Is the increased risk of liver enzyme elevation in patients co-infected with HIV and hepatitis virus greater in those taking antiretroviral therapy?
Cicconi, Paolaa; Cozzi-lepri, Alessandrob; Phillips, Andrewb; Puoti, Massimoc; Antonucci, Giorgiod; Manconi, Paolo Ee; Tositti, Giuliaf; Colangeli, Vincenzog; Lichtner, Miriamh; Monforte, Antonella d'Arminioa; for the ICoNA Study Group
From the aClinic of Infectious Diseases and Tropical Medicine University of Milan, Milan, Italy
bDepartment of Primary Care and Population Sciences, Royal Free and University College Medical School, London UK
cClinic of Infectious and Tropical Diseases, University of Brescia, Rome
dINMI L Spallanzani, Rome
eClinic of Infectious Diseases, University of Cagliari
fDepartment Infectious Diseases Vicenza Hospital, Vicenza
gDepartment Infectious Diseases, University of Bologna, Italy.
hDepartment of infectious Diseases, La Sapienza University, Rome, Italy.
Received 24 January, 2006
Revised 26 October, 2006
Accepted 7 November, 2006
Correspondence to Dr Paola Cicconi, Department of Infectious Diseases and Tropical Medicine University of Milan, Ospedale S Paolo, Via A. Rudinì, 8, 20142 Milano, Italy. E-mail: firstname.lastname@example.org
Objectives: To investigate if the risk of liver enzyme elevation (LEE) in HIV/hepatitis B or C (HBV, HCV) co-infection is altered by HAART (two or more drugs).
Methods: Analysis comprised HIV-positive patients in the ICoNA study without acute hepatitis who had ≥ 1 positive HCV antibody test and > 1 positive HBV surface antigen test. LEE was defined as > 5× baseline alanine aminotransferase (ALT) or > 3.5× baseline if the baseline was > 40 IU/l. Analysis used Poisson regression with generalized estimating equation correction to examine HBV or HCV co-infection, use of HAART, baseline ALT and demographics as LEE predictors.
Results: Of the 5272 patients, 47.6% were co-infected with HCV/HBV; 29.9% were female and 39% were intravenous drug users. There were 275 episodes of LEE during 18 259 person-years follow up. Taking HAART did not significantly increase risk of LEE [adjusted relative risk (RR), 1.19; 95% confidence interval (CI), 0.81–1.75; P = 0.37]. Co-infection increased the risk of LEE (adjusted RR, 5.07; 95% CI, 3.47–7.48; P < 0.001), with no significant differences if taking HAART (adjusted RR, 4.99; 95% CI, 3.38–7.37) or not (adjusted RR, 6.02; 95% CI, 2.02–17.98) (P = 0.74 for interaction). Females were at lower risk of LEE than males (adjusted RR, 0.59; 95% CI, 0.42–0.83; P = 0.02).
Conclusions: HIV and HBV/HCV co-infection per se is associated with increased risk of LEE that is not modified by HAART. The recommendation for caution in HAART use in co-infected patients, simply based on a high rate of LEE in people on therapy, may be questionable.
Liver enzyme elevation (LEE) has been frequently reported as a side effect of HAART [1–5]. There is evidence from randomized trials [6–7], supported by analysis of observational studies [8–12], that some antiretroviral drugs are more likely to cause LEE than others, but the risk of LEE varies between studies, as does the relative contribution of individual antiretroviral agents. This may be partly a consequence of variation in the definitions of hepatoxicity/LEE used [6,13,14].
Co-infection with either hepatitis B (HBV) or hepatitis C (HCV) has been frequently reported as a major risk factor for LEE after initiation of HAART [5,13–19]. These findings have led some authors to postulate that there could be enhanced toxicity of antiretroviral drugs in the presence of HCV or HBV co-infection and to conclude that clinicians should be careful in treating co-infected patients with antiretroviral drugs . To our knowledge, none of these previous studies has shown that the increase in risk for LEE in co-infected patients compared with HIV infection alone is higher still in those also taking antiretroviral therapy.
The aim of this study was to investigate this issue within the framework of the Italian Cohort of Naive for Antiretrovirals (ICoNA) study, an Italian observational cohort of HIV-infected patients who have been enrolled when naive for antiretroviral therapy and followed up over time in 69 treatment centres .
The ICoNA study
The ICoNA study is an Italian multicentre prospective observational study of HIV-1-positive persons that was started in 1997. Eligible patients are those who, for whatever reason, are naive to antiretroviral drugs at the time of enrolment. In all cases, the study is observational. Demographic, pre-enrolment, clinical and laboratory data and information on the specific therapies are collected for all participants and recorded online (database can be found at http:/www.icona.org). All data are updated at the occurrence of any clinical event and, in their absence, at least every 6 months. Immunovirological parameters and serologic tests for HCV [HCV-antibody (anti-HCV)] and HBV [HBV surface antigen (HBsAg), antibody to HBV surface antigen (HBsAb)] are part of the 6-monthly routine screening available for all the patients; in contrast, determinations of HCV RNA, HBV core antibody and HBV envelope antigen (and the corresponding antibody) are available for a subset of patients of the cohort but are not systematically recorded.
All patients in ICoNA with at least one determination of anti-HCV and at least two determinations of HBsAg were included. Patients were considered to have a chronic HCV infection when anti-HCV was present at any time during the follow-up, and a chronic HBV infection when HBsAg could be detected in plasma on two occasions at least 6 months apart at any time during the follow-up. Patients with signs and symptoms of acute hepatitis were excluded. HIV-positive patients who were also positive for anti-HCV and/or with HBV infection were grouped together as ‘co-infected’ patients in the main analysis.
In accordance with AIDS Clinical Trials Group criteria, severe LEE (toxicity grade 3) was defined as alanine aminotranferase (ALT) 5× upper limit of normal (ULN: 40 IU/l) . If baseline ALT was > 40 IU/l, LEE was considered to occur when patients experienced a value > 3.5× baseline ALT.
Times were recorded for the first event-defining LEE of at least grade 3 in the study population while naive for antiretroviral therapy and the time to the first event after start of HAART (defined as a regimen containing more than two antiretroviral drugs). Patients who had started antiretroviral therapy with one or two drugs were excluded as the analysis only considered those who started HAART from a therapy-naive state.
Person-years of follow-up were calculated for the various risk states. Person-years at risk were calculated from the date of recruitment in ICoNA until the last available ALT measure, the date of starting HAART or the first LEE-defining event of at least grade 3, whichever occurred first. Person-years were recounted after the start of HAART. Again, the follow up was censored at the last available ALT measure or the first LEE-defining event of at least grade 3, whichever occurred first. Person-years spent on treatment interruption were included in the analysis but patients were still classified as on treatment using an intent-to-treat approach.
A Poisson regression model with generalized estimating equation correction was fitted to account for the fact that one patient could experience repeated events (up to two: one before HAART and one after starting HAART) in order to identify the most important predictors for the development of LEE; covariates included in the model were age, body mass index, African origin, risk factor for HIV infection, co-infection, current use of HAART, baseline ALT (considered as the entry value until the start of HAART and as the ALT value at HAART for the remaining duration of follow-up). In addition, the same model was used to test formally for the presence of interaction between the co-infection and the use of HAART in order to assess whether the risk of LEE in co-infected patients differed between those on therapy and those HAART naive. A sensitivity analysis was conducted using the more stringent endpoint of mild LEE (grade 2 liver toxicity), defined as ALT elevation of at least 2.6× ULN . A further analysis was also conducted using a more stringent endpoint: very severe LEE (grade 4 liver toxicity), defined as ALT elevation of at least 10.0× ULN . If baseline ALT level was > ULN, LEE was defined when patients experienced a value > 5.0× their baseline ALT.
In order to evaluate whether the findings were affected by the choice of a cut-off to define ALT elevation, the changes of ALT from enrolment were also analysed as a continuous variable. The mean ALT values and the mean changes of ALT from enrolment and from start of HAART were calculated and plotted against fixed time points. The ALT value at each time point (every 6 months) was selected by taking a window of ±1 month; the value that was nearest to the mid-point was chosen from among the ALT measurements during this interval. A mixed regression model was fitted in order to examine the general trend in the changes in ALT, taking into account the correlation between ALT measurements coming from the same individual; HAART exposure and co-infection were investigated as factors influencing the ALT change and the interaction between the two factors was formally tested.
The study comprised 5272 patients of ICoNA with at least one determination of anti-HCV and at least two determinations of HBsAg: 47.6% were co-infected; 39% were infected with HIV by intravenous drug injection. Among those who were co-infected, 85.6% were HIV/HCV, 7.7% were HIV/HBV and 6.8% were HIV/HCV/HBV. Median age was 35 years [interquartile range (IQR), 31–40]. The median ALT of the entire population was 31 IU/l (IQR, 20–54) at enrolment, 32 IU/l (IQR, 20–56) at start of HAART and 28 IU/l (IQR, 18–52) after 1 year of HAART exposure. HIV-mono-infected patients showed significant lower levels of ALT compared with co-infected patients at each of these time points. The median levels of ALT in the HIV mono-infected and co-infected groups, respectively, were 23 IU/l (IQR, 17–35) and 45 IU/l (IQR, 28–79) (P < 0.0001) at baseline, 25 IU/l (IQR, 18–37) and 45 IU/l (IQR, 27–73) (P < 0.0001) at start of HAART, and 21 IU/l (IQR, 16–32) and 48 IU/l (IQR, 27–79) (P < 0.0001) after 1 year of HAART.
Of these 5272 patients, 3671 (69.6%) started HAART at some point during the follow-up and they were defined as to be on treatment from that time. The proportion of co-infected patients among those who started HAART (48.1%) was slightly lower than the proportion among those who remained ART naive (53.5%) (P = 0.003). In HCV/HIV-infected patients, after adjusting for current CD4 cell count, there was no evidence that the probability of starting HAART was higher in patients with current ALT > 40 IU/I compared with those with ALT of 0–40 IU/I [relative risk (RR), 0.96; 95% confidence interval (CI) 0.82–1.11; P = 0.57].
Patients spent 36% of their person years pre-therapy, 24.3% of the follow up on two nucleoside reverse transcriptase inhibitors (NRTI) plus one protease inhibitor, 17.7% on two NRTI plus one non-nucleoside reverse transcriptase inhibitor, 4.1% on two NRTI plus a boosted protease inhibitor, and 3.4% on three NRTI. Person years spent on treatment interruption were the 5.8% of the follow up (Table 1).
There was no evidence that the frequency of ALT measurements in HIV-mono-infected patients differed from that in co-infected patients: median/year 2.43 (IQR, 2.10–3.06) and 2.36 (IQR, 1.96–2.97), respectively.
Overall, there were 275 cases of grade 3 LEE [three patients (1%) experienced two events] in 18 250 person-years (15.1/1000 person-years follow-up; 95% CI, 13.4–17.0). None of the events led to drug discontinuation. The crude incidence rate of LEE among naive HIV-mono-infected patients was 3.9/1000 person-years (95% CI, 1.5–10.6). In the population of HIV-mono-infected participants, the incidence rate estimated for time spent on HAART was 5.6/1000 person-years (95% CI, 4.2–7.5). The rate of LEE among co-infected patients was high both for the time in which patients were antiretroviral drug naive (22.1/1000 person-years; 95% CI, 14.8–32.9) and for the time spent on HAART (25.4/1000 person-years; 95% CI, 22.1–29.2). In the time window 0–6 months from starting HAART, the rates of severe LEE were 18.2/1000 person-years (95% CI, 14.2–23.3) in the entire population, 5.2/1000 person-years (95% CI, 2.7–10.1) in the mono-infected patients and 31.4/1000 person-years (95% CI, 23.9–14.1) in the co-infected patients. Over the period 6–12 months from starting HAART, the overall rate was 27.7/1000 person-years (95% CI, 20.1–38.1): 7.9 (95% CI, 3.6–17.7) in mono-infected and 51.9/1000 person-years (95% CI, 36.8–73.5) in co-infected. After 12 months of HAART exposure, the crude incidences of LEE during follow-up were 15.4/1000 person-years (95% CI, 12.8–18.5), 7.0/1000 person-years (95% CI, 4.9–10.1) and 25.9/1000 person-years (95% CI, 20.9–32.0) in the entire population, the mono-infected patients and the co-infected patients, respectively.
The overall incidence rates according to the time periods considered did not show a significant linear trend (Pearson correlation for linear trend, P = 0.16) but there was evidence for higher rates in the 6–12 month period compared with the other periods (Cochran–Armitage test for deviation from linear, P = 0.005). A similar trend was seen in co-infected people when studied separately (Pearson correlation, P = 0.17; Cochran–Armitage test, P = 0.002). In contrast, in mono-infected patients, there was no evidence for a trend (Pearson correlation, P = 0.50; Cochran–Armitage test, P = 0.57).
Co-infection, compared with mono-infected, was strongly associated with the risk of LEE both in the univariable model (RR, 4.56; 95% CI, 3.36–6.19; P < 0.0001) and in the multivariable model (adjusted RR, 5.07; 95% CI, 3.44–7.48; P < 0.0001; Table 2). Having started HAART was not associated with a significant increase in the risk of LEE compared with therapy-naive state (adjusted RR, 1.19; 95% CI, 0.81–1.75; P = 0.37). The risk of LEE was 40% lower in women than in men (adjusted RR, 0.59; 95% CI, 0.42–0.83; P = 0.002). Intravenous drug users were at significant higher risk of LEE compared with heterosexuals in the univariable model (RR, 2.98; 95% CI, 2.20–4.05; P < 0.0001); however, after adjusting for co-infection, their risk was no longer significantly different from heterosexuals (adjusted RR, 1.05; 95% CI, 0.72–1.52; P = 0.79). There was no evidence that the risk associated with being co-infected as compared with HIV mono-infected was higher during person-years on HAART (adjusted RR, 4.99; 95% CI, 3.38–7.37.) or pre-therapy (adjusted RR, 6.02; 95% CI, 2.02–17.98), after having adjusted for other potential confounders [interaction: P = 0.70 (univariable) and P = 0.74 (adjusted model)].
The adjusted RR values for LEE associated to HCV, HBV and HCV/HBV co-infections individually compared with HIV mono-infection during person years on HAART were, respectively, 4.17 (95% CI, 2.7–6.4), 5.95 (95% CI, 2.4–14.7) and 6.62 (95% CI, 3.9–11.3) and pre-therapy were, respectively, 4.46 (95% CI, 1.4–13.9), 14.26 (95% CI, 3.9–52.3) and 9.71 (95% CI, 1.8–52.5). Again, there was no evidence for a difference between these set of rates (interaction P = 0.90, P = 0.26 and P = 0.52 for HCV, HBV and HCV-HBV co-infections, respectively; Fig. 1).
The sensitivity analysis conducted considering a less-stringent criteria, mild LEE (grade 2 liver toxicity) as the outcome, showed comparable results: overall adjusted RR of co-infection versus mono-infection was 5.04 (95% CI, 4.1–6.28); the RR associated with co-infection among those receiving HAART was 4.92 (95% CI, 3.9–6.1) and among those off HAART was 6.36 (95% CI, 3.6–11.1), (interaction: P = 0.83). There was likewise no evidence for a significantly greater effect on ALT elevation by co-infection in those on HAART compared with those who were HAART naive in the analysis considering very severe LEE as the outcome. In patients on HAART, the overall adjusted RR of grade 4 liver toxicity for co-infection compared with mono-infection was 5.8 (95% CI, 3.4–9.9; P < 0.0001); the risk associated with co-infection among patients on HAART was 5.9 (95% CI, 3.4–10.4) and among HAART naive was 4.3 (95% CI, 1.1–16.2), (interaction: P = 0.62).
In order to evaluate whether these findings were affected by the use of a cut-off to define ALT elevation, ALT changes from baseline were also analysed (enrolment for the off-HAART period and pre-HAART value for the on-HAART period) as a continuous variable. The 6-monthly mean ALT values and mean changes from baseline are shown in Figs 2 and 3. Co-infected patients had higher mean ALT values compared with mono-infected patients at each time point over both the periods of pre-HAART and those of on-HAART (Fig. 2). During the pre-HAART period, there was no remarkable difference in mean ALT changes from baseline between mono- and co-infected patients; over the first 12 months of HAART, a mean increase of ALT from pre-HAART levels in co-infected patients and a mean decrease in mono-infected patients were observed (Fig. 3). According to the mixed regression model, the overall mean change in ALT values after starting HAART was 3.67 IU/l (95% CI, 2.33–5.01) lower than that estimated in pre-HAART period (P < 0.0001); in the co-infected population, the mean change was 5.77 IU/l (95% CI, 4.51–7.03) higher than that estimated in mono-infected patients (P < 0.0001) (interaction between co-infection and HAART: P = 0.0047).
LEE has frequently been described in patients receiving HAART, and in particular there is evidence that several antiretroviral drugs are more likely to cause LEE than others [6–12]. The majority of cases involve asymptomatic LEE, but severe liver diseases have been also reported [4,5] and HAART-related hepatotoxicity is of increasing concern in the management of HIV-positive patients. Some authors have suggested that the frequency of LEE may be increased in individuals co-infected with hepatitis viruses during HAART and that antiretroviral therapy is the main cause for this [13–19]. Also, frequent therapy interruption, potentially linked to LEE, has been reported especially in co-infected individuals . However, serum liver enzymes, as other biomarkers alone, cannot fully predict the risk of severe liver disease and they are complementary to biopsy . Therefore, there may be some over-concern regarding the increase of ALT levels during treatment with HAART. Furthermore, we believe that it needs to be clarified whether the higher rate of LEE in co-infected patients on HAART is because patients with HCV/HBV generally tend to have higher ALT levels or is one of the consequences of HAART. Most previous studies could not evaluate the relative impact of co-infection and antiretroviral treatment as they have been conducted on treated populations including intravenous drug users or co-infected patients, who tend to have higher levels of ALT in any case [4,5,19].
Our results suggest that HAART per se is not a risk factor for LEE and it seems not to modify the association between co-infection and the risk of LEE. The increased risk of raised ALT levels in co-infected patients appeared to be similar over periods of time in which patients were off or on treatment, suggesting, in terms of ALT elevation at least, that being co-infected is no more of a special concern for people on HAART than it is for therapy-naive patients. Of note, the proportion of person-years spent on boosted protease inhibitors was low in our cohort, but a considerable part of the follow up studied here was spent on treatments containing antiretroviral drugs that have been specifically associated with ALT elevation, such as nevirapine (16.9% of follow-up) or ritonavir (6.4%) [6,7]. We obtained similar findings after redoing the analysis using other definitions of LEE and after having excluded person-years spent on these drugs (data not shown). Other studies in patients treated with antiretroviral therapy  have shown that hepatitis co-infection has an effect on the risk of LEE that is independent of other co-factors, but to our knowledge, this is the first study that directly addresses the risk of LEE in antiretroviral drug-naive patients and showing that the use of HAART and the occurrence of HBV/HCV co-infection do not synergistically increase this risk.
Our main conclusion was supported by an analysis of the changes of ALT from enrolment (or from pre-HAART levels after the initiation of HAART) as a continuous variable. Co-infected patients had higher ALT levels than mono-infected patients irrespective of whether they were untreated or treated with HAART. However, there was evidence that the difference in ALT changes from baseline between co-infected and mono-infected patients was higher over periods in which they have been receiving HAART compared with the period when they remained untreated. This was particularly evident in the first 12 months of observation (Fig. 3) and it is consistent with a higher incidence of LEE events detected by our Poisson analysis. Nevertheless, the magnitude of the estimated difference in mean ALT changes (i.e. 12 IU/l) is not clinically significant, considering that the pre-HAART ALT levels were quite low both for mono-infected (25 IU/l; IQR, 18–37) and for co-infected (45 IU/l; IQR, 27–73) patients. The fact that, even in the co-infected population, ALT values were, overall, remarkably normal could potentially be explained by a delay of HAART initiation in people with high values of ALT. However, an analysis of the rates of starting HAART according to the latest value of ALT showed no evidence that those with > 40 IU/I were less likely to initiate HAART than those with normal values. A possible limitation of the analysis of the continuous outcome is the fact that a reduced number of patients contributed to the calculation beyond 12 months of observation. Therefore, it is possible that our graphs are not representing the trend in ALT in an average person enrolled in our study but that of a selected subset of individuals with more favourable ALT profiles. The multivariable analysis, however, appropriately accounts for the missing data.
One limitation of the study needs to be mentioned. Our study defined HCV infection on the basis of the presence of anti-HCV and, therefore, potentially included HCV-positive patients who were non-viraemic. However, only a small proportion of patients with HCV antibodies had undetectable HCV RNA in a sample of our cohort .
Even though hepatocellular damage cannot be proven without liver biopsy, and data on alcohol intake or other potential factors that might be associated with LEE were not available in this study, our results are important for the clinical management of co-infected patients. Several previous studies observed a higher proportion of co-infected patients experiencing LEE on HAART compared with HIV-mono-infected patients and concluded that (i) HAART was causing LEE, (ii) HAART was more likely to cause LEE in co-infected patients, and (iii) drugs more commonly associated with elevation of transaminase levels should be used with caution in co-infected individuals, [6–12]. Our data suggest that there is no increase in the risk of experiencing LEE on or off HAART and this was similar in HIV-mono-infected and co-infected patients. Males seemed to be at higher risk of developing LEE than females. It could be hypothesized that males had a higher rate of alcohol consumption or lower levels of therapy adherence than females, but the reason for this gender-associated increased risk of LEE needs to be better investigated, and to our knowledge there are no studies that have specifically addressed this issue.
In conclusion, our study shows that the co-infection with hepatitis viruses per se is associated with an increased risk of experiencing LEE that is not modified by the use of HAART. Longer follow-up is needed to confirm these findings and to evaluate the long-term prognosis of liver disease of patients who do experience LEE.
ICoNA is supported by an unrestricted educational grant from GlaxoSmithKline Italy.
The ICoNA Study Group comprises M. Montroni, G. Scalise, M.C. Braschi, A. Riva (Ancona); U. Tirelli, G. Di Gennaro. Bari: G. Pastore, N. Ladisa, G. Minafra (Aviano (PN)); F. Suter, C. Arici (Bergamo) F. Chiodo, V. Colangeli, C. Fiorini, O. Coronado (Bologna); G. Carosi, G.P. Cadeo, C. Torti, C. Minardi, D. Bertelli (Brescia); T. Quirino, S. Melzi (Busto Arsizio); P.E. Manconi, P. Piano (Cagliari); L. Cosco, A. Scerbo (Catanzaro); J. Vecchiet, M. D'Alessandro (Chieti); D. Santoro, L. Pusterla (Como); G. Carnevale, P. Citterio (Cremona); P. Viganò, M. Mena (Cuggiono); F. Ghinelli, L. Sighinolfi (Ferrara); F. Leoncini, F. Mazzotta, M. Pozzi, S. Lo Caputo (Firenze); G. Angarano, B. Grisorio, A. Saracino, S. Ferrara (Foggia); P. Grima, P. Tundo (Galatina (LE)); G. Pagano, G. Cassola, A. Alessandrini, R. Piscopo (Genova); M. Toti, S. Chigiotti (Grosseto); F. Soscia, L. Tacconi (Latina); A. Orani, P. Perini (Lecco); A. Scasso, A. Vincenti (Lucca); A. Chiodera, P. Castelli (Macerata); A. Scalzini, L. Palvarini (Mantova); M. Moroni, A. Lazzarin, A. Cargnel, G. Rizzardini, L. Caggese, A. d'Arminio Monforte, D. Repetto, A. Galli, S. Merli, C. Pastecchia, M.C. Moioli (Milano); R. Esposito, C. Mussini (Modena); N. Abrescia, A. Chirianni, C.M. Izzo, M. Piazza, M. De Marco, R. Viglietti, E. Manzillo, S. Nappa (Napoli); A. Colomba, V. Abbadessa, T. Prestileo, S. Mancuso (Palermo); C. Ferrari, P. Pizzaferri (Parma); G. Filice, L. Minoli, R. Bruno, S. Novati (Pavia); F. Baldelli, M. Tinca (Perugia); E. Petrelli, A. Cioppi (Pesaro); F. Alberici, A. Ruggieri (Piacenza); F. Menichetti, C. Martinelli (Pisa); C. De Stefano, A. La Gala (Potenza); G. Ballardini, E. Rizzo (Ravenna); G. Magnani, M.A. Ursitti (Reggio Emilia); M. Arlotti, P. Ortolani (Rimini); R. Cauda, F. Dianzani, G. Ippolito, A. Antinori, G. Antonucci, S. D'Elia, P. Narciso, N. Petrosillo, V. Vullo, A. De Luca, A. Bacarelli, M. Zaccarelli, R. Acinapura, P. De Longis, A. Brandi, M.P. Trotta, P. Noto, M. Lichtner, M.R. Capobianchi, F. Carletti, E. Girardi, P. Pezzotti, G. Rezza (Roma); M.S. Mura, M. Mannazzu (Sassari); P. Caramello, G. Di Perri, M.L. Soranzo, G.C. Orofino, I. Arnaudo, M. Bonasso (Torino); P.A. Grossi, C. Basilico (Varese); A. Poggio, G. Bottari (Verbania); E. Raise, F. Ebo (Venezia); F. De Lalla, G. Tositti (Vicenza); F. Resta, K. Loso (Taranto); A. Cozzi Lepri (London, UK).
1. Wit FW, Waverling GJ, Weel J, Jurriaans S, Longe J. Incidence and risk factors for severe hepatotoxicity associated with antiretroviral therapy
. J Infect Dis
2002; 186 July.
2. Nunez M, Lana R, Mendoza JL, Carbonero L, Soriano V. Risk factors for severe hepatic injury after introduction of highly active antiretroviral therapy. J Acquir Immune Defic Syndr 2001; 27:426–431.
3. Nunez M, Soriano V. Hepatotoxicity of antiretrovirals: incidence, mechanisms and management. Drug Safety 2005; 28:53–66.
4. Puoti M, Torti C, Ripamonti D, Castelli S, Zaltron S, Zanini B, et al
, for the HIV-HCV Co-Infection Study Group. Severe hepatotoxicity during combination antiretroviral treatment: incidence, liver histology, and outcome. J AIDS 2003; 32:259–326.
5. Saves M, Vandentorren S, Daucourt V, Marimoutou C, Dupon M, Couzigou P, et al
. Severe hepatic cytolysis: incidence and risk factors in patients treated by antiretroviral combinations. Aquitaine Cohort, France, 1996–1998. Groupe dEpidemiologie Clinique de Sida en Aquitaine (GECSA). AIDS 1999; 13:F115–F121.
6. Saves M, Vandentorren S, Daucourt V, Marimoutou C, Dupon M, Couzigou P, et al
, for the NN Study Team. Comparison of first-line antiretroviral therapy with regimens including nevirapine, efavirenz, or both drugs, plus stavudine and lamivudine: a randomised open-label trial, the 2NN Study. Lancet 2004; 363:1253–1263.
7. Cameron DW, Heath-Chiozzi M, Danner S, Cohen C, Kravcik S, Maurath C, et al
. Randomised placebo-controlled trial of ritonavir in advanced HIV-1 disease. The Advanced HIV Disease Ritonavir Study Group. Lancet 1998; 351:543–549.
8. Sulkowski M. Drug-induced liver injury associated with antiretroviral therapy that includes HIV-1 protease inhibitors. Clin Infect Dis 2004; 38:S90–S97.
9. Meraviglia P, Schiavini M, Castagna A, Vigano P, Bini T, Landonio S, et al
. Lopinavir/ritonavir treatment in HIV antiretroviral-experienced patients: evaluation of risk factors for liver enzyme elevation. HIV Med 2004; 5:334–343.
10. Ena J, Amador C, Benito C, Fenoll V, Pasquau F. Risk and determinants of developing severe liver toxicity during therapy with nevirapine- and efavirenz-containing regimens in HIV-infected patients. Int J STD AIDS 2003; 14:776–781.
11. Martinez E, Blanco JL, Arnaiz JA, Perez-Cuevas JB, Mocroft A, Cruceta A, et al
. Hepatotoxicity in HIV-1-infected patients receiving nevirapine-containing antiretroviral therapy. AIDS 2001; 15:1261–1268.
12. Dieterich DT, Robinson PA, Love J, Stern JO. Drug-induced liver injury associated with the use of nonnucleosid reverse-transcriptase inhibitors. Clin Infect Dis 2004; 38(Suppl 2):S80–S89.
13. den Brinker M, Wit FW, Wertheim-van Dillen PM, Jurriaans S, Weel J, van Leeuwen R, et al
. Hepatitis B and C virus co-infection and the risk for hepatotoxicity of highly active antiretroviral therapy in HIV-1 infection. AIDS 2000; 14:2895–2902.
14. Sulkowski M, Thomas D, Chasson RE, Moore RD. Hepatotoxicity associated with antiretroviral therapy in adults infected with human immunodeficiency virus and the role of hepatitis C or B virus infection. JAMA 2000; 283:74–80.
15. Sulkowski MS, Thomas DL, Mehta SH, Chaisson RE, Moore RD. Hepatotoxicity associated with nevirapine or efavirenz-containing antiretroviral therapy: role of hepatitis C and B infections. Hepatology 2002; 35:182–189.
16. Servin-Abad L, Molina E, Baracco G, Arosemena L, Regev A, Jeffers L, et al
. Liver enzymes elevation after HAART in HIV-HCV co-infection. J Viral Hepat 2005; 12:429–434.
17. Aranzabal L, Casado JL, Moya J, Quereda C, Diz S, Moreno A, et al
. Influence of liver fibrosis on highly active antiretroviral therapy-associated hepatotoxicity in patients with HIV and hepatitis C virus coinfection. Clin Infect Dis 2005; 40:588–593.
18. Nunez M, Rios P, Carbonero L, Perez-Olmeda M, Gonzalez-Lahoz J, Soriano V. Role of hepatitis C virus genotype in the development of severe transaminase elevation after the introduction of antiretroviral therapy. J Acquir Immune Defic Syndr 2002; 30:65–68.
19. Mehta SH, Netski D, Sulkowski MS, Strathdee SA, Vlahov D, Thomas DL. Liver enzyme values in injection drug users with chronic hepatitis C. Dig Liver Dis 2005; 37:674–680.
20. d'Arminio Monforte A, Pezzotti P, Arici C, Ippolito G, Rezza G, Angarano G, et al. Determinants of initiation of antiretroviral therapy (HAART) in an italian cohort of HIV-positive patients naive from antiretrovirals (ICoNA Study)
:99. Database can be found at http://www.icona.org
21. AIDS Clinical Trials Group. Table for Grading Severity of Adult Adverse Experiences
. Rockville, MD: National Institutes of Allergy and Infectious Diseases, Division of AIDS; 1992.
22. Macroft A, Phillips AN, Soriano V, Rockstroh J, Blaxhult A, Katlama C, et al
, for the Eurosida Study Group. Reasons for stopping antiretrovirals used in an initial highly active antiretroviral regimen: increased incidence of stopping due to toxicity or patient/physician choice in patients with hepatitis C coinfection. AIDS Res Hum Retroviruses 2005; 21:527–536.
23. Kelleher T, Mehta S, Bhaskar R, Sulkowski M, Astemborski J, Thomas D, et al
. Prediction of hepatic fibrosis in HIV/HCV co-infected patients using serum fibrosis markers: The SHASTA index. J Hepatol 2005; 43:78–84.
24. Aceti A, Pasquazzi C, Zechini B, De Bac C, for the LIVERHAHAART Group. Hepatotoxicity development during antiretroviral therapy containing protease inhibitors in patients with HIV: the role of hepatitis B and C virus infection. J Acquir Immune Defic Syndr 2002; 29:41–48.
25. Antonucci G, Girardi E, Cozzi-Lepri A, Capobianchi MR, De Luca A, Puoti M, et al
, for the HepaICoNA Study Group and the Italian Cohort of Naive for Antiretrovirals Study Group. Role of hepatitis C virus (HCV) viremia and HCV genotype in the immune recovery from highly active antiretroviral therapy in a cohort of antiretroviral naive HIV infected individuals. Clin Infect Dis 2005; 40:101–109.
Liver toxicity; HAART; HIV/HCV/HBV co-infection
© 2007 Lippincott Williams & Wilkins, Inc.
What does "Remember me" mean?
By checking this box, you'll stay logged in until you logout. You'll get easier access to your articles, collections,
media, and all your other content, even if you close your browser or shut down your
To protect your most sensitive data and activities (like changing your password),
we'll ask you to re-enter your password when you access these services.
What if I'm on a computer that I share with others?
If you're using a public computer or you share this computer with others, we recommend
that you uncheck the "Remember me" box.
Highlight selected keywords in the article text.
Data is temporarily unavailable. Please try again soon.
Readers Of this Article Also Read