COHORTE DE PACIENTES CON CIRROSIS HEPÁTICA

Dr. José R Arribas

Unidad VIHServicio de Medicina Interna

Morbidity and mortality in HIV infected patients with Morbidity and mortality in HIV infected patients with compensated and decompensated cirrhosis: compensated and decompensated cirrhosis:

prospective cohort of 373 patientsprospective cohort of 373 patients

M López-Diéguez, JF Pascual, M Montes, C Quereda, MA Von Wichmann, J Berenguer, C Tural, JM Miró, F Pulido, E Ortega, A Arranz, J González-García, JR Arribas and the GESIDA 37/03-FIPSE 364665/03 Study Group.

Oral Presentation at EACS2007 PS8/4

OBJECTIVE

To evaluate morbidity/mortality in HIV-infected patients with compensated vs decompensated liver cirrhosis.

STUDY DESIGN (1) Multicenter national prospective cohort.

País Vasco H. Virgen de Aranzazu.

Valencia H. General Universitario Valencia.

Barcelona H. Clinic y Provincial. H. Germans Trias i Pujol.

Madrid H. Príncipe de Asturias. H. Gregorio Marañón. H. Ramón y Cajal. H. Doce de Octubre. H. La Paz.

STUDY DESIGN (2)

Cirrhosis Diagnosis

– Biopsy: (Cirrhosis or advanced bridging fibrosis).

– Decompensation• Gastrointestinal bleeding, ascites, hepatic

encephalopathy.

– Bonacini Score > 8 (Am J Gastroenterol 1997;92:1302).

BONACINI SCORE FOR CIRRHOSIS DIAGNOSIS

Three-parameter cirrhosis discriminant score:– Platelets – ALT/AST ratio – PT– Cutoff for cirrhosis diagnosis = 8

Sensibility 46% Specifycity 98%

Bonacini M, et al. Am J Gastroenterol 1997;92:1302.

STUDY DESIGN (3)

Total planned follow-up 48 months. Visits: baseline and then every 6 months.

– Each visit:• Personal interview.• Hematology, Biochemistry, Inmmunology, Virology, alfa-

fetoprotein.• Abdominal US.

– Each year:• Endoscopy to detect esophageal varices (according to

Schepis criteria*).

Schepis et al. Hepatology 2001; 33:471-2.

STUDY DESIGN (4)

SURVIVAL: time from the date of entry until the first endpoint occurred.

ENDPOINT: death, hepatocarcinoma or liver transplant.

STATISTICAL ANALISYS: Kaplan-Meyer analysis, log rank test (comparison of survival between different groups).

BASELINE CHARACTERISITICS (1)All Compensated Decompensated

N 373 274 99

Mean age (years ) 44 44 43

Female (%) 80 (22) 61 (22,2) 19 (19,4)

Cirrhosis diagnosis- Biopsy (%)- Bonacini Score >8 (%)- Prior decompensation (%)

234 (63)

41 (11)

98 (26)

234 (85,1)

41 (14,9)

_

_

_

98 (100)

Cirrhosis causes- Hepatitis C (%)

-Genotypes 2 or 3 (%)- Hepatitis B (%)- Prior alcohol abuse (%)

370 (99,2)

81 (21,7)

24 (6,4)

115 (31)

274 (99,7)

63 (22,9)

17 (6,2)

74 (26,9)

96 (97,9)

18 (18,4)

7 (7,1)

41 (41,8)

Median duration HIV infection (years) 15 15 15

All Compensated Decompensated

Median duration HVC infection (years) 23 23 23

HCV treatment received (%) 205 (55) 178 (64,7) 27 (27,6)

CDC stage C (%) 143 (39,3) 90 (32,8) 53 (54,1)

Receiving HAART at baseline (%) 322 (82,8) 244 (88,7) 78 (79,6)

HIV Transmission route

- IVDU (%) 328 (88) 239 (86,9) 89 (90,8)

CD4 cell count (median, IQR)-Baseline -Nadir

373 (228 - 577)

145 (70 - 255)

434 (272 - 644)

175 (76 - 270)

239 (140 - 365)

104 (58 - 180)

-HIV-RNA-Baseline (median, IQR)-% HIV RNA BLQ*

49 (49 - 398)

72,4

49 (49 - 200)

75,6

49 (49 - 1229)

65,2

BASELINE CHARACTERISITICS (2)

*Below limit of quantification (50-200) c/ml.

All Compensated Descompensated

Lost to follow-up (%) 40 (10,7) 21 (7,6) 19 (19,4)

Follow-up (median, IQR) 18 (14-20) 18 (15,7-20,2) 16 (6-19)

Endpoints, n (%)

Any

Death

Hepatocarcinoma

Transplant

63 (18,9)

55 (16,5)

2 (0,6)

9 (2,7)

20 (7,9)

17 (6,7)

_

3 (1,2)

43 (54,4)

38 (48,1)

2 (2,5)

6 (7,6)

Deaths, n (%)

Hepatic causes

Other

Unknown

33 (9,9)

14 (1,8)

6 (1,8)

6 (2,4)

6 (2,4)

5 (2)

27 (34,2)

8 (10,1)

1 (1,3)

RESULTS

RESULTS

Baseline Initially compensated

Type of Decompensations n (%)

Ascites

GI bleeding

Encephalopathy

HRS

SBP

Unknown

99 (26,5)

51 (51,1)

12 (12,1)

10 (10,1)

15 (15,2)

4 (4)

7 (7,1)

17 (6,2)

6 (2,2)

2 (0,7)

7 (2,6)

2 (0,7)

SURVIVAL

N 332 302 264 169

Cum

ulative probability of survival

Months

0.82

SURVIVALCompensated vs Decompensated

Compensated 253 241 218 141Decompensated 78 60 45 27

p<0,0001 (log-rank)

Cum

ulative probability of survival

Months

0.92

0.53

SURVIVAL (months)

COMPENSATED DECOMPENSATED

Mean (IC95%)

Median (IC95%)

1 year probability

2 years probability

3 years probability

66 (63-69)

NA

0.95

0.90

0.90

19 (15-23)

18 (12-24)

0.63

0.32

_

SURVIVAL Child Pugh Score

A

B

C

CP-A 219 213 196 128CP-B 57 46 34 17 CP-C 21 12 7 5

p<0,0001 (log-rank)

Cum

ulative probability of survival

Months

0.96

0.53

0.27

CHILD-PUGH SCORE

A B C

Mean (IC95%)

Median (IC95%)

1 year probability

2 years probability

3 years probability

68 (65-70)

NA

0.98

0.92

0.92

22 (18-26)

19 (13-25)

0.65

0.49

_

10 (6-13)

7 (5-9)

0.32

_

_

SURVIVAL (months)

N 253 237 210 147

PROBABILITY OF FIRST DECOMPENSATION

Percent wiithout decom

pensation

Months

Probability of decompensation

One year

Two years

Three years

0.04 (IC95% 0.01 – 0.07)

0.07

0.09

CONCLUSIONS HIV-infected patients with compensated liver cirrhosis had a

relatively high survival with a low per year probability of first decompensation.

HIV-infected patients with decompensated cirrhosis have a very poor prognosis. One third of our patients with decompensated liver cirrhosis died during the first year of follow-up.

Child Pugh score apears as a good prognostic score for HIV-infected patients with liver cirrhosis.

These results emphasize the critical importance of avoiding the development of end-stage liver disease in HIV-infected patients.

Analysis of factors associated to survival will be available soon

FACTORS ASSOCIATED WITH SURVIVAL AND FIRST HEPATIC FACTORS ASSOCIATED WITH SURVIVAL AND FIRST HEPATIC DECOMPENSATION IN A LARGE PROSPECTIVE COHORT OF DECOMPENSATION IN A LARGE PROSPECTIVE COHORT OF HIV-HCV CO-INFECTED PATIENTS WITH LIVER CIRRHOSIS.HIV-HCV CO-INFECTED PATIENTS WITH LIVER CIRRHOSIS.

M López-Diéguez, JF Pascual, M Montes, C Quereda, MA Von Wichmann, J Berenguer, C Tural, JM Miró, F Pulido, E Ortega, A Arranz, J González-García, JR Arribas and the GESIDA 37/03-FIPSE 364665/03 Study Group.

Poster Presentation at CROI2008 [1057]

clinicaloptions.com/hiv

METHODS

Prospective multicenter cohort of 331 HIV-HCV coinfected patients with cirrhosis. Median follow-up time: 18 months.

Cirrhosis diagnosis (n,%): biopsy (209, 63%), prior decompensation (86, 26%), Bonacini Score ≥ 8 (36, 11%).

Endpoints: death, hepatocarcinoma or liver transplant.

Survival defined as the time from entering in the cohort until first endpoint occurred.

The association of survival with different factors was explored in univariate and multivariate Cox proportional hazard models. Variables included: age, sex, time since cirrhosis/HIV diagnosis, alcohol intake, CD4 count (nadir, baseline and <100 at baseline), HIV viremia, suppressed HIV replication, history of anti-HCV therapy, HCV genotype, sustained viral response to anti-HCV therapy, concomitant chronic HBV, history of cirrhosis decompensation, Child Pugh score and HAART (at baseline, continuous/interrupted during follow-up).

For patients with no history of prior liver decompensation at baseline we explored variables associated with the development of first decompensation.

BASELINE CHARACTERISTICS

Male, N, (%) 258 (78)Age, median, (IQR) 44 (41–47)Months of follow-up, median, (IQR) 18 (12–20)Years since HIV diagnosis, median, (IQR) 16 (11–19)Years since cirrhosis diagnosis, median, (IQR) 3 (2–5)CDC C3, N, (%) 93 (28.1)IVDU, N, (%) 292 (88.2)HAART, N, (%)

at baseline 287 (87) non continuous HAART 166 (50)Alcohol abuse, N, (%) 100 (30.2)

BASELINE CHARACTERISTICS

CD4, median, (IQR)

at baseline

nadir

384

150

(232–589)

(71–258)HIV-RNA < BLQ*, N, (%) 236 (74)HVB co-infection, N, (%) 20 (6)HCV genotype 2 or 3, N, (%) 71 (27)HCV therapy, N, (%)

sustained viral response, N, (%)

still non evaluable, N, (%)

191

37

52

(57.7)

(11.2)

(15.7)

clinicaloptions.com/hiv

ENDPOINTS

Endpoints: 62 (54 deaths, 9 hepatocarcinomas, and 1 liver transplant).

Compensated cirrhosis at baseline: 19 (16 deaths, 3 Hepatocarcinomas)

Decompensated cirrhosis at baseline: 43 (38 deaths, 6 Hepatocarcinomas, 1 Liver Transplant)

Variables associated to survival. Univariate analysis

HR (CI 95%) p

Male gender 2.37 (1.078 – 5.21) 0.032Alcohol intake 0.506 (0.306 – 0.838) 0.008CD4 <100 at baseline 3.26 (1.48 – 7.19) 0.003Unsuppressed VL at baseline 2.16 (1.27 – 3.65) 0.004No HCV therapy received 3.01 (1.76 – 5.14) < 0.0001No response to HCV therapy 7.31 (1.01 – 52.81) 0.048Non-Continuous HAART during follow up

15.37 (6.15 – 38.46) < 0.0001

CD4 nadir 0.997 (0.995 – 0.999) 0.008Child Pugh score B

Child Pugh score C

14.46 (6.97 – 29.9)

39.45 (17.96 – 86.65)

< 0.0001

< 0.0001

Multivariate analysis: Hazard ratio of factors associated with decreased survival [HR, (CI), p]

Survival according to Child Pugh Score

Child Pugh A

Child Pugh B

Child Pugh C

(N) CP-A 220 213 205 184 74 CP-B 58 48 38 30 7 CP-C 22 14 8 6 1

Probability of first decompensation according to Child Pugh Score

(N) CP-A 206 198 187 167 65 CP-B 25 19 11 8 3

Child Pugh A

Child Pugh B

clinicaloptions.com/hiv

CONCLUSIONS

Child-Pugh scores B and C are significantly associated with decreased survival in HIV-HCV coinfected patients with cirrhosis.

Maintaining HIV viral suppression and receiving continuous HAART are associated with prolonged survival. Our study supports the continuous use of HAART in this population.

Child-Pugh B is significantly associated with the short-term risk of first hepatic decompensation. HIV-HCV coinfected patients with compensated cirrhosis and a Child-Pugh B score should be followed closely for the development of decompensation.

clinicaloptions.com/hiv

RESUMEN

El estudio GESIDA 37/03 es una de las cohortes más grandes de pacientes infectados por VIH con cirrosis hepática.

Hasta el momento esta cohorte nos ha permitido caracterizar mejor la historia natural de la cirrosis hepática en esta población

Además hemos podido analizar los factores relacionados con la supervivencia y la primera descompensación.

Continuamos el seguimiento activo de esta cohorte (Dra. Marisa Montes)

clinicaloptions.com/hiv

AGRADECIMIENTOS

M López-Diéguez, JF Pascual, M Montes, C Quereda, MA Von Wichmann, J Berenguer, C Tural, JM Miró, F Pulido, E Ortega, A Arranz, J González-García, Rosario Madero, Herminia Esteban