GI Bleeding Greg Rosenfeld St. Pauls Hospital Gastroenterology - - PowerPoint PPT Presentation

gi bleeding
SMART_READER_LITE
LIVE PREVIEW

GI Bleeding Greg Rosenfeld St. Pauls Hospital Gastroenterology - - PowerPoint PPT Presentation

GI Bleeding Greg Rosenfeld St. Pauls Hospital Gastroenterology July 14, 2015 Outline Upper GI Bleeding Presentation Differential Diagnosis Medical + Endoscopic Treatment Peptic Ulcer Disease Variceal Bleeding


slide-1
SLIDE 1

GI Bleeding

Greg Rosenfeld

  • St. Paul’s Hospital

Gastroenterology

July 14, 2015

slide-2
SLIDE 2

Outline

Upper GI Bleeding

Presentation Differential Diagnosis Medical + Endoscopic Treatment Peptic Ulcer Disease Variceal Bleeding

Lower GI Bleeding

Differential Diagnosis Investigation and Treatment

slide-3
SLIDE 3

Additional Reference

Management of Acute Bleeding from a Peptic Ulcer

Ian M. Gralnek, M.D., M.S.H.S., Alan N. Barkun, M.D., C.M., M.Sc., and Marc Bardou, M.D., Ph.D. N Engl J Med 359;9 August 28, 2008

slide-4
SLIDE 4

Upper GI Bleed

Proximal to Ligament of Treitz Approximately 4 times as common as LGIB Mortality rates from UGIB are 6-10% overall

Fallah MA et al. Med Clin North Am 2000;84:1183-208

slide-5
SLIDE 5

Ligam ent

  • f Treitz
slide-6
SLIDE 6

Epidemiology of Upper Gastrointestinal Bleeding

Incidence

170 cases /100,000/year 2 Male : 1Female

Mortality:

5-10% 0.6% if age <60 40% if rebleeding

Rockall TA Gut 1996; 38: 316-21 Laine L. Gastroenterology 2002; 123: 632 Marshall JK. Am J Gastroenterol 1999; 94: 1841 Marshall JK. J Clin Gastroenterol 1999; 29: 165

  • Kupfer. Gastroenterol Clin North Am 2000; 29: 275
slide-7
SLIDE 7

Case Summary:

Hx: 84 year old male with hematemesis and melena stool for 24 hrs

Past medical history of hypertension, pacemaker for 3rd degree AVB Admitted for pacemaker change due to infection On beta blocker, ASA Non smoker, drinks “occasionally”

PE: Hemodynamically stable

No stigmata of liver disease Rectal: Melena

slide-8
SLIDE 8

1.

Outline your initial Management of this Man?

2.

How would you localize the source of the bleed?

1.

ie Is it upper or lower?

slide-9
SLIDE 9

Initial Management: ABC’s

Airway: Intubation ?

Massive upper GI bleed Decreased mental status, unstable cardioresp

Circulation: 2 large bore IVs

? Central line access

Resuscitate and Stabilize Monitored Setting: BP/O2 sat/EKG

(vital signs + Hb)

slide-10
SLIDE 10

Hypovolemia

HR > 100 BP < 100 30:20:10 Rule

Postural vitals:

HR > 30

BP Systolic > 20 mmHg and

  • Diastolic > 10 mmHg

When do you need to call ICU?

slide-11
SLIDE 11

Resuscitation

ICU Consult:

Hemodynamic instability (shock) Decrease in hematocrit > 6%, transfusion

requirement > 2 units packed rbcs

Brisk active bleeding (hematemesis, bright red

blood per NG tube, or hematochezia)

Correct Coagulopathy:

INR > 1.5, plts < 50 FFP, cryoprecipitate, platelets Dabigatran: ? PTT. Consider Octaplex

slide-12
SLIDE 12

Secondary Management

Go back and get the details once the patient is

stabilized

What are the history features you want to know? Physical Exam?

slide-13
SLIDE 13

History

Risks for UGIB: EtOH, NSAIDS, liver disease, prior PUD, prior

bleeds, Prior HP

Previous investigations, recent CBC, Cardiorespiratory Hx, prior

endoscopy

Family Hx: Gastric cancer Duration of symptoms: CP, syncope, pre-syncope, SOB, Hx of

melena, hematochezia

Meds: anticoagulation

slide-14
SLIDE 14
  • Hematemesis - 40-50%
  • Melena - 70-80%
  • Hematochezia - 20%
  • Either hematochezia or melena: 90-98%

Syncope – 14% Presyncope – 43% Dyspepsia - 18% Epigastric pain - 40% Heartburn - 20%

  • Peter DJ et al. Emerg Med Clin North Am 1999;17:239-61

UGIB: Presenting Symptoms

slide-15
SLIDE 15

Physical Exam

Posturals, Resting HR, CardioResp exam Stigmata of CLD Stigmata of hereditary syndromes for risk of UGIB RECTAL: colour of stool, rectal masses When is FOB necessary?

slide-16
SLIDE 16

Upper GI Bleed

Melena – 50-100 mL (blood in GIT >14 hrs) Hematochezia – 1000 mL ↑ BUN / Cr ratio Stool color not reliable indicator of the location of bleeding

Hematochezia from upper GI bleeding: massive

bleeding with shock and/or orthostasis

slide-17
SLIDE 17

Management

ABC’s History and Physical What next?

Bloodwork – CBC, Xmatch, LFTs, INR, Bun, Cr, IV PPI – 80 and 8 ? Octreatide (50ug/hr)

Endoscopy?

slide-18
SLIDE 18

Timing of Endoscopy

Early Endoscopy

But what is early?

< 24 hours

Early Endoscopy and risk classification

Safe and prompt D/C of low risk pts Improves outcomes for high risk pts Reduces use of resources for high and low

risk pts.

Barkun et al., Ann Intern Med. 2010; 152: 101-113.

slide-19
SLIDE 19

Goals of Management

Resuscitation prevent shock/death Stop the bleeding Prevent recurrent bleeding (risk highest in first 24 hours) Treat underlying risk factors Avoidance of precipitants

slide-20
SLIDE 20

Outcome of UGI Bleeding:

80% stop spontaneously 20% continue to bleed or re-bleed

Bulk of resource consumption Highest morbidity/mortality

5-10% die

… must identify early those at increased risk of rebleeding, morbidity, mortality to in order to optimize management and improve outcomes.

slide-21
SLIDE 21

Risk Stratification

Clinical parameters (Blatchford)

Blood urea nitrogen Hemoglobin Systolic blood pressure Pulse Presence of melena Syncope Hepatic disease, and cardiac failure

Score 0 to 14: risk of requiring endoscopic

intervention increases with higher score

Blatchford et al. Lancet. 2000 Oct 14;356(9238):1318-21

slide-22
SLIDE 22

Blatchford Scoring System

No endoscopic findings Identifies patients at low

and high risk of needing an intervention

Transfusion or Surgery or Endoscopic therapy

Blatchford O. Lancet 2000; 356: 1318-21

slide-23
SLIDE 23

Variable`` 1 2 3 Age (yr) <60 60–79 >80 Shock No shock (HR <100; SBP >100) Tachycardia (HR >100; SBP >100) Hypotension (SBP <100) Co-morbidity Nil major Cardiac failure (e.g., IHD) Renal failure, liver failure, disseminated malignancy Diagnosis (Endoscopy) M-W tear, no lesion, no SRH* All other diagnosis including ulcer etc. Malignancy of upper GI tract Major Stigmata (Endoscopy) None or dark spot Blood in upper GI tract, adherent clot, visible or spurting vessel

Rockall Scoring System for Severity of Acute UGIB

Score <2 ( low risk ) excellent prognosis (1 in 744 patients) Score >8 high risk of death

Rockall TA. Gut 1996; 38: 316-21 Vreeburg EM. Gut 1999: 44: 331-5

slide-24
SLIDE 24

Rockall Score: Risk of Rebleed and Death

%

Rockall TA. Gut 1996;38:316-21

slide-25
SLIDE 25

Rockall Score

Clinical:

Age Pulse > 100 bpm BP < 100 Comorbidity: CHF, CAD, renal/liver disease,

disseminated malignancy

Endoscopic Diagnosis:

High risk Stigmata, blood in upper GI tract

Rockall et al. Lancet 1996; 347:1138-40

slide-26
SLIDE 26

IV PPI initiated in ER Bloodwork/Crossmatch:

Hb 110, BUN 17.1, Cr 99

Endoscopy:

4 clean based ulcers PLUS 2cm ulcer 2nd part of duodenum Adherent clot washed away No active bleeding

Case Sum m ary:

slide-27
SLIDE 27

Forrest JA. Lancet 1974; 2: 394-7 Laine L, Peterson WL. NEJM 1994; 331: 717-27

2 0 4 0 6 0 8 0

Clean base Flat spot Adhere nt clot Active bleeding % w ith Rebleeding Non- bleeding visible vessel

5 1 0 2 2 4 3 5 5

Low -risk lesions High-risk lesions

Rate of Rebleeding by Forrest Classification

III IIc IIb IIa Ia/ b

slide-28
SLIDE 28

Causes of UGI Bleeding

Peptic ulcer disease — 55% Esophagogastric varices — 14% AV malformations — 6% Mallory-Weiss tears — 5% Tumors and erosions — 4% each Dieulafoy's lesion — 1%

  • Jutabha R et al. Med Clin North Am. 1996;80(5):1035
slide-29
SLIDE 29

Source of UGI Bleeding

CURE

14% Variceal 86% Non-variceal

55% Peptic ulcer 6% Angiodysplasia 5% MW tear 4% Neoplasm 4% Erosions 1% Dieulafoy

RUGBE

100% Non-variceal

56% Peptic ulcer

47% GU 42% DU

10% Erosions 9% Esophagitis 25% Other

MW tear, angiodysplasia,

Dieulafoy, neoplasm

Savides TJ. Endoscopy 1996; 28: 244-8 Barkun A. Am J Gastroenterol 2004; 99: 1238-9

slide-30
SLIDE 30

Summary

Most patients will have visible signs of blood loss Stool color not a reliable indicator of location of bleeding Most common cause of UGI bleed is Peptic Ulcer disease:

incidence may be declining

ABC’s Stabilize before Endoscopy Medical Management: Pantoloc or Octreotide

slide-31
SLIDE 31

Summary

Risk Stratification: Blatchford and Rockall Forrest Classification: Endoscopic Stigmata Evidence for IV Proton Pump Inhibitors After Ulcer with high risk stigmata x 72 hrs Lau et al. NEJM 2000 Before Endoscopy to downstage Ulcer Lau et al. NEJM 2007

slide-32
SLIDE 32

High Risk Patients

Shock – hematochezia Cause of bleed: Varices, UGI cancer Older age, Co-morbid diseases Onset in hospital

Severe coagulopathy, Recurrent bleeding

slide-33
SLIDE 33

Case Summary

slide-34
SLIDE 34

Case Summary

slide-35
SLIDE 35

Had Enough Yet?

slide-36
SLIDE 36

Variceal Hemorrhage

slide-37
SLIDE 37

Natural History

10 – 30% of all UGIB’s. Occurs in 40% of cirrhotics & in as many as 60% of

pts with cirrhosis & ascites.

Varices – Portosystemic collaterals, usually in distal

esophagus.

slide-38
SLIDE 38

The risk of death with acute variceal bleeding is 5% to 8% at

  • ne week and about 20% at six weeks.

Patients who rebleed early, have a MELD score >18, require

>4 units of packed red blood cell transfusions, and in whom renal failure develops have the highest risk of death.

slide-39
SLIDE 39

Pathogenesis

Varices are portosystemic collaterals formed

after pre-existing vascular channels have been dilated by portal hypertension.

Most common in the distal 2-5 cm of the

esophagus, as it contains superficial veins that lack support from surrounding tissues, a feature consistent with the occurrence of prominent bleeding at this site.

slide-40
SLIDE 40

Control of acute bleeding

Acute esophageal variceal bleeding constitutes a life-

threatening emergency.

Treatment is aimed at resuscitating the patient, correcting

coagulopathy, administration of Abx, controlling the bleeding, and preventing complications.

slide-41
SLIDE 41

A combination of endoscopic therapy and

pharmacologic therapy of variceal bleeding may be superior to pharmacologic treatment alone & able to control bleeding in up to 90% of cases at intial treatment.

slide-42
SLIDE 42

Pharmacotherapy

Start ASAP and continue for up to 5 days Choice of agents with evidence:

Somatostatin Octreotide Terlipressin Vasopressin plus nitroglycerin

  • Causes splanchnic vasoconstriction, and lower portal pressure

Antibiotics have been shown to reduce mortality and

bacterial infections in patients with variceal bleeds

slide-43
SLIDE 43

Variceal Banding

slide-44
SLIDE 44

Varix banding 6 mon

slide-45
SLIDE 45
slide-46
SLIDE 46
slide-47
SLIDE 47
slide-48
SLIDE 48
slide-49
SLIDE 49

Bleeding cannot be controlled in approx 10% of patients, defined

by any of the following three factors:

(1) transfusion of four units of red blood cells or more

to maintain the hematocrit value between 25% and 30%;

(2) inability to increase the systolic blood pressure by

20 mm Hg or to greater than 70 mm Hg; or

(3) persistence of a heart rate greater than 100 beats

per minute.

slide-50
SLIDE 50

Uncontrolled bleeding = 2 sessions of

endoscopic treatment in 24 hours

Balloon tamponade (Blakemore) Salvage therapy

TIPPS or Surgical Shunt

slide-51
SLIDE 51

Sengstaken-Blakemore tube

slide-52
SLIDE 52

TIPSS

slide-53
SLIDE 53

Surgical shunt

slide-54
SLIDE 54

Case 2:

Hx: 80 year old male on a cruise to Alaska with sudden onset hematochezia followed by melena stool for 24 hrs

Past medical history of hypertension, MI, CAD and CKD On beta blocker, ASA, Plavix Non smoker, drinks “occasionally”

PE: Hemodynamically stable

No stigmata of liver disease Rectal: Maroon Colour stool

slide-55
SLIDE 55

Case 2

Transfused 2 U PRBC on the cruise Hgb 8857 given 4 U PRBC in ER but still has

bloody stool

Hgb now 78 How would you manage this man?

slide-56
SLIDE 56

Management

ABC’s Transfusion (evidence of ongoing brisk bleeding) EGD – no blood nor pathology to suggest upper

source

Colonoscopy

Maroon blood but unable to get beyond Splenic Flexure Repeat after prep – no source found

slide-57
SLIDE 57

Lower GI Bleeding

Distal to ligament of Treitz 11% of patients with hematochezia: from an

upper GI source

Clues hemodynamic instability, elevated

BUN, NGT aspirate bloody Melena – can be seen with proximal lower

GI bleeding (Right colon, distal small bowel)

slide-58
SLIDE 58

Intermittent – up to 85% stop spontaneously LGIB frequency: increases 200-fold from 3rd to 9th decades of

life

Massive bleed usually age >60:

Mortality 30% 0.7% admissions to hospital

Lower GI Bleeding

slide-59
SLIDE 59

Etiology of LGIB

Diverticulosis — 33%

slide-60
SLIDE 60

Diverticula (with injection)

slide-61
SLIDE 61

Etiology of LGIB

Diverticulosis — 33% Cancers/polyps — 19%

slide-62
SLIDE 62

Adenocarcinoma

slide-63
SLIDE 63

Etiology of LGIB

Diverticulosis — 33% Cancers/polyps — 19% Colitis/ulcers (IBD, infectious, ischemic, radiation) — 18%

slide-64
SLIDE 64

UC

Ischemic colitis

UC

slide-65
SLIDE 65

Etiology of LGIB

Diverticulosis — 33% Cancers/polyps — 19% Colitis/ulcers (IBD, infectious, ischemic, radiation) — 18% Angiodysplasia — 8%

slide-66
SLIDE 66

Angiodysplasia

slide-67
SLIDE 67

Radiation Proctitis

Pre-treatment Post-treatment with APC

slide-68
SLIDE 68

Etiology of LGIB

Diverticulosis — 33% Cancers/polyps — 19% Colitis/ulcers (IBD, infectious, ischemic, radiation) — 18% Angiodysplasia — 8% Miscellaneous (post-polypectomy, aorto-colonic fistula, stercoral

ulcer, anastomotic bleeding) — 8%

slide-69
SLIDE 69

Clues From The History

Painless

Diverticula Angiodysplasia Neoplasms

Painful

Ischemia Inflammatory Bowel Disease Infectious (E. coli, Campylobacter, Shigella etc) Hemorrhoids/Fissures

slide-70
SLIDE 70

Massive LGIB

Age < 60:

Diverticula Neoplasms IBD

Age > 60:

Diverticula Neoplasms Angiodysplasia

slide-71
SLIDE 71

Colonoscopy – Difficult to localize during active bleeding CT angiography

Detects rate of ≥ 1.0 mL/min Variable sensitivity with high specificity

RBC scan

Sensitive (80-90%) Detects ≥ 0.1 to 0.5 mL/min Relatively low specificity Poor localization

Diagnosis of LGIB

slide-72
SLIDE 72

Arterial Phase CT Scanning

Overall, multi-detector row CT

scanning has a location-based sensitivity of 90.9%, specificity of 99%, accuracy of 97.6%, PPV of 95% and NPV of 98% for detection of GI bleeding

  • Radiology. 2006 Apr;239(1):160-7
slide-73
SLIDE 73

Algorithm

Resuscitate and stabilize Colonoscopy +/- EGD Angiography (+/- CT) – If unable to

localize @ colonoscopy and brisk bleeding

RBC scan – If unable to localize @ colonoscopy Surgery – Last resort if ongoing hemorrhage with no answers

from the above investigations

slide-74
SLIDE 74

Questions?