Assessment of nutrition intervention for patients with...
Transcript of 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.
Literature Review 11
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).
Literature Review 12
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).
Literature Review 13
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
Literature Review 14
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)
Literature Review 15
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
Literature Review 16
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.
Literature Review 17
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.
Literature Review 18
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
Literature Review 19
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).
Literature Review 20
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
Literature Review 21
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
Literature Review 22
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
Literature Review 23
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.
Literature Review 24
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
Literature Review 25
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
Literature Review 26
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
Literature Review 27
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
Literature Review 28
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
Literature Review 29
(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
Literature Review 30
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
Literature Review 31
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
Literature Review 32
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).
Literature Review 33
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.
Literature Review 34
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
Literature Review 35
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
Literature Review 36
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
Literature Review 37
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
Literature Review 39
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
Literature Review 40
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).
Literature Review 41
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.
Literature Review 42
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
Literature Review 43
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).
Literature Review 44
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
Literature Review 45
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.
Literature Review 46
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.
50 Methods – Aims 1&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).
Methods – Aims 1&2 51
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.
52 Methods – Aims 1&2
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
Methods – Aims 1&2 53
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.
54 Methods – Aims 1&2
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.
Methods – Aims 1&2 55
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:
intervention independent variable dependent variable
ONS – oral nutrition support, QoL – quality of life, BMI – body mass index, PS – performance status, T0 –
baseline, T8 – week 8.
56 Methods – Aims 1&2
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,
Methods – Aims 1&2 57
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
58 Methods – Aims 1&2
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.
Methods – Aims 1&2 59
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
60 Methods – Aims 1&2
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.
Methods – Aims 1&2 61
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
62 Methods – Aims 1&2
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
Methods – Aims 1&2 63
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.
64 Methods – Aims 1&2
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
66 Methods – Aims 1&2
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
Results – Aims 1&2 69
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).
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.
Results – Aims 1&2 70
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
Results – Aims 1&2 71
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
Results – Aims 1&2 72
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
Results – Aims 1&2 73
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.
intervention independent variable dependent variable
Results – Aims 1&2 74
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)
Results – Aims 1&2 75
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
Variable WL WS p value
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
Results – Aims 1&2 76
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.
Variable WL WS p value
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).
Results – Aims 1&2 77
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
Results – Aims 1&2 78
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
Results – Aims 1&2 79
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.
Results – Aims 1&2 80
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
Methods – Aim 3 84
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)
Methods – Aim 3 86
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.
5.3.2 Nutrient intake
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
Methods – Aim 3 88
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
Methods – Aim 3 90
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.
Methods – Aim 3 92
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
Table 6.1 Comparison of baseline characteristics of patients with unresectable
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.
96 Results – Aim 3
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).
Results – Aim 3 97
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)
98 Results – Aim 3
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)
Results – Aim 3 99
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.
100 Results – Aim 3
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
Results – Aim 3 101
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
102 Results – Aim 3
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
Results – Aim 3 103
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.
104 Results – Aim 3
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
Results – Aim 3 105
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).
106 Results – Aim 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”.
Results – Aim 3 107
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)
108 Results – Aim 3
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)
Results – Aim 3 109
“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”.
110 Results – Aim 3
“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.
Results – Aim 3 111
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.
112 Results – Aim 3
“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)
Results – Aim 3 113
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,
114 Results – Aim 3
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.
Results – Aim 3 115
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
116 Results – Aim 3
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
Results – Aim 3 117
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.
118 Results – Aim 3
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.
7.3.4 Summary - Aim 2
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.
7.4.2 Nutrient intake
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
7.4.5 Summary - Aim 3
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.
154
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)
BH80 trial (subprotocol) Patient Initials___________
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
BH80 trial (subprotocol) Patient Initials___________
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
Appendices 185
Appendix 14: Oral presentation – Dietitians Association of Australia 21st National Conference
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
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.