Effect of Liver Fibrosis on Long-Term Mortality in HIV/Hepatitis C Virus-Coinfected Individuals Who Are Evaluated to Receive Interferon Therapies in the Highly Active Antiretroviral Therapy Era

Abstract

The factors associated with overall mortality and liver decompensation in HIV and hepatitis C virus (HCV)-coinfected patients who are evaluated to receive HCV antiviral therapy with a known liver histological fibrosis stage were evaluated in a prospective cohort study. A total of 387 consecutive HIV/HCV-coinfected patients attending an outpatient clinical unit between January 1997 and December 2007 who fulfilled criteria to be treated with interferon and to whom liver biopsy was performed were included and followed every 6 months from time of liver biopsy to death or to December 2008. The follow-up period was 6.2 years (IQR: 3.5–9.2). The median age at time of liver biopsy was 38 years. This included 73% men; 28% had advanced liver fibrosis (F3–F4) and a CD4 cell count of 556 cells/mm³, 72% had HIV RNA <400 copies/ml and a mean CD4 nadir of 207 cell/mm³, 21% had a previous diagnosis of AIDS, and 92% were on antiretroviral therapy. During follow-up 48% underwent HCV antiviral therapy, with a sustained virological response in 33%. The overall mortality rate and the incidence of liver decompensation or liver-related death were 1.17 and 0.72 per 100 patients-year, respectively. End stage liver disease (9/28 patients) and non-AIDS-related cancer (6/28) were the main causes of death. F3–F4 (HR: 3.74, 95% CI: 1.69–8.26, p=0.001) and previous AIDS diagnosis (HR: 3.04, 95% CI: 1.36–6.81) were the factors independently associated with death. Mortality rates in patients who received and who did not receive HCV antiviral therapy were 0.44 and 2.04 per 100 patients-year, respectively (p=0.003). In addition to the low mortality rate observed, HIV/HCV-coinfected patients with poor predictors of survival are candidates for intensive clinical management.

Introduction

T he prolonged survival of HIV and hepatitis C virus (HCV)-coinfected patients mainly due to the worldwide distribution of highly active antiretroviral therapy (HAART) has led to the emergence of a new clinical challenge.¹ The increased prevalence of HIV/HCV coinfection in the HAART era in certain geographic areas has led to the existence of a high proportion of patients with a controlled HIV replication but with long-standing chronic HCV infection, which makes this population susceptible to developing liver complications such as cirrhosis and hepatocellular carcinoma and liver-related death.^2
–4

Studies aimed at evaluating the predictors of survival in HIV/HCV-coinfected patients published so far have shown end-stage liver disease to be the main cause of mortality in these patients³ and have also shown the benefit of HAART^5,6 and of the achievement of sustained virological response to HCV^7
–9 therapy. However, the effect of fibrosis stage on mortality, which is the main prognostic factor of survival in HCV-infected patients,¹⁰ has been poorly evaluated in this group of patients.

There is currently a profile of an HIV/HCV-coinfected patient who regularly attends the HIV outpatient clinic of certain hospitals not specialized in drug addition treatment, characterized by having stable HIV infection, being on HAART, and not having a large number of comorbidities that accelerate progression of liver disease, in which the natural history of coinfection is largely unknown.

The main objective of this study is to analyze the mortality in this subgroup of patients with a known liver histological fibrosis stage. Secondary objectives include the evaluation of immunological and virological outcomes.

Materials and Methods

Study design

This is a prospective study to describe mortality, immunological and virological outcomes of HIV and HCV-coinfected patients with known stage of liver fibrosis from time of liver biopsy to death or December 2008.

Patients and follow-up

Consecutive HIV/HCV-coinfected patients naive for HCV antiviral therapy attending an outpatient HIV Clinical Unit in a tertiary teaching hospital who underwent liver biopsy as a step in the decision-making process to receive HCV antiviral therapy between January 1997 and December 2007 were included in the study. Thus, the date required to enter in the cohort was the date when liver biopsy was performed. Only patients who fulfilled the criteria to be treated with interferon-based therapies according to most international consensus during this period were evaluated for liver biopsy performance. The main criteria to receive interferon-based therapies were a positive HCV RNA in plasma, stable HIV infection, and CD4 higher than 100 cells/ml regardless of HIV viral load. Patients also had to be without active opportunistic infections and without comorbid conditions that contraindicate treatment with interferon and ribavirin. Hemoglobin levels above 12 g/dl in men or above 11 g/dl in women, neutrophil count more than 1500 cells/μl, and platelet count above 70,000 cells/μl were also required. Moreover, patients with active illicit drug consumption or alcohol intake more than 40 g/day and cirrhotic patients with a Child–Pugh score more than 6 were excluded.

At time of liver biopsy, demographic data and data related to HIV and HCV infection were collected from each patient. After liver biopsy was performed, patients were prospectively followed every 6 months until death or until December 2008.

Liver biopsy

Liver biopsy was echo-guided and performed by an expert radiologist. Biopsies were always evaluated by the same pathologist and necroinflammatory activity and fibrosis stage and steatosis were evaluated according to the Scheuer¹¹ and Brunt¹² index, respectively.

Statistical analysis

Endpoints

We selected any-cause mortality and liver decompensation as the primary and secondary clinical endpoints, respectively. Death was classified as liver related if it was a consequence of portal hypertensive gastrointestinal bleeding, hepatic encephalopathy, spontaneous bacterial peritonitis, or hepatocellular carcinoma. Patients who died from end stage liver disease or had a liver decompensation during follow-up were considered to have reached the secondary clinical endpoint. Life status was obtained from clinical history and from the regional death registry. Causes of death were established based on the information recorded from the clinical charts and ICD-10.

To assess immunological outcome, time to a CD4 cell count below 200 cells/mm³ was considered as the immunological endpoint, so only patients with a CD4 cell count above 200 cells/mm³ at the time of liver biopsy were considered. Due to the known effect of interferon therapy on the absolute CD4 cell count,¹³ a CD4 cell count <200 cells/mm³ more than 6 months after stopping peginterferon-based therapies was required for the patients who underwent HCV antiviral therapy to considered as having reached the immunological endpoint. To analyze the virological outcome, we assessed time to virological failure defined by an HIV RNA concentration >400 copies/ml in those patients with an HIV RNA concentration below 400 copies/ml at the time of liver biopsy. Patients were censored on December 31, 2008.

Statistics

Overall mortality rates and liver decompensation rates were expressed in 100 person-years. Cox proportional-hazards regression analysis was used to assess baseline risk factors for the different endpoints: age (≤38 or >38 years), sex, liver fibrosis stage (F0–F2 vs. F3–F4), necroinflammatory activity (<4 or ≥4), steatosis (Brunt score 0 vs. 1–3), CD4 cell count (≤500 or >500 cells/mm³), CD4 cell count nadir (≤200 or >200 cells/mm³), HIV RNA (≤400 or >400 copies/ml), antiretroviral therapy (yes or no), protease inhibitor (PI), nonnucleoside reverse transcriptase inhibitor (NNRTI), number of previous antiretroviral treatment regimens (≤7 or >7), and calendar period when the liver biopsy was performed (1997–2002 and 2003–2007).

Advanced liver fibrosis was defined as liver fibrosis stage ≥F3. Coinfection with hepatitis B virus was not included in the analysis as data on HBV DNA at time of liver biopsy were not available. Multivariate Cox regressions were adjusted by gender, age, and CD4 at time of liver biopsy. Given the fact that HCV antiviral therapy during follow-up was not a baseline factor, this variable was not included in the COX analysis, but the relative risk for death in patients who had received and not received HCV antiviral therapy in our cohort was calculated. Sustained virological response (SVR) was defined by the presence of HCV RNA below 50 IU/ml 6 months after stopping peginterferon-based therapy.

Factors associated with the different endpoints in the univariate analysis (p<0.10) were selected for a multivariate Cox proportional-hazards regression model. Kaplan–Meier survival was performed to compare time to reach the clinical endpoint. Continuous variables are presented as median±standard deviation and/or median (interquartile range, IQR) when appropriate. Categorical variables are presented as percentages. Student's t test for continuous variables was used when normal distribution was proven and the Mann-Whitney U test otherwise. The Chi-square or Fisher's test was used to compare categorical variables.

Statistical analysis was performed using SPSS version 12 (SPSS Inc., Chicago, IL) and EPI INFO version 6.04 (Centers for Disease Control and Prevention, USA and WHO, Geneva, Switzerland). We considered p values lower than 0.05 as statistically significant.

Results

Follow-up time

Three hundred and eighty-seven HIV/HCV-coinfected patients who underwent a liver biopsy between 1997 and 2007 were included in this study and prospectively followed until death or the end of follow-up (31 December 2008).

Baseline characteristics

The study population was predominantly male, including former intravenous drug users with well-controlled HIV replication and good immunological status at time of liver biopsy (Table 1). There was a significant difference in CD4 cell count nadir between patients with previous AIDS diagnosis (21%) [median 97 (IQR: 23–205)] and patients without [median 211 (IQR: 133–312)] (p<0.0001). Time between nadir and liver biopsy data was 21.5 (IQR: 15.7–59.5) months. Time between previous AIDS diagnosis and liver biopsy was 6.0 (±4.0) years. Twelve patients (4%) were HBV and HCV coinfected. Data on baseline HBV DNA quantification were not available for these patients.

Table 1.

Characteristics of the Cohort According to Survival State at the End of Follow-up

Baseline characteristics	Total (n=387) n (%)	Dead (n=28) n (%)	Alive (n=359) n (%)	p-value
Sex
Men	282 (73)	19 (68)	263 (73)	0.51
Women	105 (27)	9 (32)	96 (27)
Age at liver biopsy, median [IQR]	38 [35–42]	38 [33.5–41.7]	38 [35–42]
≤38 years	172 (44)	16 (57)	156 (52)	0.69
>38 years	185 (56)	12 (43)	173 (48)
Risk factors for HIV acquisition
IVDU	292 (84)	21 (78)	271 (85)	0.40
Other	54 (16)	6 (22)	48 (15)
Year of liver biopsy performance
1997–2002	215 (56)	22 (79)	193 (54)	0.01
2003–2007	172 (44)	6 (21)	166 (46)
Liver fibrosis stage (n=378)^a
F0–F2	268 (71)	11 (42)	257 (73)	0.003
F3–F4	110 (29)	15 (58)	95 (27)
Necroinflammatory activity (n=378)
≤4	275 (73)	17 (65)	258 (73)	0.37
>4	103 (27)	9 (35)	94 (27)
Steatosis (n=384)
No	313 (82)	24 (86)	289 (80)	0.80
Yes	71 (18)	4 (14)	67 (18)
HBsAg (n=306)
Negative	294 (96)	18 (100.0)	276 (96)	1
Positive	12 (4)	0 (0.0)	12 (4)
CD4 cell count (n=386), mean (±SD)	553 (±263)	512 (±271)	556 (±263)	0.39
≤500 cells/mm³	197 (51)	13 (46)	184 (51)	0.69
>500 cells/mm³	190 (49)	15 (54)	174 (49)
CD4 nadir (n=364), mean (±SD)	207 (±137)	209 (±138)	178 (±131)	0.25
≤200 cells/mm³	191 (49)	19 (68)	172 (48)	0.05
>200 cells/mm³	196 (51)	9 (32)	187 (52)
HIV viral load (n=362)
≤400 copies/ml	262 (72)	18 (70)	244 (73)	0.82
>400 copies/ml	100 (28)	8 (30)	92 (27)
Previous AIDS diagnosis (n=387)
No	297 (79)	13 (52)	284 (80)	0.006
Yes	81 (21)	12 (48)	69 (20)
HCV antiviral therapy
No	200 (52)	24 (86)	176 (49)	<0.001
Yes	187 (48)	4 (14)	183 (51)
HCV genotype (n=339)
1a/b	179 (53)	13 (57)	166 (53)	0.83
No 1a/b	160 (47)	10 (43)	150 (47)
HCV RNA viral load (n=339)
≤600,000 copies/ml	142 (43)	3 (100)	139 (42)	0.08
>600,000 copies/ml	197 (57)	0 (0)	197 (58)
SVR (n=181)
No	122 (67)	4 (100)	118 (67)	0.35
Yes	59 (33)	0 (0.0)	59 (33)
ART (n=387)
HAART including PIs (1)	211 (54)	16 (57)	195 (54)	0.49
HAART including NNRTIs	104 (27)	5 (18)	99 (27)
Combined NRTI	42 (11)	5 (18)	37 (10)
No ART	30 (8)	2 (7)	28 (8)
Number of previous ART schemes
<7	280 (72)	20 (72)	260 (72)	0.87
≥7	107 (28)	8 (28)	99 (28)

Liver biopsy sample was insufficient for staging in nine patients of whom two died during follow-up.

IQR, interquartile range; IDU, intravenous drug users; SVR, sustained virological response; ART, antiretroviral therapy; HAART, highly active antiretroviral therapy; PI, protease inhibitor; NNRTI, nonnucleoside reverse transcriptase inhibitor; NRTI, nucleoside reverse transcriptase inhibitor; IVDU, intraverous drug use.

Most of the patients were on antiretroviral therapy, of whom 315 were on HAART (88%); the remaining were on antiretroviral combinations including two or more nucleoside analogs. Among patients on PIs only 40% were on lopinavir, atazanavir, or darunavir, while the remaining were on indinavir, nelfinavir, saquinavir, ritonavir, and amprenavir. Among patients on nonnucleoside analogs, 64% were on nevirapine and 36% on efavirenz. The proportion of patients on ddI and/or d4T was 53%.

Liver biopsy

The size of liver biopsy sample was 1.7 (±0.8) cm [median number of portal tracts 8 (IQR: 6–10)]. The proportion of patients with F0–F2 and F3–F4 was 72 and 28%, respectively (Table 1). There were no statistical differences in the number of liver biopsies by calendar year (p=0.55). The proportion of patients with advanced liver fibrosis did not differ between calendar year either (p=0.153). Younger patients and patients with >500 cells/mm³ had less advanced liver fibrosis stages (F3–F4 in patients ≤38 and >38 years old was 24% and 35%, respectively, p=0.013; F3–F4 in patients with ≤500 and >500 cells/mm³ was 34 and 25%, respectively, p=0.05). There were no significant differences in liver fibrosis stage and necroinflammatory activity between HBsAg⁺ and HBsAg⁻ patients.

No differences in the prevalence of F3–F4 were found between patients taking PI, NNRTI-based regimens, nucleoside analog combinations, and without antiretroviral therapy at time of liver biopsy. In addition, we have found no differences in the prevalence of F3–F4 between patients with and without ddI and/or d4T.

Liver steatosis was similar in patients with F3–F4 and F0–F2 (66% vs. 34% in F3–F4 and F0–F2, respectively). No histological features of nonalcoholic steatohepatitis or liver storage diseases were found.

HCV antiviral therapy during follow-up

During follow-up, nearly half of the patients received interferon (IFN)-based therapies for the treatment of chronic HCV infection. Time between liver biopsy and the initiation of HCV antiviral therapy was 10 (IQR: 2–24) months. Forty-five percent (N=120) of patients with F0–F2 and 59% (N=64) of patients with F3–F4 received HCV antiviral therapy (p=0.012). Moreover, patients who were treated were significantly older [39 (IQR: 35–42) vs. 37 (IQR: 34–41) years, p=0.011] than patients who did not receive IFN. No other significant differences were found between IFN-treated and nontreated patients. Sustained virological response (SVR) was observed in 33% of the patients (53% and 27% in HCV genotype 3c/d and 1a/b, respectively, p=0.01) and was higher in older patients (24% and 41% in ≤38 years and >38 years, respectively). There were no differences in SVR according to liver fibrosis (34 vs. 31% in F0–F2 and F3–F4, respectively).

Mortality rate and causes of death

The follow-up time was 6.2 (IQR: 3.5–9.2) years. Twenty-eight patients died during follow-up (0.7%). The mortality rate during follow-up was of 1.17 per 100 patient-years (95% CI: 0.81–1.70). The most frequent cause of death was end-stage liver disease (nine patients), followed by non-AIDS-related cancer (three lung, one rectum, one cervical cancer, and one Hodgkin lymphoma), AIDS (three non-Hodgkin lymphoma and two toxoplasmosis), cardiovascular disease (two acute myocardial infarction and one nontraumatic cerebral hemorrhage), non-AIDS-related infection (two gram-negative septicemia), heroin overdose (one), homicide (one), and suicide (one). In addition to the nine patients who died from cirrhosis, 17 more showed hepatic decompensation (eight patients with F0–F2 and nine patients with F3–F4). The incidence rate of liver decompensation or death due to liver decompensation was 0.72 per 100 patient-years (95% CI: 0.42–1.16). Ascites was the first manifestation of liver decompensation in 10 patients, upper gastrointestinal bleeding due to portal hypertension in four, hepatocellular carcinoma in two, and hepatic encephalopathy in one.

Risk factors for death from any cause

Univariate analysis showed that liver fibrosis stage, CD4 cell count nadir, and previous AIDS diagnosis were factors associated with death from any cause but multivariate analysis demonstrated that only liver fibrosis stage and AIDS diagnosis before liver biopsy were the independent factors of survival (Table 2). Figure 1 shows the cumulative survival according to liver fibrosis and previous AIDS diagnosis.

FIG. 1.

KM of overall mortality according to liver fibrosis (A) and previous AIDS diagnosis (B).

Table 2.

Univariant and Multivariant Analysis of the Predictors of Mortality

	COX univariate analysis			COX multivariate analysis
Baseline characteristics	HR	95% CI	p-value	HR	95% CI	p-value
Sex	0.36	0.31–1.5	0.369	0.70	0.31–1.58	0.398
Men (1)
Women
Age at liver biopsy	1.17	0.54–2.51	0.68	1.19	0.52–2.68	0.674
≤38 years (1)
>38 years
Year of the liver biopsy performance	1.45	0.55–3.83	0.44
1997–2002
2003–2007
Fibrosis stage (N=378)^a	3.57	1.63–7.78	0.003	3.74	1.69–8.26	0.001
F0–F2 (1)
F3–F4
Necroinflammatory activity (N=378)	1.49	0.66–3.337	0.330
≤4(1)
>4
Steatosis (N=384)	1.57	0.54–4.53	0.373
No (1)
Yes
CD4 cell count (N=386)	0.86	0.41–1.84	0.819	0.60	0.27–1.36	0.674
≤500 cells/mm³
>500 cells/mm³ (1)
CD4 nadir (N=364)	2.32	1.04–514	0.038
≤200 cells/mm³
>200 cells/mm³ (1)
HIV viral load (N=362)	1.01	0.13–2.33	0.979
≤400 copies/ml
>400 copies/ml (1)
Risk factors for HIV	1.58	0.63–3.93	0.320
IDUs
Other (1)
Previous AIDS daignosis (N=387)	3.36	1.65–7.97	0.001	3.04	1.36–6.81	0.007
No (1)
Yes
HVC genotype (n=339)	0.77	0.31–1.77	0.864
1a/b (1)
No 1a/b
HCV viral load (N=181)	0.013	0.145.5	0.360
≤600,000 copies/ml
>600,000 copies/ml (1)
ART (n=387)
HAART including PIs (1)
HAART including NNRTI	1.28	0.29–5.61	0.73
Combined NRTI	0.95	0.18–4.93	0.95
No ART	1.86	0.36–9.60	0.45
Number of previous ART schemes	1.35	0.59–3.08	0.470
<7 (1)
≥7

Liver biopsy sample was insufficient for staging in nine patients of whom two died during follow-up.

IDU, intravenous drug users; SVR, sustained virological response; ART, antiretroviral therapy; HAART, highly active antiretroviral therapy; PI, protease inhibitor; NNRTI, nonnucleoside reverse transcriptase inhibitor; NRTI, nucleoside reverse transcriptase inhibitor.

Patients who received HCV antiviral therapy had a lower relative risk of death from any cause than patients who had never received HCV antiviral therapy.

Table 3 shows the overall mortality rates and relative risk of death according to the independent factors of overall mortality and in those patients who had and had not received HCV antiviral therapy during follow-up.

Table 3.

Overall Mortality Rates According to Liver Fibrosis, AIDS Previous Diagnosis, and Hepatitis C Virus Antiviral Therapy

	Incidence rate of mortality per 100 patients-year	95% CI	RR (95% CI)	p-value
Overall mortality	1.17	0.81–1.70
F0–F2 (N=11/268; 4%)^a	0.65	0.33–1.17	3.68 (1.63–8.31)	0.0014
F3–F4 (N=15/160; 14%)	2.33	1.30–3.84
Non-AIDS diagnosis (N=13/297; 5%)	0.84	0.48–1.36	3.79 (1.66–8.69)	0.0085
AIDS diagnosis (N=12/81; 15%)	2.57	1.33–4.49
HCV antiviral therapy (N=4/187; 2%)	0.44	0.12–1.13	4.63 (1.60–13.35)	0.003
No HCV antiviral therapy (N=24/200, 12%)	2.04	1.31–3.04

Liver biopsy sample was insufficient for staging in nine patients of whom two died during follow-up.

There were no significant differences in the causes of death between patients with F0–F2 and F3–F4, with or without previous AIDS diagnosis, or those who had or had not received HCV antiviral therapy (Table 4).

Table 4.

Causes of Death According to Liver Fibrosis, AIDS Previous Diagnosis, and Hepatitis C Virus Antiviral Therapy

	ESLD, n (%)	Non-AIDS, n (%)	AIDS, n (%)	p-value
All deaths (N=28)	9 (35)	14 (50)	5 (15)
Liver fibrosis^a
F0–F2	2 (18)	7 (64)	2 (18)	0.31
F3–F4	7 (47)	6 (40)	2 (13)
AIDS previous diagnosis
No	6 (38)	7 (44)	3 (19)	0.72
Yes	3 (25)	7 (58)	2 (16)
HCV antiviral therapy
Yes	1 (25)	3 (75)	—	0.47
No	8 (33)	11 (46)	5 (21)

Liver biopsy sample was insufficient for staging in nine patients of whom two died during follow-up.

ESLD, end stage liver disease.

Immunological and virological outcomes

Immunological outcome was evaluated only in patients with CD4 cell count at time of liver biopsy above 200 cells/mm³ (N=374). Eighty-four patients (22.5%) had a CD4 cell count <200 cells during follow-up. No independent risk factors for achievement of the immunological endpoint were found. Virological outcome was evaluated only in patients with HIV RNA at time of liver biopsy <400 copies/ml (N=262). One hundred and ten patients (42%) had an HIV RNA above 400 copies/ml during follow-up. The Cox multivariate analysis showed that age ≤38 years (HR: 2.29, 95% CI: 1.19–4.41, p=0.013) and PI treatment at time of liver biopsy (HR: 3.27, 95% CI: 1.50–7.16, p=0.013) were the only independent risk factors for virological failure. These results were confirmed when the analysis included only patients who had been treated with the same antiretroviral family from time of liver biopsy performance to the end of follow-up [N=168 (45%); PI: N=108 (64%); NNRTI N=60 (36%)].

Discussion

We conducted a study to evaluate the incidence of mortality and the main causes of death in a cohort of HIV and HCV-coinfected patients selected to be treated with interferon and with known liver fibrosis stage, who were prospectively followed for 6.2 years. Our most important results are the low mortality rate observed and also the demonstration that end-stage liver disease and non-AIDS-related diseases are the most frequent causes of death in this cohort. Patients with advanced liver fibrosis and those with previous AIDS diagnosis were at higher risk of death from any cause. On the other hand, patients not receiving HCV antiviral therapy were also at higher risk of death in our cohort.

Mortality in individuals with HCV chronic infection has been described extensively in the past decade.^14

–18 Some cohort studies conducted in chronic HCV infection have shown a mortality rate of 0.8 to 1.5 per 100 patient-years in monoinfected patients followed between 5 and 14 years and 1.4 to 2.5 per 100 patient-years in HIV and HCV-coinfected patients followed between 5 and 6 years.^14
–16 In one of these studies, the proportion of related deaths in HCV-monoinfected patients was 6%, while this figure was 15% in the HIV/HCV-coinfected population after a median follow-up period of nearly 8 and 6 years, respectively.¹⁴ In the absence of case-control studies comparing the mortality between HCV-monoinfected and HIV/HCV-coinfected individuals, the above mentioned results suggest a higher incidence rate of mortality in HIV/HCV-coinfected than HIV/HCV-monoinfected patients.

Our results are consistent with those found in other studies¹⁹ but more similar to those observed in the HCV-monoinfected populations than in HIV/HCV coinfection. Several reasons can explain the hypothetical lower mortality rate and also the low liver decompensation rate in our cohort, but the most important is probably the criteria used to select our study population. In our study, patients with comorbidities, including active alcohol or drug abuse, were not considered as candidates to receive interferon therapy and hence did not undergo liver biopsy, which was the main criterion to enter the cohort. The exclusion of the patients with additional predictors of poor mortality outcomes due to the presence of concomitant diseases, advanced HIV infection, low adherence to antiretroviral medications, indirectly measured by the poor HIV replication suppression, or faster liver fibrosis progression due to active drug abuse prevents generalizing our results to the entire HIV/HCV-coinfected population. However, given the low incidence of HIV/HCV coinfection in individuals newly diagnosed with HIV infection in certain geographic areas, mainly in southern Europe due to the decrease in intravenous drug addiction as the major risk factor for acquisition of HIV infection²⁰ and the increased survival of patients with HIV/HCV coinfection diagnosed in the early 1990s due to HAART,^1
–3 the majority of HIV/HCV-coinfected patients who regularly attend outpatient clinical units of certain hospitals have characteristics similar to those of our study population, so our data are perfectly applicable to an important subgroup of patients with HIV/HCV coinfection. Other features of our study mainly related to the characteristics of the patient population such as the high CD4 cell count, the evaluation of liver fibrosis, necroinflammatory activity and steatosis according to the most commonly used histological scores, and also high antiretroviral treatment exposure contribute to explaining the differences between other studies and ours.

Despite the low mortality in this cohort, it is important to stress that the main cause of death was complications of end-stage liver disease (ESLD) followed by non-AIDS-defining cancers. ESLD has been shown to be the main cause of death in unselected cohorts of HIV/HCV-coinfected patients, as also occurs in our study.^3,4,19 On the other hand, the fact that six of our patients died from non-AIDS-related cancers is also consistent with the steady increase in the cumulative incidence of non-AIDS-defining cancers over time since HAART has been widely available.^21

–27

The fact that in our study liver fibrosis stage was an independent factor for death from any cause is noteworthy for various reasons. On the one hand, our findings add to current evidence recommending that antiretroviral therapy be started early in coinfected patients²⁸ to prevent progression to advanced forms of immunosuppression, which could irreversibly accelerate progression of liver fibrosis in some patients, though it is important to stress that this strategy makes no sense without control of other viral coinfections and promotion of abstinence from alcohol and other drugs that by themselves further increase liver damage. On the other, we should point out the need to include regular assessment of liver fibrosis by biopsy or other noninvasive methods such as elastography in the clinical management of HIV/HCV-coinfected patients to improve early detection of liver complications and add information that can help to decide the best time to start HCV treatment.²⁹

The finding in our study that prior AIDS diagnosis before liver biopsy performance was an independent factor for death does not appear to indicate only the association of severe immunosuppression with death from end-stage liver disease or the occurrence of new AIDS diseases, since only 25% and 17% of patients with a prior AIDS diagnosis die from liver complications or AIDS, respectively. Our study also shows that patients in our cohort who received HCV treatment had a lower mortality over the follow-up, though we were unable to demonstrate an association between the achievement of sustained virological response and lower mortality, as has been shown in other studies,⁷ probably due to the low number of patients who received HCV treatment. We also cannot rule out that this lower mortality was related more to a selection bias of patients who finally are treated who receive or do not receive interferon treatment.

Various studies of characteristics similar to ours have also shown that liver fibrosis does not affect virological or immunological response of HIV/HCV-coinfected patients. In our study, patients who were receiving PIs at the time of liver biopsy or who were treated with this drug family during the follow-up had a greater risk of virological failure than those who received nonnucleoside analogs. Probably the fact that more than 60% of the patients who were receiving PIs were receiving first-generation PIs associated with high rates of adverse effects that required them to discontinue treatment could have influenced these results.

In summary, our study indicates that clinical management of this subgroup of patients should be focused on preventing advanced grades of immunosuppression and control of other factors that accelerate progression of fibrosis and on screening of complications of end-stage liver disease and non-AIDS-related cancers. Specifically HIV/HCV-coinfected patients with advanced liver fibrosis and with previous AIDS diagnosis are candidates for more intensive clinical management. Despite the limitations of our study but taking into account the benefit of HCV antiviral therapy in our cohort, this treatment should be offered at least to patients with poor predictors of survival.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Palella

Jr , Delaney

, Moorman

, Loveless

, Fuhrer

, Satten

et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. N Engl J Med, 1998; 338:853–860.

Mocroft

, Brettle

, Kirk

, Blaxhult

, Parkin

, Antunes

et al. Changes in the cause of death among HIV positive subjects across Europe: Results from the EuroSIDA study. AIDS, 2002; 16:1663–1671.

Weber

, Sabin

, Friis-Møller

, Reiss

, El-Sadr

, Kirk

et al. Liver-related deaths in persons infected with the human immunodeficiency virus: The D:A:D study. Arch Intern Med, 2006; 166,15:1632–1641.

Salmon-Ceron

, Rosenthal

, Lewden

, Bouteloup

, May

, Burty

et al. Emerging role of hepatocellular carcinoma among liver-related causes of deaths in HIV-infected patients: The French national Mortalité 2005 study. J Hepatol, 2009; 50:736–745.

Qurishi

, Kreuzberg

, Lüchters

, Effenberger

, Kupfer

, Sauerbruch

et al. Effect of antiretroviral therapy on liver-related mortality in patients with HIV and hepatitis C virus coinfection. Lancet, 2003; 362:1708–1713.

Merchante

, Girón-González

, González-Serrano

, Torre-Cisneros

, García-García

, Arizcorreta

et al. Survival and prognostic factors of HIV-infected patients with HCV-related end-stage liver disease. AIDS, 2006; 20:49–57.

Berenguer

, Alvarez-Pellicer

, Martinez

, López-Aldeguer

, Von-Wichmann

, Quereda

et al. Sustained virological response to interferon plus ribavirin reduces liver-related complications and mortality in patients coinfected with human immunodeficiency virus and hepatitis C virus. Hepatology, 2009; 50:407–413.

Pineda

, Aguilar-Guisado

, Rivero

, Girón-González

, Ruiz-Morales

, Merino

et al. Natural history of compensated hepatitis C virus-related cirrhosis in HIV-infected patients. Clin Infect Dis, 2009; 49:1274–1282.

Marcellin

, Pequignot

, Delarocque-Astagneau

, Zarski

, Ganne

, Hillon

et al. Mortality related to chronic hepatitis B and chronic hepatitis C in France: Evidence for the role of HIV coinfection and alcohol consumption. J Hepatol, 2008; 48:200–207.

10.

Poynard

, Bedossa

, Opolon

. Natural history of liver fibrosis progression in patients with chronic hepatitis C. Lancet, 1997; 349:825–832.

11.

Scheuer

. Classification of chronic hepatitis: A need for reassessment. J Hepatol, 1991; 13:372–374.

12.

Brunt

, Janney

, Di Bisceglie

, Neuschwander-Tetri

, Bacon

. Nonalcoholic steatohepatitis: A proposal for grading and staging the histological lesions. Am J Gastroenterol, 1999; 94:2467–2474.

13.

Torriani

, Rodriguez-Torres

, Rockstroh

, Lissen

, Gonzalez-García

, Lazzarin

et al. Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection in HIV-infected patients. N Engl J Med, 2004; 351:438–450.

14.

Walter

, Thein

, Amin

, Gidding

, Ward

, Law

et al. Trends in mortality after diagnosis of hepatitis B or C infection: 1992–2006. J Hepatol, 2011; 54:879–886.

15.

El-Kamary

, Jhaveri

, Shardell

. All-cause, liver-related, and non-liver-related mortality among HCV-infected individuals in the general US population. Clin Infect Dis, 2011; 53:150–157.

16.

Younossi

, Stepanova

. Hepatitis C virus infection, age, and Hispanic ethnicity increase mortality from liver cancer in the United States. Clin Gastroenterol Hepatol, 2010; 8:718–723.

17.

Omland

, Krarup

, Jepsen

, Georgsen

, Harritshøj

, Riisom

et al. Mortality in patients with chronic and cleared hepatitis C viral infection: A nationwide cohort study. J Hepatol, 2010; 53:36–42.

18.

Wise

, Bialek

, Finelli

, Bell

, Sorvillo

. Changing trends in hepatitis C-related mortality in the United States, 1995–2004. Hepatology, 2008; 47:1128–1135.

19.

Pineda

, García-García

, Aguilar-Guisado

, Ríos-Villegas

, Ruiz-Morales

, Rivero

et al. Clinical progression of hepatitis C virus-related chronic liver disease in human immunodeficiency virus-infected patients undergoing highly active antiretroviral therapy. Hepatology, 2007; 46:622–630.

20.

Pérez Cachafeiro

, Del Amo

, Iribarren

, Salavert Lleti

, Gutiérrez

, Moreno

et al. Decrease in serial prevalence of coinfection with hepatitis C virus among HIV-infected patients in Spain, 1997–2006. Clin Infect Dis, 2009; 48:1467–1470.

21.

Data Collection on Adverse Events of Anti-HIV drugs (D:A:D) Study Group. Smith

, Sabin

, Lundgren

, Thiebaut

, Weber

et al. Factors associated with specific causes of death amongst HIV-positive individuals in the D:A:D Study. AIDS, 2010; 24:1537–48.

22.

Kesselring

, Gras

, Smit

, van Twillert

, Verbon

, de Wolf

et al. Immunodeficiency as a risk factor for non-AIDS-defining malignancies in HIV-1-infected patients receiving combination antiretroviral therapy. Clin Infect Dis, 2011; 52:1458–1465.

23.

Shiels

, Pfeiffer

, Gail

, Hall

, Li

, Chaturvedi

et al. Cancer burden in the HIV-infected population in the United States. J Natl Cancer Inst, 2011; 103:753–762.

24.

Simard

, Pfeiffer

, Engels

. Cumulative incidence of cancer among individuals with acquired immunodeficiency syndrome in the United States. Cancer, 2011; 117:1089–1096.

25.

Zucchetto

, Suligoi

, De Paoli

, Pennazza

, Polesel

, Bruzzone

et al. Excess mortality for non–AIDS-defining cancers among people with AIDS. Clin Infect Dis, 2010; 51:1099–1101.

26.

Simard

, Pfeiffer

, Engels

. Spectrum of cancer risk late after AIDS onset in the United States. Arch Intern Med, 2010; 170:1337–1345.

27.

Reekie

, Kosa

, Engsig

, Monforte

, Wiercinska-Drapalo

et al. Relationship between current level of immunodeficiency and non-acquired immunodeficiency syndrome-defining malignancies. Cancer, 2010; 116:5306–5315.

28.

Thompson

, Aberg

, Cahn

, Montaner

, Rizzardini

, Telenti

et al. Antiretroviral treatment of adult HIV infection: 2010 recommendations of the International AIDS Society-USA panel. JAMA, 2010; 304:321–333.

29.

Bräu

, Salvatore

, Ríos-Bedoya

, Fernández-Carbia

, Paronetto

, Rodríguez-and Orengo

. Slower fibrosis progression in HIV/HCV-coinfected patients with successful HIV suppression using antiretroviral therapy. J Hepatol, 2006; 44:47–55.