1. Varices Management: Current State of the Art Atif Zaman, MD MPH Associate Professor of Medicine Director of Clinical Hepatology Oregon Health & Science University Atif Zaman, MD MPH Associate Professor of Medicine Director of Clinical Hepatology Oregon Health & Science University

2. Portal pressure Resistance to portal flow Cirrhosis Splanchnic arteriolar resistance Portal blood inflow Varices Variceal Growth VARICES AND VARICEAL HEMORRHAGE

3. Small varices Large varices No varices 7-8%/year Varices Increase in Diameter Progressively Merli et al. J Hepatol 2003;38:266 VARICES INCREASE IN DIAMETER PROGRESSIVELY

4. Prevalence of Esophageal Varices in Cirrhosis % % 100 60 40 20 0 0 Overall Child A Child B 80 Child C Pagliaro et al., In: Portal Hypertension: Pathophysiology and Management, 1994: 72 PREVALENCE OF ESOPHAGEAL VARICES IN CIRRHOSIS

5. Prevalence and Size of Esophageal Varices in Patients with Newly-Diagnosed Cirrhosis % Patients with varices % Patients with varices 100 60 40 20 0 0 Overall n=494 Overall n=494 Child A n=346 Child A n=346 Child B n=114 Child B n=114 80 Child C n=34 Child C n=34 Large Medium Small Pagliaro et al., In: Portal Hypertension: Pathophysiology and Management, 1994: 72 PREVALENCE AND SIZE OF ESOPHAGEAL VARICES IN PATIENTS WITH NEWLY DIAGNOSED CIRRHOSIS

6. Predictors of hemorrhage:  Variceal size  Red signs  Child B/C Predictors of hemorrhage:  Variceal size  Red signs  Child B/C NIEC. N Engl J Med 1988; 319:983 Variceal hemorrhage Varix with red signs PROGNOSTIC INDICATORS OF FIRST VARICEAL HEMORRHAGE

7. Large Varices Are More Likely To Rupture % Patients without bleeding % Patients without bleeding 100 50 25 0 0 0 0 12 24 75 36 12 24 36 Large Varices * * p<0.01 * 2-year probability of first bleed:  Small varices: 7%  Large varices: 30% 2-year probability of first bleed:  Small varices: 7%  Large varices: 30% Time (months) No Varices Small Varices * Merli et al., Hepatol 2003; 38:266, ** Conn et al., Hepatology 1991; 13:902 LARGE VARICES ARE MORE LIKELY TO RUPTURE

8. Screening for Varices  Current recommendations for screening  AASLD:  All patients with mod-severe cirrhosis (Child B/C)  Child A with signs of portal hypertension (plts 13mm, or evidence of collaterals)  ACG: All patients with cirrhosis upon diagnosis of cirrhosis  A number of studies have attempted to determine risk factors for presence of large esophageal varices (LEV)  Current recommendations for screening  AASLD:  All patients with mod-severe cirrhosis (Child B/C)  Child A with signs of portal hypertension (plts 13mm, or evidence of collaterals)  ACG: All patients with cirrhosis upon diagnosis of cirrhosis  A number of studies have attempted to determine risk factors for presence of large esophageal varices (LEV)

9. Prevalence and Predictors of LEV in Patients with Cirrhosis AuthorNo. patientsChild A/B/C (%)LEV (%)Predictors Pagliaro49670/23/79Low plt count Lavergne5249/35/1619Ascites Barcia9534/49/1736 Ascites Advanced CPC Chalasani34622/48/3020 Plts <88,000 Splenomegaly Madhotra18443/34/2325 Splenomegaly, Plts <68,000 Zaman30011/71/1831 Plts <80,000 Advanced CPC

10. Capsule Endoscopy for Screening for Varcies Grade 3 Varices Grade 1 Varices

11. Capsule Endoscopy vs. EGD for Variceal Screening  If EGD is the gold standard for variceal detection:  Sensitivity of CE = 100% (84%)  Specificity of CE = 89% (88%)  PPV = 96% (92%)  NPV = 100% (77%)  If EGD is the gold standard for variceal detection:  Sensitivity of CE = 100% (84%)  Specificity of CE = 89% (88%)  PPV = 96% (92%)  NPV = 100% (77%) Eisen et al, Endoscopy 2006 38:31-35 DeFranchis et al, Hepatology 2008;47:1595-1603

12. Treatment of Varices / Variceal Hemorrhage No varices Varices No hemorrhage Varices No hemorrhage Variceal hemorrhage Variceal hemorrhage Recurrent hemorrhage Recurrent hemorrhage Prevention of variceal development PREVENTION OF VARICEAL DEVELOPMENT

13. Pre-Primary Prophylaxis  Multicenter, randomized, placebo-controlled trial of timolol (non-selective beta-blocker) vs. placebo in patients  Beta-blockers did not prevent the development of varices and were associated with a higher rate of serious adverse events  Hepatic venous pressure gradient was the strongest predictor of the development of varices  Multicenter, randomized, placebo-controlled trial of timolol (non-selective beta-blocker) vs. placebo in patients  Beta-blockers did not prevent the development of varices and were associated with a higher rate of serious adverse events  Hepatic venous pressure gradient was the strongest predictor of the development of varices Groszmann, et al., NEJM 2006 NON-SELECTIVE BETA BLOCKERS DO NOT PREVENT DEVELOPMENT OF VARICES

14. Treatment of Varices / Variceal Hemorrhage No varices Varices No hemorrhage Varices No hemorrhage Variceal hemorrhage Variceal hemorrhage Recurrent hemorrhage Recurrent hemorrhage No specific therapy Repeat endoscopy in 2-3 yrs* No specific therapy Repeat endoscopy in 2-3 yrs* * Sooner with cirrhosis decompensation MANAGEMENT OF PATIENTS WITHOUT VARICES

15. Treatment of Varices / Variceal Hemorrhage No varices Varices No hemorrhage Varices No hemorrhage Variceal hemorrhage Variceal hemorrhage Recurrent hemorrhage Recurrent hemorrhage Prevention of first variceal hemorrhage PREVENTION OF FIRST VARICEAL HEMORRHAGE

16. Treated better Treated worse Porto-caval shunt  -blockers Sclerotherapy 0.1 1 1 10 * * * * Prevention of First Variceal Hemorrhage * Significantly heterogeneous D’Amico et al., Hepatology 1995; 22;332 Bleeding Death Encephalopathy Relative risk PREVENTION OF FIRST VARICEAL HEMORRHAGE

17. Non-Selective Beta-Blockers Prevent First Variceal Hemorrhage Bleeding rateControlBeta-blockerAbsolute rate (~2 year) difference All varices25%15%-10% (11 trials)(n=600)(n=590)(-16 to -5) Large varices30% 14% -16% (8 trials)(n=411)(n=400) (-24 to -8) Small varices7%2%-5% (3 trials)(n=100)(n=91)(-11 to 2) Bleeding rateControlBeta-blockerAbsolute rate (~2 year) difference All varices25%15%-10% (11 trials)(n=600)(n=590)(-16 to -5) Large varices30% 14% -16% (8 trials)(n=411)(n=400) (-24 to -8) Small varices7%2%-5% (3 trials)(n=100)(n=91)(-11 to 2) D’Amico et al., Sem Liv Dis 1999; 19:475 NON-SELECTIVE BETA-BLOCKERS PREVENT FIRST VARICEAL HEMORRHAGE

18. Variceal Band Ligation (VBL) vs. Beta-Blockers (BB) in the Prevention of First Variceal Bleed Khuroo, et al., Aliment Pharmacol Ther 2005; 21:347 First hemorrhage Survival Chen 1998 Sarin 1999 De 1999 Jutabha2000 De la Mora 2000 Lui 2002 Lo 2004 Schepke 2004 Total Chen 1998 Sarin 1999 De 1999 Jutabha2000 De la Mora 2000 Lui 2002 Lo 2004 Schepke 2004 Total Relative risk 0 0 1 1 10 0 0 1 1 40 Favors VBL Favors BB VARICEAL BAND LIGATION (VBL) VS. BETA-BLOCKERS (BB) IN THE PREVENTION OF FIRST VARICEAL HEMORRHAGE

19. Treatment of Varices / Variceal Hemorrhage Recurrent hemorrhage Recurrent hemorrhage Variceal hemorrhage Variceal hemorrhage Varices No hemorrhage Varices No hemorrhage No varices Management depends on the size of varices MANAGEMENT OF PATIENTS WITH VARICES WHO HAVE NEVER BLED

20. Treatment of Varices / Variceal Hemorrhage Recurrent hemorrhage Recurrent hemorrhage Variceal hemorrhage Variceal hemorrhage Medium/ large varices No hemorrhage Medium/ large varices No hemorrhage Small varices No hemorrhage Small varices No hemorrhage No varices 1)  -blockers (propranolol, nadolol) indefinitely 2) Endoscopic variceal ligation in patients intolerant to  -blockers 1)  -blockers (propranolol, nadolol) indefinitely 2) Endoscopic variceal ligation in patients intolerant to  -blockers MANAGEMENT OF PATIENTS WITH MEDIUM/LARGE VARICES WITHOUT PRIOR HEMORRHAGE

21. Treatment of Varices / Variceal Hemorrhage Recurrent hemorrhage Recurrent hemorrhage Variceal hemorrhage Variceal hemorrhage Medium/ large varices No hemorrhage Medium/ large varices No hemorrhage Small varices No hemorrhage Small varices No hemorrhage No varices ? Prevention of variceal growth MANAGEMENT OF PATIENTS WITH SMALL VARICES WITHOUT PRIOR HEMORRHAGE

22. Merkel et al., Gastroenterology 2004; 127:476 Nadolol May Prevent the Growth of Small Varices Nadolol % Probability of variceal growth % Probability of variceal growth 100 60 40 80 10 0 0 Placebo 20 24 36 48 60 p<0.001 Time (months) NADOLOL MAY PREVENT THE GROWTH OF SMALL VARICES

23. Treatment of Varices / Variceal Hemorrhage Recurrent hemorrhage Recurrent hemorrhage Variceal hemorrhage Variceal hemorrhage Small varices No hemorrhage Small varices No hemorrhage No varices  Repeat endoscopy in 1-2 years*  Beta-blockers?  Repeat endoscopy in 1-2 years*  Beta-blockers? Medium/ large varices No hemorrhage Medium/ large varices No hemorrhage * Sooner with cirrhosis decompensation MANAGEMENT OF PATIENTS WITH SMALL VARICES WITHOUT PRIOR HEMORRHAGE

24. Treatment of Varices / Variceal Hemorrhage Control of hemorrhage Recurrent hemorrhage Recurrent hemorrhage Variceal hemorrhage Variceal hemorrhage Medium/ large varices No hemorrhage Medium/ large varices No hemorrhage Small varices No hemorrhage Small varices No hemorrhage No varices CONTROL OF ACUTE VARICEAL HEMORRHAGE

25. Treatment of Acute Variceal Hemorrhage General Management:  IV access and fluid resuscitation  Do not overtransfuse (hemoglobin ~ 8 g/dL)  Antibiotic prophylaxis (IV ceftriaxone 1gm daily) Specific therapy:  Pharmacological therapy: terlipressin, somatostatin and analogues, vasopressin + nitroglycerin  Endoscopic therapy: ligation, sclerotherapy  Shunt therapy: TIPS, surgical shunt General Management:  IV access and fluid resuscitation  Do not overtransfuse (hemoglobin ~ 8 g/dL)  Antibiotic prophylaxis (IV ceftriaxone 1gm daily) Specific therapy:  Pharmacological therapy: terlipressin, somatostatin and analogues, vasopressin + nitroglycerin  Endoscopic therapy: ligation, sclerotherapy  Shunt therapy: TIPS, surgical shunt TREATMENT OF ACUTE VARICEAL HEMORRHAGE

26. Prophylactic Antibiotics Improve Outcomes in Cirrhotic Patients with GI Hemorrhage ControlAntibioticAbsolute rate (n=270)(n=264) difference (95% CI) Infection45%14%-32% (-42 to –23) SBP / Bacteremia27%8%-18% (-26 to –11) Death24%15%-9% (-15 to –3) ControlAntibioticAbsolute rate (n=270)(n=264) difference (95% CI) Infection45%14%-32% (-42 to –23) SBP / Bacteremia27%8%-18% (-26 to –11) Death24%15%-9% (-15 to –3) Bernard et al., Hepatology 1999; 29:1655 PROPHYLACTIC ANTIBIOTICS IMPROVE OUTCOMES IN CIRRHOTIC PATIENTS WITH GI HEMORRHAGE

27. Probability of Remaining Free of Recurrent Variceal Hemorrhage Hou M-C et al., Hepatology 2004; 39:746 Prophylactic antibiotics (n=59) % free of variceal hemorrhage % free of variceal hemorrhage 1.0 0.6 0.2 0.8 1 1 0 0 No antibiotics (n=61) 0 0 2 2 3 3 12 30 Follow-up (months) 18 24 0.4 PROPHYLACTIC ANTIBIOTICS PREVENT EARLY VARICEAL REBLEEDING

28. Bañares R et al., Hepatology 2002; 35:609 Combination Drug / Endoscopic Therapy is More Effective Than Endoscopic Therapy Alone in Achieving Five-Day Hemostasis Sclero + OctreotideBesson, 1995 Ligation + Octreotide Sung, 1995 Sclero + Octreotide / STSignorelli, 1996 Sclero + Octreotide Ceriani, 1997 Sclero + Octreotide Signorelli, 1997 Sclero + STAvgerinos, 1997 Sclero + Octreotide Zuberi, 2000 Sclero / ligation + VapreotideCales, 2001 TOTAL Sclero + OctreotideBesson, 1995 Ligation + Octreotide Sung, 1995 Sclero + Octreotide / STSignorelli, 1996 Sclero + Octreotide Ceriani, 1997 Sclero + Octreotide Signorelli, 1997 Sclero + STAvgerinos, 1997 Sclero + Octreotide Zuberi, 2000 Sclero / ligation + VapreotideCales, 2001 TOTAL Favors endoscopic therapy alone 1 1 1.6 1.8 2 2 1.2 1.4 0.8 Favors endoscopic plus drug therapy Relative Risk COMBINATION DRUG/ENDOSCOPIC THERAPY IS MORE EFFECTIVE THAN ENDOSCOPIC THERAPY ALONE

29. Endoscopic Variceal Band Ligation  Bleeding controlled in 90%  Rebleeding rate 30%  Compared with sclerotherapy:  Less rebleeding  Lower mortality  Fewer complications  Fewer treatment sessions  Bleeding controlled in 90%  Rebleeding rate 30%  Compared with sclerotherapy:  Less rebleeding  Lower mortality  Fewer complications  Fewer treatment sessions ENDOSCOPIC VARICEAL BAND LIGATION

30. TIPS in the Treatment of Variceal Hemorrhage  TIPS is rescue therapy for recurrent variceal hemorrhage (at second rebleed for esophageal varices, at first rebleed for gastric varices)  TIPS is indicated in patients who rebleed on combination endoscopic plus pharmacologic therapy  In patients with Child A/B cirrhosis, the distal spleno-renal shunt is as effective as TIPS (dependent on local expertise)  TIPS is rescue therapy for recurrent variceal hemorrhage (at second rebleed for esophageal varices, at first rebleed for gastric varices)  TIPS is indicated in patients who rebleed on combination endoscopic plus pharmacologic therapy  In patients with Child A/B cirrhosis, the distal spleno-renal shunt is as effective as TIPS (dependent on local expertise) TIPS IN THE TREATMENT OF VARICEAL HEMORRHAGE

31. Recurrent hemorrhage Recurrent hemorrhage Medium/ large varices No hemorrhage Medium/ large varices No hemorrhage Small varices No hemorrhage Small varices No hemorrhage No varices Treatment of Varices / Variceal Hemorrhage Variceal hemorrhage Variceal hemorrhage 1) Safe vasoactive drug + endoscopic therapy + antibiotic prophylaxis 2) TIPS / Shunt (rescue therapy) 1) Safe vasoactive drug + endoscopic therapy + antibiotic prophylaxis 2) TIPS / Shunt (rescue therapy) MANAGEMENT OF PATIENTS WITH ACUTE VARICEAL HEMORRHAGE

32. Lowest Rebleeding Rates are Obtained in HVPG Responders and With Ligation +  -Blockers (19 trials) (26 trials) (54 trials) % % Rebleeding 80 60 40 20 0 0 Untreated  -blockers Sclero- therapy Sclero- therapy (18 trials) Ligation (6 trials) HVPG- Responders* (6 trials)  -blockers + ISMN  -blockers + ISMN (2 trials) Ligation +  -blockers Ligation +  -blockers Bosch and García-Pagán, Lancet 2003; 361:952 *  HVPG 20% from baseline LOWEST REBLEEDING RATES ARE OBTAINED IN HVPG RESPONDERS AND IN PATIENTS TREATED WITH VARICEAL BAND LIGATION + BETA-BLOCKERS

33. Treatment of Varices / Variceal Hemorrhage No varices Varices No hemorrhage Varices No hemorrhage Variceal hemorrhage Variceal hemorrhage Recurrent hemorrhage Recurrent hemorrhage 1)  -blockers + ISMN or EVL 2)  -blockers + EVL may be preferable 3) TIPS / shunt surgery 1)  -blockers + ISMN or EVL 2)  -blockers + EVL may be preferable 3) TIPS / shunt surgery PREVENTION OF RECURRENT VARICEAL HEMORRHAGE

34. Evolution of Varices Level of Intervention Management Recommendations Cirrhosis with no varices  Repeat endoscopy in 2-3 years  No specific therapy  Repeat endoscopy in 2-3 years  No specific therapy Pre-primary prophylaxis SUMMARY OF MANAGEMENT OF VARICES AND VARICEAL HEMORRHAGE

35. Evolution of Varices Level of Intervention Management Recommendations Cirrhosis with no varices Small varices No hemorrhage Small varices No hemorrhage Medium / large varices No hemorrhage Medium / large varices No hemorrhage  Repeat endoscopy in 2-3 years  No specific therapy  Repeat endoscopy in 2-3 years  No specific therapy Small varices  Repeat endoscopy in 1-2 years  No specific therapy  ? beta-blocker to prevent enlargement Medium/Large varices  Non-selective beta-blockers  EVL in those who are intolerant to drugs Small varices  Repeat endoscopy in 1-2 years  No specific therapy  ? beta-blocker to prevent enlargement Medium/Large varices  Non-selective beta-blockers  EVL in those who are intolerant to drugs Pre-primary prophylaxis Primary prophylaxis SUMMARY OF MANAGEMENT OF VARICES AND VARICEAL HEMORRHAGE

36. Evolution of Varices Level of Intervention Management Recommendations Cirrhosis with no varices Small varices No hemorrhage Small varices No hemorrhage Medium / large varices No hemorrhage Medium / large varices No hemorrhage Variceal hemorrhage  Repeat endoscopy in 2-3 years  No specific therapy  Repeat endoscopy in 2-3 years  No specific therapy Small varices  Repeat endoscopy in 1-2 years  No specific therapy  ? beta-blocker to prevent enlargement Medium/Large varices  Non-selective beta-blockers  EVL in those who are intolerant to drugs Small varices  Repeat endoscopy in 1-2 years  No specific therapy  ? beta-blocker to prevent enlargement Medium/Large varices  Non-selective beta-blockers  EVL in those who are intolerant to drugs  Endoscopic/pharmacologic therapy  Antibiotics in all patients  TIPS or shunt surgery as rescue therapy  Endoscopic/pharmacologic therapy  Antibiotics in all patients  TIPS or shunt surgery as rescue therapy Pre-primary prophylaxis Primary prophylaxis SUMMARY OF MANAGEMENT OF VARICES AND VARICEAL HEMORRHAGE

37. Evolution of Varices Level of Intervention Management Recommendations Cirrhosis with no varices Small varices No hemorrhage Small varices No hemorrhage Medium / large varices No hemorrhage Medium / large varices No hemorrhage Variceal hemorrhage Recurrent variceal hemorrhage Pre-primary prophylaxis Primary prophylaxis Secondary prophylaxis  Repeat endoscopy in 2-3 years  No specific therapy  Repeat endoscopy in 2-3 years  No specific therapy Small varices  Repeat endoscopy in 1-2 years  No specific therapy  ? beta-blocker to prevent enlargement Medium/Large varices  Non-selective beta-blockers  EVL in those intolerant to drugs Small varices  Repeat endoscopy in 1-2 years  No specific therapy  ? beta-blocker to prevent enlargement Medium/Large varices  Non-selective beta-blockers  EVL in those intolerant to drugs  Endoscopic/pharmacologic therapy  Antibiotics in all patients  TIPS or shunt surgery as rescue therapy  Endoscopic/pharmacologic therapy  Antibiotics in all patients  TIPS or shunt surgery as rescue therapy  Beta-blockers + nitrates or EVL  Beta-blockers + EVL ?  TIPS or shunt surgery as rescue therapy  Beta-blockers + nitrates or EVL  Beta-blockers + EVL ?  TIPS or shunt surgery as rescue therapy SUMMARY OF MANAGEMENT OF VARICES AND VARICEAL HEMORRHAGE

38. The Future: Directed Therapy Using Portal Pressure Measurements

39. Portal Pressure Measurements  Definitive method to establish the diagnosis of portal hypertension  Direct methods (percutaneous, transjugular) are cumbersome and may be associated with complications  The safest and most reproducible method is measurement of the hepatic venous pressure gradient (HVPG)  Definitive method to establish the diagnosis of portal hypertension  Direct methods (percutaneous, transjugular) are cumbersome and may be associated with complications  The safest and most reproducible method is measurement of the hepatic venous pressure gradient (HVPG) PORTAL PRESSURE MEASUREMENTS

40. Portal Pressure Measurements  The hepatic venous pressure gradient (HVPG) is obtained by subtracting the free hepatic venous pressure (FHVP) from the wedged hepatic venous pressure (WHVP):  The FHVP acts as an internal zero to correct for extravascular, intraabdominal pressure increases (e.g. ascites)  The hepatic venous pressure gradient (HVPG) is obtained by subtracting the free hepatic venous pressure (FHVP) from the wedged hepatic venous pressure (WHVP):  The FHVP acts as an internal zero to correct for extravascular, intraabdominal pressure increases (e.g. ascites) HVPG = WHVP - FHVP PORTAL PRESSURE MEASUREMENTS

41. A Threshold Portal Pressure of ~12 mmHg is Necessary for Varices to Form P<0.01 5 5 10 12 15 25 30 35 20 Hepatic Venous Pressure Gradient (mmHg) Hepatic Venous Pressure Gradient (mmHg) Garcia-Tsao et. al., Hepatology 1985; 5:419 Varices Present (n=72) Varices Present (n=72) Varices Absent (n=15) Varices Absent (n=15) A THRESHOLD PORTAL PRESSURE OF ~12 mmHg IS NECESSARY FOR VARICES TO FORM

42. % Rebleeding % Rebleeding Decrease In Hepatic Venous Pressure Gradient (HVPG) Reduces Risk of Variceal Bleeding 0 0 20 40 60 80 100 HVPG decrease > 20% from baseline HVPG decrease > 20% from baseline HVPG decrease to < 12 mmHg 0% 46-65% 7-13% No change in HVPG No change in HVPG Bosch and García-Pagán, Lancet 2003; 361:952 DECREASE IN HEPATIC VENOUS PRESSURE GRADIENT (HVPG) REDUCES THE RISK OF VARICEAL BLEEDING

43. Abraldes et al., Hepatology 2003; 37:902 Probability of Survival (%) Probability of Survival (%) 100 50 25 0 0 75 Time (months) 96 84 72 60 48 36 24 12 0 0 HVPG non-responders HVPG responders p=0.003 95% 52% Survival Improves in Patients in Whom HVPG Decreases (HVPG Responders) SURVIVAL IMPROVES IN PATIENTS IN WHOM HVPG DECREASES (HVPG RESPONDERS)