| |
Coinfection Should Be Addressed Immediately
|
| |
| |
'Survival and prognostic factors of HIV-infected patients with HCV-related end-stage liver disease"
AIDS: 2 January 2006
Merchante, Nicolasa; Giron-Gonzalez, Jose Ab; Gonzalez-Serrano, Mercedesc; Torre-Cisneros, Juliand; Garcia-Garcia, Jose Aa; Arizcorreta, Anab; Ruiz-Morales, Josefac; Cano-Lliteras, Pilard; Lozano, Fernandoa; Martinez-Sierra, Carmenb; Macias, Juana; Pineda, Juan Aa; for the Grupo Andaluz para el Estudio de las Enfermedades Infecciosas (GAEI)
From the aUnidad de Enfermedades Infecciosas, Servicio de Medicina Interna, Hospital Universitario de Valme, Sevilla
bUnidad de Enfermedades Infecciosas, Servicio de Medicina Interna, Hospital Universitario Puerta del Mar, Cadiz
cUnidad de Enfermedades Infecciosas, Servicio de Medicina Interna, Hospital Universitario Virgen de la Victoria, Malaga
dSeccion de Enfermedades Infecciosas, Hospital Universitario Reina Sofia, Cordoba, Spain.
".....In summary, this study demonstrates that survival of HIV/HCV-co-infected patients with decompensated cirrhosis is very poor. Immunosuppression seems to play a key role in the outcome of liver disease. In contrast, HAART reduces liver-related mortality in this setting. If these data are confirmed, HAART should be strongly recommended in these patients. Although no specific recommendations can be made only on the basis of this study, HAART probably should be started earlier in HIV/HCV-co-infected patients than in the HIV-monoinfected population. In fact, preventing advanced immunosuppression may have an important impact in avoiding liver disease progression....." note from Jules Levin: (1) everyone with HIV should be tested for HCV (and HBV); (2) proper diagnostic & lab tests should be performed to evaluate liver disease in people with HCV or HBV; (3) treatment with peginterferon/RBV should be considered; (4) HAART should be considered to be started perhaps earlier in coinfected patients since depleted immune system appears to play a role in faster progression of HCV and perhaps HBV.
Objective: To find the survival and the predictors of death of HIV-infected patients with hepatitis C virus (HCV)-related end-stage liver disease (ESLD).
Design and methods: A prospective cohort study set in the infectious diseases units of four tertiary care public hospitals in Andalucia, Spain. From a multicentric cohort of 2664 HIV/HCV-co-infected patients, all consecutive patients with HCV-related cirrhosis who presented with the first hepatic decompensation from January 1997 to June 2004 were followed-up and 153 patients were included. The survival and the demographic, HIV-related and liver-related factors associated with death were evaluated.
Results:
Ninety-five (62%) patients died during the follow-up. In 79 (85%) individuals, the cause of death was liver related. The median survival time was 13 months. Independent predictors of survival were Child score [hazard ratio (HR), 1.2; 95% confidence interval (CI), 1.08-1.37; P = 0.001], CD4+ cell count at decompensation lower than 100 cells/ml (HR, 2.48; 95% CI, 1.52-4.06; P < 0.001) and hepatic encephalopathy as the first hepatic decompensation (HR, 2.45; 95% CI, 1.41-4.27; P = 0.001). HAART was prescribed to 101 (66%) patients. The cumulative probability of survival in patients under HAART was 60% at 1 year and 40% at 3 years, versus 38 and 18%, respectively, in patients not treated with HAART (P < 0.0001). The HR (95% CI) of death in patients on HAART was 0.5 (0.3-0.9), (P = 0.03).
Conclusions: The survival of HIV/HCV-co-infected patients with ESLD is extremely poor. Immunosuppression and markers of severe liver disease predict liver-related mortality in these patients. HAART seems to be associated with a reduced liver-related mortality.
DISCUSSION
Discussion TOP
This study shows that the survival of HIV-infected patients harbouring HCV-related ESLD is extremely poor, with roughly half of the patients dying within 1 year following the first hepatic decompensation. Predictors of survival in this population are the markers of severe liver disease, such as Child-Pugh score or HE. Moreover, the degree of immunosuppression, as measured by the CD4 cell count at the time of the first hepatic decompensation, was another strong predictor of liver-related mortality. In agreement with that, HAART seems to reduce the mortality due to ESLD.
Independent factors associated with liver mortality in the present study were some markers of liver disease severity, such as Child-Pugh score and HE. It is well known that HE has a negative impact on the survival of cirrhotic subjects [12,13], including those with HCV-related cirrhosis [6,14]. Moreover, the frequency of HE as both the first event and the cause of death was higher in HIV-co-infected patients in a previous retrospective survey, in which some of the patients analyzed here were also included [6]. According to this, HE was the leading cause of mortality among HIV/HCV-co-infected patients with ESLD in the present study. The reasons for this increased incidence of HE in the HIV-co-infected population are unclear. Consumption of depressors of central nervous system, which is frequent among HIV-infected individuals, or a higher frequency of bacterial infections could explain these differences. However, whether these topics contribute to the development of HE more frequently in the HIV-co-infected population needs to be elucidated.
As stated above, Child-Pugh score was another independent predictor of survival in the present study. Surprisingly, MELD score was not associated with survival after the multivariate analysis. Although MELD score is a good predictor of mortality in patients with chronic liver disease [15], its prognostic worth in HIV-infected patients with ESLD has not been evaluated. According to our results, the Child-Pugh score has a higher prognostic value than MELD score to predict mortality in HIV-infected patients with decompensated cirrhosis when considered together. These differences are probably due to the fact that MELD score does not include the presence of HE, a frequent adverse clinical outcome with an ominous significance in this population.
The present study has a few limitations. Although the study cohort was prospectively designed and followed-up to assess mortality, there were no fixed ART interventions in the patients who developed ESLD. Thus, the analysis of the impact of HAART on liver-related mortality was performed retrospectively. For this reason, potential bias due to lack of homogeneity in patients management can not be excluded, as HAART could be more frequently prescribed to patients with a better clinical situation or those more compliant with the follow-up schedule. Due to this, the results of this study concerning the effect of HAART on liver-related mortality in HIV/HCV-co-infected patients with ESLD should be interpreted cautiously. However, the convergence of results showing benefits of HAART in reducing the liver-related mortality in such a population [7,8] suggest that the data reported here reflect the clinical reality. Furthermore, the effect of HAART on survival was independent of the CD4 cell count in the multivariate analyses, which suggest that this bias should not have been important. On the other hand, many physicians prescribed HAART to the patients of this cohort. Prescriptions were free, provided that they complied with the most commonly used guidelines, mainly the Department of Health and Human Services (DHHS) guidelines. This fact resulted in a wide variety of regimens, which prevented the analysis of the relationship between specific drugs or combinations and survival. Only a prospective study with randomized allocation of patients to HAART and control groups would avoid these problems. This kind of study would be ethically acceptable only in patients with asymptomatic HIV infection and high CD4 cell count, which is a bias itself, along with the fact that it would take years until an answer is given. Since survival of HIV/HCV-co-infected patients after the first hepatic decompensation is short, this answer is urgently needed.
This is the first work to our knowledge that has prospectively evaluated the survival of HIV/HCV-co-infected patients with ESLD. As stated above, the probability of survival was as low as 50% at 1 year after the first hepatic decompensation. This prognosis was dramatically worse in the absence of HAART. In this sense, less than 20% of patients who did not receive ART were alive after 3 years following the first hepatic decompensation in the present study. Due to this, therapy against HCV infection should be a priority. In fact, the combination of pegylated interferon plus ribavirin leads to a substantial rate of sustained virological responses or transient control of HCV viremia during therapy [16]. Both circumstances are associated with a reduced incidence of decompensated liver disease, HCC and liver-related deaths [17]. Although HCV therapy should be given as early as possible, in order to prevent ESLD, it might be also worth giving it in patients with advanced liver disease [18].
As stated above, the positive impact of HAART on the liver-related mortality of HIV/HCV-co-infected patients with ESLD found in the present study is in agreement with that found in previous surveys that have evaluated the effect of HAART on the clinical course and outcome of HCV-related liver disease in HIV-infected subjects [7,8]. However, they were retrospective studies based on HIV/HCV-co-infected populations irrespective of the stage of liver disease. For this reason, the specific effect of HAART in ESLD was not analysed in these studies. Moreover, in the Bonn cohort, 81% of the individuals were haemophiliacs [7]. This extreme does not represent the typical picture of the HIV/HCV-co-infected populations, mainly in terms of treatment adherence.
The favourable effect of HAART was independent in our study of a marker of good control of HIV disease, such as achieving an undetectable plasma HIV viral load. This suggests that ART itself, rather than its virological consequences, delays the evolution of liver disease. This figure may be explained on the basis of the immune reconstitution associated with HAART. In fact, immunosupression has been associated with the evolution of chronic hepatitis C in the setting of HIV co-infection, as low CD4 cell count correlates with faster fibrosis progression [1,19,20]. In this regard, CD4 cell count lower than 100cells/ml at the first hepatic decompensation were associated with increased liver-related mortality in the present work. HAART induces promptly immune system recovery, even in the absence of complete virological response [21]. Consequently, HAART-associated immune restoration could positively influence the evolution of liver disease via CD4 cell recovery, regardless of whether complete virological control is achieved.
Liver transplantation has recently become a proper treatment option in HIV-infected patients with HCV-related ESLD [9,10]. However, the protocols currently used to include HIV-infected patients in liver transplantation programmes consider the same hepatic criteria as those for the general population [9,22], as these criteria are based on survival studies conducted in HIV-uninfected patients. As survival of HIV-infected patients with decompensated cirrhosis is extremely short, specific protocols for this population aimed to identify high-risk patients are needed. According to our results, low CD4 cell count, Child score and the presence of HE are factors that must be taken into account to decide the appropriate transplantation timing.
In summary, this study demonstrates that survival of HIV/HCV-co-infected patients with decompensated cirrhosis is very poor. Immunosuppression seems to play a key role in the outcome of liver disease. In contrast, HAART reduces liver-related mortality in this setting. If these data are confirmed, HAART should be strongly recommended in these patients. Although no specific recommendations can be made only on the basis of this study, HAART probably should be started earlier in HIV/HCV-co-infected patients than in the HIV-monoinfected population. In fact, preventing advanced immunosuppression may have an important impact in avoiding liver disease progression.
|
|
| |
| |
|
|
|
