide range of definitions and measures used. In the

study by Krech, depression was reported by 39% of patients with pancreatic cancer

(Krech & Walsh, 1991).

Bruera et al (Bruera et al., 1984) compared levels of depression, using Hamilton’s

psychometric test, in malnourished and well-nourished patients with advanced solid


tumours prior to chemotherapy. A high incidence of depression was found in both

groups (> 80%). There were, however, significantly more malnourished patients with

severe depression (59% vs 20%, p = 0.026).

As commented by Ottery; “psychologic effects may be the result of (1) dealing with

the diagnosis of cancer, treatment side effects, feelings of helplessness, financial

concerns etc, (2) side effects of agents used for symptom management (eg sedatives,

antiemetics, analgesics) or (3) an effect of malnutrition…” (Ottery, 1995).

Pain and depression may therefore not only limit food intake for patients with

unresectable pancreatic cancer, but also decrease their quality of life.

2.4 Effect of nutrition support

2.4.1 Studies of nutritional support in cancer

The wasting that frequently accompanies advanced cancers, especially pancreatic

cancer, has been well described. What constitutes effective nutrition support for these

patients is less clear.

The term “nutrition support” is often used in the literature to refer to the use of tube

feeding or parenteral nutrition. This reflects the fact that these are seen as medical

interventions which can be more easily controlled and monitored than oral intake.

Few studies examine oral nutrition support (ONS) for cancer patients even though

this is the primary mode of dietetic intervention for patients who are capable of oral

intake.

Efforts to reverse the weight loss process have had limited success and are

summarised in Table 2.6. Interventions with supplements and/or counselling have

usually achieved statistically significant increases in intake, however goal intakes are

only achieved by a small proportion of the study patients and mean energy intakes

are often such that weight gain would not be expected (Evans et al., 1987; Ovesen et

al., 1993a).


Table 2.6 Summary of studies of oral nutrition support in cancer

Author, year,

country

Patient population Study design Outcome

parameters

Results Comments

McCarthy 1999 USA

40 patients with breast, prostate etc cancers & radiotherapy (curative intent)

PRCT Grp1 =weekly counselling & ONS (500-600 kcal/d) Grp2 = counselling only

24hr recall - 1 food record/d x 3d/wk x 4wks

Total E & Pr intake greater in Grp1. No difference in E & Pr derived from food between grps

NB - 6 dropouts because “supps ruined appetite” Weight not reported

Macqueen 1998 UK

10 patients with Ca larynx (Grp1), and 10 with Ca pharynx (Grp2) & radiotherapy x 6 wks

Case series all pts received ONS (& enteral or parenteral if needed)

Weight, VAS for app/solids/liq/taste BMR Intake (3d diary x 4)

Grp1 = steady Wt, App & EI Grp2 =↓ Wt, App & EI App correl with EI (food only)

NB. Major barrier to intake = side effects from treatment not the Ca itself Unable to meet E needs for Grp2 even with a range of NS measures– “due to pt refusal etc

Ottery 1996 USA

186 Ca pts

Case series All patients received ONS & aggressive symptom management

Patient-generated Subjective Global Assessment (PG-SGA)

claims 50-80% success in maintaining or increasing weight (if exclude patients with expected survival <6wks)

No clear data - symposium abstract

Ovesen 1993 Denmark

105 patients with cancer of lung, ovary or breast & chemotherapy

PRCT Grp1 = counselling (±supps with goals 1.5-1.7 x est BEE & 1-1.2 g Pr/kg/d) Grp2 = no counselling, ad lib diet

Intake (3 d diaries x 6 over 5 mths), weight, anthropometry, QoL, response rate

Grp1 1.4 → 1.5 x estBEE (↑ 1MJ/d) 0.9 → 1.1 g Pr/kg/d (↑ 10g/d) Grp2 Intake unchanged 1.3 x estBEE & 0.9 g Pr/kg/d

Greater energy and protein intakes for Grp 1 but insignificant weight gain no ∆ P:F:CHO proportions Both groups ↑ QoL

Cont.


Table 2.6 cont. Author,Year, Country

Patient population Study design Outcome parameters

Results Comments

Ollenschlager 1992 Germany

29 leukemia inpatients & chemotherapy

PRCT ~22wks (median) Grp1 = intensive ONS (nutrition education & daily review by dietitian) Grp2 = no ONS, ad lib diet all pts - selected menus daily offered 1-2g Pr/kg & 30-50kcal/kgBW

Intake measured by weighing tray items pre/post meal, Body weight

Both grps lost wt initially but Grp1 regained BW earlier and more often than Grp2 (69% Grp1 vs 31% Grp2) mean intakes 23.2 kcal/kg/d = wt losing 30.9 kcal/kg/d = wt stable 39.3 kcal/kg/d = wt gaining

low intake correlated with Tx side effects (anorexia was main barrier)

Arnold 1989 USA

50 patients with head & neck cancer & radiotherapy

PRCT Grp1 = counselling & 1000kcal ONS/d x 10wks Grp2 = counselling only

Weight, albumin, transferrin, 24hr dietary recall

Grp1 greater energy & protein intake than Grp2 (1929 vs 1624 kcal/d, p=0.035) but no significant difference in weight

Mean intake supplement ~60% goal amount Both grps lost weight Both grps received intensive nutrition counselling weekly

Evans 1987 USA

180 patients with metastatic lung & colorectal cancers & chemotherapy

PRCT Grp1 = + active NS with goals 1.7 – 1.95 x estBEE Grp2 = + active NS (↑Pr, Zn & Mg) Grp3 = controls, ad lib diet

Intake (24hr recall & food frequency), weight, survival time, mortality, treatment response

No difference in weight change, survival time, mortality or treatment response Greater intake in Grps 1&2 vs Grp 3 (90% vs 60-70% TCI) TCI = targetted caloric intake

Positive association between weight change and energy intake for both cancer types 60% pts required enteral feeding (by protocol) but did not receive

Moloney 1983 Ireland

97 patients with head & neck, lung, breast, bladder etc cancer & radiotherapy

PRCT Grp1 = counselling & ONS Grp2 = control, ad lib diet

Intake – 3 day food diaries in wk1 and final wk, treatment response, mortality

Grp1 had greater intake than Grp2 No difference in mortality or treatment response

71% inpatients problems with compliance referred to but not in detail

PRCT; prospective randomised controlled trial; estBEE = basal energy expenditure estimated from Harris Benedict equation; ONS, oral nutrition support; VAS, visual analogue scale; EI,

energy intake; Pr, protein; App, appetite; BMR, basal metabolic rate; NS, nutrition support (enteral/parenteral/oral as required); QoL, quality of life


Studies of nutrition support often pool cancers of different types (eg breast and

gastrointestinal cancers) in order to obtain sufficient numbers of subjects. Cancers at

different sites have different nutritional impacts. Even for a single cancer type, there

can be quite different reasons for weight loss amongst a group of patients as

discussed in section 2.2. Studies often focus on support during chemotherapy,

radiotherapy or surgery. Limited data is available on the effectiveness of oral

nutrition support for cachectic patients for whom no further active treatment is

available. Little detail is usually provided to describe the nature of any nutrition

counselling. Frequency of contact is often all that is reported.

A summary of the single arm studies of nutrition support in pancreatic cancer that led

up to the BH80 trial is included in Table 2.7. Weight loss was stabilised in patients

with unresectable pancreatic cancer who received eicosapentaenoic acid (EPA),

either as part of a protein and energy dense oral supplement or alone (Barber et al.,

1999b; Wigmore, Barber, Ross, Tisdale, & Fearon, 2000a). Information regarding

the BH80 study is provided in Chapter 3. Benefits have also been reported for more

aggressive use of pancreatic enzyme supplements (Bruno et al., 1998).

2.4.2 Research on oral nutrition support

A systematic review of oral and enteral protein and energy supplementation in a

range of diseases by Potter et al (Potter, Langhorne, & Roberts, 1998) showed that in

the 26 trials (1600 patients) for which weight change data were available, absolute

weight change tended to be negative, especially when surgery or cancer were

involved. In most cases, however, the supplemented patients had a lesser weight loss

than the control group (weighted mean difference 2.06%, 95% CI 1.63-2.49%).

Similarly, a review of the use of oral nutrition supplements for patients with chronic

conditions including cancer (Stratton & Elia, 2000), concluded that supplement use

was associated with improved clinical outcomes, especially for patients with BMI <

20 kg/m2.

A Cochrane review by Baldwin et al (Baldwin, Parsons, & Logan, 2002) was unable

to find clear evidence that dietary advice for adults with illness-related malnutrition,

over and above the provision of nutrition supplements, resulted in improved


Table 2.7 Summary of nutrition-intervention studies in pancreatic cancer

Author, year, country

Patient population Study design Outcome parameterss Results Comments

Wigmore 1997 UK

6 patients with PC Case series Supplemented with 1→6g EPA/d orally over 1 mth

Weight, CRP, IL-6 (serum)

CRP decreased (11→0.8 mg/L) no significant change IL-6

Not controlled Wt↓ assoc with persistent elevation of CRP

Bruno 1998 Netherlands

21 patients with unresectable PC (head), suspected duct obstruction

PRCT Grp1 = enzyme supplementation x 8wk Grp2 = placebo all had dietary counselling (visits Wk0,4,8,12 and weekly phone contact)

weight, symptoms faecal fat test(FAC) wk8, intake (2d record at wk8)

Grp1 ↑ 0.7kg &Grp2 ↓2.2kg (p=0.02) Change in FAC Grp1↑12 vs Grp2↓8 (p=0.13) Grp1 had greater wk8 intake than Grp2 for energy (8.42 vs 6.66 MJ) (p=0.04) & protein (75g vs 54g) but not fat or CHO No sig difference in severity/occurrence of complaints of steatorrhoea

Limited detail on nutrient intake methods

Barber 1999 UK

20 wt losing unresectable PC pts (expected survival >2mths)

Case series + oral supplement with 2g EPA, 610kcal, 32g protein x 7 wks

weight, EI (3d record) REE, BIA, CRP KPS, appetite (LAS)

Wt loss 2.9 kg/mth (baseline) → wt gain 1kg (3wks p=0.024) and 2kg (7wks p=0.033) EI (3wks) ↑ (1450→1798kcal) (p=0.0016) REE (3wks)↓ (34.0 →31.8 kcal/kg LBM) KPS ↑ (at 3 & 7wks) App ↑ (3 wks only) No sig change CRP

Not controlled High attrition rate, n = 13 at wk7 Median survival from Dx 8.5mths

Wigmore 2000 UK

26 advanced wt losing PC pts

Case series Supplemented with up to 6g EPA/d, orally

weight Weight stabilised

PC, pancreatic cancer; CRP, C reactive protein; FAC, fat absorption coefficient; REE, resting energy expenditure (indirect calorimetry); EI, energy intake; EPA, eicosapentaenoic acid; IL-6,

interleukin 6; PRCT, prospective randomised controlled trial; BIA, bioelectrical impedance analysis; KPS, Karnofsky performance status; LAS, linear analogue scale


outcomes in terms of mortality, morbidity or weight change. Supplement use,

however, was associated with weight gain and greater energy intake. The reviewers

recommended that adequately powered randomised controlled trials (RCT) be

conducted to clarify optimal treatment methods. A valuable research question would

be whether individualised ONS improves patient outcomes. The authors however

reported that “there was almost no useable data from which to draw conclusions

about the effect of dietary advice or dietary advice with supplements if required

compared with no advice…”.

It is very difficult to conduct randomised, controlled trials of dietetic intervention

because it is ethically difficult to control for ONS when this is considered part of

standard care. The “standard” practice provided for the control arm of a study may

also become modified due to the increased awareness of nutrition that occurs for staff

and patients simply as a result of being involved in a research study. This change in

behaviour is not a problem for most drug trials or for the more medicalised forms of

nutrition support but can certainly complicate studies of ONS.

Overall, from the many studies investigating nutrition and cancer, there is agreement

that weight loss is associated with poorer outcomes. There is only limited evidence,

however, that slowing the decline in nutritional status is achievable or that nutrition

intervention can improve clinical outcomes for cancer patients. There are insufficient

well designed studies available to answer these questions.

2.5 Evidence-based goals for nutrition support

2.5.1 Current dietetic practice in oncology

The American Dietetic Association (McCallum & Polisena, 2000) has published

recommendations for nutritional support specific to cancer. There are, however,

many gaps in the evidence on which to base practice and little has been documented

about actual current practice in the area of oncology.

38 Literature Review

Figure 2.2 Ottery’s algorithm of optimal nutritional intervention

Source: Ottery, 1996b

thompsdc

This figure is not available online. Please consult the hardcopy thesis available at the QUT Library.


Ottery has proposed an algorithm (Figure 2.2) to guide decision making in the

nutrition support of oncology patients (Ottery, 1996a).

Medical management of oncology patients can be divided into two streams –

treatment with curative intent and palliative care. Palliative support is often

associated solely with the terminal phase of a disease, the final weeks of life. It does,

however, also refer to all treatment that while not curative, includes aggressive

symptom management. Ottery states that although “attempts to reverse severe

nutritional depletion are generally unsuccessful” there is still a role for supportive

care for the severely malnourished cancer patient (Ottery, 1995).

The American Society for Parenteral and Enteral Nutrition have published guidelines

on specialised nutrition support including a section on oncology [ASPEN Board of

Directors, 2002 #497]. Details are provided on the level of evidence for each

recommendation but these guidelines relate principally to the enteral and parenteral

feeding of patients undergoing active treatment rather than the broader issue of

nutrition intervention.

The first recommendation in the ASPEN guidelines is;

“Patients with cancer are at nutrition risk and should undergo nutrition screening to identify

those who require formal nutrition assessment with development of a nutrition care plan.”

[ASPEN Board of Directors, 2002 #497]

Early detection of treatable symptoms and early nutrition intervention have been

recommended in preference to referral of patients who are already malnourished

(Ottery, 1996a). This is not easily achieved given the large number of patients being

treated by oncology units, making screening methods an important part of ONS.

Screening tools have been developed, such as the Scored Patient Generated –

Subjective Global Assessment (McMahon, Decker, & Ottery, 1998; Ottery, 1996b)

and one for head and neck cancer patients undergoing radiotherapy (Macqueen &

Frost, 1998). The first section of the PG-SGA is intended for completion by the

patient and can act as a screening tool, but it incorporates the Subjective Global


Assessment and therefore functions also as an assessment tool. The validated

Malnutrition Screening Tool is not cancer-specific but is simple enough to be used

routinely on hospital admission (Ferguson, Capra, Bauer, & Banks, 1999).

Endpoints for oncology treatment studies have typically been treatment response rate

and survival. It has been recognised however, that treatment goals need to also

include symptom control and optimal QoL, especially for diseases such as pancreatic

cancer (Donnelly et al., 1995; Rothenberg et al., 1996).

“…the multiplicity of symptoms in individuals with pancreatic cancer suggests the

need for a holistic approach to psychological and physical symptom management, as

well as therapeutic intervention” (Krech & Walsh, 1991)

Nutrition interventions have usually focussed on intermediate outcomes such as

improved nutrient intake and increased anthropometric measurements. Patient

outcomes of symptom control, patient satisfaction and improved QoL are now

considered important endpoints and are more recently being included (Splett, 1996).

2.5.2 Increased survival time

Weight loss in cancer has been demonstrated to be associated with reduced survival

duration (Andreyev et al., 1998; DeWys et al., 1980), but evidence has been less

consistent for pancreatic cancer. Malnutrition has the potential to reduce survival

time either by muscle wasting that could lead to pulmonary complications or through

impaired immune response and resulting infections. In cases of cancer where

resection is feasible, removal of the tumour is expected to have the greatest impact

on survival duration.

2.5.3 Improved quality of life

Quality of life relates to those things that give worth, meaning, purpose and

satisfaction to life (Grindel, Whitmer, & Barsevick, 1996). It is “linked closely to

nutrition in terms of appetite, ability to carry on daily activities, self-image, control

and overall aspects of satisfaction” (Ottery, 1995).


The importance of including health-related quality of life as an outcome measure for

studies of patients with advanced cancer is acknowledged (Aaronson, 1990; Cella,

1992; Wade & Jain, 1984) although a wide range of tools are used making

comparison of studies difficult (Sanders, Egger, Donovan, Tallon, & Frankel, 1998).

Many early studies that include measures of QoL suffer from a lack of

standardization of the tools. The European Organisation for Research and Treatment

of Cancer quality of life questionnaire EORTC QLQ-C30 has been shown to be valid

and reliable (Aaronson et al., 1993) and has been used in many clinical trials. It

includes symptom scales as well as broader measures of quality of life.

In the study by Ovesen et al, the subgroup of cancer patients who had lost weight

prior to the study reported poorer QoL than those who had not lost weight (Ovesen et

al., 1993a). It is interesting to note that an increase in QoL was reported in both arms

of that study suggesting an affect of active treatment (chemotherapy) or perhaps of

participating in a research study.

Gastrointestinal symptoms can not only limit nutrient intake, they can also reduce

QoL, as described in a British survey of the carers of 2074 people who had died of

cancer. Seventy-eight percent of patients were reported as having experienced loss of

appetite in the year prior to their death and for nearly a quarter of these, this

symptom was considered to be “very distressing”. Nausea and vomiting, as well as

constipation, were regarded as “very distressing” in the majority of cases where they

were reported (Addington-Hall & McCarthy, 1995).

Anorexia has also been reported as a source of anxiety and conflict between patients

and their families (Holden, 1991). Detailed information on family interactions and

factors that affect patient satisfaction and QoL may require qualitative studies

involving semi-structured interviews (Holden, 1991; McGrath, 2002; Meares, 1997).

Seligman, in a discussion of the use of appetite stimulants in patients with terminal

cancer, acknowledged the difficulty of measuring their effect when the expected

benefit is increased enjoyment of food rather than weight gain (Seligman, Fink, &

Massey-Seligman, 1998). He therefore recommended qualitative studies to consider

aspects of quality of life in addition to the usual quantitative measures.


Table 2.8 Recommendations regarding energy requirements for cancer patients.

Author Medical or nutritional status Recommended daily intake

Bloch, 1993

nonambulatory or sedentary

slightly hypermetabolic or for weight

gain

hypermetabolic, severely stressed or

signif. malabsorption

20-25 kcal/kg BW

30-35 kcal/kg BW

>35 kcal/kg BW

Martin, 2000

initial re-feeding of malnourished or

depleted pt

obese pt for maintenance

maintenance or standard

malnourished and/or extensive

treament or bone marrow transplant

depleted and/or hypermetabolic

20 kcal/kg CBW

21-25 kcal/kg IBW

25-30 kcal/kg CBW

30-35 kcal/kg CBW

35-45 kcal/kg CBW

Jeejeebhoy, 1976

basal requirements

ambulatory & wt maintenance

25-30 kcal/kg IBW

30-35 kcal/kg IBW

Robuck, 1992

malnourished cancer pts

3000-4000 kcal

(12600-16800 kJ)

Nitenberg, 2000

well nourished

malnourished

REE

20-25 kcal/kg UBW

25-35 kcal/kg UBW

BW body weight, CBW current body weight, UBW usual body weight, IBW ideal body weight

20, 25, 30 and 35 kcal approximate to 85, 105, 125 and 145 kJ respectively


2.5.4 Nutritional requirements

Energy intakes of cancer patients reported in the literature have been described in

section 2.3. Tables 2.8 and 2.9 describe goals for energy and protein intake for

cancer patients that have been recommended in the literature. Unfortunately little

evidence is usually presented for such recommendations. One of the barriers to

determining requirements is that the metabolic changes that can occur in cachexia

would need to be resolved before a person could effectively utilise nutrients (Tisdale,

1997a).

Inpatients with leukemia undergoing chemotherapy who were able to gain weight

consumed a mean of 39.3 kcal/kg/d compared to 30.9 kcal/kg/d for those with steady

weight and 23.2 kcal/kg/d for those patients who were losing weight (Ollenschlager

et al., 1992). This suggests that some of the recommendations for energy intake in

Table 2.8 may be less than that required for some people with cancer.

While there is little strong evidence regarding optimal protein intakes for weight

losing cancer patients, it has been considered a crucial component of nutrition

support (Ottery, 2000). Increased protein requirements would be expected for the

rebuilding of lost skeletal muscle, to meet the demands on amino acid supplies

during an APPR and to compensate for proteolytic factors often present in cachexia

(Baracos, 2000; MacDonald et al., 2003; Tisdale, 1997a).

Results from the few studies that have reported on the protein intake of cancer

patients, suggest that the levels of protein intake recommended in Table 2.9 are

difficult to achieve. Chemotherapy patients, for example, who received nutrition

counselling in the study by Ovesen et al, increased their intake of protein, but only to

1.1 g/kg/d (Ovesen et al., 1993a).


Table 2.9 Recommendations regarding protein requirements for cancer patients.

Author Medical or nutritional status Recommended daily intake

Bloch, 1993

normal maintenance level

hypermetabolism, protein-losing

enteropathy, extreme wasting

0.8-1.0 g/kg BW

1.5-2.5 g/kg BW

Martin, 2000

most cancer patients

bone marrow transplant patients

depleted cancer patient

1.0-1.5 g/kg BW

1.5 g/kg BW

1.5-2.0 g/kg BW

Ottery, 2000

malnourished or catabolic 1.5 g/kg IBW

Robuck, 1992

malnourished cancer pt 1.5-2 g/kg BW

Nitenberg, 2000

cancer patients 0.25-0.35 g N/kg

BW, body weight; IBW, ideal body weight; N, nitrogen.

2.5.5 Body weight & lean body mass

DeWys has cautioned against equating weight gain with gain in lean body mass

(LBM) in studies of nutrition support with enteral or parenteral nutrition (DeWys &

Kubota, 1981). Body mass is simple to measure and is used as a basis for nutrition

assessment, however, it is not a measure of lean body mass. Weight is affected by

hydration, tumour mass, and the presence of oedema and ascites - all relevant issues

in unresectable pancreatic cancer.

Loprinzi et al, using dual-energy x-ray absorptiometry and tritiated water, have

demonstrated that increases in weight in patients with advanced breast cancer who

took the appetite stimulant megesterol acetate were due mainly to increases in

adipose tissue and fluid (Loprinzi, Schaid, Dose, Burnham, & Jensen, 1993). Lean

body mass has been estimated by bioelectrical impedance analysis (BIA) in

unresectable pancreatic cancer patients (Barber et al., 1999b; Wigmore et al., 1997b).

The predictive equations used, however, have not been developed in equivalent


populations and the extensive muscle wasting and fluid shifts are of concern with this

method (Ellis et al., 1999).

2.5.6 Functional improvements

One of the features of cancer that reduces QoL and increases healthcare costs is a

decline in functional capacity. Two tools commonly used to measure functional

status are the Eastern Cooperative Oncology Group scale (ECOG) and the Karnofsky

Performance Status (KPS) scale (Appendix 5). The latter has been shown to be an

easily used, valid and reliable measure of patient independence (Mor, Laliberte,

Morris, & Wiemann, 1984; Yates, Chalmer, & McKegney, 1980) but is not a quality

of life tool. It has been used as both a criterion for study eligibility and an outcome

measure in studies of cancer patients (Barber, Ross, & Fearon, 1999a; Bruno et al.,

1998; Gogos et al., 1998; O'Gorman et al., 1998). The ECOG scale has been used in

a study of advanced pancreatic cancer to show that patients with poorer performance

status had a reduced length of survival (Krech & Walsh, 1991).

Loss of muscle mass, rather than loss of fat alone, would be expected to decrease a

patient’s activity level or performance status, however LBM (determined by three-

site skinfold technique) was only weakly correlated with muscular strength and

endurance in healthy women, suggesting that maintaining functional capacity

through exercise is at least as important as absolute mass of lean body tissue

(Roubenoff, 2000). Patients with cancer associated weight loss often have reduced

activity levels due to their many symptoms or treatment, which can exacerbate

reductions in LBM and functional status.

Hand grip strength has been used as a measure of functional status (Hunt, Rowlands,

& Johnston, 1985) and a prognostic indicator for surgical outcomes. It is dependent

upon age, gender, body mass and training and therefore any comparison of the results

of individuals needs to incorporate these (Brown & Miller, 1998). Testing protocols

vary and normative data are limited, with the American College of Sports Medicine

claiming in 1995 that available data were unreliable.


Fatigue is commonly reported by patients with cancer, however, associations

between weight loss and fatigue in the literature are inconsistent (Beach, Siebeneck,

Buderer, & Ferner, 2001; Kalman & Villani, 1997; Keele, Bray, Emery, Duncan, &

Silk, 1997).

2.6 Summary

Progressive weight loss is clearly a problem for patients with unresectable pancreatic

cancer. It occurs frequently, has multiple causes and is associated with poorer

outcomes. What has been less clear is whether ONS for such patients is able to

reverse or delay this nutritional decline and whether a delay, if possible, would result

in improved outcomes. Identifying patients who would benefit most from ONS

would assist in an effective use of clinical resources. Health outcomes, especially in

the palliative setting, should include ones related to symptom control and overall

quality of life.

The following chapter describes the methods used to clarify these issues in the

analysis of data from a large international trial involving ONS for pancreatic cancer

patients

3 Methods – Aims 1 &2

The data analysed in this study were from an international nutrition intervention trial

(BH80). The first section of this chapter summarises the objectives and methods of

the BH80 trial. The second section describes the preparation of data for analysis

relating to Aims 1 and 2. Statistical methods used to analyse the data are described in

the third section. Methods relating to Aim 3 are described separately in Chapter 5.

3.1 BH80 trial

3.1.1 BH80 study design

The international, multicentre, randomised, double-blind BH80 trial was conducted

between January 1998 and January 2001 (recruitments between January 1998 and

April 2000). The aim of the study was to compare the benefits of an n-3 fatty acid

(fish oil) and antioxidant enriched oral supplement with that of an isonitrogenous

isocaloric oral supplement in weight-losing pancreatic cancer patients. The primary

objectives were to compare the experimental and control supplements for their

effects over an eight week period on the following variables.

Table 3.1 Outcome variables of the BH80 study.

Primary Outcomes Secondary Outcomes

Body weight Appetite, satiety and dietary intake

Quality of life Body composition

Acute phase proteins Functional status

Survival time

Number of unplanned hospital admissions

Participants were stratified by study site and histological confirmation of cancer then

allotted, randomly and in a double-blinded manner, to either the experimental or

control group. For the Australian sites (Princess Alexandra Hospital – PAH, and The

48 Methods – Aims 1&2

Wesley Hospital – TWH) this was performed by staff at The Wesley Hospital

Pharmacy. Unblinding did not occur until after the final data collection.

Patients were asked to consume two x 237ml cans per day of either the experimental

or the control supplement for an eight week period. The supplements were

isonitrogenous and isocaloric with protein, fat and carbohydrate contributing 21, 18

and 62% of total energy respectively. The experimental supplement contained fish

oil and higher levels of antioxidants to protect the n-3 fatty acids from peroxidation.

The major nutrient features of the supplements are shown in Table 3.2. The rationale

for pooling data from the two study arms for the secondary analysis is covered in

3.2.1.

Table 3.2 Major components of BH80 study supplements

Nutrient/feature Quantity per 237ml can

energy 1300 kJ (310 kcal)

protein 16.1 g

fat 6.5 g

carbohydrate 49.7 g

fibre 5 g

eicosapentaenoic acid ± 1.1 g

energy density 1.3 kcal/ml

The range of data collected at each of the time points is shown in Table 3.3. The

main data collection points were baseline (T0), week 4 (T4) and week 8 (T8). After

the eight week period, patients were provided with the supplement for as long as they

wished to continue taking it. Hospital admissions and adverse reactions were

recorded as they occurred.

To optimise patient retention in the trial, home visits at data collection times were

provided if judged necessary by study coordinators. Dietetic intervention included

weekly contact by phone between data collection points. The phone calls

Methods – Aims 1&2 49

incorporated monitoring of symptoms, reminding patients to commence diet diaries,

and also gave patients the opportunity to ask questions.

Table 3.3 Schedule of data collection in BH80 study

Data collection point

Description Intitial Baseline

(T0)

Interim

(T4)

Final

(T8)

Follow

Up

Inclusion/exclusion criteria X

Medical/physical history X

Medications X X X

Anthropometry X X X X

Gastrointestinal tolerance X X

Sip feed consumption X X X

Biochemical markers X X X

Appetite/satiety X X X

Quality of life X X X

Performance scale X X X X

Dietary diary X X X

T0, baseline; T4, week 4; T8, week 8

3.1.2 Subjects

Two hundred people with unresectable pancreatic cancer were enrolled in the BH80

trial between January 1998 and April 2000. European, Canadian and Australian sites

were involved. A list of sites is provided in Appendix 1. Most referrals were made by

specialist medical officers at large teaching hospitals, however advertisements for the

trial resulted in some self-referrals and referrals via general practitioners.

Eligibility criteria for the trial included weight loss of at least 5% over six months, an

expected survival of at least two months and no chemotherapy, radiotherapy or

surgery during the study or for four weeks prior to baseline. For the full range of

inclusion/exclusion criteria see Appendix 2.


3.1.3 Ethical issues

The trial was performed as for drug trials, in accordance with the International

Committee for Harmonization, Good Clinical Practices and the Helsinki Declaration.

The protocol was approved by research ethics committees of each of the participating

centres and all participants provided written informed consent (Appendix 3). PAH

research protocol # 27/99 was approved on 2 March 1999 and Therapeutic Goods

Administration approval was given for the Australian trial sites on 31 March 1999.

For patient confidentiality each patient was allocated a four digit code number. That

number, along with the patient’s initials was the only identifier on data collection

sheets. All records were stored in locked cabinets.

The poor prognosis and frequently poor medical condition of participants meant that

patient burden needed to be minimised to assist compliance as well as for ethical

reasons.

3.1.4 Anthropometry

Self-reported pre-illness weight and duration of weight loss was obtained at

recruitment and corroborated where possible from hospital medical charts, general

practitioner records or family members. Body weight was recorded, without shoes

and in light clothing, on spring balance scales (Tanita Solar Powered Scale Model

1618, Tanita, Uxbridge, Middlesex, UK). These scales measured in units of 0.2 kg up

to 100 kg and 0.5 kg from 100 to 150 kg. The scales were compared to reference

scales on a monthly basis and were checked for accuracy early in the study. For

example, scales at the Brisbane sites were checked by Queensland Weighing

Machines. The same scales were used for each patient on subsequent visits.

The presence of oedema or ascites was recorded based on clinical observation.

Height was measured to the nearest 0.1cm using a portable stadiometer (Harpenden,

Holtain Ltd, Crosswell, Dyfed, UK).


Body composition was determined using a four-terminal multiple frequency Xitron

Hydra bioelectrical impedance analyser (Xitron Technologies, San Diego, California,

USA). Electrodes were positioned on the wrist and ankle joints and metacarpal and

metatarsal heads. The same limbs were used for consequent measurements.

Measurement was made with patients in a supine position and with limbs slightly

abducted from the body. Resistance at 5 kHz and 200 kHz was recorded.

Total body water was calculated using equations derived following measurement of

total body water (by dilution of tritiated water) on a heterogeneous surgical patient

population, which included a large proportion of cancer patients (Hannan, Cowen,

Plester, Fearon, & deBeau, 1995). For calculations of lean body mass, lean tissue

was assumed to contain 73% water.

3.1.5 Nutrient intake

Food intake was estimated and recorded by the participants (assisted by their carers if

necessary) as three-day diet diaries. The diaries covered one weekend day and two

week days prior to assessment at baseline, T4 and T8 when entries were clarified by

the study dietitian as necessary. The patients/carers received instruction by a study

dietitian at enrolment on how to complete the food record. Diary booklets in the

relevant language were provided. Methodological issues concerning measurement of

nutrient intake are discussed further in section 5.3.1.

Throughout the study patients also kept a daily record of intake of study supplement

on a Sip Feed Form as an estimation of the amount consumed to the nearest half can.

Measuring intake of experimental or control products in studies of nutrition

supplements poses difficulties greater than those faced in drug trials. Estimates of

part cans consumed are likely to be inaccurate as the cans are opaque and there was

no instruction to measure the remainder in any way. In drug trials, compliance is

often measured by asking participants to return blister packs of tablets. It would be

impractical to retain and return opened cans of supplement for similar tallying. It

would also be uncommon for people to take only part of a tablet or capsule, whereas

it is understandable for a person commencing a can of supplement to feel unable to

consume the full amount.


Mean daily values were calculated for macronutrients using country-specific

computerised nutrient analysis programs. The Australian sites used the FoodWorks

software programme (FoodWorks 2.04 Professional Edition 1998,99 Xyris Software,

Brisbane, Australia).

3.1.6 Biochemical analysis

Blood samples (baseline, T4 and T8) were collected by staff from accredited

pathology companies (Sullivan & Nicolaides for the Brisbane sites), at the patient’s

home if necessary. Complete blood counts along with urea, creatinine, glucose,

albumin, electrolytes and liver function tests were performed locally by the

pathology companies.

Samples were also stored and sent to a central laboratory (Aston University,

Birmingham, UK) for measurement of plasma fatty acid profiles (gas

chromatography) and serum levels of C-reactive protein (enzyme-linked

immunosorbent assay). Plasma phospholipid levels of EPA from samples taken

before commencement of the supplement and at T4 and T8 were used to assess

patient compliance.

3.1.7 Quality of Life

At baseline, T4 and T8 patients completed two validated quality of life tools. The

EORTC-QLQ-C30 (version 3) is a multidimensional, cancer-specific quality of life

questionnaire developed by the European Organisation for Research and Treatment

of Cancer Study Group on Quality of Life for use in international clinical trial

settings (Aaronson et al., 1993). This provides a global quality of life measure

(global QoL) as well as function and symptom scores including nutrition-related

symptom scales of pain, nausea and vomiting and appetite loss (Appendix 4).

The pain scale was formed from two questions; “During the past week have you had

pain?” and “During the past week did pain interfere with your daily activities?”. The

nausea and vomiting scale was also composed of two questions; “During the past


week have you felt nauseated?” and “During the past week have you vomited?”. The

four possible responses for each question are “not at all” (1), “a little” (2), “quite a

bit” (3) and “very much” (4). Raw data for each scale in this questionnaire were

converted to a score out of 100 by study data managers as per EORTC protocol

(Fayers, Aaronson, Bjordal, & Sullivan, 1999).

EuroQol (EQ5D) is a generic quality of life measure of health status which provides

a single index score (EQ-5Dindex) and the patient’s assessment of their overall health

state (EQ-5D vas) (Kind, 1998).

3.1.8 Appetite & Satiety

The appetite scale included in the EORTC QLQ-C30 involved a single question;

“During the past week have you lacked appetite?”. The four possible responses are

“not at all” (1), “a little” (2), “quite a bit” (3) and “very much” (4). Responses were

converted to a score out of 100 by study data managers as per EORTC protocol

(Fayers et al., 1999).

Patients were also instructed at enrolment how to complete five 100mm linear

analogue scales (LAS) relating to appetite and satiety, and then completed them

unassisted prior to their midday and evening meals on the second day of each three

day food diary. The result was recorded as the distance from the left hand anchor

point of the mark made by the patients.

The five questions and their anchors are:

How hungry do you feel right now? (not at all hungry – extremely hungry)

How thirsty do you feel right now? (not at all thirsty – extremely thirsty)

How much food do you think you could eat right now? (not at all– a large amount)

How full do you feel right now? (not at all full – extremely full)

How pleasant would it be to eat right now? (not at all pleasant – extremely pleasant)

These LAS were developed for the study by Barbara Rolls (Fearon, 1998). LAS have

been described in the literature for use in measuring a patient’s perception of the

degree of symptoms, such as pain, that are typically difficult to measure objectively.


3.1.9 Functional status

The Karnofsky Performance Status (KPS) was documented at recruitment, baseline,

T4 and T8 (Yates et al., 1980). The KPS criteria are described in Appendix 5.

Grip strength was also recorded as a measure of functional status. It was measured in

the dominant hand using a spring-loaded dynamometer (Takei Grip-A 5001, Takei

Scientific Instruments Co.). The patient was seated with the hand hanging freely by

the side. The handle of the dynamometer was adjusted to suit the hand size so that

when grasped, the second joint of the forefinger was at a 90 degree angle. The best of

three grips was recorded with a rest of at least two minutes between attempts

(Schroeder & Hill, 1991).

3.1.10 Survival duration

Date of death was obtained from hospital records, the patient’s general practitioner or

family members. The death certificate was sighted when possible. Survival duration

was measured from baseline. The study question was whether nutrition intervention

improved outcomes for patients with unresectable pancreatic cancer irrespective of

the position in their disease trajectory when they sought dietetic counselling. As the

patients entered the study at differing periods of time from diagnosis, stage of disease

was included as a variable to reflect extent of disease.

3.2 Preparation of data for secondary analysis

3.2.1 Rationale for pooling subjects

Intent to treat analysis of the BH80 data (Fearon et al., 2003) showed no significant

difference in weight change between the two supplement groups. There was,

however, a marked attenuation of weight loss in both groups with mean weight

changes of – 0.25 and – 0.37 kg/month for experimental and control groups

respectively, compared to reported pre-study weight loss of 3.3 kg/month.


This stabilisation is in contrast to the progressive weight loss usually found in

patients with unresectable pancreatic cancer (Wigmore et al., 1997b) suggesting that

many patients in each study arm, benefited from oral nutrition support (ONS).

To examine the weight stabilisation in these patients, the data from the two arms of

the BH80 study were pooled, while maintaining both randomisation grouping and T8

plasma phospholipid EPA levels as variables in order to retain the ability to

distinguish the experimental groupings. The study hypotheses were focused on

determining whether weight stabilisation over eight weeks of ONS was effective and

so baseline and T8 were the only time points (other than date of death) used for this

analysis.

Figure 3.1 Model representing the first two aims of this study of nutrition

intervention in patients with unresectable pancreatic cancer

Aim 1

Weight change

QoL

Survival Aim 2

Weight

change Energy

intake T8BMI T0

PS T0

Stage of disease

Symptoms T0

Energy intake T0

ONS

Key to symbols:


ONS – oral nutrition support, QoL – quality of life, BMI – body mass index, PS – performance status, T0 –

baseline, T8 – week 8.


3.2.2 Creation of weight losing or weight stable groups

Evaluable patients were divided into two groups based on whether they lost more

than 1kg over the eight week study period (weight losing group = WL), or lost no

more than 1kg, gained weight or were weight stable (weight stable group = WS). The

WS group was a pooling of all “non-weight losing” patients because the research

question was whether preventing further weight loss was a suitable dietetic goal.

The cut-off of 1kg over eight weeks was considered to be a clinically meaningful

change in weight (Rosenbaum, Wang, Pierson, & Kotler, 2000), ie. change of 1kg

over two months would not be considered true alteration in weight, over and above

the usual fluctuations of body weight.

3.2.3 Inclusion criteria

Patients were considered eligible for data analysis if weight data were available for

both baseline and T8 so that the variable, weight change, could be determined.

Oedema is common in the final two weeks of life (Wigmore et al., 1997b), and

would have considerably confounded weight change data. No objective measure of

fluid weight gain was available. To reduce this effect, the decision was made, prior to

commencement of data analysis, to exclude patients from analysis if survival

duration was less than 70 days from baseline (ie. the eight week study period plus

two weeks) or if oedema or ascites had been reported in the 70 days from baseline.

In studies with weight change as an outcome measure in which the effect of two

treatments are being compared (eg megesterol acetate vs megestrol acetate and

ibuprofen (McMillan et al., 1999) or nutrition supplement with or without EPA in the

BH80 study), the randomisation process has been assumed to reduce any

confounding effect of fluid weight gain from ascites or oedema. In this analysis

however, the groups being compared were defined by weight change, so it was

important to reduce any possible effect of fluid accumulation. This has meant a

reduction in “generalisability” in that patients closest to death have been excluded,


however the number of patients with T8 data who were excluded is small and the

nature of the exclusions are well defined.

Patient characteristics at baseline for those who were included in data analysis versus

those who were excluded were compared using Student’s unpaired t-test for

continuous variables with normal distributions. For categorical variables the Chi-

square test of independence was used with Fisher’s Exact test (two-tail). Pearson’s

Chi-square was used when Fisher’s was not appropriate.

3.2.4 Confirming normal distribution

Descriptive statistics and Kolmogorov-Smirnov Goodness of Fit Test were used to

determine whether variables were normally distributed based on the following

guidelines;

Median is within ±10% mean

Standard deviation <50% mean

Skewness falls between –3 and +3

Kurtosis falls between –3 and +3

Minimum and maximum values fall within ± 3 standard deviations of the mean

Histogram appears visually “normal”

Kolmogorov-Smirnov Goodness of Fit Test (p >0.05)

3.3 Data analysis

Statistical analyses were performed using the SPSS software package (version 10,

SPSS Inc. Chicago). Two-sided tests and a significance level of 0.05 were used.

3.3.1 Outcomes of dietetic intervention

Survival duration and quality of life were selected as outcome measures for dietetic

intervention due to the aggressive nature of unresectable pancreatic cancer.

Intermediate outcomes such as improved anthropometric measures are not the

ultimate goals of nutrition support in the palliative setting. While there is ample


evidence to show the inverse associations between weight loss and both quality of

life and survival time in cancer in general, the evidence is less convincing for

pancreatic cancer where there is a short life expectancy. The variables used to assess

outcomes for weight stabilisation are described in Table 3.4.

Survival analysis

Survival duration (from baseline) for the two weight change groups (WL and WS)

was compared using the Kaplan-Meier log rank test, to assess whether weight

stabilisation was an appropriate goal for these patients.

Kaplan-Meier analysis compares the distribution of times between either baseline

and death or baseline and the time when the patient was last seen (ie. then withdrawn

from study and/or follow-up ceased). Data for each patient was classified as “status

0” (when time to death was known), or “status 1” (for those with time until last

observed). This maximises the amount of information used at each time point

because data for patients lost to follow up are not “censored” until the last time they

were observed.

Median survival time and 95% confidence intervals were obtained. The log rank

statistic was used to test for equality of survival distribution for the two groups.

Results were considered to be statistically significant if the p value was less than

0.05.

Quality of Life

The quality of life measure chosen was the global health status/QoL scale (global

QoL) which is derived from items 29 and 30 of the EORTC QLQ-C30 questionnaire

and recommended as the overall quality of life measure from this tool (Fayers et al.,

1999). Possible scores range from 0-100 with higher scores representing higher

quality of life. This QoL measure has been used extensively in cancer studies and

therefore was selected in preference to the generic EQ5D health related QoL

measure.


Mean T8 global QoL scores, were compared for WS and WL groups using Student’s

unpaired t-test. Baseline scores were also compared to determine whether the two

groups were similar at the start of the study.

Table 3.4 Description of variables used for assessing outcomes for weight

stabilisation in patients with unresectable pancreatic cancer receiving

nutrition intervention over eight weeks (Aim 1) Name of variable Continuous or

categorical

Units /

categories

Comments

Weight change1

categorical WL, WS Difference between T8 and T0

Survival duration

continuous days Time from baseline

Global QoL T8 continuous Score 0-100 EORTC QLQ-C30 QL2

(comprised of questions 29 &30) 1 Weight change is an independent variable for Aim 1

WL, weight losing; WS, weight stable; QoL, quality of life ; T0, baseline; T8, week 8

3.3.2 Determinants of weight stabilisation

The choice of variables was limited to those included in the BH80 trial. Appetite

loss, pain, nausea and vomiting, energy and protein intake, BMI, KPS, CRP,

pancreatic enzyme use, diabetes and stage of disease were included because from

clinical experience and review of the literature they were considered to potentially

have an impact on food intake and/or weight changes. Randomisation grouping and

plasma EPA at T8 were included in case the addition of fish oil to the nutrition

supplement was a major determinant of weight stabilisation. Age and gender were

also included.

Data excluded from use

Data was available on BIA and grip strength as well as appetite and satiety linear

analogue scales but these were not included for the following reasons. BIA results

were not included because of concerns regarding the suitability of the predictive

equations used for this patient group. The appetite and satiety LAS’s were not


included because no validation studies of the selected tools have been reported for

this patient group. Grip strength was not included because in healthy individuals it is

directly related to body mass as well as being affected by gender, age and training

(Brown & Miller, 1998; Martin, Neale, & Elia, 1985).

Some continuous variables used in the univariate analysis were converted to

categorical variables for logistic regression analysis because this analysis requires the

independent variables to be either interval or categorical. The rationale for the

selection of categories is described below and the variables that were selected for

analysis are listed in Tables 3.5 and 3.6.

Body Mass Index

The usual cut-off points of 20, 25 and 30 kg/m2 were used to create categories of

body mass index both pre-illness and at baseline. Categories at the extremes were

merged where necessary to avoid having categories with too few cases for statistical

analysis. For example if only a few patients had BMI’s of 30 or greater at baseline,

then the upper category for that variable would become “≥ 25 kg/m2”.

Age

Age was divided by decade, merging decades at each extreme as for BMI, to avoid

categories with insufficient cases. The resultant categories reflect the fact that

pancreatic cancer is predominantly a disease of older people.

Karnofsky Performance Status (KPS)

The KPS (Appendix 5) is an ordinal variable so the pre-defined levels were used as

the categories for analysis. In order to eliminate categories with insufficent numbers,

the categories at each extremity were merged.


Table 3.5 Description of baseline variables used to identify determinants of

weight stability in patients with unresectable pancreatic cancer

receiving nutrition intervention over eight weeks (Aim 2) Name of variable Continuous /

categorical

Units /

categories

Comments

Age

continuous years

Age categorical ≤60, 61-70,

>70 yrs

Gender

categorical M,F

BMI pre-illness continuous kg/m2 from measured Ht & reported pre-

illness WtBMI pre-illness categorical ≤24, 25-29, ≥30

kg/m2

from measured Ht & reported pre-

illness WtBMI T0

continuous kg/m2 from measured Wt & Ht

BMI T0 categorical <20, 20-24, ≥25

kg/m2

from measured Wt & Ht

KPS T0 categorical 50-60, 70, 80,

90-100

Pain T0 categorical present, absent EORTC QLQ-C30 (Q 9 &19)

Appetite loss T0 categorical present, absent EORTC QLQ-C30 (Q 13)

Nausea & vomiting T0 categorical present, absent EORTC QLQ-C30 (Q 14 & 15)

Energy intake T0

continuous kJ/d three day food diary

Energy intake T0 continuous kJ/kg/d three day food diary & actual body

weightProtein intake T0

continuous g/d three day food diary

Protein intake T0 continuous g/kg/d three day food diary & actual body

weightC-reactive protein T0 categorical <10mg/L,

≥10mg/L

ELISA

Randomisation

categorical Experimental,

Control

As per BH80 study

Pancreatic enzyme use

categorical Yes, No At baseline

Diabetes

categorical Yes, No At baseline

Stage of disease

categorical I-II, III, IV At baseline

KPS, Karnofsky performance status; T0, baseline


Table 3.6 Description of Week 8 variables used to identify determinants of

weight stability in patients with unresectable pancreatic cancer

receiving nutrition intervention over eight weeks (Aim 2)

Name of variable Continuous /

categorical

Units / categories Comments

Weight change 1

categorical WL, WS Difference between T8 and

T0Energy intake T8

continuous kJ/d three day food diary

Energy intake T8 continuous kJ/kg/d three day food diary &

actual body weightEnergy intake T8 categorical ≤100, 101-125, 126-

150, >150 kJ/kg/d

three day food diary &

actual body weight

Protein intake T8

continuous g/d three day food diary

Protein intake T8 continuous g/kg/d three day food diary &

actual body weightPlasma EPA T8

categorical <3%, ≥3%

1 Weight change is a dependent variable for Aim 2

WL, weight losing; WS, weight stable; T0, baseline; T8, week 8 ; EPA, eicosapentaenoic acid

Symptoms

The raw scores for symptom scales in the EORTC QLQ-C30 are converted to scores

out of 100. These scores were then dichotomised into absence (score of 0) or

presence (any score greater than 0) in order to maintain sufficient numbers in each

category and to compare symptom categories that are practical for dietitians in the

clinical setting. For example it is relatively easy to determine whether patients are

having “pain” or “no pain”, whereas the degree of pain experienced is highly

subjective.

Energy Intake

Energy expenditure is related to lean body mass. Weight is included in predictive

equations for energy requirements, the simplest of which use weight alone (Reeves &

Capra, 2003). It can therefore be misleading to compare absolute values for energy


intake between individuals of different weights. Energy intake was presented as a

proportion of body weight (kJ/kg BW/d) to attempt to standardise for the wide range

in body sizes and reflect expected energy requirements. Actual body weight at the

time of the reporting of food intake was used rather than baseline weight or “ideal”

weight because of the degree of weight loss experienced by many patients. Energy

intake is also presented as kJ/d to provide the reader with a sense of the quantities of

food consumed. This is also the only way in which intake is reported in many studies

and so it is included for comparison.

Energy intake at T8 (as a function of body weight) was divided into four categories

with cut-offs at 100, 125 and 150 kJ/kg/d. These “rounded” cut-off figures are

roughly equivalent to the values of 25, 30 and 35 kcal/kg/d often referred to in the

literature (Bloch, 1993).

C Reactive Protein

A cut-off of 10 mg/L was chosen to define “raised” CRP levels as has been used in

the literature (Falconer et al., 1994; O'Gorman et al., 1998; Wigmore et al., 1997c).

Cut-off points in earlier studies appear to have been largely based on the limits of

detection of the test used. CRP is often undetectable in healthy subjects (Wigmore et

al., 1995).

Stage of Disease

There are four categories for stage of pancreatic cancer (Appendix 6). Many Stage I

(ie. early) tumours are resectable therefore only small numbers of patients with Stage

I cancer would be expected to be recruited to this study and a category comprising

both Stage I and II was created.

Plasma eicosapentaenoic acid (EPA)

Achieving raised levels of EPA was an expected outcome for patients in the

experimental arm of the BH80 trial. As EPA was potentially a determinant of weight

stability or regain, this characteristic (plasma EPA at T8), was included as a variable.


Without supplementation, EPA comprises approximately 1% of total plasma

phospholipids. Studies involving supplementation with at least 2 g EPA/day have

demonstrated increases to 5 – 10%. Three percent was chosen as the cut-off for the

dichotomous variable because it was greater than the levels reported in

unsupplemented people but less than the proportions reached by the majority of

people in supplementation studies (~ 25th percentile) (Table 3.7).

Comparison of weight groups – univariate analysis

Chi-square tests were used for comparison of categorical data for the WS and WL

groups. Continuous data with normal distributions were compared using Student’s

unpaired t-tests. Likelihood ratio (exact significance, two-sided) was used for the Chi

square test in preparation for the logistic regression analysis.

Determinants of WS - multivariate analysis

Variables that were significantly different (or approached significance) between the

two groups were then analysed by logistic regression to identify variables that were

independent determinants of weight stabilisation. Logistic regression was used

because the dependent variable (weight change) was dichotomous.

WS was nominated as the response category. The odds ratio (OR) represents the

increase in odds (or decrease if OR<1) of being in the weight stable group compared

to the referent category of an independent variable. Dichotomous independent

variables were coded as 0 (referent category) for the category thought likely to be

related to weight loss or as 1 for the category thought to be related to weight stability.

For example, if, from clinical experience, pain is expected to make weight stability

less likely, then presence of pain would be coded as 0 (the referent category) and

absence of pain would be coded as 1. This makes it more likely that the odds ratios

will all be >1 (rather than some >1 and some <1), making interpretation simpler.

Categories of independent variables with small cell numbers were combined where

necessary to avoid “overfitting” which is demonstrated by wide confidence intervals.

Methods – Aims 1-2 65

Table 3.7 Summary of studies of both healthy subjects and cancer patients describing plasma eicosapentaenoic acid (EPA) as a percentage

of total phospholipids, with and without fish oil or EPA supplementation. Supplementation Author, Year Study population Baseline

EPA Schedule Resultant EPA

Wigmore, 1996 18 patients

unresectable pancreatic cancer (UK)

Undetectable 2→8g EPA/d over 4 wks

(fish oil capsules)

5.3 % (3.5-6.2) 1

Barber, 1999b 18 patients advanced pancreatic

cancer (UK)

0.5 % (0.2-1.6)1 ~2g EPA/d x 3wks

(fish oil enriched supp.)

5.2 % (2.6-6.7) 1

Wigmore, 2000a 26 patients

unresectable pancreatic cancer (UK)

0 % (0-0) 1 2→6g EPA/d over 4 wks

(high purity EPA capsules)

10.0 % (9.0-14.0) 1

Mantzioris, 2000 15 healthy males (Australia) 0.98 % (0.39)2 0.8g EPA x 4wks

(foods enriched with n-3 FA’s)

2.98 % (1.08) 2

Blonk, 1990

45 healthy men (Netherlands) 0.67 % (0.40) 2 1.5, 3 or 6g marine n-3 FA’s x 12 wks

Stark, 2000 18 postmenopausal women 1.18 % (0.05) 2 2.4g EPA x 4wks (RCT) 6.44% (0.10)2

Phinney, 1990 healthy subjects (USA) 0.6 % N/A

Cabre, 1992 healthy subjects (Spain) 0.3 % N/A

Adler, 1997 50 men with moderate

hypercholesterolaemia (Canada)

1.2 % (0.67) 2 N/A

Sinclair, 1994 108 healthy subjects (Australia) 1.1 % (0.6)2 N/A

Zuijdgeest-Van Leeuwen, 2002 healthy subjects and cancer patients 0.68 – 1.28 % N/A 1 median (IQR) 2 mean (SD) FA, fatty acid; RCT, randomised controlled trial; N/A, not applicable


The inclusion of covariables, ie. variables that are closely related to each other, was

also avoided to reduce the likelihood of “overfitting”. Each variable was entered into

the model after adjusting for each of the other variables.

4 Results – Aims 1 & 2

The first section of this chapter describes the reasons for excluding patients from the

data set and compares the excluded patients with those who were included in the

analysis. A description of the included patients is also provided.

The first aim of this study was to determine whether weight stabilisation for patients

with unresectable pancreatic cancer is associated with beneficial outcomes. The

second aim was to then identify any characteristics that are associated with weight

stabilisation. Results relating to Aims 1 and 2 are shown in sections 4.2 and 4.3

respectively.

4.1 Included and excluded patients

4.1.1 Reasons for excluding patients from the study

Of the 200 patients enrolled in the multicentre study, 107 were eligible for this

secondary analysis (Figure 4.1).

Figure 4.1 Reasons for exclusion of some BH80 participants from secondary

analysis

Excluded 2 –

survived ≤ 70days

Enrolled in BH80

200

Excluded 90 –

withdrawn before week 8

Reason for withdrawal before week 8:

death 27

patient/family requested withdrawal 26

change in medical status 33

adverse event or no longer eligible 4

Eligible patients

107

Excluded 1 –

ascites/oedema

Results – Aims 1&2 68

Weight change data was not available for ninety patients who had been withdrawn

from the study prior to Week 8. One patient was excluded from analysis because

ascites/oedema was reported within 70 days of baseline, and two were excluded due

to death within 70 days of baseline. The primary reason for withdrawal related to

death or disease progression which is expected in this population.

Complete data were not available for all variables for these 107 patients. Missing

data comprised only 0 - 4%, except for energy intake at Week 8 (12% missing) and

plasma EPA at Week 8 (7% missing).

4.1.2 Comparison of included and excluded patients

The characteristics of patients who were included in the analysis and those who were

excluded are compared in Table 4.1 for continuous variables, and Table 4.2 for

categorical variables.

Table 4.1 Comparison of baseline characteristics of patients with unresectable

pancreatic cancer who were included in analysis with those who were

excluded (continuous variables). 1

Variable

Included Excluded p value

Age (years)

66.9 ± 8.9 [107] 67.8 ± 9.9 [93] 0.454

Height (cm)

1.68 ± 0.09 [107] 1.65 ± 0.10 [90] 0.125

Weight pre-illness (kg)

75.4 ± 12.5 [107] 71.8 ± 13.7 [93] 0.055

Weight T0 (kg)

62.4 ± 11.2 [107] 58.7 ±11.1 [78] 0.026

BMI pre-illness (kg/m2)

26.8 ± 3.9 [107] 26.2 ± 3.9 [90] 0.275

BMI T0 (kg/m2)

22.2 ± 3.6 [107] 21.5 ± 3.0 [78] 0.143

Energy intake T0 (kJ/d) 2

7191 ± 1883 [105] 5649 ± 2000 [74] <0.001

Energy intake T0 (kJ/kg/d) 2

117.5 ± 32.2 [105] 98.9 ± 40.0 [73] 0.001

Global QoL T0

56.4 ± 21.0 [106] 43.4 ± 18.8 [75] <0.001

1 t test for independent samples mean ± SD [n] 2 three day food diaries and sip feed forms T0, baseline; QoL, quality of life



pancreatic cancer who were included in analysis with those who were

excluded (categorical variables).

Variable Included

n [%]

Excluded

n [%]

p value

Gender

Male 62 [58] 48 [52] 0.3951

Female 45 [42] 45 [48]

Randomisation

Experimental 50 [47] 45 [48] 0.8871

Control 57 [53] 48 [52]

Recurrence

Yes 16 [15] 11 [12] 0.5421

No 91 [85] 82 [88]

Enzyme usage

Yes 33 [31] 26 [28] 0.7561

No 74 [69] 67 [72]

Diabetes

Yes 15 [14] 8 [9] 0.2711

No 92 [86] 85 [91]

Stage of disease

I-II 51 [49] 33 [37]

III 22 [21] 15 [17] 0.0822

IV 32 [30] 41 [46]

KPS

50-60 13 [12] 23 [29]

70 34 [32] 35 [44] <0.0013

80 34 [32] 14 [18]

90-100 26 [24] 7 [9] 1 Fischers exact test (two-tailed) 2 Pearson χ2

(2, 0.05) = 5.01 3 Pearson χ2(3, 0.05) = 18.26

KPS, Karnofsky performance status

The patients who were included in the study did not differ from the excluded patients

with respect to gender, age, reported pre-illness weight, height, randomisation,

pancreatic enzyme usage, stage of disease, pre-illness BMI or baseline BMI.


There were significant differences however for Karnofsky Performance Status and

global QoL. Excluded patients also weighed less at baseline and had lower energy

intake both in absolute terms and when expressed as a proportion of current body

weight.

These differences are consistent with the excluded patients being a subgroup of

people further along the disease trajectory. Common reasons for lack of T8 data and

therefore exclusion, were early death or being too ill to continue in the study. The

short expected survival duration for patients with unresectable pancreatic cancer also

explains the high attrition rate in the BH80 trial.

4.1.3 Description of included patients

Sixty-two (58%) of the patients included in the study were male. Self-reported pre-

illness body weight and BMI varied greatly (52 - 117 kg, and 20 - 41 kg/m2 ).

Eighteen percent of patients were obese (BMI ≥ 30 kg/m2 ) prior to the weight loss

associated with the pancreatic cancer. Reported loss of body weight at enrolment

ranged from 5 to 40 % (median 16%).

Even though all patients were considered to have unresectable tumours, 51 (48%)

had Stage I or II disease. Sixteen (15%) patients were reported to have diabetes at the

beginning of the study. All cases of diabetes needed to be well-controlled to be

eligible for enrolment.

Eligibility criteria excluded patients who had received surgery, chemotherapy or

radiotherapy within four weeks of baseline data collection. While all patients met the

criteria, nearly all had received some medical intervention prior to study. Biliary

stenting and gastric bypass were the most common procedures.

Sixteen (15%) patients had recurrent pancreatic cancer and had previously had a

pancreatic resection (Whipple’s procedure). Thirty-three (31%) patients were

receiving pancreatic enzyme replacement at baseline. There was some variation in


frequency in prescribing, possibly related to differences in clinical practice at the

different centres.

4.2 Outcomes of nutrition intervention

This section describes the results for aim 1, ie. to determine whether weight

stabilisation for patients with unresectable pancreatic cancer is associated with

beneficial outcomes (Fig 4.2).

Figure 4.2 Model representing the first aim of this study of nutrition intervention

in patients with unresectable pancreatic cancer

QoL

Survival

Weight change

QoL – quality of lifedependent variable

Key to symbols:

independent variable

The 107 evaluable patients were divided into two categories based on whether they

lost more than 1kg over the 8 week study period (WL), or lost no more than 1kg,

gained weight or were weight stable (WS). There were 44 WL patients (41%) and 63

WS patients (59%). Baseline variables for the two groups are compared in section

4.3. All patients had been losing weight at baseline. Weight loss over the study

period for the group as a whole (n=107) was 0.6 ± 3.7 kg (mean ± SD).

4.1.4 Survival Duration

Known survival time from baseline for the patients included in the study ranged from

72 to 614 days and, as would be expected, did not have a normal distribution.

Survival duration from baseline was significantly greater for the weight stable group

than the weight losing group (Fig 4.3). Median survival for the weight stable patients


was 259 days (95% CI: 229-289 days) compared to 164 days (95% CI: 97-231 days)

for the patients who continued to lose weight.

Figure 4.3 Comparison of survival time from baseline for weight losing (n=44)

and weight stable (n=63) pancreatic cancer patients – Kaplan Meier

survival plot 1

Time from baseline (days)

7006005004003002001000

Cum

ulat

ive

surv

ival

1.2

1.0

.8

.6

.4

.2

0.0

-.2

weight stable

weight losing

1 Kaplan Meier log rank statistic 5.53, df 1, p=0.019

Data for 19% (12/63) WS and 7% (3/44) WL patients were censored during Kaplan-

Meier survival analysis at the time patients were last observed, because date of death

was not available. As described in section 3.3.1 all data is used for each point in time

until information is no longer available (either due to death or loss to follow-up) at

which point it is censored. The majority of patients with censored data in the WS

group (9/12) attended the final official monthly follow-up that occurred six months

after the eight week study period. While actual survival times are not known for

these people, this censored data reflects patients who had lived long enough to


continue through to the official BH80 final follow up which was at a timepoint

beyond the expected survival time for most patients.

4.1.5 Quality of Life

Global QoL scores, as measured by the EORTC QLQ-C30 global health status/QoL

scale, were normally distributed at both baseline and T8 (Appendix 16). At T8, the

WS group had better global QoL scores than the WL group (WS 55.0 ± 19.5 versus

WL 47.1 ± 17.4 , p = 0.037). The groups had not been significantly different at

baseline (WS 58.2 ± 21.9 versus WL 53.7 ± 19.7 , p = 0.280).

4.3 Determinants of weight stabilisation

Aim 2, which was to identify any characteristics that are associated with weight

stabilisation, is depicted in Figure 4.4.

Figure 4.4 Model representing the second aim of this study of nutrition

intervention in patients with unresectable pancreatic cancer

Weight

change Energy

intake T8BMI T0

PS T0

Stage of disease

Symptoms T0

Energy intake T0

ONS

Key to symbols:

ONS – oral nutrition support, QoL – quality of life, BMI – body mass index, PS – performance status, T0 –

baseline, T8 – week 8.



Table 4.3 Comparison of baseline characteristics of 107 weight losing (WL) and

weight stable (WS) patients with unresectable pancreatic cancer

receiving eight weeks of nutrition intervention.

Variable WL WS p value

BMI (kg/m2) 1 23.4 ± 3.1 [44] 21.4 ± 3.6 [63] 0.003

BMI distribution 2

<20 kg/m2 4 (10%) 19 (30%)

20-24 kg/m2 23 (52%) 33 (52%) 0.007

≥25 kg/m2 17 (39%) 11 (17%)

Energy intake (kJ/kg/d) 1,4 107 ± 30 [44] 125 ± 32 [61] 0.004

Protein intake (g/kg/d) 1,4 1.02 ± 0.35 [44] 1.17 ± 0.41 [61] 0.054

Age (years) 1 65.7 ± 9.3 [44] 67.7 ± 8.6 [63] 0.245

Age distribution 2

≤ 60 15 (34%) 15 (24%)

61-70 15 (34%) 25 (40%) 0.514

> 70 14 (32%) 23 (37%)

Female gender 2 15 (34%) 30 (48%) 0.172

C reactive protein ≥10 mg/L 2 17 (41%) 13 (21%) 0.045

Pancreatic enzyme supplementation 2 13 (30%) 20 (32%) 0.835

Diabetes mellitus 2 4 (9%) 12 (19%) 0.180

BH80 experimental group 2 19 (43%) 31 (49%) 0.561

Absence nausea/vomiting 2,4 14 (32%) 42 (67%) <0.001

Absence appetite loss 2,4 8 (18%) 26 (41%) 0.012

Absence pain 2,4 7 (16%) 21 (33%) 0.048

Stage of disease 2

I-II 22 (52%) 29 (46%)

III 7 (17%) 15 (24%) 0.704

IV 13 (31%) 19 (30%)

Karnofsky performance status 2

50-60 4 ( 9%) 9 (14%)

70 16 (36%) 18 (29%) 0.739

80 13 (30%) 21 (33%)

90-100 11 (25%) 15 (24%)

Global QoL1 53.7 ± 19.7 58.2 ± 21.9 0.280 1 Unpaired t test, mean ± SD, [n] 2 Chi-square test, n, (%) Likelihood ratio (exact significance, two-sided) 3 three day food diaries and sip feed forms 4 EORTC QLQ-C30 symptom scales (score of 0 = absence)


Baseline age, weight, BMI, energy and protein intake (both absolute and per

kilogram of body weight) and global QoL were normally distributed (according to

criteria in section 3.2.4). Baseline variables that had non-normal distributions, as

would be expected, were CRP levels, and the EORTC QLQ-C30 symptom scales for

appetite loss, nausea/vomiting and pain. EORTC QLQ-C30 scores are ordinal but

can be used as continuous variables (Fayers et al., 1999). Energy intake at T8 had a

normal distribution but plasma EPA did not. Descriptive statistics are shown in

Appendix 16.

4.1.6 Comparison of the WL and WS groups

The differences between the two weight groups are shown in Tables 4.3 to 4.7.

The groups did not differ significantly at baseline for age, gender, randomisation

grouping, pancreatic enzyme supplementation, Karnofsky Performance Status,

presence of diabetes or stage of disease. There were significant differences at

baseline for BMI, the presence of symptoms of appetite loss, pain and nausea and

vomiting and the presence of an acute phase response (CRP ≥10mg/L). Self-reported

pre-illness body mass index also differed between the WS and WL groups.

Table 4.4 Comparison of self-reported pre-illness body mass index (BMI) for

weight losing (WL) and weight stable (WS) pancreatic cancer patients


pre-illness BMI 1 28.0 ± 4.1 [44] 26.1 ± 3.5 [63] 0.01

pre-illness BMI distributions2

≤24 kg/m2 6 (14%) 24 (38%)

25-29 kg/m2 28 (64%) 30 (48%) 0.018

≥ 30 kg/m2 10 (23%) 9 (14%) 1 Unpaired t test, mean ± SD [n] 2 Chi-square test, n (%) Likelihood ratio (exact significance, two-sided) χ2 (2, 0.05) = 8.30


Table 4.5 Comparison of plasma eicosapentanoic acid (EPA) distributions for

weight losing and weight stable pancreatic cancer patients following

eight weeks of nutrition intervention

EPA week 8 Weight losing

n (%)

Weight stable

n (%)

<3%

≥3%

31 (76%)

10 (24%)

35 (60%)

23 (40%)

Total 41 58

χ2 (1, 0.05) = 2.57 p = 0.133 Likelihood ratio (exact significance, two-sided)

At T8 there were no significant differences for plasma EPA levels (Table 4.5),

however there were for energy and protein intake (Table 4.6). Mean energy intake

(kJ/d) was significantly different at T8 but not at baseline. When expressed as a

function of actual body weight, the WS and WL groups differed at both time points.

Table 4.6 Comparison of characteristics of weight losing (WL) and weight

stable (WS) pancreatic cancer patients at baseline and week 8 1,2.


Energy intake T0 (kJ/d) 7020 ± 2090 [44] 7314 ± 1727 [61] 0.433

Energy intake T8 (kJ/d) 6806 ± 2470 [39] 8455 ± 1849 [55] <0.001

Energy intake T0 (kJ/kg/d) 107 ± 30 [44] 125 ± 32 [61] 0.004

Energy intake T8 (kJ/kg/d) 110 ± 39 [39] 141 ± 35 [55] <0.001

Protein intake T0 (g/d) 66.5 ± 21.4 [44] 68.7 ± 24.2 [61] 0.633

Protein intake T8 (g/d) 66.4 ± 27.4 [40] 83.4 ± 21.8 [57] 0.001

Protein intake T0 (g/kg/d) 1.02 ± 0.35 [44] 1.17 ± 0.41[61] 0.054

Protein intake T8 (g/kg/d) 1.08 ± 0.42 [40] 1.39 ± 0.36 [57] <0.001 1Unpaired t tests, mean ± SD [n] 2 three day food diaries and sip feed forms T0, baseline; T8, week 8

The differences for energy intake at T8 remained significant when the variable was

converted to a categorical form to prepare for logistic regression analysis (Table 4.7).


Table 4.7 Comparison of energy intake distributions at week 8 for weight losing

and weight stable pancreatic cancer patients

Energy intake T8 (kJ/kg/d) 1 Weight losing

n (%)

Weight stable

n (%)

≤100

101-125

126-150

>150

13 (33%)

13 (33%)

9 (23%)

4 (10%)

8 (15%)

12 (22%)

15 (27%)

20 (36%)

Total 39 55

χ2 (1, 0.05) = 11.67 p = 0.010 Likelihood ratio (exact significance, two-sided) 1 three day food diaries and sip feed forms T8, week 8

4.1.7 Determinants of weight stability - multivariate analysis

Rationale for variables included in logistic regression

The following variables were included in the logistic regression: age, gender, stage

of disease, baseline BMI, CRP, and the presence of pain, loss of appetite, nausea and

vomiting, as well as plasma EPA level and energy intake at T8.

All variables hypothesised to be determinants of effective nutrition intervention (Aim

2) were included with the following exceptions. KPS was excluded because no

association was found between performance status and weight change group by Chi

square test. Energy intake at Week 8 was used rather than the baseline value because

it was a major focus of nutrition intervention. Including both baseline and T8 values

is likely to have resulted in overfitting.

Stage of disease, gender and age were included even though they showed no

significant difference, because they might be expected as confounders. Blood levels

for CRP (baseline) and EPA (T8) were included; the first as a measure of acute phase

response and the latter because it was the focus of the BH80 trial. It was important to

include these in case the ability of EPA to diminish inflammatory response (a


Table 4.8 The effect of different variables on the likelihood of patients with

pancreatic cancer being weight stable following eight weeks of nutrition

intervention (n=83) – logistic regression analysis.

Variable Odds ratio 95% CI p

Nausea/vomiting T0 Presence 1.0 referent Absence 6.5 1.6-27.2 0.010 Gender Male 1.0 referent Female 5.2 1.3-21.0 0.020 BMI T0 (kg/m2) <20 5.2 0.6-44.2 0.127 20-24 4.8 1.0-24.0 0.058 ≥25 1.0 referent C-reactive protein T0 (mg/L) <10 3.2 0.8-12.4 0.092 ≥10 1.0 referent Appetite loss T0 Presence 1.0 referent Absence 3.6 0.8-16.5 0.096 Energy intake T8 (kJ/kg/d) ≤ 100 1.0 referent 101-125 2.1 0.4-10.9 0.372 126-150 2.5 0.5-12.7 0.267 >150 3.5 0.5-24.7 0.200 Stage I-II 1.0 referent III 3.1 0.6-15.7 0.177 IV 2.6 0.5-12.8 0.245 EPA T8 < 3% 1.0 referent ≥ 3% 2.2 0.6-8.1 0.251 Age (years) ≤ 60 0.6 0.1-3.4 0.583 61-70 2.4 0.5-11.4 0.288 > 70 1.0 referent Pain T0 Presence 1.0 referent Absence 1.4 0.3-6.7 0.713

1 Logistic regression – each variable entered into the model after adjusting for all other variables.

T0, baseline; T8, week 8; EPA, eicosapentanoic acid; CI, confidence interval


hypothesis of the BH80 study) was a determinant of weight stability. Randomisation

grouping was not included however, because no difference was evident between the

groups and also because it was expected to be closely related to EPA levels at T8.

Results of logistic regression analysis

Eighty-three patients were included in logistic regression analysis; 49 (78%) WS and

34 (77%) WL. Only patients with complete data for each one of the variables being

analysed could be included.

Logistic regression analysis shows that after adjusting for the variables listed in

Table 4.8, the absence of nausea or vomiting at baseline and female gender were

independent determinants of weight stability.

The absence of nausea and vomiting at baseline increased the likelihood of being in

the WS group by a factor of 6.5 (odds ratio) compared to the referent category

(presence of nausea and vomiting). The 95% confidence interval is quite wide,

reflecting the small numbers in some categories and the large number of variables,

however the CI does not encompass “1” giving strength to the argument that lack of

nausea and vomiting increases the chances of weight stabilisation (ie. the probability

that the true odds ratio falls between 1.6 and 27.2 is 0.95). Similarly, women were

5.2 times more likely to be in the WS group than men after adjusting for the other

variables.

Variables that approached statistical significance (odds ratios > 3 and p values < 0.1)

were lack of appetite loss, CRP less than 10 mg/L and BMI 20-24 kg/m2, whereas

stage of disease, age group and absence of pain were not independent predictors of

weight outcome, nor were energy intake or plasma EPA level at T8.

4.4 Summary

Patients who were able to stabilise their weight loss or gain weight, lived longer from

baseline and reported better quality of life than those who continued to lose weight.


This difference in median survival time of three months is clinically significant for

patients with a disease that progresses rapidly. The absence of nausea and vomiting

at baseline and female gender were independent determinants of successful weight

stabilisation.

5 Methods - Aim 3

Aims 1 and 2 involved the examination of the large amount of quantitative data

available from the BH80 trial to compare the patients who continued to lose weight

with those whose weight stabilised following ONS. Aim 3 focuses on more detailed

information for the patients enrolled at the Australian sites. This includes both

quantitative data collected as part of the BH80 trial and qualitative data collected as

part of a sub-protocol (Figure 5.1).

Figure 5.1 Model representing Aim 3.

Weight

change Intake T8

ONS

pain, n&v, app,

taste, depression

Symptoms T0

E, Pr, F,

CHO Intake T0

BMI T0

PS T0

Stage of disease Psychosocial factors

E, Pr, F,

CHOfood texture

Patient’s

perceptions

QoL

beyond T8

Survival

ONS – oral nutrition support, QoL – quality of life, BMI – body mass index, PS – performance status, n&v –

nausea & vomiting, app – appetite, E – energy, Pr – protein, F – fat, CHO – carbohydrate, T0 – baseline, T8 –

week 8.

This aim includes the description of changes in taste perceptions, appetite and food

intake and an exploration of patients’ perceptions regarding changes in weight,

appetite and quality of life through case studies and patient narratives.

5.1 Comparison of qualitative and quantitative methods

Methods – Aim 3 82

Qualitative research methods have been a predominant part of the social sciences for

decades but are not as well established as quantitative methods in health research.

Essentially quantitative methods involve the collection of measurable data and

hypothesis testing in a controlled setting while qualitative methods involve studying

people in their natural social setting and deal more with words. Table 5.1 compares

features of the two methods to demonstrate the difference in approaches.

Table 5.1 Comparison of quantitative and qualitative research methods.

Quantitative Qualitative

Perception of the subject

matter

Reductionist; identification

and operational definition of

specific variables

Holistic; observing people

in the context of their social

environments

Positioning of researcher Objective; detached

observation and precise

measurement of variables

Subjective; close personal

interaction with subjects

Data base Quantitative; inter-

relationships among specific

variables

Qualitative; descriptions of

actions and related personal

meanings in context

Theories Normative; general

propositions explaining

causal relationships among

variables

Interpretive; providing

insights into the nature and

social contexts of personal

meanings

Theory testing Controlled; empirically

supporting or falsifying

hypotheses deduced from

theories

Consensual; matching

researcher’s interpretations

with those of subjects and

other observers

Application To collect data on incidence

of disease/symptoms and

evidence for efficacy of

interventions or validity of

assessment methods

To determine causes of

illness

To gain an understanding of

people’s perceptions of how

disease or health care affect

their lives.

Suited to situations with little

preexisting knowledge;

sensitive or complex issues

Adapted from Polgar & Thomas (Polgar & Thomas, 1995)

Methods - Aim 3 83

The main techniques used in qualitative inquiry are:

interviews with individuals or groups

direct observation of people’s behaviour

examination of documents

One of the major limitations of quantitative methods is that they involve “context

stripping” (Guba & Lincoln, 1998). The process of recording defined, objective

variables removes that information from the context in which it occurs, eg we can

measure weight change without recording information on whether the individual

perceives that change as distressing, beneficial or of no concern to them. It has been

suggested (Seligman et al., 1998) that in measuring the effect of appetite stimulants

in end-stage disease, quality of life factors such as an increased enjoyment of food

should be studied in addition to the quantitative measurements such as weight gain.

The failure of most studies to include this data may relate to the lack of valid and

objective measurement tools as well as a reluctance to use qualitative methods, either

due to distrust of the techniques or lack of skills.

Qualitative methodology has been used to examine clinical consultation processes

(Martin, Banwell, Broom, & Nisa, 1999; Tapsell, 1997) and there have been many

studies into patients’ perceptions of disease, disability and the delivery of health care

(Gaskill, Henderson, & Fraser, 1997; Grande, Todd, & Barclay, 1997; Little,

Jordens, Paul, Montgomery, & Philipson, 1998; Northouse, Schafer, Tipton, &

Metivier, 1999; Russell, 1997; Staniszewska & Ahmed, 1999). There are, however,

few examples in the literature of the use of qualitative methods in nutritional care of

cancer patients. The semi-structured interview method has been used to describe

caregivers’ perceptions of cessation of food intake for terminally ill cancer patients

(Meares, 1997), the emotional impact of anorexia in terminally ill cancer patients

(Holden, 1991) and nutrition related problems experienced by patients undergoing

peripheral blood stem cell transplant (McGrath, 2002).

Interest in the use of qualitative methods in health research is increasing as

demonstrated by articles in main-stream medical journals describing the potential

uses of non-quantitative techniques (Greenhalgh & Hurwitz, 1999; Holman, 1993;

Pope & Mays, 1995). The Australian National Health and Medical Research Council


has also addressed this change by providing guidelines for Institutional Ethics

Committees when assessing research proposals that involve qualitative methods

(National Health and Medical Research Council (Australia), 1995).

The complex nature of human health and disease and the particular strengths and

limitations of any one research method have led to the use of a combination of

methods. For example focus groups or individual interviews may be used to optimise

the content and wording of a survey questionnaire. Randomised controlled trials may

identify the most effective treatment for a disease but qualitative research may be

needed to identify reasons why that treatment isn’t successfully implemented.

The research question should determine the method employed and dietitians should

be addressing the complex range of issues that impact on patient outcomes. To do

this the profession needs to be skilled in the appropriate use of both qualitative and

quantitative methodology. While Aims 1 & 2 have relied solely on the traditional

quantitative methods I have included qualitative methods (see section 5.4 - sub-

protocol) with the aim of gaining a deeper insight into the nutrition-related

experiences of patients with unresectable pancreatic cancer.

5.2 Subjects

The detailed dietary information described in this chapter was obtained from patients

enrolled at the two Australian sites - Princess Alexandra Hospital (PAH) and The

Wesley Hospital (TWH) - as part of the BH80 study. Data was analysed from each of

the patients for whom at least two time points were available.

Patients recruited to the BH80 trial from one of the Brisbane sites (PAH) were

invited to participate in the additional study. There were expected to be

approximately 10-15 patients recruited to the sub-protocol.

5.3 Data from the BH80 study

Methods - Aim 3 85

Methods used to collect data in the BH80 study have been described in the earlier

methods chapter (section 3.1). Additional aspects related to Aim 3 are discussed

below.

5.3.1 Accuracy of dietary recording methods

All dietary reporting methods (eg. food records both weighed and estimated, food

frequency questionnaires, 24 hour recall and dietary histories) are subject to errors

when compared to energy expenditure as measured by the doubly-labelled water

technique (Black et al., 1993; Trabulsi & Schoeller, 2001). Under-reporting is

common (Black et al., 1991; de Vries, Zock, Mensink, & Katan, 1994; Livingstone et

al., 1990; Martin et al., 1996; Tomoyasu, Toth, & Poehlman, 1999) and in some

studies is associated with obesity (Black et al., 1993; Johnson, Goran, & Poehlman,

1994; Price, Paul, Cole, & Wadsworth, 1997; Voss, Kroke, Klipstein-Grobusch, &

Boeing, 1998). The ratio of reported energy intake to predicted basal metabolic rate

(EI/BMR) has been suggested as a method for detecting under-reporting in weight

stable study populations, enabling the exclusion of implausibly low EI/BMR values

that fall below specified cut-off levels (Goldberg et al., 1991).

In studies of weight losing cancer patients however, this is not useful, both because a

feasible EI/BMR cannot be estimated in the presence of weight loss and because of

the difficulties inherent in estimating BMR in cancer patients many of whom may

have alterations in metabolic rate. In a study of dietary intake (4 day food records)

and REE (indirect calorimetry) in 297 patients with solid tumours (Bosaeus,

Daneryd, Svanberg, & Lundholm, 2001) no systematic reporting errors were evident.

Cancer patients may respond differently to most study populations in that the focus is

usually on increasing food intake and weight rather than the more common concerns

regarding the health consequences of obesity.

Factors that may reduce the validity of self-reported food diaries as representative of

intake over the study period include:

the burden of completing the diary for patients who have fatigue, pain or

depression (frail patients were usually assisted by family members)


adjusting food types and quantity for the data collection period to make it easier

to record. Low energy reporting in some studies has been related to under-eating

(Goris & Westerterp, 1999).

wishing to please the researcher by appearing to meet study goal intakes

daily fluctuations in food intake due to nausea, vomiting, hospital admissions,

pain etc.

Factors that might increase the validity of self-reported food diaries include:

eating is a more “visible” process for many people in this group (and/or their

carers) because it becomes such an effort

in some cases such small amounts of food are consumed that they may be more

easily quantified

there are fewer inaccuracies when there is less variety in eating habits (Barnard,

Tapsell, Davies, Brenninger, & Storlien, 2002). Many of the study participants

were elderly with set eating routines and fairly simple meal patterns

there are fewer inaccuracies when there are reduced and less varied activity

levels (Barnard et al., 2002). Many of the study participants had low activity

levels as a result of their age and/or illness.

Under-reporting is associated with increased BMI in healthy populations, as

discussed above, and many pancreatic cancer patients may have been obese prior to

their illness-related weight loss. It is unknown whether the food reporting behaviour

of previously overweight people who have lost weight through illness, reflects older

patterns or whether they act more like healthy, weight-stable non-overweight people

and are less likely to be under-reporters. One of the benefits of additional qualitative

data is to examine the circumstances which may lead to these potential errors.


Data on macronutrient intake was obtained from the three-day diet diaries and daily

record of intake of study supplement (Sip Feed Form) described in section 3.1.5.

Average daily intake for energy (kJ), protein (g), fat (g) and carbohydrate (g) for the

Australian sites were calculated using the FoodWorks software programme

Methods - Aim 3 87

(FoodWorks 2.04 Professional Edition 1998,99 Xyris Software, Brisbane, Australia)

with data from all diaries entered by one person (WD).

Diary entries were clarified with patients at each data collection point, both to assist

in correct interpretation and to check that low reported intakes were not due to

incomplete diaries. No adjustments were made for unusually low or high intakes as

fluctuating intakes are expected in this patient group, however diaries that were

illegible or determined to be incomplete were excluded from data analysis.

Changes in total protein and energy intake were examined as well as proportions of

energy contributed by each of the macronutrients, before and during ONS.

Proportions of energy and protein contributed by meals versus supplements, and

general use of milk drinks and non-study supplements before and during the study

were described. Only the study supplements were considered as “supplements” in

analysis of the RCT (Chapters 3 & 4). Some patients however were already

consuming commercially available supplements or home-prepared nutrient dense

drinks at baseline. These have been identified as supplements through examination of

the food diaries to provide a more detailed description of alterations to choices of

food type and texture.

5.3.3 Anthropometry

Body weight was measured as per section 3.1.4 at enrolment, baseline, T4 and T8 as

well as at monthly reviews for as long as patients continued to participate in the

study, unless it was inappropriate due to nearness to death or if the patient was too

far away to enable home visits.

The patient’s actual body weight for each time point was used for calculating ratios

of energy or protein intake to body weight. No adjustments were made for extremes

of body weight. If it were expected that principally fat stores were being depleted

then there would be an argument for using a single weight for each person

throughout the study. One of the features of cancer associated weight loss, however,

is the loss of lean body mass in addition to fat loss. The dynamic nature of body


composition in this patient group and the many difficulties faced in trying to measure

it, mean that at present there is no clear answer to this question.

5.3.4 Quality of life questionnaire

At baseline, T4 and T8, patients completed the EORTC-QLQ-C30 (version 3) as

described in section 3.1.7. This was to be compared with qualitative data obtained

from patient narratives.

5.3.5 Appetite

Raw scores were used for describing responses to the appetite question in the

EORTC QLQ-C30 rather than transforming them to the standardised scale score (0-

100) as was done for the BH80 study in section 3.1.8. Either method is an accepted

means of reporting this data (Fayers et al., 1999).

Five linear analogue scales (LAS) regarding appetite and satiety were also completed

as described in section 3.1.8. These scales have not been validated, but as they were

part of the BH80 data set the results were examined to see whether they appeared to

provide useful information for this clinically important issue. No tool was included in

the BH80 study to attempt to measure early satiety. Responses to the third LAS

question, however, “How much food do you think you could eat right now?”, might

reflect a patient’s experience of early satiety at previous meals.

5.3.6 Functional status

The Karnofsky Performance Score (Appendix 5) (Yates et al., 1980) was

documented at recruitment, baseline, T4 and T8.

5.3.7 Statistical methods

Descriptive statistical analyses were performed using SPSS (version 10, SPSS Inc.

Chicago) software package. Results are expressed as percentage or median (range).

Methods - Aim 3 89

5.4 Sub-protocol

All patients recruited to the BH80 trial through Princess Alexandra Hospital from

September 1999 were invited to participate in an additional study which included a

structured interview with the patient at baseline, reviews of qualitative changes to

food intake at regular intervals, and a recorded, semi-structured interview with the

patient (and spouse/carer) following T8. Recruitment of ten to fifteen patients was

planned.

The aims were to describe changes in taste perceptions, appetite and food intake and

to explore patients’ perceptions regarding changes in their weight, appetite and

quality of life.

5.4.1 Ethical issues

This research proposal was approved by the Queensland University of Technology

Human Research Ethics Committee (QUT 1766H), Princess Alexandra Hospital

Research Ethics Committee (subprotocol for protocol # 27/99 approved 4 May

1999), and the sponsor, Abbott Laboratories. Patients who agreed to participate in the

additional study completed a separate patient information and consent form for the

sub-protocol (Appendix 7).

Patient confidentiality was maintained by using the same four digit code number

allocated for the BH80 trial. That number, along with the patient’s initials was the

only identifier on data collection sheets. For ease of interpretation of the data from

the Australian sites, those codes have been transferred to smaller consecutive

numbers (Appendix 8).

Because the BH80 trial already required frequent contact with the participants, the

collection of additional data for the sub-protocol had little impact on patient burden.

Patients were asked whether additional information could be sought from carers as

part of the consent process. The focus of the study was on the patient’s experience


rather than care-giver issues, however carers could provide valuable additional

insight and, as another source of information, this would assist with validation.

5.4.2 Structured Interview

Qualitative information on pre-illness dietary habits, taste and appetite changes and

barriers to food intake was obtained at baseline during a structured interview using

open-ended questions. Responses were recorded on prepared forms (Appendix 9 T=0

structured interview).

5.4.3 Recent Food Intake Report

At weeks 2, 4, 6 and 8 and at the monthly reviews that followed the main eight week

study period for the BH80 trial, a food intake report (Appendix 10 recent food intake

report) covering the previous 1-2 weeks were taken in order to determine any major

changes in the food types eaten as the disease progressed. Patients’ perceptions of

barriers to food intake were also recorded.

Patients were routinely contacted at these times as part of the BH80 trial to discuss

intake & symptoms and so there was minimal added patient burden. These contacts

were either in person or by phone and information was recorded in writing using the

forms described.

5.4.4 Semi-structured Interview (Narrative)

After the completion of the eight week BH80 study period, a semi-structured

interview was conducted with each patient (and also with the spouse/carer if

possible) and audio-taped. The timing for the narrative (at week 9 or 10) was chosen

because it was beyond the main BH80 study period and therefore would not interfere

with the administration of the quality of life questionnaires.

The patients were encouraged to describe their thoughts and feelings regarding the

ways in which changes in body weight, taste and appetite had affected their daily life

Methods - Aim 3 91

during their illness. Guide questions (Appendix 11 T=9/10 semi-structured

interview) were developed to help focus the interview (McCracken, 1988).

The interviews were tape recorded and transcribed verbatim. This is a time

consuming process but facilitates analysis of the conversation, retaining detail and

reducing the influence of preconceptions. Copies of both the original tapes and the

transcripts have been retained for further scrutiny if required. The transcription

notation used is described in Appendix 12 (Maykut & Morehouse, 1994; Mishler,

1986; Silverman, 1993).

The interviews were to have been analysed for recurring themes (categories) using

the “constant comparative method” described by Maykut and Morehouse (Maykut &

Morehouse, 1994) which involves coding data pages to their sources, unitizing the

data into “chunks of meaning”, further categorizing and developing rules for

inclusion to these categories. There were, however, insufficient patients enrolled into

the sub-protocol to make thematic analysis of the narratives feasible. Instead, the

narratives have been used as a basis for descriptive case studies as they provide

examples of many of the issues faced by patients with unresectable pancreatic

cancer. The findings that come from this form of data collection and analysis are not

generalizable but rather provide an insight into the experience of the particular

informants.

Validation methods for qualitative data include “triangulation” in which data

obtained using different methods are compared to see if they corroborate one

another, and “respondent validation” in which research findings are taken back to

informants to see if they consider the conclusions to be valid (Miles & Huberman,

1994). Kirk and Miller (Kirk & Miller, 1986) suggest that “reliability depends

essentially on explicitly described observational procedures”. The issue of reliability

was addressed by carefully documenting all stages of the research process, audio-

taping and transcribing interviews and keeping a research journal.

If sufficient interviews had been obtained to proceed to content analysis, independent

assessment of the analysis by a person experienced in qualitative methods would

have been pursued to enhance reliability.


The additional information collated for Aim 3 was used to examine how changes in

weight and appetite are perceived by the patient, provides two case studies and

informs the algorithm for medical nutrition therapy for unresectable pancreatic

cancer presented in the discussion.

6 Results – Aim 3 Both quantitative and qualitative data are used in this chapter to describe more fully

the disease process experienced by the patients recruited to the BH80 trial by the

Australian sites.

6.1 Description of Australian participants

Eighteen patients were enrolled in the study – nine through The Wesley Hospital and

nine through the Princess Alexandra Hospital. Five patients died or withdrew from

the study before week 4 data could be collected. The remaining thirteen patients are

discussed below and have been designated patient ID numbers 1-13 (decoded in

Appendix 8). While recruitment of 10-15 patients to the sub-protocol had been

anticipated, only three patients were able to be recruited before the BH80 trial closed.

This was due to lower than expected overall recruitment numbers as well as delays in

obtaining ethics approval. No patients who were invited to participate in the sub-

protocol declined.

The median age of the thirteen patients was 71yrs with a range of 51-90yrs. Four

patients (3F:1M) were randomised into the experimental group and received the high

protein, high calorie supplement containing fish oil (group E). The nine patients

(3F:6M) in the BH80 control arm received the isocaloric, isonitrogenous supplement

(group C).

Six patients were recruited within 1-2 months of diagnosis but the other patients had

been diagnosed 5 -14 months earlier. One patient entered the study following the

diagnosis of recurrent disease. Two patients had been diagnosed with diabetes

mellitus; one 18 years and another 4 months prior to the study.

Eligibility criteria for the BH80 trial excluded patients who had received surgery,

chemotherapy or radiotherapy within four weeks of baseline data collection. While

all patients met these criteria, all but two patients had received some medical

94 Results – Aim 3

intervention prior to the study. Biliary stenting and gastric bypass were the most

common procedures and the one patient with recurrent cancer had undergone a

partial pancreatic resection (Whipple Procedure) two years earlier.

Three patients received pancreatic enzyme replacement from the time of diagnosis,

one commenced 2 months prior to the study and one during the study period. 62%

patients (8/13) did not receive enzyme replacement at any stage. There were no study

guidelines for the brand or dosage of pancreatic enzyme supplement prescribed.

Self-reported pre-illness body weight varied greatly (47-117kg). 70% patients (9/13)

were overweight or obese prior to the weight loss associated with the pancreatic

cancer. Pre-illness Body Mass Index (BMI) ranged from 19 to 35kg/m2, with the

following number of patients fitting into each BMI category; 2 = BMI<20, 2 = BMI

20-25, 5 = BMI 25-30, 4 = BMI>30. These were based on reported weights with

supporting evidence obtained from family, general practitioner and medical charts.

Patients were recruited to the BH80 trial at different stages in their disease trajectory,

and it was very difficult even for experienced clinicians to predict survival duration.

Three of the thirteen patients included in this analysis died within the eight week

study period, however one of the patients survived more than a year from enrolment.

Survival time from diagnosis (or diagnosis of recurrent disease in one case) ranged

from 93 - 794 days (median 213 days).

6.1.1 Comparison with BH80 participants.

The patients recruited from the Australian sites appear to be similar to the BH80

patients who were eligible for analysis in Chapters 3 and 4 except for disease

recurrence (fewer) and baseline weight (higher) (Table 6.1). This was not tested

statistically due to the small numbers from the Australian sites.

Results – Aim 3 95


pancreatic cancer from the Australian sites versus the combined sites.

Baseline variable Australian pts

(n=13)

All sites

(n=107)

Age (yrs)

median (range)

71 (51-90) 67 (47-87)

Gender (M:F)

7:6 62:45

Weight (kg)

median (range)

70.6 (37.0-94.6) 64.0 (36.0-94.6)

BMI (kg/m2)

median (range)

25.0 (16.9-28.2) 22.0 (13.4-33.2)

Stage of disease

(I-II:III:IV)

5:2:6 51:22:32

Recurrence

n (%)

1 (8%) 16 (15%)

Diabetes

n (%)

2 (15%) 16 (15%)

Enzyme Use

n (%)

5 (38%) 33 (31%)

6.2 Body weight changes

Weight loss, as is typical for pancreatic cancer, was dramatic. Figure 6.1 shows the

percentage of pre-illness body weight lost by these patients up until the last available

weight measurement. The greatest proportion of weight loss occurred prior to the

study. Three patients lost more than one third of their body weight during the course

of their illness and an additional six patients lost at least 20% of their body weight.


Figure 6.1 Percentage of pre-illness body weight lost by each Australian patient

with unresectable pancreatic cancer until final weight measurement. 1

05

10152025303540

1 2 3 4 5 6 7 8 9 10 11 12 13

Patient ID#

Perc

enta

ge b

ody

wei

ght l

ost

1 % weight loss = (pre-illness weight – final weight)/pre-illness weight x 100 ; weights were censored

if ascites or oedema were recorded, in which case the prior weight was used; duration of recorded

weight loss (Pt ID# - weeks): 1 - 64, 2 - 32, 3 - 123, 4 - 20, 5 - 44, 6 - 26, 7 - 37, 8 - 14, 9 - 28,

10 - 32, 11 - 22, 12 - 84, 13 - 8.

Body weight was measured monthly beyond the eight week study period for most

patients. It was common for body weight to increase in the weeks before death as a

result of oedema or ascites. Any weights recorded when these were clinically evident

were excluded from analysis (Table 6.2).


Table 6.2 Changes in body weight (kg) over time, of patients with unresectable

pancreatic cancer recruited into the BH80 study from Australian sites.

Pt

ID

Preillness

weight1

T0

weight

T4

weight

T8

weight

Final

weight

Weeks

T0 to

final wt

Weeks

final wt

to death

1 95 75.8 72.2 68.4 62 16 4

2 75 65.2 63.8 62 56.4 20 3

3 67 43.6 44 41.82 42.4 11 22

4 50 37 37.2 n/a 37.2 4 1

5 117 94.6 94 91.8 73.2 3 32 10

6 83 70.6 70.8 71.2 66.2 4 19 6

7 52 48 46.4 43.4 40 13 3

8 82.6 77 77.4 73.4 73.4 8 3

9 70.7 59 56 n/a 56 4 2

10 47 43.8 42.8 n/a 42.8 4 17

11 95 86 79 (82.8) 5 79 5 4 5

12 92 82.2 82.8 81.8 80 32 12

13 97 75.6 (79.2) 6 n/a 75.6 6 0 6 T0, baseline; T4, week 4; T8, week 8 1 self-reported weight 2 measured T8 wt but patient was dehydrated

Weights in brackets were censored (ie. not used in calculations) if oedema or ascites were recorded 3 actual final

wt (week 38, 75kg) censored due to oedema 4 actual final wt (week 24, 75kg) censored due to oedema/ascites 5 actual final wt (T8, 82.8kg) censored due to oedema 6 actual final wt (T4, 79.2kg) censored due to oedema

Some patients had initially been pleased with the loss of weight (prior to diagnosis)

but most were concerned by it.

“I don’t think I’d like to go back to that weight before…… and I think for my age

that I don’t want to be carrying a big weight around” (Pt #5 Week 12)

“Well he was upset last month when he lost – dropped that two kilos” (Wife of Pt #5

Week 12)

“..things don’t fit and I look terrible in the mirror – scrawny…….it’s what depresses

me next after the pain is the weight loss…………..the boys at first used to comment

on it – now they realise not to say it anymore.” (Pt #7)


6.3 Changes in dietary intake

Food intake data for four of the Australian patients was only available to T4. These

patients died within ten weeks of baseline (Group 1) and have been considered

separately to the eight patients who lived longer than ten weeks (Group 2). Food

intake data was not available beyond baseline for one patient (Pt #10).

Energy intakes, reported as kJ/kg body weight/day, for the eight patients who lived

more than ten weeks beyond baseline, varied widely throughout the study period (44

- 188kJ/kg BW/d). Data from patients who died within ten weeks of baseline were

not considered here because body weights were often affected by oedema.

Participants who maintained their weight at T4 and T8 to within 1 kg of the previous

value, had an intake of at least 100 kJ/kg/d. One person, however, was unable to

maintain their weight on a reported intake of 160 kJ/kg/d. (NB. These were

calculated on the patient’s actual body weight at each data collection point and was

not adjusted if outside the “Healthy Weight Range”.)

6.3.1 Energy intake.

The range of energy intakes remained wide over the eight week study period – from

3 to 13 MJ/d (median 7 MJ/d). Total energy intake decreased between baseline and

T4 for each of the four patients who survived less than ten weeks from baseline

(Group 1). Half of the other eight patients (Group 2) increased their total energy

intake by at least 1700 kJ/d. (Figure 6.2)


Figure 6.2 Change in energy intake1 between baseline and week 4 for patients with

unresectable pancreatic cancer recruited into the BH80 study from

Australian sites.

94

13

111

2 3

6

12 5 8 7

-3000

-2000

-1000

0

1000

2000

3000

Patient ID#

Cha

nge

in to

tal i

ntak

e (k

J/d)

Group 1 Group 2

Group 1 = patients who died within 10 weeks of baseline

Group 2 = patients who lived longer than 10 weeks from baseline 1 three day food diaries and sip feed forms

The energy intake for each patient at each available time point is shown in the

following two figures.- Figure 6.3 for the group who survived less than ten weeks

and Figure 6.4 for the longer lived patients. The uncoloured area represents energy

from supplements. At baseline this refers to commercially available supplements

such as “Sustagen” or “Ensure” in addition to any milk used to make the drinks.

Patients were requested to consume no supplements other than the study product

during the eight week BH80 study period so for T4 and T8 the uncoloured area

represents the study supplement only.


Figure 6.3 Energy intake1 at baseline (T0) and week 4 (T4) for patients with

pancreatic cancer recruited into the BH80 study from Australian sites

who survived less than 10 weeks from baseline (Group 1).

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

9 4 13 11

Patient ID# - T0 and T4

Ener

gy (k

J/d)

food only supplements

T0

T4

T0

T4

T0

T4

T0 T4

Supplements at T0 = any commercial product, and at T4 &T8 = BH80 study product 1 three day food diaries and sip feed forms


Figure 6.4 Energy intake 1 at baseline, week 4 and week 8 for patients with unresectable pancreatic cancer recruited into the BH80 study from

Australian sites who survived more than 10 weeks from baseline (Group 2).

0

1 0 0 0

2 0 0 0

3 0 0 0

4 0 0 0

5 0 0 0

6 0 0 0

7 0 0 0

8 0 0 0

9 0 0 0

1 0 0 0 0

1 1 0 0 0

1 2 0 0 0

1 3 0 0 0

1 4 0 0 0

1 5 0 0 0

1 2 3 6 1 2 5 8 7

P a tien t ID # - T 0 , T 4 & T 8

Ener

gy in

take

(kJ/

d)

fo o d o n ly su p p le m e n ts

Supplements at T0 = any commercial product, and at T4 &T8 = BH80 study product only 1 three day food diaries and sip feed forms


Table 6.3 Macronutrient contributions to total energy1 for patients with

unresectable pancreatic cancer recruited into the BH80 study from

Australian sites.

% total energy intake Pt ID Time

protein Fat CHO alcohol

T=0 22 32 46 0T=4 20 31 49 0

1

T=8 21 27 52 0

T=0 28 19 53 0T=4 18 23 59 0

2

T=8 17 26 57 0

T=0 17 31 52 0T=4 18 26 56 0

3

T=8 20 21 59 0

T=0 21 31 48 0T=4 21 26 53 0

4

T=8 - - - -

T=0 15 43 41 1T=4 19 37 43 1

5

T=8 23 31 45 1

T=0 18 43 39 0T=4 19 39 42 0

6

T=8 19 40 41 0

T=0 22 42 36 0T=4 22 32 46 0

7

T=8 20 33 47 0

T=0 16 25 59 0T=4 18 24 58 <1

8

T=8 17 30 53 0

T=0 16 32 52 0T=4 22 29 49 0

9

T=8 - - - -

T=0 17 28 55 0T=4 - - - -

10

T=8 - - - -

T=0 27 29 44 0T=4 18 29 53 0

11

T=8 - - - -

T=0 22 32 36 10T=4 19 28 48 5

12

T=8 20 30 44 6

T=0 19 33 48 0T=4 23 24 53 0T=8 - - - -

13

1 three day food diaries and sip feed forms T=0, baseline; T=4, week 4; T=8, week 8; CHO, carbohydrate


6.3.2 Macronutrients

Table 6.3 shows the relative contributions of macronutrients to total energy for each

patient. The proportion of energy provided by fat decreased over the study period by

at least 5 percentage points in more cases than it increased, whereas the opposite was

true for carbohydrate. There was no pattern to the changes for percentage of energy

contributed by protein. Only one patient reported an alcohol intake greater than 5%

of energy. The majority of patients (10/13) consumed no alcohol.

The similarities in macronutrient distribution between patients disguises the dramatic

differences in both the quantity and nature of foods consumed. Some patients

continued to eat a wide range of foods during the study, but others limited their foods

considerably. At baseline, patient #4 was eating very small amounts of soft foods

such as lasagne and omelette, but mainly relied upon soups, yoghurt and high protein

drinks. This was reported to be due to extreme fatigue. Similarly, patient # 11

reported only ten different foods in his T4 three day food diary compared to more

than 20 foods at the commencement of the study. Both of these patients died within

ten weeks of baseline.

Patient #2 selected a wide range of foods during the eight week study period, but by

week 10 he reported eating mostly fruit and soups due to chronic abdominal pain and

loss of appetite. In contrast, patient #6 continued to enjoy his meals, including bacon

and eggs for breakfast, until close to his death.

6.3.3 Protein intake

Protein intake for each patient is shown in Table 6.4. Daily intakes at baseline ranged

from 51 – 136 g (median 64 g) with four patients consuming less than 1 g protein/kg

body weight. Forty-two percent of patients (5/12) managed to increase their protein

intake during the first four weeks of ONS, but intake declined for an equal number.

None of the patients who died within ten weeks of baseline had been able to increase

their protein intake.


Table 6.4 Total protein intake1 for patients with unresectable pancreatic cancer

recruited into the BH80 study from Australian sites.

Baseline Week 4 Week 8 Pt ID# g/d g/kg BW/d g/d g/kg BW/d g/d g/kg BW/d

1 55 0.7 36 0.5 47 0.7

2 123 1.9 75 1.2 57 0.9

3 54 1.2 54 1.2 55 1.3

4 62 1.7 35 0.9 - -

5 79 0.8 119 1.3 141 1.5

6 136 1.9 147 2.1 151 2.1

7 68 1.4 96 2.1 89 2.1

8 60 0.8 88 1.1 75 1.0

9 62 1.1 55 1.0 - -

10 51 1.2 - - - -

11 64 0.7 39 0.5 - -

12 89 1.1 94 1.1 120 1.5

13 83 1.1 85 - - - 1 three day food diaries and sip feed forms BW, body weight

None of the patients described themselves as following a vegetarian diet before or

during the study. Half of the patients (7/13), however, recorded at least one day in

their food diaries on which they ate less than 30 g of meat, fish or chicken. Dairy

foods and eggs appeared to be used frequently as a substitute.

One patient (Pt #1) reported limiting cow’s milk intake due to a belief that it

provoked sinusitis and patient #2 drank very little cow’s milk but used soy milk and

“Sustagen”. Patients # 3 and 11 chose to continue using low fat milk, whereas patient

#8 switched to full cream milk after discussion with the study dietitian. Most patients

used milk of some type on a daily basis but only for cereal, tea or coffee.

For 69% of patients (9/13) dairy foods and supplements provided more protein than

did meat, fish, chicken or eggs based on food diaries for at least one data collection

point. Supplements provided as much as 77% total protein at T4. For the patients


surviving longer than 10 weeks beyond baseline, supplements provided between 19-

43% total protein at T4 and T8.

6.4 Nutrition supplements

6.4.1 Supplement intake

Half the patients (6/13) were already consuming 500 – 4100 kJ/d from liquid

nutritional supplements (“Ensure” or “Sustagen”) before the study commenced. Such

supplements contributed 8-80% of their total energy intake at baseline. Energy intake

from food and supplements is shown in Figures 6.3 and 6.4.

At T4, intake of the study supplement ranged from 0-2500 kJ/d, contributing 0-74%

total energy. At T8, intake of the supplement, for the remaining eight patients, ranged

from 800-2500 kJ/d, contributing 17-33% total energy. The patients who survived

less than 10 weeks from baseline (Group 1) took a greater proportion of intake as

supplement.

Some patients reported great difficulty in consuming the study supplement. In the

first four weeks of the study only 38% patients (5/13) were able to consume the goal

of at least 1.5 x 237ml cans of supplement each day. By T8, five of the remaining

nine patients averaged at least 1.5 cans/day. Only three patients (Pts #5,6 and 8) were

able to consistently meet the goal intake of supplement. These patients consumed

total energy intakes of 8-13 MJ/d. Patients #5 and #6 consistently reported little or no

loss of appetite during the eight week study period, however patient # 8 consumed at

least 1.5 cans each day in spite of reporting some loss of appetite at each data

collection point.

Supplements became increasingly relied upon by some patients as liquids became

easier to consume than solid foods. For each of the patients who survived less than

10 weeks from baseline, both total energy intake and intake of energy contributed by

food had decreased at T4 with the supplement supplying at least 30% total energy

(Figure 6.3).


For the longer-lived group, two patients increased their food intake by more than

20%, two decreased food intake by more than 20% and food intake for the other four

was reduced to a lesser extent (<20%) (Figure 6.4). The inaccuracies inherent in the

recording and analysis of diet diaries make variations less than 20% unreliable.

6.4.2 The role of supplements

The study supplement and other supplements prior to the study, appeared to play a

number of roles. Patients who had already commenced high protein drinks before

enrolling in BH80, identified them as a standard method for regaining lost weight

due to illness.

Once the eight week study period was over, some patients stopped taking it, some

continued to struggle on with the supplement “just in case” and others continued

enthusiastically. One patient required frequent reassurance that the supply would

continue;

“You know that’s very important to me! You know I worry if I don’t have it there…

it’s keeping me going.” (Pt#5)

Another patient (#2) continued to take the product in hospital in his final days of life

even though he was aware of his imminent death and in spite of discouragement by

nursing staff. Suffering from fatigue, anorexia, abdominal pain and extreme early

satiety, he saw the study supplement as a way to maximise the value of each

mouthful. He also used the expression “it’s keeping me going”.

Patients often stated that they had difficulty with the taste or richness of the

supplement and demonstrated strong individual preferences for their temperature and

texture. Individuals tended to be consistent with their method for consuming the

supplement, whether drinking a can first thing in the morning before breakfast or

sipping on it throughout the day. Sometimes it was diluted with water or milk, or

flavoured with “Milo” or “Sustagen”.


While most drank the supplement cold, others preferred it at room temperature,

warmed slightly or semi-frozen. There were often strong preferences for one of the

two flavours (vanilla or chocolate-hazelnut). None of the patients at the Australian

sites who were provided with a range of flavouring sachets, chose to continue with

them, suggesting that people struggling with the supplement may have been having

problems such as nausea and early satiety that reached beyond taste issues. Some

patients disliked the supplement in spite of enjoying sweet milky drinks.

Despite the problems that some patients found in meeting the study goal for intake of

the supplement, a liquid source of nourishment was reassuring for patients and carers

when patients were unable to eat solid foods.

6.5 Changes in appetite and taste perceptions

6.5.1 Appetite changes

At baseline no patient reported a score of 4 (“lacked appetite very much”) for

appetite loss on the EORTC QLQ-C30, but by T4 three of the patients had shifted

from scores of 3 to 4 (Table 6.5). This could reflect a worsening of their condition as

part of the disease process, or the effect of including supplement between meals

when appetite was already poor. Of these patients (# 4,7,11) one died one week later,

one died five weeks later and the other reported “forcing down” the supplement but

survived a further twelve weeks from T4. Only three patients reported scores of 1 or

2 (“no lack” or “little lack of appetite”) throughout the eight week period – these

patients lived for four to eleven months beyond T8. They did however report

reductions in appetite as their disease progressed.

One patient identified factors other than cancer as contributing to poor appetite;

“I’ve never had a terrific appetite and I think smoking is part of the problem

there…” (Pt#7)


Some patients reported better appetite in the earlier part of the day and others the

latter. One patient (#12), who reported no loss of appetite, ceased taking the

supplement after the study period, preferring instead to focus on enjoying food.

Table 6.5 Self-reported lack of appetite scores1 for patients with unresectable

pancreatic cancer at the Australian sites immediately prior to baseline,

week 4 and week 8 assessments over eight weeks nutrition intervention.

Patient ID Baseline Week 4 Week 8

1 +++ +++ ++

2 +++ ++ ++

3 ++ +++ +++

4 +++ ++++ n/a

5 + ++ +

6 + + +

7 +++ ++++ ++++

8 +++ ++ ++++

9 ++ +++ n/a

10 +++ + n/a

11 +++ ++++ n/a

12 + + +

13 ++ +++ n/a 1 EORTC QLQ-C30 Q13 “During the past week have you lacked appetite?”

+ = not at all, ++ = a little, +++ = quite a bit, ++++ = very much

6.5.2 Early satiety

Some patients found they were rarely hungry. Others found that in spite of looking

forward to a meal they would often eat much less than they expected.

“….he’d just eat, and then it was though it was turning off a tap. That was it, he just

couldn’t eat anymore.” (Wife of Pt #5)


“I just don’t know what it is. I just say to (his wife) that’s it. I’ve just had enough. …

but really enjoying everything up till that moment..” (Pt #5)

Patient #2 complained of a “lack of space”, especially in the first half of the day, and

this was reflected in responses above 50 at each time point for the pre-lunch LAS

question “how full do you feel right now?”. This patient was troubled by abdominal

pain and distension throughout the study. It is difficult to know how patients

interpreted this LAS question however because Patient #6 regularly reported

responses of 70 to 80 and yet he also reported having a very good appetite and his

food records indicated energy intakes of 13,000 kJ/day. He reported abdominal

discomfort but stated that it did not limit his food intake.

“Oh I could eat anything – anything you put in front of me.” (Pt#6)

6.5.3 Taste changes

Alterations to taste were common and specific to individuals. Some patients reported

a diminished sense of taste while others could identify specific foods that had

become unpleasant.

“It’s hard to explain - sort of no taste, no flavour you know….like nothing was sweet

or anything like that.” (Pt#5)

“(I) went off certain foods. In the beginning it seemed to be coffee. Why, I don’t

know because I’ve always enjoyed it. And my favourite dessert used to be

“Vienetta” but I was eating it one night and I felt really sick so that – I didn’t buy

that anymore. So that was the end of that.” (Pt#7)

Tastes sometimes normalised in spite of the progression of disease, for example

Patient#7 regained her enjoyment of coffee and chocolate.

Meat was noted by some patients as being a food they still enjoyed, with Pt#2

reporting his favourite smell as “fatty meat cooking”.


“Like savoury mince I love at the minute you know because the onion taste is a

stronger taste ….and I find I can eat that with no problems.” (Pt#7)

Aversions to the smell of food being prepared were reported. This was a particular

difficulty for this patient who lived alone;

“when I got home (from hospital) I found that smelling the cooking would

sometimes make me feel nauseous. Not always but sometimes .. and having to cook

it myself didn’t help.” (Pt#7)

In contrast to these, patient #6 did not experience any taste changes and continued to

report a “good” appetite in the week before his death.

6.6 Symptoms and quality of life

The ability to continue living independently was an important issue for some

patients.

“I won’t be beaten ……... I’m still going to go an’ – well lead as normal a life as I

can for as long as I can” (Pt#7)

Functional status varied widely. Some patients were confined to bed within weeks of

diagnosis and others continued with their usual occupation or activities in retirement.

Patient #6, who was used to a physically active life on cattle properties, walked for

hours each day and felt that this eased his abdominal pain. While some patients with

good performance status had few symptoms, this was not always the case. Patient #1

continued to work and travelled to visit relatives in spite of major problems with

pain, nausea and vomiting.

Although some patients reported no barriers to food intake in the eight week study

period, they all experienced symptoms that impacted on their nutritional status as

their disease progressed.


6.6.1 Pain

Pain was common and a major barrier to intake. Some patients were reluctant to ask

for better pain relief or to use narcotic analgesic medication even once it had been

prescribed. Frequent review by the dietitian enabled the identification of inadequate

pain control. In some cases patients needed hospital admissions for pain

management.

Abdominal pain was sometimes associated with gastrointestinal disturbances;

“wind”, “indigestion” or constipation making patients reluctant to eat until it was

resolved.

6.6.2 Nausea & vomiting

Nausea with or without vomiting was a common symptom. Vomiting sometimes

indicated gastric outlet obstruction caused by the enlarging tumour and required

surgical intervention (Pt #6). It was also sometimes a chronic symptom that was

difficult to control. Vomiting occurred frequently for patient #1 throughout the study,

confounding food intake data and causing major day to day fluctuations in food

intake.

Nausea was sometimes triggered by the smell of food being cooked which was

especially a problem for people cooking for themselves. It was also associated with

morphine use. Nausea and the resulting reduction in food and fluid intake, as well as

frequent vomiting, meant that some patients required hospital admission for

dehydration.

6.6.3 Asthenia

Patients often reported becoming more easily fatigued than usual and needing to rest

during the day. This made shopping and meal preparation difficult for patients living

alone.


“I find I just can’t do what I used to, right. Heavy shopping .. I’m findin’ a way

round it if you like, you know.” (Pt#7)

“..he hasn’t got the strength he had, you know, before his op.” (Wife of Pt#5)

Patients who were able to resume or continue household tasks, gardening, walking

and travel, appeared to value this ability.

6.6.4 Constipation/diarrhoea

Pancreatic cancer and medication for pain relief often led to alterations in bowel

function. Abdominal pain and constipation were ongoing and distressing problems

for patient #2 who reported that the discomfort of constipation limited his food

intake.

Fibre intakes varied from negligible levels for patient #4 who was consuming only

liquids and very small amounts of mashed foods, to more than 30 g/day (Pts #5 and

6). Patients consuming larger amounts of fibre were eating a greater overall amount

and range of food. They were also able to consume the two cans of supplement

which provided 8 g of fibre. While dietary fibre may have been a contributing factor

to changes in bowel function there was no consistent connection between fibre intake

and gut symptoms.

6.6.5 Xerostomia

Patient #2 found some foods such as potato “getting stuck” because of his dry mouth

and adapted to this by increasing the use of gravies. Liquids such as the study

supplement were sometimes found to be easier to consume than solids;

“sometimes I’ve had that instead of a sandwich…. ‘cause I used to have the dry

mouth and it was easier to get it down.” (Pt#7)


6.7 Summary

The patients recruited by the Australian sites appeared representative of the BH80

participants. The majority had been overweight or obese prior to their illness but

there was a wide range of body weights. All lost a large proportion of their weight

during the course of their disease. Protein and energy intakes also varied widely with

intakes usually decreasing as the disease progressed. The range of foods, as well as

the amount, declined in the final weeks of life. Use of commercial supplements was

common even before study entry, but some patients found them difficult to consume.

These findings demonstrate that people with unresectable pancreatic cancer not only

have to deal with the awareness of having an incurable disease, but also with an array

of symptoms that affect their quality of life on a daily basis. The range and degree of

problems were highly individual. Symptoms such as pain, taste changes, early

satiety, nausea, anorexia, alterations to bowel habit and xerostomia severely hindered

food intake. New symptoms developed during the course of the disease requiring

ongoing reassessment of symptom management strategies and liaison with other

members of the health care team.

The following case studies demonstrate the varied nature of nutrition-related

symptoms experienced by people with unresectable pancreatic cancer.

6.8 Case Studies

6.8.1 Case 1 - Mrs A

Mrs A, a widow in her sixties, had been recruited to the study one month after being

diagnosed with unresectable pancreatic cancer. She had experienced back pain,

nausea, anorexia and fatigue but was not jaundiced and so had not undergone any

invasive medical procedures.

Only minor weight loss had occurred at diagnosis but weight loss continued and was

a concern to her. On occasions when she had unintentionally lost weight in the past,


she had been able to regain weight with an increase in high calorie foods. Mrs A had

a preillness BMI of 19 kg/m2, and felt very self-conscious as she became markedly

underweight. Over the course of her illness she lost 12 kg (23 % of her usual body

weight), dropping to a BMI of only 15 kg/m2 a few weeks before she died.

Mrs A lived alone and was keen to maintain her independence. She had the support

of a neighbour as well as family members. Pain and loss of appetite were the main

factors that limited her food intake, along with nausea, depression, dry mouth and

taste changes.

Her appetite improved on occasions when she visited her family, surprising herself

with the amount of food she could eat, however she avoided visits if her pain was not

under control. Her pain was such that she was commenced on morphine soon after

diagnosis. She found that her bowel function shifted from loose stools to constipation

and then to diarrhoea in the final weeks.

Mrs A was able to increase her overall energy intake in spite of her poor appetite,

from 5400 to 7600 kJ/d and maintained this at week 8. This was achieved with use of

the supplement as well as strategies to overcome taste changes and xerostomia.

Consuming the goal amount of at least 1.5 cans of supplement each day proved a real

challenge for her. She would sometimes substitute a glass of the study supplement

for a sandwich at lunchtime if she felt very tired but she found it difficult to consume

the supplement in addition to food. She found the supplement very rich and diluted it

two parts supplement, one part milk and one part water.

Mrs A acknowledged that her heavy smoking habit probably contributed to the dry

mouth and poor appetite. Her appetite was best in the earlier part of the day and

worst in the evening.

Taste changes were present at diagnosis and were not constant such that she was able

to reintroduce some foods. Meat tasted normal but vegetables, especially potatoes,

tasted “wrong”. Some of her favourite foods were no longer enjoyable, eg. coffee,

white sauce, chocolate and icecream. They either caused nausea, tasted “wrong”, or

tasted correct but no longer appealed to her.


These taste alterations and feeling nauseated by cooking smells, as well as increasing

fatigue, meant that she often ate soup or a bowl of cereal in place of her usual meat

and vegetables, and sometimes slept through a mealtime.

Fatigue was present at diagnosis and continued to worsen. Family members assisted

with house-hold tasks, however she persevered with cooking and shopping for

herself most of the time. Increasing pain led to her admission to a palliative care unit

in week 14. By this time she was spending most of each day in bed or a chair and had

peripheral oedema. Mrs A passed away 16 weeks after enrolling in the study.

6.8.2 Case 2 - Mr B

Mr B was severely jaundiced at diagnosis. A palliative biliary bypass was performed

which provided relief and a gradual return of appetite. By the time he joined the

study five weeks later, he was reporting a good appetite and found no difficulty

drinking two cans of the supplement each day which increased his oral intake from

8900 to 10600 kJ/d.

He was in his late seventies and lived with his wife. At first he had been pleased

when his weight dropped from a preillness BMI of 35 kg/m2 and it was only the

onset of jaundice that took him to his GP. His weight stabilised initially but then

started dropping again. By week 32 (10 weeks before his death), he had lost 44 kg,

37% of his pre-illness body weight. He appeared cachectic and yet had ironically

reached the “Healthy Weight Range” with a BMI of 22 kg/m2.

Mr B described himself as someone who usually had a very good appetite and

enjoyed nearly all foods. He reported no barriers to oral intake for the first 17 weeks

of the study by which time early satiety, anorexia and pain were identified as

problems. Early satiety became less important over time possibly due to dramatically

reduced intake, with pain becoming a major barrier. Anorexia increased with time.

Even when his appetite was very poor, however, Mr B had no difficulty drinking

both cans of supplement each day. He preferred it cold and undiluted and would only

take five minutes to consume each can; one on rising and the other prior to going to


bed. He continued this until troubled by nausea and vomiting in the week before his

death.

His GP, on the advice of a gastroenterologist, prescribed a bile salt sequestrant for

mild diarrhoea in the first few weeks of the study. Mr B had reported what appeared

to be oil in the toilet bowl but denied any floating, foul smelling or pale stools.

Pancreatic enzymes were not tried. His use of the medication was patchy due to the

unpleasant texture and ceased after a few weeks as the diarrhoea had resolved.

At week 17, when his appetite started to decline, Mr B reported that the food he

tolerated best was corned meat. He occasionally experienced nausea and vomiting

following high fat meals. By week 24, meat along with porridge, were the best

tolerated foods and he was avoiding high fat foods and eating fewer vegetables. By

week 27 his response regarding best tolerated foods was “none” other than the

supplement.

He experienced no taste changes until week 21, then reported a general diminution of

taste. By week 27 everything was tasteless. Mr B consistently reported a better

appetite in the morning than the evening.

Mr B found in week 38 that his extensive weight loss meant that his dentures no

longer fitted. Attempts to wear them led to feeling “bilious”. Rather than battle with

the nausea and gagging, he chose to leave the dentures out and concentrate on very

soft foods and liquids such as cereal, egg-flips and well-cooked stews.

Pain was adequately controlled with paracetamol until week 24 when morphine was

commenced. Mr B and his wife needed encouragement to make contact with the

local palliative care service – they didn’t want to be a nuisance. For the last few

months of his life Mr B’s functional capacity was poor, and he remained in bed for

most of the day.

Mr B’s appetite declined to the point where he had very little interest in food and

increasing concern was expressed by his wife. At week 27 he stated that he became

annoyed even by the idea of food and by family members constantly pushing him to


eat. After a discussion with Mr B’s wife regarding the changing role of nutrition

support as disease progresses, she appeared more comfortable with a less aggressive

approach. At the following review Mr B stated that he was relieved that his wife had

“backed off” and was now offering small amounts without coaxing him to eat all the

time.

Improved pain control, home visits by the palliative care team and his GP, and strong

family supports contributed greatly to Mr B’s ability to remain at home until shortly

before his death.

7 Discussion

Patients with unresectable pancreatic cancer clearly experience an array of nutrition-

related symptoms that can affect both their ability to maintain weight and their

quality of life. This study has shown that weight stabilisation is not only feasible for

some of these patients, it is also associated with improved outcomes.

The first section of this chapter (7.1) will address issues related to study design.

Sections 7.2, 7.3 and 7.4 will examine the results for Aims 1,2 and 3 respectively

with respect to the initial model and the literature. In the final section an algorithm

will be proposed for clinical nutrition care of patients with unresectable pancreatic

cancer.

7.1 Study design issues

Most studies into oral nutrition support in cancer have involved patients undergoing

chemotherapy (Evans et al., 1987) (Ovesen et al., 1993a). A limitation of this study is

that it was a retrospective analysis of data collected for another purpose and therefore

conclusions need to be made with caution. It is rare however to obtain good quality

nutrition-related data on such a large number of cancer patients. A particularly

valuable feature of this study was that the subjects were a relatively homogeneous

group and received supportive care only.

It is clear that outcomes were better for the weight stabilised (WS) patients than

those who continued to lose weight (WL). What is not clear is why this happened.

This study cannot clarify whether nutrition intervention per se slowed the weight

loss, or whether the WL group simply represents a subgroup with more aggressive

disease. On the other hand the progressive wasting of patients with unresectable

pancreatic cancer has been well described (Wigmore et al., 1997b). Every patient had

been losing weight as a criteria for study entry. There was also no difference at

baseline between the WS and WL groups for pre-study weight loss or other

prognostic factors such as stage of disease or performance status (Davidson, Ash,

Discussion 120

Capra, & Bauer, 2003). This supports the argument that weight stabilisation was not

simply a marker of less aggressive disease.

A randomised study comparing nutrition intervention and “no intervention” as the

control arm would be required to investigate causation. This would be a very difficult

study to conduct, for both ethical and logistic reasons. Compliance issues in the

BH80 trial have reminded us of the difficulty of expecting a patient with a profound

diagnosis to comply to a control protocol after being informed of the nature of the

trial. Similarly, it seems likely that changes in eating behaviour might occur in the

control group of a study comparing the presence or absence of ONS. The fact that the

BH80 trial was a randomised study excluded patients who did not want to take the

50% chance of not receiving the experimental supplement. Data on reasons for

declining participation in the study was not routinely recorded, however such

responses have been reported verbally by study coordinators. In the context of having

a disease with such a short life expectancy it is not surprising that some patients,

having been given information on the study rationale, chose to find alternative ways

to supplement with fish oil.

This study had limited power to identify differences between the groups, particularly

in the second aim which involved multivariate analysis. The patient numbers

available were relatively small for examining complex interactions between

numerous characteristics, hence the large confidence intervals and the failure to

reach statistical significance in spite of high odds ratios for some variables.

As was anticipated, oedema and ascites occurred frequently in the final weeks of life.

This supports the a priori decision to exclude patients with fluid changes within the

study period and those who died within two weeks of T8.

The short expected survival duration for patients with unresectable pancreatic cancer

explains why many of the BH80 patients were not eligible for the secondary analysis

and, as would be expected, the excluded patients had lower Karnofsky Performance

Status scores. Common reasons for lack of T8 data and therefore exclusion, were

early death or being too ill to continue in the study. The patients in this post hoc

Discussion 121

analysis of the BH80 data, therefore, would not be representative of all patients with

unresectable pancreatic cancer but rather those who are at least 10 weeks from death.

The majority of patients included in the data analysis were able to stabilise their

weight over the eight week study period. This, however, is not generalisable to all

patients diagnosed with unresectable pancreatic cancer. Sixty-nine percent of the

patients enrolled in the BH80 study were either excluded from this analysis or were

in the WL group. Approximately one third of the recruited patients lived for at least

ten weeks and were able to stabilise their weight. A large proportion of patients with

unresectable pancreatic cancer deteriorate so rapidly that it seems unlikely that

aggressive ONS would provide any benefits. Nutrition intervention should be aimed

at the subgroup anticipated to live at least two to three months. Unfortunately it is not

easy to determine to which individuals this criterion applies.

7.2 Outcomes for weight stabilisation

A marked attenuation of weight loss was achieved in the BH80 trial (Fearon et al.,

2003) in contrast to the natural history of unresectable pancreatic cancer in which the

decline in nutritional status continues (Wigmore et al., 1997b). All patients had been

losing weight at baseline.

All patients had received nutrition intervention involving a protein and energy dense

oral supplement with or without n-3 fatty acids and additional antioxidants. Patients

who were able to stabilise their weight came from both arms of the BH80 trial

suggesting that features common to both forms of ONS, whether components of the

supplements or other aspects of intervention, were leading to weight stabilisation for

many patients.

Outcome measures for nutrition intervention in a palliative context differ from those

in which cure of disease is the focus. A short-term decrease in quality of life may be

acceptable during treatment with a curative intent. Weight stabilisation or gain would

be of little value, however, in the palliative setting if they did not reflect

Discussion 122

improvements in quality of life or extended survival time compared to patients who

continued to lose weight.

7.2.1 Survival duration

This study has shown a significantly greater survival time from baseline for the

group whose weight stabilised, compared to those who continued to lose weight.

Kaplan Meier analysis demonstrated a statistically significant extension in median

survival of three months (p = 0.019, Figure 4.2, Table 4.7). This is also a clinically

significant improvement in this patient group for whom diagnosis is sudden and

unexpected, and progression of disease is usually rapid.

Associations between weight loss (or continued weight loss), and reduced survival

time have been demonstrated in cancer patients but these findings have been less

consistent for pancreatic cancer (Andreyev et al., 1998; DeWys et al., 1980). This

may relate to the large patient numbers required to demonstrate significant

differences in short overall survival times. Patients in the previous studies all

received chemotherapy whereas the present study involved intensive ONS. The

frequent fluid shifts that complicate weight measurements in advanced cancer may

have masked survival differences in the earlier studies, whereas we excluded patients

who showed clinical evidence of oedema or ascites during the study period.

Median survival duration for patients with unresectable pancreatic cancer who have

been treated with chemotherapeutic drugs and novel antineoplastic agents has been

reported to range from four to six months (Bramhall et al., 2001; Burris et al., 1997;

Johnson, Puntis, Davidson, Todd, & Bryce, 2001). It is difficult to compare survival

times between unrelated studies because of potential selection biases, but these

figures do demonstrate the short life expectancy for this disease. Median survival for

the entire BH80 group (130 days) was comparable to groups treated with

gemcitabine (Fearon et al., 2003). The patients who were able to halt their weight

loss survived a median of 259 days (approximately nine months) from the start of the

study.

Discussion 123

7.2.2 Quality of life

An increase in survival time might not be a desirable outcome if the time gained was

of poor quality due to intractable pain or other symptoms. In this study however,

patients who were able to stabilise their weight after eight weeks of ONS, had

significantly higher global QoL scores than patients who continued to lose weight.

In an examination of fourteen studies in which the EORTC QLQ-C30 had been used,

King (King, 1996) interpreted “small” and “large” differences in global QoL scores

to be two and sixteen respectively. The statistically significant difference of eight in

this study, combined with an improved survival time, is therefore also clinically

meaningful.

This supports earlier findings that weight loss is associated with poorer QoL in

patients with advanced gastrointestinal cancer (Andreyev et al., 1998; O'Gorman et

al., 1998) and cancer of the breast, ovary or lung (Ovesen et al., 1993b).

7.2.3 Summary - Aim 1

The first aim of this study was to determine whether achieving weight stabilisation is

an appropriate goal of nutrition intervention for people with unresectable pancreatic

cancer.

It had been hypothesised that patients who lost no more than 1kg over eight weeks of

nutrition intervention would have increased survival time compared to those who lost

more than 1kg and also that they would have better quality of life. In both cases the

null hypothesis was rejected because there was a statistically significant difference

between the WL and WS groups.

Weight stabilisation can be considered a measure of “effective nutrition intervention”

and therefore an appropriate goal for medical nutrition therapy for this patient group.

Discussion 124

7.3 Determinants of weight stabilisation

As discussed in Chapter 2, patients with unresectable pancreatic cancer experience a

range of symptoms and metabolic alterations that could contribute to weight loss.

The symptoms that patients present with at diagnosis vary considerably, for example

some have severe anorexia and early satiety while others retain their appetites. Data

on potential nutrition impact factors that were available from the BH80 trial and

would be readily available in a clinical setting, were analysed to identify

characteristics of patients whose weight stabilised.

Features of the WL and WS groups were compared using chi-square or unpaired t-

tests. This showed that more of the patients who continued to lose weight

experienced symptoms of nausea and vomiting, appetite loss and pain at baseline and

more of them had raised CRP levels. They had lower protein and energy intakes at

both baseline and T8, and their usual and baseline BMI’s were higher.

Inter-relationships would be expected between these variables, for example loss of

appetite would be anticipated to result in reduced energy intake. Logistic regression

analysis was therefore performed to identify whether any variables were predictive of

weight stabilisation independently of the other variables. Only the absence of nausea

or vomiting at baseline and female gender significantly increased the likelihood of a

patient being in the weight stable group after controlling for other variables.

Admittedly, statistical power was limited for these analyses. In a bivariate setting,

there was adequate power (ie. 80% or more) to detect as statistically significant (ie.

two-sided α= 0.05), three to four-fold differences for the more common (prevalence

of approximately 50%) and less common characteristics (prevalence of

approximately 10%) respectively. For the multivariate analyses statistical power was

more limited as revealed by the wide confidence intervals.

The determinants are discussed in the following order; those that were independently

predictive of weight stabilisation, followed by those identified by bivariate analysis

alone, and finally those which were not predictive of weight stabilisation.

Discussion 125

7.3.1 Independent determinants

Nausea and vomiting

Fewer people in the WS group suffered from nausea and vomiting. It is conceivable

that nausea and vomiting could impact directly on weight outcomes by reducing both

the intake and availability of nutrients. The fact that the presence of nausea and

vomiting at baseline reduced the likelihood of weight stability independently of

energy intake at T8, suggests that it may also be a marker of aggressive disease, or

forms of pancreatic cancer that impede anabolism.

The minimisation of nausea and vomiting is a major aspect of palliative care for

patients with unresectable pancreatic cancer (Held-Warmkessel et al., 1998; Rhodes

& McDaniel, 2001). These symptoms can be due to a combination of factors.

Increasing tumour size or enlargement of the liver or spleen can cause

gastrointestinal obstructions or stimulate mechanoreceptors in the bowel wall. The

use of narcotic analgesics can cause nausea through delayed gastric emptying,

constipation or the accumulation of metabolites. Nausea can also be due to uremia or

electrolyte imbalances and is an early sign of hypercalcaemia. Many of these are

related to advancing disease and may explain why patients with nausea and vomiting

are less likely to be able to respond to ONS.

Cytokines and growth factors produced by the tumour can also be emetogenic. It

may be that more aggressive tumour types cause nausea in this way, or that tumours

that produce PIF, or other as yet unidentified catabolic factors, also produce nausea-

inducing products.

Vomiting would also confound the dietary intake data, because the amount of food

ingested would no longer represent the amount available in the gut for digestion.

The results from this study relate only to the reported presence or absence of nausea

and/or vomiting in the week preceding enrolment in the study. It does not provide

information on whether nausea and vomiting persisted over the eight week study

Discussion 126

period. It was also a simple all or nothing measurement and did not record whether

the patient found the nausea or vomiting distressing. Whether more active

management of nausea and vomiting for this patient group would result in better

patient outcomes warrants further investigation. At very least it would be expected to

provide immediate gains in quality of life and therefore warrants addressing in

clinical practice.

The primary strategy for minimising nausea and vomiting is the use of antiemetics

such as the seratonin receptor antagonists and metoclopramide (Rhodes & McDaniel,

2001). Dietitians can identify when symptom control is not adequate and discuss this

with medical staff. A combination of drugs is sometimes needed and while some

drugs commonly used in the palliative setting, such as opioids, are emetogenic,

others, such as corticosteroids, can increase the effect of antiemetics.

Dietary strategies to support the pharmacological management of nausea and

vomiting focus on reducing triggers specific to each patient. Small, frequent meals of

cool, bland foods and beverages are often best tolerated and the smell of food

cooking may need to be avoided (McCallum & Polisena, 2000).

Gender

Female gender was identified as a factor that independently increased the likelihood

of being in the weight stable group. This difference failed to reach statistical

significance in the bivariate analysis, however this could be due to inadequate

numbers to detect a difference. Women comprised 48% of the WS group but only

34% of the WL group (p=0.172).

Similar gender differences in outcomes for people with cancer have been reported in

other studies. Gender was one of the variables included in a prognostic tool for

cancer of the head of pancreas and periampullary region (Terwee et al., 2000). It was

found that the relative risk of death for gender, depended on age, with women

surviving longer than men in the older age groups. Van den Bosch et al also found, in

a similar patient group, that male gender was an unfavourable prognostic factor (van

den Bosch et al., 1994). In the study by Andreyev et al of 1555 patients with

Discussion 127

advanced gastrointestinal cancer, weight loss occurred in a larger proportion of the

men presenting for chemotherapy than the women (Andreyev et al., 1998).

It has been suggested (Chlebowski et al., 1996) that the shorter survival and

increased risk of developing weight loss for men with non-small cell lung cancer

compared to women, may be related to the fact that men with lung cancer can

develop hypogonadism even before chemotherapy commences. These gender

differences in body composition alterations have been confirmed in 21 patients with

NSCLC undergoing chemotherapy (Harvie, Campbell, Thatcher, & Baildam, 2003).

The male patients lost LBM and gained body fat but no changes occurred in the

women. Decreased serum testosterone levels are associated with reduced muscle

mass, and testosterone analogues have been shown to increase weight and lean body

mass (Langer et al., 2001). The reduction in androgens that can occur in men with

cancer may, therefore, explain the greater proportion of men in the WL group.

The gender difference found in this study may be related to an effect of sex steroids

on the cancer itself. Both oestrogen and testosterone appear to stimulate the growth

of pancreatic cancer cells (Lillemoe et al., 2000) although clinical trials with

tamoxifen, a drug which has an anti-oestrogen effect, have failed to show any

benefit.

The favourable prognostic feature of female gender in this study, may also be an

artefact relating to factors that might enable patients with advanced cancer to

participate in research studies. If elderly men are more likely than women to have a

spouse to care for them, then men with advanced disease may also be more likely

than women with advanced disease to enrol in a study. A person with severe illness

may consider involvement in a study too burdensome unless supported and

encouraged to do so by a carer or spouse. Information on such social supports was

not gathered in the BH80 study. Poorer outcomes for men may also reflect the fact

that they tend to present later in the course of disease than women (Moynihan, 2002).

Discussion 128

7.3.2 Determinants significant by bivariate analysis

Energy intake

At baseline, the WS group already had a greater mean energy intake than the WL

group. Their intake also increased over the period of ONS, from 125 to 141

kJ/kg/day, whereas there was little improvement in the intake of the WL group.

Increased energy requirements have been demonstrated using indirect calorimetry in

weight-losing pancreatic cancer patients (Falconer et al., 1994; Wigmore et al., 1995)

particularly those with an APPR. Resting energy expenditures of 108 kJ/kg/day have

been reported. It is not surprising then that the WL group, most of whom were

ambulatory, achieved a mean energy intake at T8 of only 110 kJ/kg/day. In contrast,

the mean energy intake of 141 kJ/kg/d achieved by the WS group would be expected

to allow weight maintenance at low levels of physical activity. This suggests that the

average requirements for weight maintenance in weight losing pancreatic cancer

patients is around 140 kJ/kg/d or 33 kcal/kg/d, which fits within the 30-35 kcal/kg/d

range recommended for slightly hypermetabolic patients (Bloch, 1993).

Energy intake, however, did not reach significance as an independent determinant of

weight stabilisation in logistic regression analysis. This may be due to insufficient

patient numbers to demonstrate an effect but may also reflect the multiplicity of

barriers to anabolism in cancer cachexia.

The mean energy and protein intakes at T8 for the WL group (110 kJ/kg/d and 1.1

g/kg/d) are very similar to those reported for weight losing cancer inpatients by

Levine et al (100 kJ/kg/d and 0.9 g/kg/d) (Levine & Morgan, 1998). Intake of the

WS group (140 kJ/kg/d and 1.4 g/kg/d) resemble more those of the weight stable

control patients who did not have cancer (140 kJ/kg/d and 1.2 g/kg/d). This supports

the assertion that compliance with a dietary prescription to raise energy and protein

intake to estimated requirements, is important for success in halting weight loss

(Capra, Bauer, Davidson, & Ash, 2002).

Discussion 129

The increase in mean energy intake of 1100 kJ/d achieved by the WS group is similar

to the increase of 1000 kJ/d reported by Ovesen et al for patients with ovarian, breast

or lung cancer receiving dietary counselling during chemotherapy (Ovesen et al.,

1993a). In both studies these increases failed to achieve weight gain. In a study of the

effect of megesterol acetate (MA) on appetite and body weight in weight losing

patients with advanced cancer (some of whom were receiving chemotherapy),

patients receiving MA who were evaluable at two months, had increased their energy

intake by 2500 kJ/d and patients on placebo had increased by 700 kJ/d resulting in

respective weight changes of +1.0 kg and +0.0 kg (Tchekmedyian et al., 1992). Once

again, an increase of nearly 1000 kJ/d was not sufficient to increase body weight.

While there are concerns that weight gained with MA is predominantly adipose

tissue, these results do suggest that appetite stimulants or some other means of

overcoming the anorexia experienced in cancer cachexia, may be needed to allow the

magnitude of increases required for weight gain to occur. Weight gain may be an

unrealistic goal for many cancer patients. We have demonstrated, however, that

weight stabilisation is associated with improved outcomes.

Energy requirements for patients with advanced cancer would be expected to vary

due to both differing physical activity levels as well as varied metabolic alterations.

Some of the WL patients were unable to maintain their weight in spite of reporting

high energy intakes. At T8, 36% patients in the WS group reported intakes > 150

kJ/kg/d but so did 10% patients in the WL group. If the T8 results accurately

represent intake over the study period, then those individuals would appear to be

hypermetabolic. A closer examination of food intake and assessments of absorptive

capacity and vomiting would be needed to clarify this.

It is possible that the increase in nutrient intake would have been greater, or may

have occurred in a greater number of patients, if the intensive nutrition intervention

had not been in the setting of a research trial. The purpose of the BH80 study was to

determine whether the fish oil containing supplement was more effective than the

control supplement in improving weight, therefore the focus was on achieving a

minimum of 1.5 cans of supplement per day. In a non-research setting, a dietitian

would be able to utilise a wider range of supplements and foods eg. the addition of

protein powder to savoury foods such as soup, to adapt to individual taste

Discussion 130

preferences. Patients with poor appetite, however, may be more motivated to take a

specialised research product and less inclined to persevere with “ordinary” foods.

This study cannot clarify whether providing oral supplements leads to weight

stability, however, it does demonstrate that many weight losing patients with

unresectable pancreatic cancer were able to be assisted to increase oral intake and

minimise their weight loss, whereas those who were unable to increase their intake

continued to lose weight.

Measurement of dietary intake

One of the difficulties in determining whether nutrition intervention is effective for

weight losing cancer patients, is the estimation of nutrient intake. The act of

recording food intake has been shown to affect intake even in healthy well-motivated

subjects (Rebro, Patterson, Kristal, & Cheney, 1998). Under-reporting of food intake

has frequently been observed in studies of healthy, weight stable populations (Black

et al., 1991; Livingstone et al., 1990; Martin et al., 1996).

The ratio of EI/BMR, a tool for identifying under-reporting, was developed using

data from weight stable populations (Goldberg et al., 1991). It is based on the

concept that ratios of energy intake to basal metabolic rate can be estimated below

which it is highly unlikely that an individual could maintain energy balance and

therefore at that level it is likely that under-reporting has occurred. It is not useful for

detection of under-reporting in this situation because every study participant was

losing weight on study entry. Potential alterations to metabolic rate due to cancer

would further complicate the situation.

Over-reporting may actually be more of an issue for this group of patients many of

whom were struggling with early satiety and nausea, while being encouraged to

increase intake to meet study goals. Another limitation of the use of food diaries may

be the day to day variation in intake that occurs for patients with fluctuating

symptoms such as pain and nausea. Recording of food intake can be further

complicated by malabsorption or episodes of vomiting, reducing the availability of

nutrients.

Discussion 131

Compliance to the dietary prescription is a challenge in the presence of the many

symptoms experienced by patients with advanced cancer, and recording food intake

is an added burden. There was more data missing for energy intake at T8 than any of

the other variables examined (12% vs less than 4% missing for other variables). In

some cases diaries had been kept but contained insufficient detail to be of use. This

missing data would be expected to bias the results towards higher food intakes

assuming that sicker patients were the ones who were less likely to complete the food

records and would be likely to consume less.

Tracers such as deuterated water can be included to confirm intake of supplements,

but unfortunately no objective marker is available for overall nutrient intake. The

double labelled water technique allows the comparison of reported energy intake

with expenditure but is not practical for general clinical use. Patient-recorded food

diaries, despite their limitations, provide a balance between the need to gather

sufficient information and the minimisation of patient burden in the palliative care

setting. The individual therapist/client relationship formed in this study was felt to

improve the accuracy of the nutrient intake data. Patients were reminded by phone

when the recordings were due to commence and any unclear or unusual diary entries

were cross-checked with patients at each data collection point.

Appetite

A greater proportion of the WS group reported no loss of appetite at baseline than the

WL group (41% vs 18%). After adjusting for the other variables in logistic

regression, however, it did not reach statistical significance as an independent

determinant of weight stability. Interrelationships would be expected between

appetite, food intake and symptoms such as nausea and pain. Loss of appetite has

been shown to be a strong negative prognostic indicator in advanced lung and

colorectal cancer (Sloan et al., 2001) and hospice patients with various cancer types

(Reuben, Mor, & Hiris, 1988). Krech et al, however, were unable to show any

relationship between appetite and survival time in unresectable pancreatic cancer

(Krech & Walsh, 1991).

Discussion 132

Appetite, as with all symptoms, is subjective and therefore difficult to measure. The

measure of appetite used in this study was a simple one, taken from the EORTC

QLQ-C30 questionnaire and related to the week prior to commencement of the study.

It cannot be used as a surrogate for food intake because appetite is not the only factor

involved in determining the amount eaten. Some patients find it less difficult than

others to override a loss of appetite, at least initially, and are able to eat more than

they feel like eating (DeWys, 1977).

Acute phase protein response

The WL group contained a higher proportion of patients with raised CRP levels

(41% vs 21% for WS), supporting the findings by Falconer et al that an acute phase

response is associated with hypermetabolism (Falconer et al., 1994) and weight loss

(Falconer et al., 1995). Lack of APPR, however, was not found to be an independent

predictor of weight stability.

The majority of patients in this study did not have elevated CRP levels at baseline

despite the fact that they had all lost at least 5% of body weight on entry to the study.

This highlights the fact that the increased energy requirement that occurs during an

APPR is only one of a range of factors likely to be responsible for weight loss in

unresectable pancreatic cancer (Inui, 2002; Tisdale, 2002). Further studies examining

the effects of EPA-fortified supplements on patients demonstrating an APPR or the

presence of cachectic factors such as PIF may provide information on the effect of

EPA on these metabolic changes.

Body mass index

The majority of patients included in this study (72%) were overweight or obese prior

to weight loss from their cancer, and none had been underweight. This supports

previous findings that BMI may be a risk factor for pancreatic cancer (Michaud et

al., 2001b; Silverman et al., 1998). Pre-illness and baseline BMI values but did not

differ for those who were or were not included in analysis so this observation reflects

the entire BH80 study group.

Discussion 133

Mean BMI values, both pre-illness and at baseline, were greater for patients in the

WL group than the WS group – 28.0 vs 26.1 kg/m2 pre-illness and 23.4 vs 21.4

kg/m2 at baseline. Baseline BMI produced a high odds ratio in logistic regression

analysis but did not reach statistical significance (P = 0.058). The odds ratio

(likelihood of being in the WS group) for those who commenced the study with a

BMI in the healthy weight range (20-25 kg/m2) was 4.8 (95% CI 1.0-24.0) compared

to those who were overweight or obese (> 25 kg/m2).

Pancreatic cancer patients with measurable levels of PIF in the urine have been

reported to have had greater pre-illness weight than those who were PIF negative (76

vs 65 kg, p=0.009) (Wigmore et al., 2000b). They also went on to have greater

weight loss, leading the authors to suggest that production of PIF by the tumour

might be related to body mass. This may explain why overweight/obese was more

common in the WL group in our study.

These results suggest that being overweight or obese may be a disadvantage, not only

as a risk factor for development of pancreatic cancer, but also in terms of poorer

outcomes during the course of the disease. Further investigations would be needed to

confirm this, determine the mechanisms involved and then to identify intervention

strategies.

Pain

Pain has been identified as a one of the symptoms frequently occurring in cancer that

can limit nutrient intake (Feuz & Rapin, 1994; Gallagher-Allred, 1989; Ottery,

1995). The treatment of pain can also lead to further nutrition impact symptoms such

as nausea and constipation.

At baseline, pain was experienced by the majority of patients as has been reported in

other studies (Krech & Walsh, 1991; Lillemoe, 1998; Ottery, 1996a). It was

significantly more common amongst the patients who continued to lose weight

(84%), than those who stabilised their weight loss (67%) but lack of pain was not

independently predictive of weight stability.

Discussion 134

The measure of pain used was a simple all-or-nothing scale, derived from the QoL

questionnaire as for the other symptom scales, and does not reflect whether pain was

well managed during the study. Effective pain relief measures have been shown to be

associated with prolonged survival in advanced pancreatic cancer (Lillemoe et al.,

2000), possibly, at least in part, by facilitating adequate food and fluid intake.

7.3.3 Characteristics which were not predictive

Some of the features that were not significantly different between the weight groups,

are discussed below.

Stage of disease

The lack of significant difference in disease staging between the groups supports the

proposal that patients in the WL group did not simply have more extensive disease

than those in the WS group. The four level staging was provided by medical officers

at enrolment and does not account for any progression that may have occurred over

the eight week study period, however not even a trend was apparent.

There was a trend towards more advanced disease, however, in those patients who

were excluded from secondary analysis compared to those who were included (p =

0.082) (Table 4.2). Thirty percent of included patients had Stage IV disease

compared to 46% of excluded patients and this is supported by the fact that death

within two months of enrolment was one of the main reasons for exclusion. The

presence of metastases (Stage IV disease), particularly liver metastases, has been

reported as having prognostic value (Falconer et al., 1995; Terwee et al., 2000; van

den Bosch et al., 1994). These same studies acknowledge difficulties in staging

pancreatic cancer, especially in distinguishing between Stages II and III.

Nevertheless there was clearly no difference between the WL and WS groups in the

proportion of Stage IV disease (Table 4.16).

Randomisation & plasma EPA

Discussion 135

Two methods were used for examining the relationship between consumption of

EPA and weight stability. The simplest was whether patients had been randomised to

either the control or experimental arm of the BH80 study but because of known

compliance problems in both arms, plasma EPA was included as a more direct

measure of the effect of EPA.

There was no difference in the proportion of patients from each arm of the BH80 trial

in the two weight groups (Table 4.12). This supports the decision to pool the two

randomisation groups and consider them as one large group, all of whom had

received ONS.

The number of patients who were able to reach plasma EPA levels ≥3% by T8 was

also not significantly different for the two groups (p= 0.133), although there did

appear to be a trend towards a greater proportion in the WS group (40% vs 24% for

the WL group) (Table 4.24).

A plasma level of 3% EPA is well above levels found in unsupplemented people

(Barber et al., 1999b; Sinclair, O'Dea, & Johnson, 1994; Wigmore et al., 2000a;

Zuijdgeest-Van Leeuwen et al., 2002) and is approximately the 25th percentile for

studies involving supplementation with 2g EPA per day (Barber et al., 1999b;

Wigmore et al., 1996); a dose not achieved by the majority of patients in this study.

Higher levels of EPA may be capable of producing significant increases in weight

and LBM, however strategies for improving compliance would be required for

median plasma EPA levels to reach the 5% achieved in the pilot study for the BH80

trial (Barber et al., 1999b).

EPA may later be found to provide beneficial outcomes for specific subgroups of

cancer patients or at higher plasma levels, but the BH80 study was unable to show

any benefit over and above that of intensive nutrition intervention which included a

protein and energy dense supplement (Fearon et al., 2003).

Discussion 136

Functional status

Measures of performance status such as The Eastern Cooperative Oncology Group

performance status (ECOG-PS) and the Karnofsky Performance Status scale (KPS)

have been shown to have prognostic value (Krech & Walsh, 1991; Reuben et al.,

1988; Sloan et al., 2001). The fact that KPS scores for patients who were excluded

from this secondary analysis were significantly lower than those who were included

(p<0.001, Table 4.2), reflects the fact that any patient surviving less than ten weeks

from baseline was excluded. Fifty-six percent of patients included in the analysis had

KPS scores of 80 or above (ie. relatively normal activity levels), compared to only

27% of excluded patients.

There was, however, no difference in KPS scores at baseline between the WS and

WL groups (Table 4.17). Decreasing performance status affects the ability to shop

and cook for oneself, however, it was not useful in this study in predicting who

would be able to stabilise their weight loss. The frequent contact with health

professionals during this study, with a focus on nutrition, meant that community

services and strategies for simplifying meal preparation could be identified for

patients with reduced functional status who lived alone.

Pancreatic Insufficiency

One of the barriers to achieving adequate energy intake for weight stability for some

patients with pancreatic cancer is suboptimal absorption of nutrients. This is due to

insufficient pancreatic enzymes and possibly bicarbonate, especially with tumours of

the head of pancreas (Bruno et al., 1998).

The WL and WS groups did not differ significantly in terms of use of enzyme

supplements at baseline (30% and 32% respectively). We cannot, however, assume

that this means that there was no difference in prevalence of pancreatic insufficiency.

While less than one third of all patients were receiving pancreatic enzyme

supplements, it has been suggested that approximately 60% of patients with

unresectable pancreatic cancer will have pancreatic insufficiency (Gouma D., 2001

personal communication). This higher figure is supported by a small study in which

Discussion 137

nine of twelve patients with ductal pancreatic cancer had fat malabsorption as

determined by faecal fat tests (Perez et al., 1983).

Tests to determine the proportion of consumed fat that is excreted in the faeces over

a three day period, may be considered an unwarranted burden on patients in the

palliative setting, and clinical signs are usually relied upon instead. The usual signs

of steatorrhoea may be disguised by the reduced bowel motility that often results

from analgesic use (Ottery, 1996a). This, along with the poor prognosis for patients

with pancreatic cancer, may lead to an under-prescribing of pancreatic enzyme

supplements.

Steatorrhoea can lead to symptoms of anorexia, early satiety and bloating, due to the

effect of unabsorbed fat on gastrointestinal motility (Bruno et al., 1998). Patients

with unrecognised and untreated pancreatic insufficiency, may reduce their fat intake

to control gastrointestinal symptoms and therefore limit total energy intake, further

contributing to weight loss. If abdominal cramping can be reduced with enzyme

replacement then this would play an important role in symptom management quite

apart from its nutritional impact.

Although diabetes is commonly found in patients with pancreatic cancer, weight loss

in a group of patients prior to resection, did not correlate with insulin resistance

(Permert et al., 1993). In the present study, there was a trend towards more cases of

diabetes in the WS group (p=0.180) but diabetes was reported in only 15% of

patients overall. People with unrecognised and therefore uncontrolled diabetes would

be expected to have more difficulty maintaining their weight. Any patients with

raised blood glucose levels, however, should have been identified and treated during

the study.


The second aim of this study was to identify determinants of weight stabilisation in

patients with unresectable pancreatic cancer.

Discussion 138

We had hypothesised that patients who would be likely to benefit from nutrition

intervention, ie. those for whom weight stabilisation was feasible, could be identified

using a variety of features on presentation. These baseline characteristics included a

higher BMI, greater energy intake, better performance status, less pain, less loss of

appetite and less nausea and vomiting. Similarly it was hypothesised that patients

who would be likely to benefit from nutrition intervention would achieve greater

energy intake by week eight.

Figure 7.1 Model representing Aims 1 & 2 of this study of nutrition intervention

in unresectable pancreatic cancer

Weight

change

QoL

Survival duration

Energy

intake T8BMI T0

PS T0

Stage of disease

Symptoms T0

Energy intake T0

ONS

ONS – oral nutrition support, QoL – quality of life, BMI – body mass index, PS – performance status,

T0 – baseline, T8 – week 8.

Fewer of the patients whose weight stabilised had experienced symptoms of nausea

and vomiting, appetite loss or pain at baseline and they had higher protein and energy

intakes at both baseline and T8. Patients in the WS group, however, had lower rather

than higher BMI’s compared to the WL group and there was no difference in

performance status.

Female gender and absence of nausea and vomiting were the only independent

determinants of weight stabilisation identified in this study. No strong predictors

emerged. This may be because of the lack of a reliable method of determining

Discussion 139

prognosis and because there was so much interaction between the variables chosen.

While there remains no way to identify confidently which patients will be able to

stabilise their weight with ONS, this study has identified risk factors that should be

considered as a part of individualised nutrition intervention.

7.4 Additional nutrition related factors

Additional nutrition-related issues, examined in detail for thirteen Australian

patients, were the focus of the third aim of this study. Findings for these patients

were reported in detail in Chapter 6 and are discussed below. Baseline characteristics

for this sub-group appear similar to those of the group as a whole but were not

compared statistically due to the small numbers.

There are likely to be other characteristics linked to a person’s ability to stabilise

their weight that were not examined in the BH80 study, such as social supports, taste

changes, tumour phenotype, personality and motivation to eat, as well as metabolic

alterations that are not yet measurable.

7.4.1 Bodyweight changes

The majority of the Australian patients (9/13) reported usual weights that placed

them in the overweight or obese categories. Even at baseline, after considerable pre-

study weight loss, the median BMI was 25 kg/m2. Increased BMI has been reported

to be a risk factor for pancreatic cancer (Michaud et al., 2001b; Silverman et al.,

1998), however, these patients had pre-illness BMI’s typical of the Australian

population, with 63% of Australians of a similar age (65 years and over) being

overweight or obese (McLennan & Podger, 1998).

Some patients reported a period of deliberate weight loss merging with the onset of

disease-related weight loss. Diagnosis can sometimes be delayed because overweight

patients may initially welcome the unintentional weight loss, if more unfamiliar signs

such as jaundice are not also present.

Discussion 140

All patients, regardless of initial BMI, underwent major changes to their appearance,

with the majority of patients (9/13) losing more than 20% of their body weight

during the course of the disease. The dramatic loss of weight made their illness

visible, and was very confronting for the patient and people around them.


Nutrient analysis was limited to the macronutrients due to BH80 study design. Issues

concerning use of the three day food record method in this patient population have

been discussed in section 5.3.1. Interindividual variation for energy and nutrient

intakes was great, as was the case for the larger group. Patients with the highest

intakes were more physically active and intakes declined for all patients who died

within ten weeks of baseline.

Dietary advice focussed on consuming as wide a range of foods as tolerated,

however, the proportion of total energy appeared to increase for carbohydrate and

decrease for fat. Some patients may have limited their fat intake to reduce symptoms

of malabsorption or nausea. The fact that the study supplement was relatively low in

fat and high in carbohydrate may also have contributed to those minor changes.

In a study of 105 chemotherapy patients (described in Table 2.4), Ovesen et al found

that while nutrition counselling increased dietary intake it made no difference to the

proportions of energy contributed by the various macronutrients (Ovesen et al.,

1993a). Levine et al compared the ad lib food intake of ten inpatients with advanced

cancer and anorexia who had lost at least 15% body weight, to that of patients

hospitalised for other conditions (Levine & Morgan, 1998). The cancer patients ate

less (6 vs 9.5 MJ/day, p<0.001), but there was no difference in the proportions of

energy contributed by protein, fat or carbohydrate for the two patient groups.

These results do not support the suggestion that cancer patients selectively eat less

protein. They appear to eat less of all the macronutrients. When meat was a problem

Discussion 141

due to taste changes or texture difficulties, an increase in dairy products often

compensated for the reduced intake of protein via meat.

7.4.3 Taste & appetite changes

Taste and appetite changes were common and idiosyncratic in this group of patients.

Such alterations contribute to inadequate nutritional intake (DeWys, 1977) but are

difficult to measure. Information on taste change was obtained qualitatively and

appetite was assessed from the self-reported scale within the EORTC QLQ-C30.

Data from the unvalidated linear analogue scales for appetite that were available

from the BH80 study did not appear to be useful.

A newly developed distaste for meat or coffee was common but not predictable, with

some patients continuing to enjoy them. Aversions are more likely to develop to

foods that have distinctive odours as well as tastes, and are especially likely when

gastrointestinal symptoms are a common feature of the disease (Bernstein, 1999).

Once taste alterations occurred for any one individual, they did not necessarily

remain constant, suggesting they may not be true conditioned responses.

The fact that some of the patients reported meat as the food they tolerated best

highlights the need to not make assumptions about which foods cancer patients may

prefer. While it may well be true that “a lot of tumour patients refuse beef and pork

first, and poultry and fish later” (Ollenschlager et al., 1988) and “alteration in taste is

often manifested by a metallic taste” for pancreatic cancer patients (Held-

Warmkessel et al., 1998), the results of this study have demonstrated the individual

nature of responses.

Bruera et al found a much higher incidence of an elevated glucose threshold in

cancer patients with advanced solid tumours compared to controls (61% vs 5%)

(Bruera et al., 1984). From this finding, the authors recommended increasing the

sweetness of foods for cancer patients as a way of increasing nutrient intake. The

results from the present study, however, do not support this is as a strategy for

routine use as some patients specifically reported a distaste for sweet foods.

Discussion 142

Mouth dryness is frequently reported in advanced cancer, particularly in weight

losing patients (Grosvenor et al., 1989), and lack of saliva is known to diminish taste

perceptions (Schiffman, 1997). Positive taste stimulation triggers the release of

salivary, gastric, pancreatic and intestinal secretions as part of the cephalic phase

response (Schiffman, 1997). The dysfunction of taste perception, in addition to

altering food choices and reducing intake, may hamper the digestive process and

contribute to early satiety.

Loss of appetite became more common with disease progression and remains a major

barrier to increasing oral intake. Nutrition counselling can address ways to maximise

nutrient intake when the usual sensory rewards are reduced or absent.

7.4.4 Symptoms and quality of life

Data for the Australian patients confirm the frequent occurrence of symptoms for

unresectable pancreatic cancer reported in the literature, ie. pain, nausea, loss of

appetite and the range of nutrition impact factors that can contribute to weight loss.

The two case studies (sections 6.8.1 and 6.8.2) demonstrate the individual nature of

patient experiences and the need to tailor nutrition intervention to each patient’s

needs. Attitude towards the study supplement, for example, was highly individual.

Some patients enjoyed the supplement, whereas others found it difficult to consume

or even nauseating, and ceased taking it as soon as the eight week study period

ended. Some patients appeared to put great faith in the supplement whereas others

appeared to be participating in research for altruistic reasons. Many patients seemed

to continue in the study because of the frequent contact with supportive health care

workers.

As patients became frailer, liquids were often preferred to solid meals because less

effort was involved in their consumption. Eating a “normal” meal became less

appealing when it required great effort. Frailty will have differing affects on intake

depending on the nature and degree of support from family and carers.

Discussion 143


Nutrition-related features have been described for patients with unresectable

pancreatic cancer who were recruited by the Australian sites. Both quantitative and

qualitative data were used to provide a clearer picture of the illness as experienced by

patients. A clinical path for medical nutrition therapy in unresectable pancreatic

cancer has been developed based on these insights and is described in the following

section.

7.5 Implications for nutrition intervention

This study has demonstrated that weight loss can be reduced in many patients with

unresectable pancreatic cancer, at least in the short term, and that this is associated

with improved outcomes. These outcomes occur at all levels of the cascade of events

described in Splett’s model of nutrition intervention (Splett, 1996) which categorises

the range of outcomes to effective dietetic practice (Appendix 17). Intermediate

outcomes relate to changes in dietary intake, biochemical or anthropometric

measures but ideally intervention should produce beneficial clinical, cost and patient

outcomes as well. This study has shown that when appropriate screening and

intervention are implemented, intermediate outcomes such as increased nutrient

intake and absorption can lead on to weight stabilisation and then on to important

patient outcomes such as improved quality of life and increased survival time.

Nutrition intervention for these patients may become of even more importance in the

future as treatment options increase. Minimising weight loss has an important role in

optimising tolerance of and response to treatment. Weight-losing patients undergoing

chemotherapy for advanced gastrointestinal cancers experienced more frequent and

more severe side effects resulting in shorter and less effective treatment (Andreyev et

al., 1998). Side effects from chemotherapy could result in reduced QoL, outweighing

any benefits in survival duration. ONS should therefore be included in any

chemotherapy protocols for unresectable pancreatic cancer in an attempt to limit

weight loss and optimise nutritional status.

Discussion 144

It is likely that multiple strategies will be required to overcome the various metabolic

alterations that can occur in cancer cachexia (MacDonald et al., 2003). As effective

remedies become available an array of dietary, pharmacological and exercise

interventions, relevant to each individual, may permit considerable improvements in

quality of life and perhaps increased survival times for patients with advanced

cancers.

Intent-to-treat analysis of the BH80 study was unable to detect an additional benefit

for EPA over and above a nutrient dense supplement. Some studies (Barber et al.,

1999b; Bauer, 2003; Gogos et al., 1998), however, suggest improved outcomes in

terms of body weight, lean body mass, functional measures, survival duration or

quality of life providing pharmacological doses of EPA are consumed. Other

potential strategies include supplementation with essential amino acids (May, Barber,

D'Olimpio, Hourihane, & Abumrad, 2002), anabolic agents (Tchekmedyian, Fesen,

Price, & Ottery, 2003) or exercise (Zinna & Yarasheski, 2003) to promote regain of

LBM, progestational agents to improve appetite (Maltoni et al., 2001) and blocking

of catabolic mediators (Todorov et al., 1996). All such methods require further

clinical studies.

The variables considered in the present study were ones that are currently readily

available to a dietitian in a clinical setting. The effective and easily measured goal of

maintaining body weight will be achievable for many of the patients presenting with

unresectable pancreatic cancer. The two main elements of nutritional care for these

patients, described below, are ongoing assessment of the appropriate level of

intervention and intensive, individualised ONS.

7.5.1 Appropriate level of nutrition intervention

Aggressive symptom management is appropriate for all patients with unresectable

pancreatic cancer because of its beneficial impact on quality of life. All patients with

pancreatic cancer should be referred for nutrition assessment as they are a high risk

group. Exceptions would be patients who are in the final days of life and patients

who are weight stable and display no nutrition impact factors. There should also be a

Discussion 145

regular nutrition screening process in place so that these well nourished patients are

referred as soon as problems develop.

Aggressive ONS is unlikely to benefit those patients who are within two to three

months of death. In the present study, the patients who did not reach T8 had poorer

performance status, poorer QoL scores and ate less, than those who lived longer. A

judgement regarding the appropriate level of nutrition intervention should be based

upon the patient’s immediate prognosis as well as the patient’s own goals.

Weight stabilisation was achievable for 60% of patients for whom data was available

at T8. Similarly, Ottery (Ottery, 1996b) reports success with aggressive symptom

management in the stabilisation of weight or weight regain for 50-80% weight losing

cancer patients if those with expected life expectancy of less than six weeks are

excluded from analysis.

Early intervention is important. Factors that affect weight loss are common early in

the course of disease (Grosvenor et al., 1989) and attempts to reverse severe

nutritional depletion are unlikely to succeed (Ottery, 1996a). The present study has

demonstrated that weight stabilisation, rather than a return to premorbid weight, is a

realistic goal.

Nutrition intervention must also adapt to the progression of the disease. As the

patient approaches the terminal phase of disease, the role of ONS becomes one of

comfort (Gallagher-Allred, 1989; MacDonald et al., 2003). The appetite and food

intakes of hospice patients have been reported to vary greatly (McCann, Hall, &

Groth-Juncker, 1994).

Carers often need “permission” to stop pushing the patient to eat once they are in the

terminal phase of their illness. Family may perceive reduced food intake as a form of

“giving up” and there may be conflict within a family concerning the appropriate

response to anorexia, especially if they do not share a common sense of the patient’s

prognosis. An acknowledgment that it is appropriate to gradually withdraw from

eating may also play a role in the preparation for death.

Discussion 146

It is important though that acute deterioration due to a remediable cause such as

biliary stent obstruction or poor pain management is not misinterpreted as the

terminal stage of disease. Working closely with medical and nursing staff facilitates

appropriate dietetic support.

Frequent weighing can be discouraging for patients who are continuing to lose

weight and may not be in their best interest. Certainly for patients in the final stage of

pancreatic cancer it would be inappropriate. Such judgement needs to be made as

part of the dynamic process of dietetic intervention in palliative care.

7.5.2 Individualised and intensive intervention

Intensive oral nutrition support is required for increased nutrient intake to occur in

the face of barriers such as dysguesia and loss of appetite. It involves much more

than the prescribing of a commercial supplement although that remains one of the

most effective ways of increasing nutrient intake. Compliance with the dietary

prescription is crucial to successful outcomes (Capra et al., 2002). Close follow-up is

needed to facilitate compliance as well as to adjust the intervention as symptoms

develop and the disease progresses.

The patients in this study who were able to stabilise their weight were able to

increase their energy and protein intake to levels close to those recommended in the

literature (Bloch, 1993; Martin, 2000; Robuck & Fleetwood, 1992). This required a

combination of prescribing the amount to be consumed (in this case, a nutrition

supplement), good symptom control and close follow-up.

Pancreatic cancer occurs at an age which is often accompanied by other medical and

social conditions – recent retirement, death or illness of spouse, living alone,

diabetes, cardiovascular disease. The appropriateness of previous dietary restrictions,

such as cholesterol-lowering diets, should be reassessed and clarified for the patient.

This can reduce tension between the patient and family members and expand the

range of food options.

Discussion 147

Nutrition-related symptoms such as nausea, vomiting, pain, appetite loss, taste

changes and early satiety can be identified using tools such as the PG-SGA described

in section 2.5.1. Close liaison with the patient’s consultant, general practitioner or

palliative care team facilitates symptom relief.

Gastrointestinal problems can be related to the disease process itself, as well as a

roller-coaster effect of laxative and antidiarrhoeal medication. Constipation

exacerbates abdominal discomfort, reducing both QoL and food intake. If pancreatic

enzyme supplementation is required, the dietitian can identify the times when intake

of fat is greatest and recommend modifications to either food or enzyme supplement

timing to optimise their utilisation.

Even strong associations such as the link between nausea and vomiting at baseline

and weight loss, do not guarantee outcomes for any one patient. One third of WS

patients reported nausea and vomiting and one third of WL patients did not. The case

studies also remind us that generalisations need to be kept in perspective when

dealing with each individual patient.

Mean daily energy and protein intakes of 140 kJ/kg and 1.4 g/kg (actual body

weight) respectively were associated with weight stability. Goal intakes of 125-170

kJ/kg/day and 1.5-2.0 g protein/kg/day allow for the varied metabolic rates found in

patients with unresectable pancreatic cancer and the preservation of lean body mass.

Individualised ONS allows for the most appropriate strategies to be put in place to

optimise nutrient intake. For example, a person may have an aversion to the sweet,

rich flavour and milky feel of a nutritional supplement and yet they may still enjoy

eating a large serving of meat. Similarly, taste changes and early satiety may be

limiting a person’s meal intake severely and yet they may be able to consume

considerable volumes of nutrient dense liquid supplements between meals.

Review by phone may be a useful addition to hospital-based dietetic outpatient

review as a way of reducing patient burden while maintaining the intensity of

intervention required to deal with newly arising symptoms, taste fatigue and patient

Discussion 148

Figure 7.2 Clinical pathway for medical nutrition therapy in unresectable

pancreatic cancer

Assessment for deciding treatment mode Clinical judgement based on prognosis, functional status, ascites/oedema, oral intake, patient’s goals, treatment plan

Is aggressive intervention appropriate?

Comfort cares (endstage disease) symptom management relax dietary restrictions explain changing role of nutrition Goals = optimise QoL and satisfaction

NO

YES

Assessment PG-SGA Anthropometry height, weight, weight history Nutrition E & Pr intake, appetite, texture, meal pattern, supplements, restrictions Biochemistry albumin, haemoglobin, blood glucose, C reactive protein Activity activity level, fatigue, functional status Psychosocial family supports, anxiety, depression, patient’s expectations Clinical stage, procedures, prognosis, co-morbidities, Medications analgesics, enzymes, laxatives, antiemetics, alternative therapies Symptoms pain, nausea, early satiety, dysgeusia, steatorrhoea, constipation, Supports GP, pall care team

Optimal symptom

management? NO

YES

Liaise with medical team to minimise pain, nausea,

depression etc

Optimal enzyme usage?

NO

YES

Liaise with medical team - commence/adjust pancreatic enzymes to suit meal pattern

Early referral

Discussion 149

Goals = weight stabilisation, tolerance of CT/RT,

optimal QoL and satisfaction

Intensive intervention Energy 125-170 kJ/kg/d Protein 1.5-2.0 g /kg/d via food ± commercial supplements Modify meals for early satiety & fatigue Strategies for dysguesia & loss of appetite Facilitate social aspects of eating Exercise to maintain LBM (if appropriate)

Review Weekly (face to face or by phone) Assess compliance & identify new issues

Identify further barriers to intake

Review Monthly (face to face)

Discussion 150

concerns. At least monthly, the level of intervention must be reassessed to determine

whether aggressive ONS remains appropriate.

A clinical path for the nutrition support of patients with unresectable pancreatic

cancer has been developed from the results of this study and from the literature. The

process is depicted in Figure 7.2.

8 Conclusion

This study has confirmed that unresectable pancreatic cancer is accompanied by a

range of nutrition related symptoms that are distressing to patients. It has also shown

that the dramatic weight loss common in unresectable pancreatic cancer can be

slowed for some patients, at least in the short term.

Such weight stabilisation has been shown in this study to be associated with

improved outcomes. Those patients who were able to stabilise their weight after eight

weeks of oral nutrition support lived longer from baseline and reported better quality

of life than those who continued to lose weight. Weight stabilisation is therefore a

reasonable and worthwhile goal for this patient group.

Patients who halted their weight loss had greater energy and protein intakes than

those who continued to lose weight. They also had lower BMI’s and were less likely

to have pain, appetite loss or an acute phase protein response at baseline. Female

gender and the absence of nausea or vomiting at baseline, were independently

associated with an increased likelihood of weight stabilisation.

The results of this study need to be interpreted with caution, as this was a post hoc

analysis. These results do, however, support the hypothesis that nutrition intervention

is beneficial for some patients with unresectable pancreatic cancer. Further research

designed specifically to address this question is required to confirm these results.

Suggested issues for future research

Ideally, a randomised controlled trial of individualised and intensive oral

nutrition support versus usual care in this population with the goal of weight

stabilisation.

Studies to improve methods for predicting survival time and therefore which

patients are most likely to benefit from oral nutrition support.

Conclusion 152

Assessment of whether optimal management of nausea and vomiting can

facilitate weight stabilisation.

Investigations into which features of personality or physiology allow some

patients to maintain an adequate intake despite dysguesia or nausea and whether

such features can be developed.

Assessment of the role of nutrition support in palliative chemoradiation therapy

which is becoming more common and may bring with it increased side effects,

particularly nausea.

Identification of effective exercise strategies to maintain lean body mass and

function for patients with cancer-induced weight loss.

Development of objective methods for measuring intake and confirming

compliance with dietary prescriptions.

Development of methods for determining changes in lean body mass and

functional status that are inexpensive, valid for this patient group, and limit

patient burden.

Recommendations for dietetic practice

A screening process should be in place so that patients with nutrition impact

symptoms can be identified and assessed at an early stage.

Each patient requires an individualised approach. If life expectancy is no more

than two months then symptom control and comfort cares will be the priority,

whereas more active oral nutrition support is warranted for patients who are

expected to live longer.

Weight stabilisation is an appropriate and effective goal rather than the regaining

of lost weight.

Conclusion 153

Multifaceted approaches are needed to achieve nutrition goals. Intervention needs

to not only deal with problems specific to each individual but also to adapt as

issues change with time.

Symptom management is of major importance and necessary for compliance with

the dietary prescription. Symptoms of particular concern are pain, nausea and

other gastrointestinal symptoms.

Working in conjunction with other health professionals is especially important

given the palliative focus of treatment.

This process is summarised in a clinical path for the nutritional care of patients with

unresectable pancreatic cancer (Figure 7.2) and it is recommended that it be adopted

into practice.

Appendices

Appendix 1: BH80 study sites

Primary Principal Investigator: Prof. K.C.H. Fearon

Univ Dept Surgery, Royal Infirmary,

Edinburgh University, UK

Sponsor: Ross Products Division

Abbott Laboratories

Columbus, Ohio, USA

Study Sites

Royal Infirmary of Edinburgh, Edinburgh, UK

University Hospital of Maastricht, Maastricht, The Netherlands

Centre Hospilalier de l’Universite de Montreal (CHUM), Montreal, Canada

Academisch Medisch Centrum, Amsterdam, The Netherlands

Istituto Nazionale per la Ricerca sul Cancro, Genoa, Italy

Universite Libre de Bruxelles Erasme, Brussels, Belgium

The Wesley Hospital, Brisbane, Australia

Princess Alexandra Hospital, Brisbane, Australia

Manchester Royal Infirmary, Manchester, UK

Russells Hall Hospital, Dudley, UK

Hammersmith Hospital, London, UK

Leeds Hospital, Leeds, UK

156 Appendices

Appendix 2: BH80 Inclusion/Exclusion Criteria

Inclusion Criteria Exclusion Criteria

1. male or non-pregnant, non-lactating female 18

years or older

1. surgery, radiotherapy or chemotherapy less than

four weeks prior to baseline

2. diagnosed unresectable adenocarcinoma of the

pancreas or ampulla and papilla Vateri,

confirmed by histology, unequivocal radiological

or operative findings

2. obstruction of the gastrointestinal tract

precluding ingestion of the product

3. unintentional weight loss > 5% within a 6

month period

3. history of allergy to any of the ingredients in the

experimental or control products

4. willing to be randomised to receive 2 x 237ml

servings per day of either the experimental or the

control nutritional sip feed in addition to their

regular diet

4. history of uncontrollable diabetes and/or

chronic renal failure

5. willing to disclose current usage of all

nutritional supplements including oral

supplements, vitamin and mineral supplements

and/or any health food supplements for the

clinical record and to discontinue their intake for

the duration of the study

5. known HIV infection

6. interested in participating in the study and

willing to compy with study protocol after a)

being fully informed about the experimental and

control formulae, b) informaed about the clinical

trial procedure, c) reviewing the study

methodology and d) providing written informed

consent

6. use of medications that could profoundly

modulate metabolism or weight

7. Karnofsky score of 60 or greater 7. use of fish oil or n-3 preparations (>200mg

EPA/d) within the past 90 days prior to study

enrolment

8. expected survival of at least two months 8. Crohn’s disease, inflammatory bowel disease,

short bowel syndrome or other major

gastrointestinal diseases

9. dementia, brain metastases, eating disorders or

any other psychological condition which may

interfere with product consumption

Appendices 157

10. severe diarrhoea, nausea or vomiting as

diagnosed by a physician, which may interfere

with product consumption

11. body mass index (BMI) greater than 30 kg/m2

158 Appendices

Appendix 3: BH80 patient information and consent

PATIENT INFORMATION SHEET FOR ABBOTT PROTOCOL NO BH80

A prospective multi-centre randomised double-blind trial comparing the efficacy of an n-3 fatty acid enriched oral supplement with that of an

isonitrogenous isocaloric oral supplement in weight-losing pancreatic cancer patients

You are invited to help with this research study at Princess Alexandra Hospital and this sheet gives information about it. Please read it carefully and take your time before deciding whether to take part. Discuss anything you don’t understand with your doctor and/or dietitian. Why is this research being carried out? Many patients with your condition lose weight. Doctors and dietitians try to stop this by giving patients extra food in various forms. Research has shown that giving a fish oil extract may help this extra food be of more benefit. Abbott Laboratories, who are sponsoring this study, have developed an oral liquid nutrition supplement containing the fish oil. The study will compare this new feed with a standard feed of the sort commonly given to patients losing weight. What does the study involve? If you are able to help us, you will be one of about 100 patients taking part in the study worldwide. You will be randomly allocated to receive either a standard nutrition drink or the study nutrition drink. This means that you would be allocated by chance (like the toss of a coin) and would have an equal chance of receiving either the standard drink or the study drink. Neither you, the dietitian nor the doctor will know which of these drinks you will receive. This prevents bias in the study. Your dietitian will see you in your home or at the hospital, whatever is the most convenient for you. You see your dietitian in a week or two, but between now and then you will fill in a diary for three days describing what you eat. When you see your dietitian again, your dietitian will measure your weight, the strength of your hand grip and the relative amounts of fat and muscle in your body using Bioelectrical Impedance Analysis. This is simple and painless and involves lying down for 5-10 minutes connected to the portable Bioelectrical Impedance Meter. Small electrode pads are attached with some gel to your wrists and ankles. You will then be asked to provide some blood samples and complete some forms with the help of the dietitian. A 4 week supply (72 cans) of either the standard nutrition drink or the study nutrition drink will be delivered to your home. You are to take the drink in addition to the usual food you eat. We ask that you drink two cans

Appendices 159

(about 500 ml) every day and continue to do this for the next 8 weeks. We will give you a diary to record the amount of nutrition drink you take and another diary to write down any symptoms you may have. You will need to fill in a food diary just before your dietitian visits after 4 and 8 weeks. Your dietitian will telephone you every so often to make sure everything is okay. After the 4 and 8 weeks, your dietitian will see you to repeat all the tests you had at the beginning of the study. After these tests you will have finished the main study period though your dietitian may suggest that it is best for you to carry on taking the nutrition drink for as long as you wish, free of charge. In this case you will continue to be contacted monthly by the dietitian who will measure your weight and collect any other relevant information. How will taking part in the study help me and are there any risks? Your doctor and/or dietitian will have asked you to take part in this study because you have recently lost weight. It is hoped that by taking the study drink, any further weight loss may be prevented. The various tests and measurements in this study are not harmful in any way. The blood samples will be taken in the usual method and are expected to cause only minor discomfort. Any new preparation may have side effects which are not yet known including risks to you or an unborn child. You must tell the dietitian if you are pregnant or breastfeeding, or if you intend to become pregnant in the near future. Like any liquid nutrition supplement, the drink may occasionally cause unwanted effects, mainly upset stomach, nausea or diarrhoea though they are usually mild and do not last long. None of these effects is likely to damage your health, and your dietitian and doctor will always take precautions to remove or reduce them. You should be aware that there are available alternative sources of the fish oil ingredients used in this study. Ask your dietitian if you wish to know more about these. Any new information that comes to light during the study which may affect your decision to continue in the study will be made known to you through your dietitian. Compensation for injury In the unlikely event that any injury happens to you due to your taking part in this study, you will be compensated for your reasonable medical expenses by the sponsor of this study, Abbott Laboratories. This will happen provided you have immediately told your doctor about the injury, have not deliberately caused it and have followed the medical advice given to you. Financial compensation due to other damages or losses is not available. If you would like more details on the nature of this compensation please ask your dietitian.

160 Appendices

Confidentiality Your identity will always be treated as confidential and will not be disclosed to the public. All data from the study will be passed to Abbott Laboratories but without your name and address on them. Your own doctor will be told that you are taking part in the study and kept informed of your progress and your hospital notes will state that you are in this study. This is done to make sure all doctors involved in your care are aware of the treatment you are receiving. Representatives of Abbott Laboratories will need to inspect your relevant health records; confidentiality will be assured. You can change your mind even if you agree to take part Before starting the study, you will be asked to consent in writing. However, by signing this form you are not waiving any of your legal rights. If at any time you feel that you do not wish to continue, you may withdraw from the study and this will not affect your future care by your doctor or hospital in any way. Your doctor may also withdraw you from the study if he feels it is not right for you to carry on for any reason. You may contact us (Ph 3240 5033) or a member of the ethics committee (Ph 3240 5856) about any matters of concern regarding this study. Thank you for your assistance Wendy Davidson Dr Susan Ash Research Dietitian Nutritionist Deputy Director, Nutrition & Food Princess Alexandra Hospital Princess Alexandra Hospital

Appendices 161

CONSENT FORM FOR ABBOTT PROTOCOL NO BH80

A prospective multi-centre randomised double-blind trial comparing the efficacy of an n-3 fatty acid enriched oral supplement with that of an

isonitrogenous isocaloric oral supplement in weight-losing pancreatic cancer patients

Patient initials: Patient study number: Study agreement (to be completed by the patient) Have you read the information sheet about this study? YES/NO Have you been able to ask questions about this study? YES/NO Have you received answers to all your questions? YES/NO Have you received enough information about this study? YES/NO Do you understand that you are free to withdraw from this study? At any time Without giving a reason for withdrawing Without affecting your future medical care YES/NO Do you agree to your medical records being reviewed by study personnel? YES/NO Do you agree to take part in this study? YES/NO Signed ............................................................. Date ............................... Name (Block letters) ........................................................................ Investigator ..................................................... Date ................................ Witness ........................................................... Date ..............................

162 Appendices

Appendix 4: EORTC QLQ-C30 (version 3)

Appendices 163

164 Appendices

Appendix 5: Karnofsky Performance Status (KPS) Scale

Able to carry out normal

activity; no special care

needed

100 Normal. No complaints. No evidence of

disease.

90 Able to carry on normal activity. Minor

signs or symptoms of disease.

80 Normal activity with effort. Some signs or

symptoms of disease.

Unable to work; able to

live at home and care for

needs.

70 Cares for self. Unable to carry on normal

activity or do active work. A varying

amount of assistance needed.

60 Requires occasional assistance but is able

to care for most of needs.

50 Requires considerable assistance and

frequent medical care.

Unable to care for self;

requires equivalent of

institution or hospital care;

disease may be

progressing rapidly.

40 Disabled. Requires special care and

assistance.

30 Severely disabled. Hospitalisation is

indicated although death is not imminent.

20 Very Sick. Hospitalisation necessary.

Active supportive treatment is necessary.

10 Moribund. Fatal processes progressing

rapidly.

0 Dead NB: Patients with KPS < 60 excluded from the BH80 trial.

Source: (Mor et al., 1984)

Appendices 165

Appendix 6: Staging for Pancreatic Cancer

Stage I T1 N0 M0

T2 N0 M0

Stage II T3 N0 M0

Tumour in pancreas extending directly to stomach, spleen, colon or

adjacent large vessels

Stage III any T N1 M0

Any tumour in the pancreas with evidence of lymph node metastasis

Stage IV any T any N M1

Any tumour in the pancreas with evidence of distant metastasis

T, location and size of tumour

N, evidence of metastases in lymph nodes close to the cancer

M, evidence of distant metastases

Source: Appendix D of the BH80 protocol (Fearon, 1998)

166 Appendices

Appendix 7: Sub-protocol patient information and consent (ver

2)

“Assessment of nutrition intervention for patients with unresectable pancreatic cancer

in a fish oil supplementation trial – does it make a difference?”

INFORMATION SHEET

You are invited to participate in an additional study at the Princess Alexandra Hospital

related to the BH80 trial. This study is being undertaken as part of a postgraduate study

program with the Queensland University of Technology. Please read this information sheet

carefully and take your time before deciding whether to take part. Discuss anything you

don’t understand with the dietitian.

Why is this research being carried out?

As you have read in the information sheet for the BH80 trial, we are part of a multi-centre

trial investigating whether adding fish oil to a nutritional supplement for people with

unresectable pancreatic cancer helps them to regain weight.

As an extension of this research we wish to interview some of the people participating in the

BH80 trial to find out how they feel about their weight and appetite changes and to gain

more detailed information on the type of taste and appetite changes that occur.

While the larger study will provide the large amount of data necessary to determine whether

the fish oil actually brings about weight gain for most patients, this extra study is designed to

find whether the people involved feel that their quality of life is improved by a weight

increase and/or appetite improvements.

What is being asked of you?

When you first start this study you will be asked questions on both your usual and present

food intake and how they differ, as well as questions about any taste changes you may have

noticed. Approximately every fortnight, the dietitian will ask you similar questions regarding

Appendices 167

your food intake over the previous two weeks. These questionnaires will take approximately

15-20 minutes and will be useful in discussing your progress with the dietitian.

At the end of the main study period, which lasts 8 weeks, you will be invited to describe

your experience of weight & appetite change during a tape-recorded interview. This can be

as brief or as lengthy as you wish.

Each of these discussions with the dietitian will take place during the planned follow-ups as

part of the BH80 trial – some may be by phone. The only significant increase in time asked

of you would be for the taped interview which could be conducted in your home or at the

hospital, whichever you choose.

Confidentiality

Your identity will be treated as confidential and will not be disclosed to the public.

You can change your mind even if you agree to take part

Before this information is collected, you will be asked to sign a consent form. However, by

signing this form you are not waiving any of your legal rights. If at any time you feel that

you do not wish to continue, you may withdraw from the study and this will not affect future

care by your doctor or the hospital in any way. You may also withdraw from this smaller

study without affecting your involvement in the larger trial.

You may contact me (Ph 3240 5033) or a member of the ethics committee (Ph 3240 5856)

about any matters of concern regarding this study.

Thank you for your assistance

Wendy Davidson

(Acting) Research & Development Dietitian

Nutrition & Food Services

Princess Alexandra Hospital

168 Appendices

CONSENT FORM

“Assessment of nutrition intervention for patients with unresectable pancreatic cancer

in a fish oil supplementation trial – does it make a difference?”

I ………………………………………..consent to take part in the above study. (print your name)

I have read the attached Information Sheet and understand the nature and purpose of this

study. All my questions have been answered to my satisfaction.

I acknowledge that my involvement in the study may not be of benefit to me.

The opportunity has been given to me to have a friend or relative present when the study was

explained.

I understand that taking part in the study is voluntary and I am free to

withdraw at any time if I wish and this will not affect my clinical management.

I do/do not (delete if not applicable) agree to the investigator obtaining additional

information related to this study from

…………………………………………………….

……………………………………………………..

(print name & state relationship).

I understand that all the information gained in the study will be treated confidentially.

Patient: ............................................................. Date ...............................

Witness ........................................................... Date ..............................…

(Print name of Witness) ……………………..

I have explained the nature and purpose of this study to the above patient and have answered

their questions.

Investigator ..................................................…….. Date ................................

Wendy Davidson

(Acting) Research & Development Dietitian,

Nutrition & Food Services

Princess Alexandra Hospital

Appendices 169

Appendix 8: Identification coding of patients from the

Australian sites

Subprotocol ID code BH80 ID code

1 2101

2 2102

3 2201

4 2103

5 2202

6 2203

7 2205

8 1901

9 2002

10 2004

11 2005

12 2007

13 2008

170 Appendices

Appendix 9: T=0 structured interview (ver 1)

BH80 trial (subprotocol) Patient Initials___________

Date___________________ Patient Number__________

T=0 structured interview (to be completed by dietitian as face-to-face interview with patient)

1. Has your appetite changed since your symptoms (of pancreatic cancer) started?

2. How does your present food intake differ from before your illness? (as meal plan)

Typical day’s intake before illness

Typical day’s intake in last week or two

B

B

L

L

D

D

Snacks

Snacks

Appendices 171

3. Are there any foods or drinks that you eat less of now?

Why?

4. Are there any foods or drinks that you eat more of now?

Why?

5. Do any particular foods taste different to usual?

What are they?

How are they different?

6. What are the main barriers to eating your usual amount and type of food?

7a Are any particular times of the day better for eating?

7b Are any particular times of the day worse for eating?

Other issues/comments

172 Appendices

Appendix 10: Recent food intake report (ver 1)


Date___________________ Patient Number__________

Recent Food Intake Report (to be completed by dietitian as face-to-face or phone interview with patient)

A typical day’s food intake over past 1-2

weeks

B

L

D

Snacks

Main food avoided/reduced?

Foods best tolerated?

Recent taste changes?

Main barriers to food intake (not listed to patient)

Pain

Nausea/vomiting

Early satiety

Anorexia/”just not interested”

Fatigue

Shopping difficulties

Food preparation

Dysguesia

Depression

Other…….………………….

Appendices 173

Appendix 11: T=9/10 semi-structured interview/narrative


Date___________________ Patient Number___________

Location________________ Other people present___________

T=9/10 semi-structured interview/narrative (patient’s responses to be audiotaped by dietitian)

As you know I’ve been collecting information on your weight, food intake, blood

tests etc over the past couple of months. What I’d like to do now is hear about what

the experience of changing weight and appetite has been like for you.

I’ll ask you a few questions and I’d like you to share as many of your thoughts,

experiences & feelings as you can recall about each topic until you have nothing

more to say about the situation.

1. (if patient has reported changes to appetite) Please tell me about the way

your appetite has changed and whether it has had an affect on your everyday

life.

Prompts - have meal-time routines changed? - family meals – going out/visiting –

any good aspects? – do others encourage you to eat?

2. (if patient has reported taste changes) Can you describe for me how your

sense of taste has changed and whether it has influenced your choice of food?

Prompts – what about your favourite foods? – enjoyment of food?

3. Can you tell me about the things that have most limited your intake of

food?

Prompts – pain – nausea – early satiety – GI disturbances – fatigue

174 Appendices

4. (if WT ↓) Please tell me about how it has felt to lose weight.

(if WT ↓↑) Please tell me about how it has felt to lose and then regain

some of your weight.

Prompts – does it feel physically different? - body image/sense of identity - feeling

cold - strength/fatigue

5. Was there any piece of information about eating that you found

particularly useful or wish you’d received earlier or in a different manner?

Carer questions

In what ways have ……’s appetite changes affected his/her daily life

In what ways have ……’s weight changes affected his/her daily life

Appendices 175

Appendix 12: Notation used in transcription of interviews

Transcripts were typed, double-spaced with lines numbered and speakers indicated

by first initial and colon. Header contains brief details of patient identity code, date,

time, location and context.

Symbol Meaning

Words in italics Spoken with emphasis

WORDS in upper case Spoken very loudly

[ Overlap of speech by two speakers

( ) empty brackets Couldn’t identify words

(words) Unclear speech

((words)) Transcriber’s comments

(P) Pause or hesitation

(…) Longer pause

O:kay Prolongation of the immediately prior

sound

Word - Speech has been cutoff

Hm hm etc Nonlexical speech

Adapted from (Silverman, 1993) (Mishler, 1986) (Maykut & Morehouse, 1994)

176-184 Appendices

Appendix 13: Published paper (in press) – Clinical Nutrition

thompsdc

This Appendix is not available online. Please consult the hardcopy thesis available at the QUT Library.

Appendices 185

Appendix 14: Oral presentation – Dietitians Association of Australia 21st National Conference

thompsdc


186 Appendices

Appendix 15: Poster presentation – Dietitians Association of

Australia 20th National Conference

W Davidson, S Ash, Nutrition and Operational Support Services, Princess Alexandra Hospital, QLD J Bauer, PhD Scholar, The Wesley Research Institute, PO Box 499, Toowong, QLD A/Prof S Capra, Centre for Public Health Research, Queensland University of Technology, Brisbane QLD

thompsdc


187

Appendix 16: Descriptive statistics for study participants including data to test for normality

Variable T0/T8 meanN SDmedian skew kurtosis min K-Smax 1

Age (yrs) T0 107 66.9 67.0 8.9 .091 -.634 47 87 .8004

Height (m) T0 107 1.68 1.68 .09

-.168 -.521 1.42 1.86 .3697

Weight (kg) T0 107 62.4 64.0 11.2 .177 -.162 36.0 94.6 .5319

BMI (kg/m2) T0 107 22.2 22.0 .1863.5 .404 13.4 33.2 .9301

Energy intake (kJ/d) 2 T0 105 7191 7089 1883 -.012 -.366 2320 11449 .9919

Energy intake (kJ/d) 2 T8 94 7770 7806 2268 -.455 .238 798 12320 .5688

Energy intake (kJ/kg/d) 2 T0 105 117.5 115.0 32.3 .134 .050 27.0 197.0 .9831

Energy intake (kJ/kg/d) 2 T8 94 128.6 127.0 39.7 -.083 -.044 19.4 216.1 .7235

Plasma eicosapentanoic acid T0 102 1.05 .85 .80 2.747 10.898 .00 5.18 .0030

Plasma eicosapentanoic acid T8 99 2.37 1.36 2.04 1.04 0.21 0 8.23 0.001

C-reactive protein T0 103 12.4 5.4 21.2 3.731 15.669 .2 133.8 .0000

Global quality of life 3 T0 106 56.4 58.3 21.0 -.102 .117 0 100 .2389

Global quality of life 3 T8 103 51.7 50.0 19.0 -.221 -.314 0 91.7 .0817

Appetite loss4 T0 106 42.1 33.3 36.0 .238 -1.242 0 100 .0004

Nausea / vomiting4 T0 107 12.9 0 18.6 2.146 5.830 0 100 .0000

Pain4 T0 107 33.8 33.3 28.9 .520 -.588 0 100 .00701 K-S: Kolmogorov-Smirnov Goodness of Fit Test (p value) 2 three day food diaries and sip feed forms 3 EORTC QLQ-C30 Global quality of life 4EORTC QLQ-C30 symptom scales

Appendices 188

Appendix 17: Splett’s cascade model of outcome measures

References Aaronson, N. K. (1990). Quality of life research in cancer clinical trials: a need for

common rules and language. Oncology (Huntington), 4(5), 59-66. Aaronson, N. K., Ahmedzai, S., Bergman, B., Bullinger, M., Cull, A., Duez, N. J.,

Filiberti, A., Flechtner, H., Fleishman, S. B., & de Haes, J. C. (1993). The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. Journal of the National Cancer Institute, 85(5), 365-76.

Addington-Hall, J., & McCarthy, M. (1995). Dying from cancer: results of a national population-based investigation. Palliative Medicine, 9(4), 295-305.

Alter, C. L. (1996). Palliative and supportive care of patients with pancreatic cancer. Seminars in Oncology., 23(2), 229-40.

Ames, H. G., Gee, M. I., & Hawrysh, Z. J. (1993). Taste perception and breast cancer: evidence of a role for diet. Journal of the American Dietetic Association, 93(5), 541-6.

Anderson, K. L., & Burckhardt, C. S. (1999). Conceptualization and measurement of quality of life as an outcome variable for health care intervention and research. Journal of Advanced Nursing, 29(2), 298-306.

Andreyev, H. J., Norman, A. R., Oates, J., & Cunningham, D. (1998). Why do patients with weight loss have a worse outcome when undergoing chemotherapy for gastrointestinal malignancies? European Journal of Cancer., 34(4), 503-9.

Australian Institute of Health and Welfare. (2002). Cancer in Australia 1999. Canberra.

Baldwin, C., Parsons, T., & Logan, S. (2002). Dietary advice for illness-related malnutrition in adults (Cochrane Review)., The Cochrane Library (Vol. Issue 1, ): Oxford: Update Software.

Baracos, V. E. (2000). Regulation of skeletal-muscle protein turnover in cancer-associated cachexia. Nutrition, 16, 1015-1018.

Barber, M. D., Ross, J. A., & Fearon, K. C. (1999a). Changes in nutritional, functional, and inflammatory markers in advanced pancreatic cancer. Nutrition & Cancer, 35(2), 106-10.

Barber, M. D., Ross, J. A., Voss, A. C., Tisdale, M. J., & Fearon, K. C. (1999b). The effect of an oral nutritional supplement enriched with fish oil on weight-loss in patients with pancreatic cancer. British Journal of Cancer, 81(1), 80-6.

Barnard, J. A., Tapsell, L. C., Davies, P. S., Brenninger, V. L., & Storlien, L. H. (2002). Relationship of high energy expenditure and variation in dietary intake with reporting accuracy on 7 day food records and diet histories in a group of healthy adult volunteers. European Journal of Clinical Nutrition, 56(4), 358-367.

Bauer, J. (2003). New insights into medical nutrition therapy for cancer cachexia. Unpublished Doctor of Philosophy, Queensland University of Technology, Brisbane.

Beach, P., Siebeneck, B., Buderer, N. F., & Ferner, T. (2001). Relationship between fatigue and nutritional status in patients receiving radiation therapy to treat lung cancer. Oncology Nursing Forum., 28(6), 1027-31.

Bernstein, I. L. (1999). Taste aversion learning: a contemporary perspective. Nutrition, 15(3), 229-34.

190 References

Black, A. E., Goldberg, G. R., Jebb, S. A., Livingstone, M. B., Cole, T. J., & Prentice, A. M. (1991). Critical evaluation of energy intake data using fundamental principles of energy physiology: 2. Evaluating the results of published surveys. European Journal of Clinical Nutrition, 45(12), 583-99.

Black, A. E., Prentice, A. M., Goldberg, G. R., Jebb, S. A., Bingham, S. A., Livingstone, M. B., & Coward, W. A. (1993). Measurements of total energy expenditure provide insights into the validity of dietary measurements of energy intake. Journal of the American Dietetic Association, 93(5), 572-9.

Blackburn, G., Bistrian, B., Maini, B., & al, e. (1977). Nutritional and metabolic assessment of the hospitalised patient. Journal of Parenteral and Enteral Nutrition, 1, 11-22.

Bloch, A. S. (1993). Cancer. In M. M. Gottschlich, L. E. Matarese, & E. P. Shronts (Eds.), Nutrition Support Dietetics; Core Curriculum (pp. 213-227). Silver Spring: ASPEN.

Bosaeus, I., Daneryd, P., Svanberg, E., & Lundholm, K. (2001). Dietary intake and resting energy expenditure in relation to weight loss in unselected cancer patients. International Journal of Cancer, 93(3), 380-383.

Bramhall, S. R., Rosemurgy, A., Brown, P. D., Bowry, C., Buckels, J. A. C., & Group, f. t. M. P. C. S. (2001). Marimastat as first-line therapy for patients with unresectable pancreatic cancer: a randomized trial. Journal of Clinical Oncology, 19(15), 3447-3455.

Brennan, M. F. (1977). Uncomplicated starvation versus cancer cachexia. Cancer Research, 37(7 Pt 2), 2359-64.

Brown, D., & Miller, W. (1998). Normative data for strength and flexibility of women throughout life. European Journal of Applied Physiology, 78, 77-82.

Bruera, E., Carraro, S., Roca, E., Cedaro, L., & Chacon, R. (1984). Association between malnutrition and caloric intake, emesis, psychological depression, glucose taste, and tumor mass. Cancer Treatment Reports, 68(6), 873-6.

Bruno, M. J., Haverkort, E. B., Tijssen, G. P., Tytgat, G. N., & van Leeuwen, D. J. (1998). Placebo controlled trial of enteric coated pancreatin microsphere treatment in patients with unresectable cancer of the pancreatic head region. Gut, 42(1), 92-6.

Burris, H. A., Moore, M. J., Andersen, J., Green, M. R., Rothenberg, M. L., Modiano, M. R., Cripps, M. C., Portenoy, R. K., Storniolo, A. M., Tarassoff, P., Nelson, R., Dorr, F. A., Stephens, C. D., & Von Hoff, D. D. (1997). Improvements in survival and clinical benefit with gemcitabine as first-line therapy for patients with advanced pancreas cancer: a randomized trial. Journal of Clinical Oncology, 15(6), 2403-13.

Capra, S., Bauer, J., Davidson, W., & Ash, S. (2002). Nutritional therapy for cancer-induced weight loss. Nutrition in Clinical Practice, 17, 210-213.

Carson, J. A., & Gormican, A. (1977). Taste acuity and food attitudes of selected patients with cancer. Journal of the American Dietetic Association, 70(4), 361-5.

Cella, D. F. (1992). Overcoming difficulties in demonstrating health outcome benefits. JPEN: Journal of Parenteral & Enteral Nutrition, 16(6 Suppl), 106S-111S.

Chandra, R. (1995). Nutrition and immunity in the elderly: clinical significance. Nutrition Reviews., 53(4 Pt 2), S80-3; discussion S83-5.

References 191

Chlebowski, R. T., Palomares, M. R., Lillington, L., & Grosvenor, M. (1996). Recent implications of weight loss in lung cancer management. Nutrition, 12(1 Suppl), S43-7.

Cooperman, A. M. (2001). Pancreatic cancer: the bigger picture. Surgical Clinics of North America, 81(3), 557-74.

Cooperman, A. M., Kini, S., Snady, H., Bruckner, H., & Chamberlain, R. S. (2000). Current surgical therapy for carcinoma of the pancreas. Journal of Clinical Gastroenterology, 31(2), 107-13.

Cotanch, P. H. (1992). Measuring nausea and vomiting. In M. Frank-Stromborg (Ed.), Instruments for clinical nursing research (pp. 313-321). Boston, MA: Jones and Bartlett Publishers.

Davidson, W., Ash, S., Capra, S., & Bauer, J. (2003). Weight stabilisation is associated with improved survival duration and quality of life in unresectable pancreatic cancer. Clinical Nutrition.

de Vries, J. H., Zock, P. L., Mensink, R. P., & Katan, M. B. (1994). Underestimation of eneryg intake by 3-d records compared with energy intake to maintain body weight in 269 nonobese adults. American Journal of Clinical Nutrition, 60(6), 855-860.

Demling, R., & De Santi, L. (1998). Closure of the "non-healing wound" corresponds with correction of weight loss using the anabolic agent oxandrolone. Ostomy Wound Management, 44(10), 58-66.

DeWys, W. D. (1977). Changes in taste sensation in cancer patients: correlation with caloric intake. In M. R. Kare & O. Maller (Eds.), The chemical senses and nutrition .

DeWys, W. D., Begg, C., Lavin, P. T., Band, P. R., Bennett, J. M., Bertino, J. R., Cohen, M. H., Douglass, H. O., Jr., Engstrom, P. F., Ezdinli, E. Z., Horton, J., Johnson, G. J., Moertel, C. G., Oken, M. M., Perlia, C., Rosenbaum, C., Silverstein, M. N., Skeel, R. T., Sponzo, R. W., & Tormey, D. C. (1980). Prognostic effect of weight loss prior to chemotherapy in cancer patients. Eastern Cooperative Oncology Group. American Journal of Medicine, 69(4), 491-7.

DeWys, W. D., & Kubota, T. T. (1981). Enteral parenteral nutrition in the care of the cancer patient. Jama, 246(15), 1725-7.

Donnelly, S., Walsh, D., & Rybicki, L. (1995). The symptoms of advanced cancer: identification of clinical and research priorities by assessment of prevalence and severity. Journal of Palliative Care, 11(1), 27-32.

Doyle, D., Hanks, G. W. C., & MacDonald, N., Professor. (1998.). Oxford textbook of palliative medicine (2nd ed. ed.). Oxford ; New York :: Oxford University Press,.

Ellis, K. J., Bell, S. J., Chertow, G. M., Chumlea, W. C., Knox, T. A., Kotler, D. P., Lukaski, H. C., & Schoeller, D. A. (1999). Bioelectrical impedance methods in clinical research: a follow-up to the NIH Technology Assessment Conference. Nutrition, 15(11-12), 874-80.

Ellison, N. M., Chevlen, E., Still, C. D., & Dubagunta, S. (2002). Supportive care for patients with pancreatic adenocarcinoma: symptom control and nutrition. Hematology Oncology Clinics of North America, 16(1), 105-21.

Evans, W. K., Nixon, D. W., Daly, J. M., Ellenberg, S. S., Gardner, L., Wolfe, E., Shepherd, F. A., Feld, R., Gralla, R., Fine, S., & et al. (1987). A randomized study of oral nutritional support versus ad lib nutritional intake during

192 References

chemotherapy for advanced colorectal and non-small-cell lung cancer. Journal of Clinical Oncology, 5(1), 113-24.

Falconer, J. S., Fearon, K. C., Plester, C. E., Ross, J. A., & Carter, D. C. (1994). Cytokines, the acute-phase response, and resting energy expenditure in cachectic patients with pancreatic cancer. Annals of Surgery, 219(4), 325-31.

Falconer, J. S., Fearon, K. C., Ross, J. A., Elton, R., Wigmore, S. J., Garden, O. J., & Carter, D. C. (1995). Acute-phase protein response and survival duration of patients with pancreatic cancer. Cancer, 75(8), 2077-82.

Fallon, M. T., & Hanks, G. W. (1999). Morphine, constipation and performance status in advanced cancer patients. Palliative Medicine, 13(2), 159-60.

Fayers, P., Aaronson, N., Bjordal, K., & Sullivan, M. (1999). EORTC QLQ-C30 Scoring Manual. Belgium: EORTC Data Center.

Fearon, K. (1998). A prospective multi-centre randomised double-blind trial comparing the efficacy of an n-3 fatty acid enriched oral supplement with that of an isonitrogenous isocaloric oral supplement in weight-losing pancreatic cancer patients. (Clinical Research Protocol for BH80 ). Columbus: Abbott Laboratories.

Fearon, K., von Meyenfeldt, M., Moses, A., van Geenan, R., Roy, A., Gouma, D., Giacosa, A., Van Gossum, A., Bauer, J., Barber, M., Aaronson, N., Voss, A., & Tisdale, M. (2003). The effect of a protein and energy dense, n-3 fatty acid enriched oral supplement on loss of weight and lean tissue in cancer cachexia: A randomised double blind trial. Gut, 52, 1479-1486.

Fearon, K. C. H., & Moses, A. G. W. (2002). Cancer cachexia. International Journal of Cardiology, 85(1), 73-81.

Ferguson, M., Capra, S., Bauer, J., & Banks, M. (1999). Development of a valid and reliable malnutrition screening tool for adult acute hospital patients. Nutrition, 15(6), 458-64.

Feuz, A., & Rapin, C. H. (1994). An observational study of the role of pain control and food adaptation of elderly patients with terminal cancer. Journal of the American Dietetic Association, 94(7), 767-70.

Gallagher-Allred, C. R. (1989). Nutritional care of the terminally ill. Rockville, Md: Aspen Publishers.

Gaskill, D., Henderson, A., & Fraser, M. (1997). Exploring the everyday world of the patient in isolation. Oncology Nursing Forum, 24(4), 695-700.

Gogos, C. A., Ginopoulos, P., Salsa, B., Apostolidou, E., Zoumbos, N. C., & Kalfarentzos, F. (1998). Dietary omega-3 polyunsaturated fatty acids plus vitamin E restore immunodeficiency and prolong survival for severely ill patients with generalized malignancy: a randomized control trial. Cancer, 82(2), 395-402.

Goldberg, G. R., Black, A. E., Jebb, S. A., Cole, T. J., Murgatroyd, P. R., Coward, W. A., & Prentice, A. M. (1991). Critical evaluation of energy intake data using fundamental principles of energy physiology: 1. Derivation of cut-off limits to identify under-recording. European Journal of Clinical Nutrition, 45(12), 569-81.

Goris, A. H., & Westerterp, K. R. (1999). Underreporting of habitual food intake is explained by undereating in highly motivated lean women. Journal of Nutrition, 129(4), 878-882.

Grande, G. E., Todd, C. J., & Barclay, S. I. G. (1997). Support needs in the last year of life: patient and carer dilemmas. Palliative Medicine, 11(3), 202-8.

References 193

Greenhalgh, T., & Hurwitz, B. (1999). Narrative based medicine: why study narrative? British Medical Journal., 318(7175), 48-50.

Greenlee, R. T., Hill-Harmon, M. B., Murray, T., & Thun, M. (2001). Cancer statistics, 2001. CA: a Cancer Journal for Clinicians, 51(1), 15-36.

Grindel, C., Whitmer, K., & Barsevick, A. (1996). Quality of life and nutritional support in patients with cancer. Cancer Practice: a Multidisciplinary Journal of Cancer Care, 4(2), 81-7.

Grosvenor, M., Bulcavage, L., & Chlebowski, R. T. (1989). Symptoms potentially influencing weight loss in a cancer population. Correlations with primary site, nutritional status, and chemotherapy administration. Cancer, 63(2), 330-4.

Guba, E., & Lincoln, Y. (1998). Competing paradigms in qualitative research. In N. Denzin & Y. Lincoln (Eds.), The landscape of qualitative research: theories and issues (1 ed., pp. 195-220). Thousand Oaks, California: Sage Publications.

Hall, J. C., Staniland, J. R., & Giles, G. R. (1980). Altered taste thresholds in gastro-intestinal cancer. Clinical Oncology, 6(2), 137-42.

Hannan, W. J., Cowen, S. J., Plester, C. E., Fearon, K. C., & deBeau, A. (1995). Comparison of bio-impedance spectroscopy and multi-frequency bio-impedance analysis for the assessment of extracellular and total body water in surgical patients. Clinical Science, 89(6), 651-8.

Harvie, M. N., Campbell, I. T., Thatcher, N., & Baildam, A. (2003). Changes in body composition in men and women with advanced nonsmall cell lung cancer (NSCLC) undergoing chemotherapy. Journal of Human Nurition and Dietetics, 16, 323-326.

Haslett, P. A. (1998). Anticytokine approaches to the treatment of anorexia and cachexia. Seminars in Oncology, 25(2 Suppl 6), 53-7.

Held-Warmkessel, J., Volpe, H., & Waldman, A. R. (1998). Symptom management for patients with pancreatic cancer. Clinical Journal of Oncology Nursing, 2(4), 135-9.

Hill, A. J., & Blundell, J. E. (1982). Nutrients and behaviour: research strategies for the investigation of taste characteristics, food preferences, hunger sensations and eating patterns in man. Journal of Psychiatric Research, 17(2), 203-12.

Hirai, K., Hussey, H. J., Barber, M. D., Price, S. A., & Tisdale, M. J. (1998). Biological evaluation of a lipid-mobilizing factor isolated from the urine of cancer patients. Cancer Research, 58(11), 2359-65.

Holden, C. M. (1991). Anorexia in the terminally ill cancer patient: the emotional impact on the patient and the family. Hospice Journal, 7(3), 73-84.

Holman, H. R. (1993). Qualitative inquiry in medical research. Journal of Clinical Epidemiology, 46(1), 29-36.

Hoppin, J. A., Tolbert, P. E., Holly, E. A., Brock, J. W., Korrick, S. A., Altshul, L. M., Zhang, R. H., Bracci, P. M., Burse, V. W., & Needham, L. L. (2000). Pancreatic cancer and serum organochlorine levels. Cancer Epidemiology, Biomarkers & Prevention, 9(2), 199-205.

Howe, G. R., & Burch, J. D. (1996). Nutrition and pancreatic cancer. Cancer Causes & Control., 7(1), 69-82.

Howe, J. R., & Conlon, K. C. (1997). The molecular genetics of pancreatic cancer. Surgical Oncology, 6(1), 1-18.

Hunt, D. R., Rowlands, B. J., & Johnston, D. (1985). Hand grip strength--a simple prognostic indicator in surgical patients. Jpen: Journal of Parenteral & Enteral Nutrition, 9(6), 701-4.

194 References

Inui, A. (1999). Cancer anorexia-cachexia syndrome: are neuropeptides the key? Cancer Research, 59(18), 4493-501.

Inui, A. (2002). Cancer anorexia-cachexia syndrome: current issues in research and management. CA: a Cancer Journal for Clinicians., 52(2), 72-91.

Johnson, C. D., Puntis, M., Davidson, N., Todd, S., & Bryce, R. (2001). Randomized, dose-finding phase III study of lithium gamolenate in patients with advanced pancreatic adenocarcinoma. British Journal of Surgery, 88, 662-668.

Johnson, R. K., Goran, M. I., & Poehlman, E. T. (1994). Correlates of over- and underreporting of energy intake in healthy older men and women. American Journal of Clinical Nutrition, 59(6), 1286-1290.

Kaaks, R., & Lukanova, A. (2001). Energy balance and cancer: the role of insulin and insulin-like growth factor-I. Proceedings of the Nutrition Society, 60(1), 91-106.

Kalman, D., & Villani, L. J. (1997). Nutritional aspects of cancer-related fatigue. Journal of the American Dietetic Association, 97(6), 650-4.

Karayiannakis, A. J., Syrigos, K. N., Polychronidis, A., Pitiakoudis, M., Bounovas, A., & Simopoulos, K. (2001). Serum levels of tumor necrosis factor-alpha and nutritional status in pancreatic cancer patients. Anticancer Research, 21(2B), 1355-8.

Keele, A. M., Bray, M. J., Emery, P. W., Duncan, H. D., & Silk, D. B. (1997). Two phase randomised controlled clinical trial of postoperative oral dietary supplements in surgical patients. Gut, 40(3), 393-9.

Kim, Y. I. (1999). Folate and cancer prevention: a new medical application of folate beyond hyperhomocysteinemia and neural tube defects. Nutrition Reviews, 57(10), 314-21.

Kind, P. (1998). EuroQol EQ-5D User Guide. Centre for Health Economics. York, UK: University of York.

King, M. T. (1996). The interpretation of scores from the EORTC quality of life questionnaire QLQ-C30. Quality of Life Research, 5(6), 555-67.

Kirk, J., & Miller, M. L. (1986). Reliability and validity in qualitative research. Newbury Park, Calif: Sage Publications.

Krech, R. L., & Walsh, D. (1991). Symptoms of pancreatic cancer. Journal of Pain & Symptom Management, 6(6), 360-7.

Langer, C. J., Hoffman, J. P., & Ottery, F. D. (2001). Clinical significance of weight loss in cancer patients: rationale for the use of anabolic agents in the treatment of cancer-related cachexia. Nutrition, 17(1 Suppl), S1-20.

Larsson, J., Akerlind, I., Permerth, J., & Hornqvist, J. O. (1995). Impact of nutritional state on quality of life in surgical patients. Nutrition, 11(2 Suppl), 217-20.

Lesourd, B. M. (1997). Nutrition and immunity in the elderly: modification of immune responses with nutritional treatments. American Journal of Clinical Nutrition., 66(2), 478S-484S.

Levine, J. A., & Morgan, M. Y. (1998). Preservation of macronutrient preferences in cancer anorexia. British Journal of Cancer, 78(5), 579-81.

Lillemoe, K. D. (1998). Palliative therapy for pancreatic cancer. Surgical Oncology Clinics of North America, 7(1), 199-216.

Lillemoe, K. D., Yeo, C. J., & Cameron, J. L. (2000). Pancreatic cancer: state-of-the-art care. CA: a Cancer Journal for Clinicians, 50(4), 241-68.

References 195

Little, M., Jordens, C. F., Paul, K., Montgomery, K., & Philipson, B. (1998). Liminality: a major category of the experience of cancer illness. Social Science & Medicine., 47(10), 1485-94.

Livingstone, M. B., Prentice, A. M., Strain, J. J., Coward, W. A., Black, A. E., Barker, M. E., McKenna, P. G., & Whitehead, R. G. (1990). Accuracy of weighed dietary records in studies of diet and health. British Medical Journal, 300(6726), 708-12.

Lopes, J., Russell, D. M., Whitwell, J., & Jeejeebhoy, K. N. (1982). Skeletal muscle function in malnutrition. American Journal of Clinical Nutrition, 36(4), 602-610.

Loprinzi, C. L., Kugler, J. W., Sloan, J. A., Mailliard, J. A., Krook, J. E., Wilwerding, M. B., Rowland, K. M., Jr., Camoriano, J. K., Novotny, P. J., & Christensen, B. J. (1999). Randomized comparison of megestrol acetate versus dexamethasone versus fluoxymesterone for the treatment of cancer anorexia/cachexia. Journal of Clinical Oncology, 17(10), 3299-306.

Loprinzi, C. L., Schaid, D. J., Dose, A. M., Burnham, N. L., & Jensen, M. D. (1993). Body-composition changes in patients who gain weight while receiving megestrol acetate. Journal of Clinical Oncology, 11(1), 152-4.

MacDonald, N., Easson, A. M., Mazurak, V. C., Dunn, G. P., & Baracos, V. E. (2003). Understanding and managing cancer cachexia. Journal of the American College of Surgeons, 197(1), 143-161.

Macqueen, C. E., & Frost, G. (1998). Visual analog scales: a screening tool for assessing nutritional need in head and neck radiotherapy patients. Journal of Human Nutrition & Dietetics, 11(2), 115-24.

Maltoni, M., Nanni, O., Scarpi, E., Rossi, D., Serra, P., & Amadori, D. (2001). High-dose progestins for the treatment of cancer anorexia-cachexia syndrome: a systematic review of randomised clinical trials. Annals of Oncology, 12(3), 289-300.

Markley, E. J., Mattes-Kulig, D. A., & Henkin, R. I. (1983). A classification of dysgeusia. Journal of the American Dietetic Association, 83(5), 578-80.

Martin, C. (2000). Calorie, protein, fluid and micronutrient requirements. In P. McCallum & C. Polisena (Eds.), The clinical guide to oncology nutrition (pp. 45-52).

Martin, C. M., Banwell, C. L., Broom, D. H., & Nisa, M. (1999). Consultation length and chronic illness care in general practice: a qualitative study. Medical Journal of Australia, 171, 77-81.

Martin, L. J., Su, W., Jones, P. J., Lockwood, G. A., Tritchler, D. L., & Boyd, N. F. (1996). Comparison of energy intakes determined by food records and doubly labeled water in women participating in a dietary-intervention trial. American Journal of Clinical Nutrition, 63(4), 483-90.

Martin, S., Neale, G., & Elia, M. (1985). Factors affecting maximal momentary grip strength. Human Nutrition Clinical Nutrition, 39(2), 137-147.

Mattes, R. (1990). Hunger ratings are not a valid proxy measure of reported food intake in humans. Appetite, 15(2), 103-13.

May, P. E., Barber, A., D'Olimpio, J. T., Hourihane, A., & Abumrad, N. M. (2002). Reversal of cancer-related wasting using oral supplementation with a combination of B-hydroxy-B-methylbutryate, arginine and glutamine. The American Journal of Surgery, 183, 471-479.

Maykut, P. S., & Morehouse, R. (1994). Beginning qualitative research : a philosophic and practical guide. London ; Washington, D.C: Falmer Press.

196 References

McCallum, P. D., & Polisena, C. G. (Eds.). (2000). The clinical guide to oncology nutrition. Chicago: The American Dietetic Association.

McCann, R. M., Hall, W. J., & Groth-Juncker, A. (1994). Comfort care for terminally ill patients. The appropriate use of nutrition and hydration [see comments]. Journal of the American Medical Society, 272(16), 1263-6.

McCracken, G. D. (1988). The long interview. Newbury Park, California: Sage Publications.

McGrath, P. (2002). Reflections on nutritional issues associated with cancer therapy. Cancer Practice: a Multidisciplinary Journal of Cancer Care, 10(2), 94-101.

McLennan, W., & Podger, A. (1998). National Nutrition Survey: Nutrient Intakes and Physical Measurements Australia 1995. Canberra: Australian Bureau of Statistics.

McMahon, K., Decker, G., & Ottery, F. D. (1998). Integrating proactive nutritional assessment in clinical practices to prevent complications and cost. Seminars in Oncology, 25(2 Suppl 6), 20-7.

McMillan, D. C., Preston, T., Watson, W. S., Simpson, J. M., Fearon, K. C., Shenkin, A., Burns, H. J., & McArdle, C. S. (1994). Relationship between weight loss, reduction of body cell mass and inflammatory response in patients with cancer. British Journal of Surgery, 81(7), 1011-4.

McMillan, D. C., Wigmore, S. J., Fearon, K. C., O'Gorman, P., Wright, C. E., & McArdle, C. S. (1999). A prospective randomized study of megestrol acetate and ibuprofen in gastrointestinal cancer patients with weight loss. British Journal of Cancer, 79(3-4), 495-500.

Meares, C. J. (1997). Primary caregiver perceptions of intake cessation in patients who are terminally ill. Oncology Nursing Forum, 24(10), 1751-7.

Michaud, D. S., Giovannucci, E., Willett, W. C., Colditz, G. A., & Fuchs, C. S. (2001a). Coffee and alcohol consumption and the risk of pancreatic cancer in two prospective United States cohorts. Cancer Epidemiology, Biomarkers & Prevention, 10(5), 429-37.

Michaud, D. S., Giovannucci, E., Willett, W. C., Colditz, G. A., Stampfer, M. J., & Fuchs, C. S. (2001b). Physical activity, obesity, height, and the risk of pancreatic cancer. Journal of the American Medical Association, 286(8), 921-9.

Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis : an expanded sourcebook (2nd ed.). Thousand Oaks: Sage Publications.

Mishler, E. G. (1986). Research interviewing : context and narrative. Cambridge, Mass: Harvard University Press.

Mor, V., Laliberte, L., Morris, J. N., & Wiemann, M. (1984). The Karnofsky Performance Status Scale. An examination of its reliability and validity in a research setting. Cancer, 53(9), 2002-7.

Moynihan, C. (2002). Men, women, gender and cancer. European Journal of Cancer Care, 11, 166-172.

Murphy, S., & Von Roenn, J. H. (2000). Pharmacological management of anorexia and cachexia. In P. D. McCallum & C. G. Polisena (Eds.), The clinical guide to oncology nutrition (pp. 127-133). Chicago: The American Dietetic Association.

National Health and Medical Research Council (Australia). (1995). Ethical aspects of qualitative methods in health research : an information paper for institutional ethics committees. Canberra, ACT: NHMRC.

References 197

Nelson, K., Walsh, D., Deeter, P., & Sheehan, F. (1994). A phase II study of delta-9-tetrahydrocannabinol for appetite stimulation in cancer associated cachexia. Journal of Palliative Care, 10, 14-18.

Neoptolemos, J. P., Cunningham, D., Friess, H., Bassi, C., Stocken, D. D., Tait, D. M., Dunn, J. A., Dervenis, C., Lacaine, F., Hickey, H., Raraty, M. G., Ghaneh, P., & Buchler, M. W. (2003). Adjuvant therapy in pancreatic cancer: historical and current perspectives. Annals of Oncology, 14(5), 675-92.

Nielsen, S. S., Theoglides, A., & Vickers, Z. M. (1980). Influence of food odors on food aversions and preferences in patients with cancer. The American Journal of Clinical Nutrition, 33, 2253-2261``.

Nitenberg, G., & Raynard, B. (2000). Nutritional support of the cancer patient: issues and dilemmas. Critical Reviews in Oncology-Hematology, 34(3), 137-68.

Nixon, D. W. (1996). Cancer, cancer cachexia, and diet: lessons from clinical research. Nutrition, 12(1 Suppl), S52-56.

Northouse, L. L., Schafer, J. A., Tipton, J., & Metivier, L. (1999). The concerns of patients and spouses after the diagnosis of colon cancer: a qualitative analysis. Journal of WOCN, 26(1), 8-17.

O'Gorman, P., McMillan, D. C., & McArdle, C. S. (1998). Impact of weight loss, appetite, and the inflammatory response on quality of life in gastrointestinal cancer patients. Nutrition & Cancer, 32(2), 76-80.

Ollenschlager, G., Konkol, K., & Modder, B. (1988). Indications for and results of nutritional therapy in cancer patients. Recent Results in Cancer Research, 108, 172-84.

Ollenschlager, G., Thomas, W., Konkol, K., Diehl, V., & Roth, E. (1992). Nutritional behaviour and quality of life during oncological polychemotherapy: results of a prospective study on the efficacy of oral nutrition therapy in patients with acute leukaemia. European Journal of Clinical Investigation, 22(8), 546-53.

O'Riordain, M. G., Falconer, J. S., Maingay, J., Fearon, K. C., & Ross, J. A. (1999). Peripheral blood cells from weight-losing cancer patients control the hepatic acute phase response by a primarily interleukin-6 dependent mechanism. International Journal of Oncology, 15(4), 823-7.

Ottery, F. (1996a). Supportive nutritional management of the patient with pancreatic cancer. Oncology (Huntington), 10(9 Suppl), 26-32.

Ottery, F. D. (1995). Supportive nutrition to prevent cachexia and improve quality of life. Seminars in Oncology, 22(2 Suppl 3), 98-111.

Ottery, F. D. (1996b). Definition of standardized nutritional assessment and interventional pathways in oncology. Nutrition, 12(1 Suppl), S15-9.

Ottery, F. D. (2000). Bidirectional interplay of nutrition and chemotherapy. Nestle Nutrition Workshop Series Clinical & Performance Program., 4, 183-202; discussion 203-6.

Ovesen, L., Allingstrup, L., Hannibal, J., Mortensen, E. L., & Hansen, O. P. (1993a). Effect of dietary counseling on food intake, body weight, response rate, survival, and quality of life in cancer patients undergoing chemotherapy: a prospective, randomized study. Journal of Clinical Oncology, 11(10), 2043-9.

Ovesen, L., Hannibal, J., & Mortensen, E. (1993b). The interrelation of weight loss, dietary intake, and quality of life in ambulatory patients with cancer of the lung, breast and ovary. Nutrition and Cancer, 19, 159-167.

Padilla, G. V. (1986). Psychological aspects of nutrition and cancer. Surgical Clinics of North America, 66(6), 1121-35.

198 References

Padilla, G. V., Presant, C., Grant, M. M., Metter, G., Lipsett, J., & Heide, F. (1983). Quality of life index for patients with cancer. Research in Nursing & Health, 6(3), 117-26.

Passik, S. D., & Breitbart, W. S. (1996). Depression in patients with pancreatic carcinoma. Diagnostic and treatment issues. Cancer., 78(3 Suppl), 615-26.

Perez, M. M., Newcomer, A. D., Moertel, C. G., Go, V. L., & Dimagno, E. P. (1983). Assessment of weight loss, food intake, fat metabolism, malabsorption, and treatment of pancreatic insufficiency in pancreatic cancer. Cancer., 52(2), 346-52.

Permert, J., Adrian, T. E., Jacobsson, P., Jorfelt, L., Fruin, A. B., & Larsson, J. (1993). Is profound peripheral insulin resistance in patients with pancreatic cancer caused by a tumor-associated factor? American Journal of Surgery., 165(1), 61-6; discussion 66-7.

Persson, C., Sjoden, P. O., & Glimelius, B. (1999). The Swedish version of the patient-generated subjective global assessment of nutritional status: gastrointestinal vs urological cancers. Clinical Nutrition., 18(2), 71-7.

Polgar, S., & Thomas, S. A. (1995). Introduction to research in the health sciences (3rd ed.). Melbourne, Vic. ; Edinburgh: Churchill Livingstone.

Pope, C., & Mays, N. (1995). Reaching the parts other methods cannot reach: an introduction to qualitative methods in health and health services research. British Medical Journal, 311, 42-5.

Potter, J., Langhorne, P., & Roberts, M. (1998). Routine protein energy supplementation in adults: systematic review. British Medical Journal, 317(7157), 495-501.

Price, G. M., Paul, A. A., Cole, T. J., & Wadsworth, M. E. (1997). Characteristics of the low-energy reporters in a longitudinal national dietary survey. British Journal of Nutrition, 77(6), 833-51.

Raben, A., Tagliabue, A., & Astrup, A. (1995). The reproducibility of subjective appetite scores [published erratum appears in Br J Nutr 1995 Aug;74(2):283]. British Journal of Nutrition, 73(4), 517-30.

Rebro, S. M., Patterson, R. E., Kristal, A. R., & Cheney, C. L. (1998). The effect of keeping food records on eating patterns. Journal of the American Dietetic Association, 98(10), 1163-5.

Reeds, P., Fjeld, J., & Jahoor, F. (1994). Do the differences between the amino acid composition of acute-phase and skeletal muscle proteins have a bearing on nitrogen loss in traumatic states? Journal of Nutrition, 124, 906-10.

Reeves, M. M., & Capra, S. (2003). Predicting energy requirements: are current methods evidence based? Nutrition Reviews, in press.

Reuben, D. B., Mor, V., & Hiris, J. (1988). Clinical symptoms and length of survival in patients with terminal cancer. Archives of Internal Medicine, 148(7), 1586-91.

Rhodes, V. A., & McDaniel, R. W. (2001). Nausea, vomiting, and retching: complex problems in palliative care. CA: a Cancer Journal for Clinicians, 51(4), 232-48.

Robuck, J. T., & Fleetwood, J. B. (1992). Nutritional support of the patient with cancer. Focus on Critical Care, 19(2), 129-30, 132-4, 136-8.

Rosenbaum, K., Wang, J., Pierson, R. N., Jr., & Kotler, D. P. (2000). Time-dependent variation in weight and body composition in healthy adults. JPEN: Journal of Parenteral & Enteral Nutrition, 24(2), 52-5.

References 199

Rothenberg, M. L., Abbruzzese, J. L., Moore, M., Portenoy, R. K., Robertson, J. M., & Wanebo, H. J. (1996). A rationale for expanding the endpoints for clinical trials in advanced pancreatic carcinoma. Cancer., 78(3 Suppl), 627-32.

Roubenoff, R. (2000). Sarcopenia and its implications for the elderly. European Journal of Clinical Nutrition., 54(Supplement 3), S40-S47.

Russell, S. (1997). Dignity to the end. Nursing Times, 93(48), 34-5. Sanders, C., Egger, M., Donovan, J., Tallon, D., & Frankel, S. (1998). Reporting on

quality of life in randomised controlled trials: bibliographic study. British Medical Journal., 317(7167), 1191-4.

Sauter, P. K., & Coleman, J. (1999). Pancreatic cancer: a continuum of care. Seminars in Oncology Nursing, 15(1), 36-47.

Schiffman, S. S. (1997). Taste and smell losses in normal aging and disease. Journal of the American Medical Association, 278(16), 1357-62.

Schroeder, D., & Hill, G. L. (1991). Postoperative fatigue: a prospective physiological study of patients undergoing major abdominal surgery. Australian & New Zealand Journal of Surgery, 61(10), 774-9.

Seligman, P. A., Fink, R., & Massey-Seligman, E. J. (1998). Approach to the seriously ill or terminal cancer patient who has a poor appetite. Seminars in Oncology, 25(2 Suppl 6), 33-4.

Silverman, D. (1993). Interpreting qualitative data : methods for analysing talk, text, and interaction. London ; Thousand Oaks [Calif.]: Sage Publications.

Silverman, D. T., Dunn, J. A., Hoover, R. N., Schiffman, M., Lillemoe, K. D., Schoenberg, J. B., Brown, L. M., Greenberg, R. S., Hayes, R. B., & Swanson, G. M. (1994). Cigarette smoking and pancreas cancer: a case-control study based on direct interviews. Journal of the National Cancer Institute., 86(20), 1510-6.

Silverman, D. T., Schiffman, M., Everhart, J., Goldstein, A., Lillemoe, K. D., Swanson, G. M., Schwartz, A. G., Brown, L. M., Greenberg, R. S., Schoenberg, J. B., Pottern, L. M., Hoover, R. N., & Fraumeni, J. F., Jr. (1999). Diabetes mellitus, other medical conditions and familial history of cancer as risk factors for pancreatic cancer. British Journal of Cancer., 80(11), 1830-7.

Silverman, D. T., Swanson, C. A., Gridley, G., Wacholder, S., Greenberg, R. S., Brown, L. M., Hayes, R. B., Swanson, G. M., Schoenberg, J. B., Pottern, L. M., Schwartz, A. G., Fraumeni, J. F., Jr., & Hoover, R. N. (1998). Dietary and nutritional factors and pancreatic cancer: a case-control study based on direct interviews. Journal of the National Cancer Institute, 90(22), 1710-9.

Silverstone, T. (1982). Drugs and appetite. London ; New York: Academic Press. Sinclair, A., O'Dea, K., & Johnson, L. (1994). Estimation of the n-3 PUFA status in a

group of urban Australians by the analysis of plasma phospholipid fatty acids. Australian Journal of Nutrition and Dietetics, 51(2), 53-56.

Sloan, J. A., Loprinzi, C. L., Laurine, J. A., Novotny, P. J., Vargas-Chanes, D., Krook, J. E., O'Connell, M. J., Kugler, J. W., Tirona, M. T., Kardinal, C. G., Wiesenfeld, M., Tschetter, L. K., Hatfield, A. K., & Schaefer, P. L. (2001). A simple stratification factor prognostic for survival in advanced cancer: the good/bad/uncertain index. Journal of Clinical Oncology, 19(15), 3539-46.

Splett, P. L. (1996). Cost Outcomes of Nutrition Intervention. Part 2. New York: Mead Johnson and Company.

Staal-Van, D. B. A. J., Dentener, M. A., Schols, A. M. W. J., Buurman, W. A., & Wouters, E. F. M. (1995). Increased resting energy expenditure and weight

200 References

loss are related to a systemic inflammatory response to lung cancer patients. Journal of Clinical Oncology, 13(10), 2600-2605.

Staniszewska, S., & Ahmed, L. (1999). The concepts of expectation and satisfaction: do they capture the way patients evaluate their care? Journal of Advanced Nursing, 29(2), 364-72.

Stolzenberg-Solomon, R. Z., Albanes, D., Nieto, F. J., Hartman, T. J., Tangrea, J. A., Rautalahti, M., Sehlub, J., Virtamo, J., & Taylor, P. R. (1999). Pancreatic cancer risk and nutrition-related methyl-group availability indicators in male smokers. Journal of the National Cancer Institute, 91(6), 535-541.

Stratton, R., & Elia, M. (2000). Are oral nutrition supplements of benefit to patients in the community? Findings from a systematic review. Current Opinion in Clinical Nutrition and Metabolic Care, 3(4), 311-315.

Strohl, R. A. (1992). Measuring Alterations in Taste and Smell. In M. Frank-Stromborg (Ed.), Instruments for Clinical Nursing Research (pp. 357-367). Boston, MA: Jones and Bartlett Publishers.

Stubbs, L. (1989). Taste changes in cancer patients. Nursing Times, 85(3), 49-50. Tapsell, L. (1997). Dietetics as everyday practice: an approach to research.

Australian Journal of Nutrition and Dietetics, 54(2), 82-87. Tchekmedyian, N. S., Hickman, M., Siau, J., Greco, F. A., Keller, J., Browder, H., &

Aisner, J. (1992). Megestrol acetate in cancer anorexia and weight loss. Cancer, 69(5), 1268-74.

Tchekmedyian, S., Fesen, M., Price, L. M., & Ottery, F. D. (2003). Ongoing placebo-controlled study of oxandrolone in cancer-related weight loss. International Journal of Radiation Oncology, Biology, Physics, 57(2), S283-284.

Terwee, C. B., Nieveen Van Dijkum, E. J., Gouma, D. J., Bakkevold, K. E., Klinkenbijl, J. H., Wade, T. P., van Wagensveld, B. A., Wong, A., & van der Meulen, J. H. (2000). Pooling of prognostic studies in cancer of the pancreatic head and periampullary region: the Triple-P study. Triple-P study group. European Journal of Surgery, 166(9), 706-12.

Tisdale, M. J. (1997a). Biology of cachexia. Journal of the National Cancer Institute, 89(23), 1763-73.

Tisdale, M. J. (1997b). Cancer cachexia: metabolic alterations and clinical manifestations. Nutrition, 13(1), 1-7.

Tisdale, M. J. (2002). Biochemical mechanisms of cellular catabolism. Current Opinion in Clinical Nutrition and Metabolic Care, 5, 401-405.

Tisdale, M. J., & Beck, S. A. (1991). Inhibition of tumour-induced lipolysis in vitro and cachexia and tumour growth in vivo by eicosapentaenoic acid. Biochemical Pharmacology, 41(1), 103-7.

Todorov, P., Cariuk, P., McDevitt, T., Coles, B., Fearon, K., & Tisdale, M. (1996). Characterization of a cancer cachectic factor. Nature, 379(6567), 739-42.

Tomoyasu, N. J., Toth, M. J., & Poehlman, E. T. (1999). Misreporting of total energy intake in older men and women. Journal of the American Geriatricians Society, 47(6), 710-715.

Trabulsi, J., & Schoeller, D. A. (2001). Evaluation of dietary assessment instruments against doubly labelled water, a biomarker of habitual energy intake. American Journal of Physiology Endocrinology and Metabolism, 281(5), E891-899.

Trant, A. S., Serin, J., & Douglass, H. O. (1982). Is taste related to anorexia in cancer patients? American Journal of Clinical Nutrition, 36(1), 45-58.

References 201

van Bokhorst de van der Scheuren, M. A., van Leeuwen, P. A., Sauerwein, H. P., Kuik, D. J., Snow, G. B., & Quak, J. J. (1997). Assessment of malnutrition parameters in head and neck cancer and their relation to postoperative complications. Head and Neck, 19(5), 419-425.

van den Bosch, R. P., van der Schelling, G. P., Klinkenbijl, J. H., Mulder, P. G., van Blankenstein, M., & Jeekel, J. (1994). Guidelines for the application of surgery and endoprostheses in the palliation of obstructive jaundice in advanced cancer of the pancreas. Annals of Surgery, 219(1), 18-24.

Voss, S., Kroke, A., Klipstein-Grobusch, K., & Boeing, H. (1998). Is macronutrient composition of dietary intake data affected by underreporting? Results from the EPIC-Potsdam Study. European Prospective Investigation into Cancer and Nutrition. European Journal of Clinical Nutrition, 52(2), 119-126.

Wade, E. V., & Jain, R. (1984). Nutritional support: enhancing the quality of life of the terminally ill patient with cancer. Journal of the American Dietetic Association, 84(9), 1044-5.

Walsh, D., Nelson, K. A., & Mahmoud, F. A. (2003). Established and potential therapeutic applications of cannabinoids in oncology. Supportive Care in Cancer, 11, 137-143.

Wanebo, H. J., Glicksman, A. S., Vezeridis, M. P., Clark, J., Tibbetts, L., Koness, R. J., & Levy, A. (2000). Preoperative chemotherapy, radiotherapy, and surgical resection of locally advanced pancreatic cancer. Archives of Surgery, 135(1), 81-7; discussion 88.

Wickham, R. S., Rehwaldt, M., Kefer, C., Shott, S., Abbas, K., Glynn-Tucker, E., Potter, C., & Blendowski, C. (1999). Taste changes experienced by patients receiving chemotherapy. Oncology Nursing Forum, 26(4), 697-706.

Wigmore, S. J., Barber, M. D., Ross, J. A., Tisdale, M. J., & Fearon, K. C. (2000a). Effect of oral eicosapentaenoic acid on weight loss in patients with pancreatic cancer. Nutrition & Cancer, 36(2), 177-84.

Wigmore, S. J., Falconer, J. S., Plester, C. E., Ross, J. A., Maingay, J. P., Carter, D. C., & Fearon, K. C. (1995). Ibuprofen reduces energy expenditure and acute-phase protein production compared with placebo in pancreatic cancer patients. British Journal of Cancer, 72(1), 185-8.

Wigmore, S. J., Fearon, K. C., Maingay, J. P., & Ross, J. A. (1997a). Down-regulation of the acute-phase response in patients with pancreatic cancer cachexia receiving oral eicosapentaenoic acid is mediated via suppression of interleukin-6. Clinical Science, 92(2), 215-21.

Wigmore, S. J., Plester, C. E., Richardson, R. A., & Fearon, K. C. (1997b). Changes in nutritional status associated with unresectable pancreatic cancer. British Journal of Cancer, 75(1), 106-9.

Wigmore, S. J., Plester, C. E., Ross, J. A., & Fearon, K. C. (1997c). Contribution of anorexia and hypermetabolism to weight loss in anicteric patients with pancreatic cancer. British Journal of Surgery, 84(2), 196-7.

Wigmore, S. J., Ross, J. A., Falconer, J. S., Plester, C. E., Tisdale, M. J., Carter, D. C., & Fearon, K. C. (1996). The effect of polyunsaturated fatty acids on the progress of cachexia in patients with pancreatic cancer. Nutrition, 12(1 Suppl), S27-30.

Wigmore, S. J., Todorov, P. T., Barber, M. D., Ross, J. A., Tisdale, M. J., & Fearon, K. C. (2000b). Characteristics of patients with pancreatic cancer expressing a novel cancer cachectic factor. British Journal of Surgery, 87(1), 53-8.

202 References

Windsor, J. A., & Hill, G. L. (1988). Weight loss with physiologic impairment. A basic indicator of surgical risk. Annals of Surgery, 207(3), 290-6.

Yates, J. W., Chalmer, B., & McKegney, F. P. (1980). Evaluation of patients with advanced cancer using the Karnofsky performance status. Cancer, 45(8), 2220-4.

Zinna, E. M., & Yarasheski, K. E. (2003). Exercise treatment to counteract protein wasting of chronic diseases. Current Opinion in Clinical Nutrition and Metabolic Care, 6(1), 87-93.

Zuijdgeest-Van Leeuwen, S. D., Van der Heijden, M. S., Rietveld, T., Van den Berg, J. W., Tilanus, H. W., Burgers, J. A., Wilson, J. H., & Dagnelie, P. C. (2002). Fatty acid composition of plasma lipids in patients with pancreatic, lung and oesophageal cancer in comparison with healthy subjects. Clinical Nutrition, 21(3), 225-230.

Assessment of nutrition intervention for patients with...

Documents

Transcript of Assessment of nutrition intervention for patients with...