approach to upper gi bleedings3.gi.org/wp-content/uploads/2014/09/14acg_regional_williamsburg... ·...

Bryan Sauer, MD, MSc

Approach to Upper GI BleedingApproach to Upper GI Bleeding

Bryan G. Sauer, MD, MSc (Clin Res)

Assistant Professor of MedicineAssistant Professor of Medicine

Co-Medical Director of Endoscopy

University of Virginia

Division of Gastroenterology and Hepatology

Natural HistoryNatural History

■ In 1970, during the BC (before cimetidine) era, 79% of bleeding stopped without intervention

■ Our job as gastroenterologists:St i i bl di

Schiller et al. BMJ 1970

■ Stop any remaining bleeding

■Reduce risk of rebleeding

ACG Regional Postgraduate Course - Williamsburg, VA Copyright 2014 American College of Gastroenterology

1


BackgroundBackground

■ GI bleeding is a significant medical problem 300,000 hospitalizations annually

Incidence increases with age

Rebleeding rates range from 7-16%despite endoscopic treatmentdespite endoscopic treatment Variceal rebleeding (25-29%)

PUD rebleeding (20-22%)

Mortality: 10-14%

ASGE Guideline: The role of endoscopy in the management of acute non-variceal upper GI bleeding 2012; van Leerdam ME, Best Practice & Res Clin Gastro 2008; Barkun AN et al. Ann Intern Med 2010.

Sources of UGIBSources of UGIB

■ Non-variceal UGIB ■Ulcer (33-56%)

■ Erosions (19%)

■Mallory-Weiss tear (4%)

■ Vascular lesions (3%)

■ Tumor (1%)

• Hemosuccus pancreaticus• Hemobilia• Iatrogenic (post-sphincterotomy, etc)• Dieulafoy's lesion• Fistula (aortoenteric, etc)• Polyps

■ Tumor (1%)

■ Portal-hypertension■ Esophageal varices

Enestvedt et al. Nonvariceal upper-GI hemorrhage. GIE 2008 (CORI), Barkun A et al. RUGBE, Am J Gastro 2004

• Gastric or duodenal varices• Gastric antral vascular ectasia (GAVE)


2


Scoring Systems for Risk Scoring Systems for Risk AssessmentAssessment

■ Blatchford Rockall■ Blatchford, Rockall

■ In general, high risk includes■Age >65

■ Shock

■ Poor overall health status, comorbidities,

■ Low hemoglobin

■ Fresh blood, elevated urea

Transfusion in UGIBTransfusion in UGIBRCT 921 patients

Restrictive (Hgb <7g/dl)Restrictive (Hgb <7g/dl)

vs

Liberal (Hgb <9g/dl)

Restrictive Group:

- improved survival

l bl di t

Villanueva C et al, NEJM 2013

- lower rebleeding rate (10% v 16%)

- fewer adverse events

- shorter hospital stay


3


Before EndoscopyBefore EndoscopyIV PPI Prokinetics Gastric

lavage*Early

endoscopy**

OUTCOMES CochraneReview

Meta-analysis

RCT RCT, retrospective

Mortality X (retrospective)

Rebleeding X (retrospective)

Progression to Surgery X (retrospective)

Rate of high risk stigmata XX

Sreedharan A et al. Cochrane 2010, Barkun AN et al. GIE 2010, Lee SD, et al. J Clin Gastro 2004, Spiegel et al. Arch Intern Med 2001, Anantharkrishnan AN, et al. CGH 2009

* 40 Fr tube with up to 15L of lavage ** for high risk patients, <24 hours

Repeat EGD/Visualization XX X (fundus)

Length of stay/charges XX

Transfusion requirements X


lavage*Early

endoscopy**


Meta-analysis












4



lavage*Early

endoscopy**


Meta-analysis











ACG Practice Guidelines

“Patients with UGIB should generally undergo endoscopy within

Timing of EndoscopyTiming of Endoscopy

g y g py24 hours of admission, following resuscitative efforts to optimize hemodynamic parameters and other medical problems”

“In patient with higher risk clinical features (e.g., tachycardia, hypotension, bloody emesis or NG aspirate) endoscopy within 12h may be considered to potentially improve clinical outcomes”

2012 ACG Practice Guidelines; Barkun AN et al. Ann Intern Med 2010

International Consensus Recommendations

“In patients receiving anticoagulants, correction of coagulopathy is recommended but should not delay endoscopy”


5


Endoscopic Stigmata of Ulcers & Endoscopic Stigmata of Ulcers & Rebleed RiskRebleed Risk

Stigmata Forrest Prevalence(%) Rebleed(%)*g ( ) ( )

Active bleed 1a, 1b 10-20 90

Visible vessel 2a 15-25 50

Adherent clot 2b 10-20 25

Laine L, Peterson WL. Bleeding Peptic Ulcer. NEJM 1994. Freeman ML Gastrointest Endosc Clin North Am 1997

* Without treatment

Flat spot 2c 10-20 10

Clean base 3 35 5

Results of Endoscopic TherapyResults of Endoscopic TherapyWhere it all began…Where it all began…

Sham (n=23) MPEC (n=21)* p-value( ) ( ) p

Hemostasis (%) 3 (13) 19 (90) <0.0001

Blood Transfusions 5.4 + 0.9 2.4 + 0.9 0.002

Emergency Intervention (%) 13 (57) 3 (14) 0.005

Hospital Stay (days) 7 2 + 1 1 4 4 + 0 8 0 02

Laine L. N Engl J Med 1987;316:1613

For actively bleeding lesions, * MPEC=multipolar electrocoagulation

Hospital Stay (days) 7.2 + 1.1 4.4 + 0.8 0.02

Hospital Cost ($) 7,550 + 1,480 3,420 + 750 0.001

Deaths (%) 3(13) 0 NS


6


Endoscopic TherapyEndoscopic Therapy

■ Endoscopic therapy reduces■Bleeding (active or recurrent)

■Need for surgery

■Mortality

■ Results driven by high risk stigmata:■Active bleeding (NNT 2)*

The Big Three

Sacks et al. JAMA 1990, Cook et al. Gastroenterology 1992, Laine and McQuaid, CGH 2009

■Active bleeding (NNT 2)

■ Visible vessels (NNT 5)*

■Adherent clot & flat spot—not reduced*

* recurrent bleeding

Endoscopic Stigmata of BleedingEndoscopic Stigmata of BleedingAdherent ClotAdherent Clot

Stigmata Forrest Prevalence(%) Rebleed(%)g ( ) ( )

Active bleed 1a, 1b 10-20 90

Visible vessel 2a 15-25 50

Adherent clot 2b 10-20 25

Laine L, Peterson WL. Bleeding Peptic Ulcer. NEJM Sept 15 1994. Volume 331:717-727 Freeman ML Gastrointest Endosc Clin North Am 1997; 1:229.

Flat spot 2c 10-20 10

Clean base 3 35 5


7


After clot removalAfter clot removal

59% would benefit from therapy

Bini EJ et al. GIE 2003

Adherent clot studiesAdherent clot studies■ Four RCTs published 2002-2003:

■ Endo Rx Better = 2 (Jensen (n=32) Bleu (n=56))■ Endo Rx Better 2 (Jensen (n 32), Bleu (n 56))

■ No difference = 2 (Sung (n=39), Jung (n=19))

Retrospective Study 2003

Meta-analysis 2005

Meta-analysis 2009

n 244 240, 6 studies 189, 5 studies

Location NYC US, Spain, Asia US, UK, Asia

Bini EJ, Cohn J. GIE 2003, Kahi et al. Gastroenterology 2005, Laine L, McQuaid KR, Clin Gastro Hep 2009

RCT: Jensen DM et al. Gastro 2002, Bleau BL et al. GIE 2002, Jung HK et al. Am J Gastro 2002, Sung JJY, et al. Ann Int Med 2003

, p , , ,

Endo vs. Medical Rx 138 versus 106 112 versus 128 71 versus 118

Rebleeding 0.07 (0.02-0.22) 0.39 (0.22-0.69) 0.31 (0.06-1.77)

FAVORS Endo Rx Endo Rx No Difference


8


Adherent Clot SummaryAdherent Clot Summary

■ Controversy exists

■ Only one study used continuous infusion PPI as control■ Showed no difference in rebleeding rate

■ Current Guidelines: endoscopic therapy ma be considered altho gh intensi emay be considered, although intensive PPI therapy alone may be sufficient

Sung JJY, et al. Ann Int Med 2003, Barkun AN et al. Annals of Int Medicine 2010

Endoscopic therapeutic choicesEndoscopic therapeutic choices

■ Injection( )■ Epinephrine (1:10,000) or saline

■ Sclerosant

■ Thrombin/Fibrin Glue

■ Thermal■ Bipolar electrocoagulation (heat + pressure)

■ Heater probe (heat + pressure)

■ APC (heat only)

■ Mechanical ■ Clip (theoretical advantage of no tissue injury)


9


Injection therapy Injection therapy

■ Epinephrine: 1:10,000 – 1:100,000

■ Less effective than:■Other monotherapies (NNT 9)

■When combined with 2nd therapy (NNT 5)

■ TWO IS BETTER THAN ONET d liti i + th l/ h i l■ Two modalities—epi + thermal/mechanical

■ Two procedures—2nd look endoscopy if used as monotherapy

Laine L, McQuaid KR. CGH 2009, Park WG, et al. Technological Review, GIE 2007

Thermal TherapyThermal Therapy

■ Bipolar electrocoagulation ■Coaptive coagulation: compress vessel

(pressure), then coagulate (heat) to seal

■ Low wattage (15-20W) for 5-10 seconds

■ Heater Probe

■ Argon Plasma Coagulation■ Argon Plasma Coagulation■ Less well-studied

■No difference in RCT for high risk stigmata when compared to epi + heater probe

Chau C et al. GIE 2003


10


Thermal TherapyThermal Therapy

■ When compared to no therapy, reduced:■Bleeding (NNT 4)

■ Surgery (NNT 8)

■Mortality (NNT 33)

■ Can be used as monotherapy

Laine L, McQuaid KR. CGH 2009, Chau CH et al. GIE 2003

Endoscopic HemoclipsEndoscopic Hemoclips

■ Initial hemostasis lower than other d i t t tendoscopic treatments:

■RR 0.78 (0.64 – 0.95)

■ When clips do not work well■Challenging locations

■Lesser curvature/posterior wall of stomach

Lin H et al, Am J Gastro 2002, Lin H et al. Dig and Liver Disease 2003, Saltzman JR et al. Am J Gastro 2005

■Lesser curvature/posterior wall of stomach

■Posterior duodenum

■Retroflexed view

■ Fibrotic lesions


11



■ No difference in outcomes compared to t d d d i th i (th l)standard endoscopic therapies (thermal):■Rebleeding, surgery, mortality

■Better than epinephrine monotherapy

■ SUMMARY: When able to be placed, clips appear as successful as thermalclips appear as successful as thermal therapy

Daram SR et al. Surg Endosc 2013


QuickClip2

(Olympus)

Resolution

(BSCI)

Instinct

(Cook)

Jaw span 11 mm 11 mm 16 mmJaw span 11 mm 11 mm 16 mm

Rotation Yes No Yes

Re-opening ability No Yes Yes

MRI conditional No Yes Yes


12


Ulcer with active oozingUlcer with active oozing


13


Ulcer with visible vesselUlcer with visible vessel


14



15



16


Mallory Weiss TearMallory Weiss Tear


17


Mallory Weiss TearMallory Weiss Tear

Vascular EctasiaVascular Ectasia


18


Vascular EctasiaVascular Ectasia

Portal Hypertensive BleedingPortal Hypertensive Bleeding

■ Esophageal varices

■ Gastric varices

■ Duodenal varices

■ Gastric Antral Vascular Ectasia (GAVE)


19


High Risk Esophageal VaricesHigh Risk Esophageal Varices

Esophageal Varices with Band Esophageal Varices with Band LigationLigation


20


Gastric Antral Vascular EctasiaGastric Antral Vascular Ectasia

Gastric VaricesGastric Varices■ Conventional approach

(sclerosis banding)(sclerosis, banding)

■ TIPS (Transjugular Intrahepatic Portosystemic Shunt)

■ Endoscopic glue injection: CyanoacrylateCyanoacrylate

■ BRTO (angiographic retrograde occlusion)

■ Surgical shunt: rarely


21


BRTO (BalloonBRTO (Balloon--Occluded Retrograde Occluded Retrograde Transvenous Obliteration)Transvenous Obliteration)


22


Complications of therapyComplications of therapy

■ Endoscopic therapy (8 of 1044, 0.8%) versus no endoscopy (1 of 931 0 1%): RR 2 12 (0 79 5 70)endoscopy (1 of 931, 0.1%): RR 2.12 (0.79-5.70)

Modality n Induced Bleeding Perforations Rate of Cpx

Epinephrine 958 2 0 0.2%

Sclerosant ± epi 1339 1 6 0.5%

HP ± epi 1070 2 9 1.0%

Laine L, McQuaid KR. CGH 2009

HP ± epi 1070 2 9 1.0%

BPEC ± epi 580 1 2 0.5%

Clips ± epi 373 0 0 0.0%

Cpx = complications, epi = epinephrine, HP = heater probe, BPEC = bipolar electrocoagulation

New Endoscopic TherapiesNew Endoscopic Therapies(aka non(aka non--standard therapies)standard therapies)

■ Hemospray/EndoClot■Disclaimer: Not FDA approved in the

United States

■ Over the Scope closure devices■ Over-the-Scope closure devices


23


HemosprayHemospray

•• Mechanical Mechanical tamponadetamponadeppeffecteffect

•• Absorbs waterAbsorbs water

•• Activates clotting Activates clotting cascadecascade

Gastrointestinal Endoscopy 2013 77, 692-700

HemosprayHemospray

Sung JJ et al. Endoscopy 2011


24


HemosprayHemospray

Case series of 16 patientsSulz MC et al. Endoscopy 2014

Reports using Reports using HemosprayHemospray//EndoclotEndoclot

■ Peptic ulcer bleeding, primary tx■ 20 adults, 95% hemostasis, 2 rebleed

■ Malignant bleeding■ 5 patients, 100% hemostasis, one rebleed

■ Variceal bleeding9 ti t 100% h t i bl d

Sung et al. Endoscopy 2011; Chen et al. GIE 2012; Ibrahim et al. GIE 2013, Huang R et al. Dig Endoscopy 2014

■ 9 patients, 100% hemostasis, zero rebleed

■ Post EMR■ 20 lesions, 90% hemostasis, 3 rebleed


25


SEAL SurveySEAL Survey

■ 10 pilot sites across Europe in 20112011

■ 63 patients with UGIB■ 30 ulcers, 33 “other” pathology

■ 55 (87%) treated as monotherapy

■ Primary hemostasis: 85% (47/55)

Smith LA, et al. J Clin Gastroenterol, epub ahead of print, Barkun et al GIE 2013

■ Rebleeding rate at 7d: 15%

■ Second-line therapy in 8 patients, all with hemostasis

Over the Scope ClipOver the Scope Clip

■ Retrospective study, 30 patientspatients

■ Conventional Rx failures

■ Hemostasis—97%

Rebleed— 6%

■ Reports include use in ulcers,

Manta et al. Surg Endosc 2013; Chan et al. Endoscopy 2014; Alcaide N, et al. Rev Esp Enferm Dig 2014

pMW tear, dieulafoy, GIST, anastomosis , EMR/ESD, diverticular, post polypectomy


26


PostPost--endoscopic Therapyendoscopic Therapy

PPI after endoscopic therapy

No. of patients

Lau et al. N Engl J Med 2000

■30-day rebleed rate■ 6.7 % for IV omeprazole■ 26.5 % for placebo


27


What Dose of PPI?What Dose of PPI?

■ RCT 201 patients after endoscopic Rx with epi + thermocoagulation (Forrest 1a/1b/IIa)epi + thermocoagulation (Forrest 1a/1b/IIa)

■ Compared continuous infusion versus bolus

High-dose

Stnd-dose

P-value

Recurrent bleed (72h)

5 6 0.77

Chen C et al. Alim Pharm & Ther, 2012

(72h)

Recurrent bleed (30d)

7 7 0.98

Surgery 0 0 1

Death (bleeding-related)

1 1 0.99

High dose versus nonHigh dose versus non--high dosehigh dose

■ Meta-analysis of 1157 patients in 7 studies

Hi h d ti i f i■ High dose = continuous infusion

■ NO difference:

■ Rates of rebleeding (OR 1.30, CI 0.88-1.91)

■ Surgery (OR1.49, CI 0.66-3.37)

■ Mortality (OR 0.89, CI 0.37-2.13)

Wang, C et al. Arch Int Med 2010


28


“We found that the best dose and route of

“Our results show that, with regards to deaths, rebleeding episodes, emergency surgeries and

administration of PPIs cannot yet be determined.”

g p , g y gneed for repeat endoscopic treatments, it is not certain if high intravenous dose of PPIs are more, less or equally effective compared to lower (oral or intravenous) dose of PPIs.”

Neumann I et al. Cochrane Review 2013

PPI after UGIB TreatmentPPI after UGIB Treatment

■ PPI decrease rebleed rates

■ Current recommendation is for IV PPI bolus (80mg) followed by continuous infusion 8mg/hr for 72 hours

■ Further evidence may support non-continuous infusionscontinuous infusions

Laine L, Jensen DM. Am J Gastroenterol 2012, Barkun AN et al Ann Int Med 2010.


29


When to restart aspirin?When to restart aspirin?ACG Practice Guideline: “Early resumption of antiplatelet therapy within 1-3 days y p p py yafter hemostasis, and certainly within 7 days, will be appropriate in most patients with established CV disease”

Rebleeding Mortality

Laine L, Jensen DM. Am J Gastroenterol 2012, Sung et al. Ann Int Med 2010.

What about the “highest” risk What about the “highest” risk ulcer bleed?ulcer bleed?

■ RCT (n=105) comparing ( ) p gtranscatheter arterial embolization AFTER endoscopic hemostasis in high risk ulcers (Forrest Ia-IIb)

STAESTAE ControlControl P ValueP Value

Laursen SB et al. Scand J Gastroenterol 2014

Mean transfusion 4.3 units4.3 units 4.9 units4.9 units NSNS

Rebleeding 4%4% 14%14% 0.10

Surgery 2% 0% NS

Mortality (30-d) 4%4% 14% 0.100.10


30


How to Prevent Recurrent Ulcer How to Prevent Recurrent Ulcer Bleeding?Bleeding?

Laine L & Jensen DM. Management of Patients with Ulcer Bleeding, 2012 ACG Practice Guidelines

Summary:Summary:Approach to Upper GI BleedingApproach to Upper GI Bleeding

PPI therapy should be initiated upon PPI therapy should be initiated upon presentation for upper GI bleeding

Early endoscopy (<24 hours) should be performed in most patients

Endoscopic therapy should be yperformed for actively bleeding lesions/visible vessels and considered in adherent clots


31


New therapies include hemospray

Summary:Summary:Approach to Upper GI BleedingApproach to Upper GI Bleeding

New therapies include hemosprayand over-the-scope clip

PPI after endoscopy improves outcomes


32

approach to upper gi bleedings3.gi.org/wp-content/uploads/2014/09/14acg_regional_williamsburg... ·...

Documents