Provided by NATAP
"During this year, death due to ESLD represented 28% of all non AIDS-related deaths in the HIV-positive population. These findings are in accordance with recent studies from Europe.....the progression to cirrhosis and HCC in patients with chronic HCV infection occurs over an average of two or three decades [5,6], potentially leading to a burden of disease over time......Our multi-centre, national study demonstrates that ESLD, predominantly due to HCV coinfection, is now a leading cause of mortality in the HIV-infected population. Five years after the introduction of HAART, the importance of liver disease as a cause of death has increased progressively"....(from Jules: if anything liver disease deaths & hospitalizations have gotten worse over time because coinfected response less well to peg/rbv and have a hard time with tolerability(although this study is from 2003 liver disease has gotten worse in coinfected in many ways since 2003).....HCV appears to cause inflammation & may contribute to other comorbidities....with the HIV aging patient population other comorbidities are emerging but liver disease remains a predominant cause of death & hospitalization because with aging liver disease progresses without successful treatment.....this recently published study linked to immediately below reports untreated HCV increases HIV comorbidities & consequent deaths thus these deaths or hospitalizations are NOT reported to be associated with HCV!)
"We found that failure to achieve an SVR was associated with increased risk of liver decompensation, hepatocellular carcinoma, liver transplantation, and liver-related death; findings consistent with our previous observations . Interestingly, we also found that failure to achieve an SVR was associated with an increased risk of HIV progression and non-liver-related mortality. Most non-liver-related deaths were due to non-AIDS-defining cancers, cardiovascular events, and bacterial infections. Of note, both the risk of non-liver-related death, and non-liver-related non-AIDS-related death was significantly higher for nonresponders than for responders after adjustment......may be explained by several factors, including immune activation, defective immunity, systemic inflammation, and liver disease itself"
Mortality due to hepatitis C-related liver disease in HIV-infected patients in France (Mortavic 2001 study)
AIDS: 15 August 2003
Rosenthal, Eric; Poiree, Marilyne; Pradier, Christiana; Perronne, Christianb; Salmon-Ceron, Dominiquec; Geffray, Loicd; Myers, Robert Pe; Morlat, Philippef; Pialoux, Gillesg; Pol, Stanislash; Cacoub, Patricei; for the GERMIVIC Joint Study Group
From the Service d'Hematologie Clinique, Hopital Archet, Nice, aCISIH, Hopital Archet, the bService des Maladies Infectieuses, Hopital Raymond Poincare, Garches, the cService de Medecine Interne, Groupe Hospitalier Cochin, Paris, the dService de Medecine Interne, Centre Hospitalier General Bisson, Lisieux, the eService d'Hepato-Gastro-enterologie, Groupe Hospitalier Pitie-Salpetriere, Paris, the fService de Medecine Interne, Hopital Saint-Andre, Bordeaux, the gService des Maladies Infectieuses, Hopital Tenon, the hService d'Hepatologie, Hopital Necker, and the iService de Medecine Interne, Groupe Hospitalier Pitie-Salpetriere, Paris, France. *Joint Study Group on Hepatitis C virus of the French National Society of Internal Medicine and the French Society of Infectious Diseases;
Objective: To determine mortality due to end-stage liver disease (ESLD) in a nationwide cohort of HIV-infected patients 5 years after the introduction of highly active antretroviral therapy (HAART) and to compare this with that observed before and during the early years of HAART.
Design: and methods: All departments of internal medicine and infectious diseases from the GERMIVIC Study Group prospectively recorded all deaths in HIV-infected patients during 2001. Sixty-five departments, following a total of 25 178 HIV-infected patients, participated in the study. Results were compared with those of previous surveys conducted using similar methodology in 1995 and 1997.
Results: Among 265 deaths observed during 2001, 129 (48.7%) were related to AIDS, 38 (14.3%) to ESLD, and 98 (36.7%) to other causes. Mortality due to ESLD represented 28% of non AIDS-related deaths; 36 of the 38 patients (95%) dying from ESLD had chronic hepatitis C virus (HCV) infection. In 2001, deaths due to ESLD (14.3%) were significantly more frequent than in 1995 (1.5%; P < 0.01) and 1997 (6.6%; P < 0.01). During this interval, the prevalence of hepatocellular carcinoma as a cause of death increased (1995, 4.7%; 1997, 11%; 2001, 25%; P < 0.05), as did alcohol consumption (P < 0.01).
Conclusions: In the post-HAART era, ESLD due to HCV is a growing cause of mortality in HIV-infected patients. Increased longevity attributable to HAART, and a higher prevalence of alcohol consumption, are probably involved in this trend.
Coinfection with hepatitis C virus (HCV) and HIV is common as both viruses are transmitted by parenteral routes, i.e., injecting drug use and blood transfusion. In the USA, it is estimated that 30% of the 800 000 HIV-infected individuals are coinfected with HCV [1,2]; similar rates have been reported in Western Europe . Among certain subgroups of HIV-infected patients, such as in injecting drug users and haemophiliacs, the prevalence of coinfection approaches 70-90% [2-4]. HCV infection leads to chronic hepatitis in 85% of HIV-negative patients; approximately 20% will eventually develop cirrhosis of the liver [5,6]. In HIV-positive individuals, chronic hepatitis C is more severe, particularly in those with advanced immunosuppression [7-10]. Regardless of HIV serostatus, time represents a critical factor in determining mortality due to end-stage liver disease (ESLD). In HIV-positive patients, longer survival observed since the introduction of highly active antiretroviral therapy (HAART) may permit the progression of HCV-related liver disease and increase mortality due to complications including liver failure and hepatocellular carcinoma (HCC). Moreover, HAART itself may accelerate the progression of HCV-related liver disease . Several studies have suggested an increased risk of mortality caused by ESLD in HIV-HCV coinfected patients [12-21]. However, most of these studies examined cohorts of haemophilic patients [12,13] or single-centre cohorts of HIV seropositive patients and compared mortality rates during the pre-HAART era with that during the early years of HAART.
Five years following the introduction of HAART, we have examined the mortality associated with HCV-related ESLD in a nationwide cohort of HIV-infected patients in France. The objective of the study was to assess trends in mortality by comparing current data to two previous surveys conducted by our study group in the pre-HAART era (1995) and during the first few years of this therapy (1997) .
As in our previous surveys , our current study analysed approximately half of the HIV-related deaths reported to the French Public Health Network (Institut de Veille Sanitaire) in 2001 . During this year, death due to ESLD represented 28% of all non AIDS-related deaths in the HIV-positive population. These findings are in accordance with recent studies from Europe showing the growing importance of liver disease as a cause of mortality in HIV-infected patients [24,25]. For example, Camino et al. reviewed the causes of death among 1600 HIV-infected patients in Spain during a 21-month period in 1998 and 1999 . Of the 44 deaths registered during this interval, liver disease was responsible for 25%. In another multicentre study from France, 422 deaths were reported among HIV-infected patients during the year 2000 . HCV coinfection was the most frequent non AIDS-related cause of death, representing 10% of all cases.
Combining our contemporary data with our previous surveys from 1995 and 1997 , has allowed us to evaluate temporal trends in the mortality of HIV-infected patients in France. Importantly, unlike previous studies addressing this issue [14-16,18,24], our data were derived from a large population of HIV-infected patients from a national network. All participating centres were involved in three consecutive surveys conducted with similar methodology. Although the annual incidence of death decreased between 1995 and 2001, mortality attributable to ESLD (expressed as a percentage of the total number of deaths) increased progressively from 1.5% in 1995, to 6.6% in 1997, and 14.3% in 2001. These findings confirm those of previous studies of predominantly single-centre cohorts in Europe and North America [14-16,18,24] (Table 5), emphasizing the growing importance of ESLD in HIV-infected patients. Several factors may be implicated in this evolution. First, prior to the HAART era, early mortality due to opportunistic infections probably precluded future deaths related to associated chronic viral hepatitis. Moreover, the progression to cirrhosis and HCC in patients with chronic HCV infection occurs over an average of two or three decades [5,6], potentially leading to a burden of disease over time. Increased longevity of HIV-infected patients in the era of HAART, along with accelerated progression of HCV-related liver disease  places this group of patients at extremely high risk for liver disease and its complications in future years.
In the current study, mortality due to ESLD in HIV-infected patients was predominantly related to HCV coinfection (95%). However, additional factors, such as alcohol consumption (66%), HBV coinfection (21%), and the use of hepatotoxic medications, must be considered as potential cofactors in accelerating the progression of chronic liver disease or promoting decompensation in some of these patients. In particular, the use of alcohol was more prevalent among patients who died from ESLD in 2001 (66%) than in those during previous years (1995, 38%; 1997, 33%). In parallel, the proportion of injecting drug users, whom are often co-dependent on alcohol, was higher in 2001 (76%) than in 1995 (29%) and 1997 (39%). As observed in HIV-negative patients with chronic hepatitis C , the incidence of HCC as a cause of death increased over the study interval (1995, 4.7%; 2001, 25%). Regardless of its aetiology, cirrhosis per se is an important risk factor for HCC . Thus, longer survival of HIV-infected patients with HCV-related cirrhosis may be implicated in the higher rate of HCC observed in 2001. Similarly, a higher prevalence of alcohol abuse probably played a major factor in the rising proportion of HCC-related deaths . These data emphasize the importance of abstinence from alcohol due to its role in accelerating fibrosis progression  and promoting hepatocarcinogenesis in HCV-infected patients.
Our data demonstrate that approximately three-quarters of patients who died from ESLD in 2001 were receiving HAART. This is significantly higher than in 1997 (44%), and parallels the rising prescription of HAART in France during the period. The association of chronic HCV infection with hepatotoxicity during HAART constitutes a potential concern for morbidity and mortality in HIV-HCV coinfected patients [31,32]. In particular, HCV infection is an independent risk factor for HAART-related hepatotoxicity , and it has been reported that HAART may increase the severity of hepatic necroinflammatory lesions in patients with chronic hepatitis C . However, other studies have suggested that the use of protease inhibitors may be protective with respect to the progression of HCV-related liver disease . Unfortunately, the design of our study did not allow us to determine the role of HAART in overall or liver-related mortality in our patient population. Nevertheless, a recent study demonstrating that HAART is the strongest predictor of survival in HIV-positive patients  suggests that treatment for HIV is more important in terms of overall survival than concerns regarding the potential exacerbation of HCV-related liver disease by these medications.
Our multi-centre, national study demonstrates that ESLD, predominantly due to HCV coinfection, is now a leading cause of mortality in the HIV-infected population. Five years after the introduction of HAART, the importance of liver disease as a cause of death has increased progressively. This finding probably relates to prolonged longevity attributable to HAART, and possibly due to an increased prevalence of alcohol use. These findings emphasize the importance of developing effective strategies for the prevention and treatment of chronic HCV infection and the moderation of alcohol intake in HIV-positive patients.
Eighty-two questionnaires were distributed to French departments of internal medicine and infectious diseases; a total of 65 departments (79%) responded. During the year 2001, these departments followed 25 178 HIV-infected patients. The modes of HIV infection were homosexual transmission (31%), injecting drug use (21%), heterosexual transmission (38%), transfusion of blood products (3%), and other or unknown (7%) (Table 1).
Among the 25 178 HIV-infected patients followed in 2001, 265 deaths were observed, representing a mortality rate of 1.05%. Causes of death were as follows (Table 2): AIDS (n = 129, 48.7%), ESLD (n = 38, 14.3%) and other (n = 98, 37.0%). ESLD represented 27.9% of the non AIDS-related deaths.
Mortality due to ESLD in relation to hepatitis viruses
The main characteristics of the 38 patients who died due to ESLD are outlined in Table 3. These patients were predominantly male (79%) and the mean age was 42 years (range, 32-69 years). HIV transmission occurred predominantly via injecting drug use (76%). Among the 38 cases with death due to ESLD, 19 (50%) were considered definite and 19 (50%) probable. Thirty of these patients had HCV-related cirrhosis (including HCC in seven cases), six had cirrhosis due to HBV-HCV coinfection (including HCC in one case), and two had HBV-related cirrhosis (including HCC in one case and HDV coinfection in another). Of the 30 patients who died due to HCV-related cirrhosis (without HBV coinfection), 25 (83%) consumed alcohol. In these patients, daily alcohol consumption was high (> 60 g) in 11 (44%), moderate (30-60 g) in eight (32%), and low (< 30 g) in six (24%). Ten of the 38 patients (26.3%) who died as a result of ESLD had previously received treatment with IFN-α alone (n = 6) or in combination with ribavirin (n = 3). HIV category according to the 1993 revised CDC classification system was determined in 34 patients: seven (21%) were asymptomatic (stage A), nine (26%) were symptomatic (stage B), and 18 (53%) had AIDS (stage C). At the time of death, the mean CD4 lymphocyte count was 200 x 106/l (range, 3-700 x 106/l), and 28 patients (74%) were receiving HAART.
Comparison with the 1995 and 1997 surveys
The distribution of risk factors for HIV transmission differed between 2001 and the previous cohorts (Table 1). Whereas heterosexual transmission was more common in 2001 (1995, 25% versus 2001, 38%; P < 0.01), transmission due to homosexual activity (1995, 39% versus 2001, 31%; P < 0.01) and injecting drug use decreased (1995, 25% versus 2001, 21%; P < 0.01). Although overall mortality decreased from 1995 to 2001 (8.15% versus 1.05%; P < 0.01), the proportion of deaths attributable to cirrhosis and/or HCC increased (1995, 1.5%; 1997, 6.6%; 2001, 14.3%; P < 0.01) (Table 2). During the same interval, the percentage of patients who died because of AIDS decreased from 91.6% in 1995 to 48.7% in 2001 (P < 0.01). Among the patients who died from ESLD, the percentage of patients with HCV infection (without HBV or HDV coinfection) was higher in 2001 (78.9%) than in 1995 (57.1%) and 1997 (55.5%) (P < 0.05; Table 4). HCC as a cause of death also increased over this time interval (1995, 4.7%; 1997, 11.1%; 2001, 25%; P < 0.05). The percentages of injecting drug use and alcohol consumption were also higher in 2001 (76% and 66%, respectively) when compared with that observed in the 1995 (29% and 38%, respectively; P < 0.05) and 1997 surveys (39% and 33%, respectively; P < 0.05). Finally, prescription of HAART was more frequent in 2001 (74%) than in 1997 (42%, P < 0.01), and there was a non-significant increase in median CD4 cell counts at death between these time points.