Small Bowel Obstruction, Acute Pancreatitis,Upper GI Bleeding:

An Update

Jonathan Terdiman, MDProfessor of Clinical Medicine and Surgery

University of California, San Francisco

Disclosures: Nothing to disclose

Acknowledgements: Jon Carter, MD and John Cello, MD

Abdominal plain films

CT abdomen and pelvis

bowel wall edema, collapsed colonsmall bowel feces sign present

• What is the cause of the patient’s intestinal obstruction?

• Is the obstruction strangulating or non-strangulating?

• Can any tests differentiate patients whose non-strangulating obstructionwill resolve with time?

• How long should non-operative management be tried?

• Can adhesiolysis reduce the risk of recurrent SBO?

What is the cause of the patient’s intestinal obstruction?

Etiology Incidence, %

Adhesions20% within 1 month of surgery30% within 1 year of surgery25% years 1-525% after 5 years

60

Cancer 20

Hernia 10Inflammatory Bowel Disease 5

Volvulus 3

Miscellaneous 2

Pathogenesis: adhesive disease

Procedure Incidence, %

Appendectomy 23

Colorectal surgery 21Gynecologic surgery 12Foregut surgery 9

Small bowel surgery 8

Multiple laparotomies 24

Is the obstruction strangulating or non-strangulating?

Is the obstruction strangulating or non-strangulating?

Silen et al., Strangulation obstruction of the small intestine. Arch Surg 1962;85:121-129. “The results of this study indicate that the clinical differentiation between simple and strangulating obstruction is often impossible.”

The “classic signs” of strangulating obstruction are:* continuous (rather than colicky) pain* fever* tachycardia* peritoneal signs* leukocytosis

….but alone, or in combination, sensitivity / specificity low

Clinical Study

• Retrospectively reviewed 192 cases operated on for a small bowel obstruction (1996-2006) at UCSF Medical Center.

• A predictor model was created based upon operative findings: strangulated (n=44) or non-strangulated (n=148).

• Independent Predictors of strangulation: WBC > 12K, Rebound/Guarding at PE, Reduced Enhancement of SB at CT.

Is the obstruction strangulating or non-strangulating?

The best initial study is a CT abdomen/pelviswith IV contrast and without (positive) oral contrast

“I don’t buy it. You just have to use your judgment.”

Madhulika G. Varma, MD, Chief Section of Colon and Rectal Surgery, UCSF

Can any tests differentiate patients whose non-strangulating obstruction will resolve non-operatively?

Complete obstruction = absence of significant flatus or stool for 12 hours and no colonic gas seen on KUB.

Complete obstruction = 20% success rate with non-operative treatment, 20-40% risk of strangulation

Partial obstruction = 80% success rate with non-operative treatment, low risk of strangulation (3-6%)

OLD: CLINICAL PRESENTATION

Can any tests differentiate patients whose non-strangulating obstruction will resolve non-operatively?

NEW: ORAL WATER SOLUBLE CONTRAST ADMINISTRATION

Instill 50-150cc of gastrograffin (water-soluble contrast) orally or via NGT. Obtain abdominal plain films at 4, 8, and/or 24 hours

Presence of gastrograffin in the colon at 8 hours predicts non-operative resolution with 95% sensitivity and 99% specificity. PPV = 99%, NPV =85%.

At 24 hours, 99% sensitivity, 97% specificity, 99% PPV, 97% NPV

Can any tests differentiate patients whose non-strangulating obstruction will resolve non-operatively?

Possible therapeutic role for gastrograffin:

PROPORTION NEEDING SURGERY

“I don’t buy it. You just have to use your judgment.”

Madhulika G. Varma, MD, Chief Section of Colon and Rectal Surgery, UCSF

How long should non-operative management be tried?

85-95% of patients with adhesive SBO who are destined to recover without surgery will show marked improvement within 72 hours

EAST guidelines 2009:3-5 days

Bologna guidelines 2010: 3 days

Can adhesiolysis reduce the risk of recurrent SBO, readmission, or reoperation?

Review of 500 patients Norway operated upon for adhesive SBO from 1961-1995.All patients interviewed at median follow-up of 11 years, 85% follow-up rate

For patients with > 2 prior bouts of adhesive SBO,

Surgery:

reduced risk of readmission (RR 0.55)

…but did not change risk of future operations (RR 0.79)

Can adhesiolysis reduce the risk of recurrent SBO, readmission, or reoperation?

Surgery…had no effect on total readmissions (32% vs 34%)but spaced out readmissions over time (median 0.7 vs 2 years)and had no difference in reoperation rate (14% vs 11%)

Special cases

virgin abdomen • confirm diagnosis (no ileus)• exam to rule out hernia• CT abdomen/pelvis to look for intussusception, tumor, Crohn’s, malrotation, etc.

• diagnostic laparoscopy and LOA

intra-abdominal malignancy •1/3 caused from adhesions that can be lysed.•1/3 caused from tumor than can be resected or bypassed.

•1/3 nothing can be done surgically• high failure rate / mortality with nonoperative management.

intussussception • if <3cm in length, spontaneous resolution expected.

• operate for obstruction, pain, intransigence.

hernia • operate on all hernias causing SBO.• relative contraindication for laparoscopy• zebras are: obturator, paraduodenal, transmesenteric, transomental, congenital hernias

Crohn’s • treat medically initially. Surgery for intransigence.

tumors • GIST, adenoma, lipoma, hamartoma, hemangioma, adenocarcinoma, lymphoma, carcinoid, metastasis• Rx is resection

early postoperative(within 30 days of surgery)

• risk of strangulation low• success rate of nonoperative management is 85% at 7 days, 90% at 14 days

Acute Pancreatitis

DefinitionsMorphological Definition: Tissue is rarely available.

Mild: Interstitial edema andlimited fat necrosis

Severe: Hemorrhage, extensive necrosis ofpancreas & peripancreatic tissue

Severe systemic disease - oftenmulti-organ failure

Importance: Nearly 2-3% of all patients admitted to acute care hospitals

Incidence– Acute pancreatitis 5 to 75 cases per 100,000/year

Scotland: 24 cases per 100,000Finland: 73 cases per 100,000

– Gallstone pancreatitis: 35-75% of all casesMortality– Overall for acute pancreatitis: 10%

Etiology - Acute Pancreatitis

1. Mechanical Factors – most common

Gall stones – about 75% of cases worldwideTrauma/post-operativeERCP - about 1- 4 % of proceduresParasites - Ascaris, Clonorchis, Opisthorchis,

Fasciola hepaticaDuct obstruction – tumorPancreas divisum

Etiology - Acute Pancreatitis

2. Metabolic

AlcoholHypertriglyceridemia ( usually TG > 2500 )Drugs ( thiazides, 6-MP, MTX, DDI, DDC, sulfa )Scorpion stingOrganophosphate insecticidesHeredofamilial disease – familial pancreatitis

cystic fibrosis“Tropical" pancreatitis

Etiology - acute Pancreatitis

3. Other Factors

Infections - mumpshepatitis A/B/CCMV, cryptosporidiosis, microsporidiosisHIV (AIDS papillary stenosis)

Vasculitis - lupus

Idiopathic - much may be microcalculi

Sphincter of Oddi disfunction

Gallstone Pancreatitis

Most common etiology world-wideCholelithiasis – pancreatitis risk about 2% over 30 yrs.Structural risk factors: small stones, large cystic duct diameter, common distal channel.

Acute Alcoholic Pancreatitis –Epidemiologic and Experimental DataUsually demographics – steady state alcohol usage with normal, nutritious diet Mean age of onset of “acute” pancreatitis is 10 years later than “chronic” pancreatitis.Most patients with acute pancreatitis already have “chronic pancreatitis” histologically – on autopsy.Effect of alcohol/ nutritious diet in primates and canines – increased pancreatic juice viscosity, increased protein concentration and decreased pancreatic “stone protein” and trypsin inhibitory protein.

Diagnosis of Acute Pancreatitis – use of serum enzymes

(Steinberg et al. Ann Int Med 1985; 102:576)

Sensitivity % Specificity %

Amylase 95 86

P-amylase 92 83

Trypsinogen 97 83

Lipase 87 95

Severity Assessment• Ranson criteria of historical interest• Multiple other scoring systems• BISAP (Bedside Index for Severity of Acute Pancreatitis)

is likely best• BISAP assessed in first 24 hours

– BUN > 25 mg/dl– Glasgow coma score < 15– Age > 60– Pleural effusion– SIRS present*

*2 of the following 5 present•T < 36, > 38•RR > 20 or PaCO2 < 32•HR > 90•WBC < 4 or > 12 or >10% bands

BISAPWu BU, Gut 2008;57:1698-703

CT can see fluid collections around pancreas, but also presence &

amount of

pancreatic necrosis

<30%

30-50%

> 50%More necrosis = worse prognosis

Degree of necrosis

Management: “Golden Hours”• Management in the first 24 hours:• Aggressive fluid resuscitation is the key

– 2000 ml bolus in ED– 250-300 ml/hr– Use lactated Ringer’s– UO of at least 0.5 ml/kg/hr

• Under resuscitation– Pancreatic hypoperfusion occurs– SIRS, Pancreatic Necrosis, Mortality all

increase

Early Fluid ResuscitationWarndorf MG, Clin Gastroenterol Hepatol 2011;9:705-9

• Early resuscitation– SIRS at 72 hours 10% vs 23%– Organ Failure 5% vs 10%– ICU care 6% vs 17%– Hospital Stay 8 d vs 11 d

Mortality Benefit in other studies as much as 0 versus 20%Banks PA, Am J Gastroenterol 2006;101:2379-400Gardner TB, Pancreatology 2009;9:770-6

What Doesn’t Help• NG Suction• H2 blockers or PPIs• Atropine or Glucagon• Somatostatin or Octreotide• Protease Inhibitors• TPN• Peritoneal lavage• Antibiotics and ERCP?

Nutritional SupportTotal Enteral Nutrition (TEN)Enteral feeds should be post ligament of

Treitz (not gastric or duodenal feeds!!)– Advantages:

reduced cost compared to TPN or PPNno associated risk of catheter line sepsisprevents gut atrophy and bacterial translocation and infection

– Disadvantages:risk of proximal migration of the tube tip into the duodenum or stomachpatient refusal due to discomfort

Prophylactic Antibiotics• Wittau M, Scand J

Gastroenterol 2011;46:261-70• 14 trials with 841 patients• Infected pancreatic necrosis:

– RR 0.78 (95% CI, 0.6-1.1)

• Non pancreatic infections:– RR 0-.70 (95% CI, 0.46-1.1)

• Mortality: – RR 0.74 (95% CI, 0.5-1.1)

Bai Y, Am J Gastroenterol 2008;103:104-10

Early Treatment of Acute Biliary Pancreatitis

Acute ERCP (97 pts.) vs. Conservative (98 pts) Acute ERCP Converv. P value

mortality (overall) 5% 9% NS

overall morbitity 18% 29% NS

morbidity (patients 16% 33% 0.03 with stones)

systemic complications 10% 14% NS

local complications 10% 12% NS

biliary sepsis 0 12% 0.003

Fan ST et al. NEJM 1993;328:228-32.

20-50% biliary pancreatitis recurrence within 8 weeks, so prior to discharge or soon after ERCP with sphincterotomy or cholecystectomy is needed

Early ERCP/Papillotomy compared with Conservative Treatment for Acute Biliary

Pancreatitis

Parameter ERCP Conservative Treatment P Value

Patients randomized 126 112

ERCP performed 121 22

ES/ Stone removal 58 13

Overall deaths 14 7 NS

Pancreatitis deaths 10 4 NS

Respiratory failure 15 5 0.03

Renal failure 9 4 NS

Cholecystitis 13 20 NSJaundice 1 12 0.02Cholangitits 17 13 NS

Folsch, Nitsche, Ludtke et al NEJM 1997;336:237-42.

ERCP for all versus only for ongoing obstruction and sepsis

UK 1995(N=4137)

US-CORI 2004

(N=7822)

Canada2002

(N=1869

Hong Kong2007

PUD 35% 21% 50% 60%Erosive disease 21% 43% 10% 13%

MW Tear 5% 4% 3%Varices 4% 11% 4%Malignancy 4% 1% 4%AVM 5% 0.3%Esophagitis 15% 17% (w/erosive)Other 6% 4%No Dx Made 25% 4% 11%

Upper GI Bleeding

Definition of Low Risk: < 5 % Rebleeding, < 1 % Mortality

BlatchfordSystolic BP > 110 Pulse < 100 Hgb > 12 (women), 13 (men)BUN < 6.5 mmol/LComorbities: none

Clinical RockallSystolic BP > 100Pulse < 100Age < 60 yearsComorbidities: none

Early Endoscopy: Risk Assessment and Therapy

Stigmata Prevalence (%) Rebleed (%)

Arterial bleed 10 90

Visible vessel 25 50

Adherent clot 10 25Clean ulcer base 35 < 5

Endoscopic Treatments for PUD with High-risk Stigmata (active bleeding / visible

vessels)Injection:• Epinephrine (1:10000)Thermal:• Contact

thermocoagulation: MPEC, heater probes

Mechanical:• Endoclips: loops, bandsCombination: • Epi injection + thermal

or mechanicalOutomes: Reduction in rebleed, LOS, Tx, Surgery, Mortality by >/= 50%

Combination Monotherapy # Trials / N OR Rebleeding (95 % CI)

Epi + Thermal Epi alone 3 / 376 0.363 (0.177-0/727)

Epi + Clips Epi alone 4 / 362 0.334 (0.177-0.629)

Epi + Thermal Thermal alone 3 / 425 0.699 (0.405-1.206)

Epi + Clips Clips alone 3/ 234 1.045 (0.447-2.449)

Pooled Rebleeding

0.597 (0.444-0.802)

Marmo R, Am J Gastro 2007;102:279

Combination Therapy vs Monotherapy: Is It Better?

This image cannot currently be displayed.

Early EGD Reduces LOSCooper GS et al. Medical Care 1998;4:462

Cooper GS et al. Gastroinest Endos 1999;49:145

PPIs• > 25 randomized controlled trials• Considerable heterogeneity with respect

to intervention and outcomes measured• Study outcomes: PPI > placebo, EGD >

PPI, PPI + EGD > PPI, EGD + PPI > EGD + placebo– rebleeding, transfusion need– surgery

Sung, J. J.Y. et. al. Ann Intern Med 2009;150:455-464

Kaplan-Meier estimate of the cumulative percentage of patients with recurrent bleeding within 30 days

Meta-Analysis: RCTs of PPI vs Placebo After Endoscopic Therapy of Ulcers with

High-risk StigmataEndoscopic Therapy Followed by High-dose IV PPI

PPI Placebo OR (95% CI)

Rebleeding 5.2%* 12.8% 0.39 (0.18-0.87)

Surgery 5.2%* 9.4% 0.53 (0.31-0.89)

Mortality 5.7% 5.6% 0.98 (0.25-3.77)

Endoscopic Therapy Followed By Oral or Low-dose IV PPI vs Placebo

Rebleeding 10.2%* 17.2% 0.52 (0.35-0.78)

Surgery 6.6% 9.9% 0.59 (0.33-1.05)

Mortality 2.4% 2.4% 1.00 (0.42-2.35)

Leontiadis G, BMJ 2005• 21 RCTs: 2915 pts

IV PPI Infusion vs Placebo Before EGD

• Endoscopic Tx:– Peptic Ulcers: 22.5%

(PPI) vs 36.8% (placebo) (p< 0.05)

– Other bleeding sources: no difference

• Hospitalized < 3 d– 61 % PPI vs 49%

placebo (P< 0.05)• No effect of PPI on:

– Urgent endoscopy– Transfusions– Rebleeding– DeathLau J, 2007;356:1631.

• N = 638 UGIB: 377 PUD (60%)• Omeprazole 80 mg bolus, 8 mg/hr X 72 hr

Date of download: 9/3/2012

Copyright © The American College of Physicians. All rights reserved.

From: International Consensus Recommendations on the Management of Patients With Nonvariceal Upper Gastrointestinal Bleeding

Ann Intern Med. 2010;152(2):101-113. doi:10.7326/0003-4819-152-2-201001190-00009

Summary of Consensus Recommendations for the Management of Patients With Nonvariceal Upper Gastrointestinal Bleeding

Figure Legend:

Endoscopic Therapy Initial Control

Permanent Control

80-90%

Rebleeding

Endoscopic TherapyPermanent

Control75%

Rebleeding

Angiography

25%

ER Endoscopy

IV PPIOctreotide

Erythromycin

Treatment Algorithm

At discharge: 1) test and treat and test again for HP, 2) if ASA or plavix needed, resume with PPI, 3) if NSAID needed, change to cox-2 inhibitor with PPI