Hepatitis C infection: Dying with or from the infection

[Guest guest]

Hepatitis C infection: Dying with or from the infection

Hashem El-Serag1, , and R. Kramer2 1Houston Center for Quality of Care and Utilization Studies, E. DeBakey VA Medical Center and Sections of Health Services Research and Gastroenterology and Hepatology, Department of Medicine, Baylor College of Medicine, 2002 Holcombe Boulevard (152), Houston, TX 77030, USA2Houston Center for Quality of Care and Utilization Studies, E. DeBakey VA Medical Center and Section of Health Services Research, Department of Medicine, Baylor College of Medicine, 2002 Holcombe Boulevard (152), Houston, TX 77030, USA Associate Editor: M.P. Manns. Available online 4 December 2007. Article Outline

In this issue of the Journal, Marcellin et al. reported mortality associated with hepatitis C virus (HCV) infection using national death certificates in France in 2001 [1]. They selected death certificates based on four groups of medical terms related to HCV infection (Group 1: hepatitis C or HCV infection; Group 2: other hepatitis; Group 3: liver disease, cirrhosis, hepatocellular cancer (HCC), or possible complications of cirrhosis; Group 4: HIV, AIDS, or renal transplantation) to ensure that all deaths that may be associated or directly related to HCV were captured, and one group with general bacterial infection to verify there was no additional HCV. A random sample of these (n = 999) was obtained and a survey was sent to the physicians who filled out the death certificates to evaluate the presence of HCV, its possible complications, and its role in cause of death. Using French population census estimates, they computed that 3618 deaths were associated with HCV infection and 2646 deaths were directly related to (i.e., from) HCV infection. On average these deaths occurred 10 years earlier than the mean age of death of the general population.

Taken on face value, these numbers may be interpreted to mean that 42% (2646/6264) of people whose death certificates showed positive HCV testing died of a cause directly related to HCV. If indeed true, the reported fatality rate indicates a very grave prognosis for HCV-infected individuals. For the 1% of HCV-infected of the French general population, the study confirms the findings of a previous model based study by Deuffic et al. that estimated 3300 HCV-related deaths in France and suggests that HCV-related deaths may actually increase further [2].

HCV was first identified in 1989 [3], at which time the course of chronic hepatitis C was generally believed to be indolent, with fibrosis and hepatic morbidity developing only slowly and in a small proportion of infected patients. Recent evidence indicates that rates of progression to serious and life-threatening hepatic conditions such as cirrhosis, HCC or hepatic decompensation are markedly higher than earlier estimates. This possible discrepancy may have to do with study design issues (longitudinal vs. cross-sectional), and possibly cohort and period effects; all of which will be discussed below.

True prospective cohort studies of HCV natural history are limited, with more expedient designs such as retrospective or concurrent prospective methods constituting the majority of published cohort studies. The early studies found either no or a small increase in the overall mortality of HCV-infected cohorts compared to age and sex matched controls without HCV infection [4]. It was however noted that liver related complications as well as mortality started to increase after 10 years of the onset of HCV infection. Cohort studies reported incidence rates for cirrhosis to be 8–24% over 2–3 decades [5] and [6]. However, the failure of most cohort studies to show any cases of HCC, despite the fact that once cirrhosis is established HCC develops at a rate of 1–4% per year [7] and [8], highlights the limitation of the available cohort studies. These studies rarely employ sufficiently long term follow-up (more than 30 years) and examine relatively small number of patients.

On the other hand, cross-sectional studies or cohort studies in referral centers indicate that a considerable proportion of patients with HCV presenting to these centers had HCV-related liver disease and occasionally death. For example, recent longitudinal studies in patients with paired liver biopsies showed an increase of fibrosis stage in about 30% of cases followed up over 3 years [9] and [10]. Among untreated HCV-infected patients with two liver biopsies at a minimum interval of 1 year, 40% had documented histological progression 3.2 years after their first biopsy [11]. In addition to histological progression, clinical decompensation has also been shown to be high. Analysis of data from 1000 HCV patients with mild-to-advanced fibrosis at baseline completing 5–7 years of follow-up determined an incidence rate of 43.4/10,000 person-years for development of decompensated cirrhosis [12]. While selection as well as diagnostic bias (ascertainment bias) may play a role, a more sinister picture for HCV-related prognosis has emerged.

Other factors that partly explain these seemingly disparate findings between old and more recent studies including the one by Marcellin et al. are “cohort” as well as “period” effects related to the HCV epidemic in a certain geographical area. In general, the spread of the HCV epidemic can be traced to a 1–2 decade period within defined geographical areas. For example, spread time occurred in Japan in the 1920s, in the 1940s in European countries, and in the 1960s in the United States [13] and [14]. Studies conducted more recently are likely to examine HCV-infected patients who are further along in the HCV epidemic (i.e., longer period of chronic infection) and hence more likely to have developed complications. In addition, increased awareness of HCV-related complications especially in high risk patients, and improved accuracy as well as availability of screening and diagnostic tests might have produced a “period effect” during which HCV-related complications are more likely to be diagnosed and reported.

Marcellin et al. employed an efficient way of assessing the impact of viral hepatitis by examining national death records. There are several advantages including the population based sampling frame that is less susceptible to selection bias, the large sample size, and the relative ease of conducting the study. Several conditions have to be fulfilled for these studies to yield valid results. First, the accuracy of the coding for HCV and liver related deaths on death certificates in France is not known. Marcellin et al. used a physician survey as well as having two hepatologists review the returned questionnaires to more accurately determine the cause of death. The positive as well as negative predictive values of HCV codes on the death certificate are unclear, but could have been estimated from this study. Second, many more patients likely died from other causes, but were also positive for HCV. Examining these patients as well would have given us a more comprehensive picture of what patients with HCV are dying from. These limitations tend to bias the study toward finding a higher proportion of death caused by HCV. However, a back of the envelope calculation shows that a very large degree of misclassification is needed to be present to drop the proportion of deaths from HCV to 30% (which is still very high) of deaths with HCV.

In conclusion, while the study by Marcellin et al. might have presented a somewhat exaggerated estimate of deaths from HCV, it is becoming clear that in recent years death from HCV constitutes a major cause of death among those with HCV. These findings are likely to reflect maturation of a cohort effect in addition to increased recognition, and indicate potentially grave consequences to patients and health policy makers.

References

[1] P. Marcellin, F. Pequignot, E. Delarocque-Astagneau, J.-P. Zarski, N. Ganne and P. 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).

[2] S. Deuffic, L. Buffat, T. Poynard and A.J. Valleron, Modeling the hepatitis C virus epidemic in France, Hepatology 29 (1999), pp. 1596–1601. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (129)

[3] Q.L. Choo, G. Kuo, A.J. Weiner, L.R. Overby, D.W. Bradley and M. Houghton, Isolation of a cDNA clone derived from a blood-borne non-A, non-B viral hepatitis genome, Science 244 (1989), pp. 359–362. View Record in Scopus | Cited By in Scopus (2616)

[4] L.B. Seeff, Z. Buskell-Bales, E.C. , S.J. Durako, H.J. Alter and F.L. Iber et al., Long-term mortality after transfusion-associated non-A, non-B hepatitis. The National Heart, Lung, and Blood Institute Study Group, N Engl J Med 327 (1992), pp. 1906–1911. View Record in Scopus | Cited By in Scopus (468)

[5] M. Colombo, Hepatitis C virus and hepatocellular carcinoma, Best Pract Res Clin Gastroenterol 13 (1999), pp. 519–528. Abstract | Abstract + References | PDF (156 K) | View Record in Scopus | Cited By in Scopus (16)

[6] V.K. Rustgi, The epidemiology of hepatitis C infection in the United States, J Gastroenterol 42 (2007), pp. 513–521. View Record in Scopus | Cited By in Scopus (0)

[7] H.B. El Serag, A.C. Mason and C. Key, Trends in survival of patients with hepatocellular carcinoma between 1977 and 1996 in the United States, Hepatology 33 (2001), pp. 62–65. Abstract | PDF (129 K) | Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (135)

[8] K. Ikeda, S. Saitoh, I. Koida, Y. Arase, A. Tsubota and K. ma et al., A multivariate analysis of risk factors for hepatocellular carcinogenesis: a prospective observation of 795 patients with viral and alcoholic cirrhosis, Hepatology 18 (1993), pp. 47–53. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (417)

[9] S.D. Ryder, W.L. Irving, D.A. , K.R. Neal and J.C. Underwood, Progression of hepatic fibrosis in patients with hepatitis C: a prospective repeat liver biopsy study, Gut 53 (2004), pp. 451–455. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (72)

[10] J.P. Zarski, J. McHutchison, J.P. Bronowicki, N. Sturm, R. -Kennedy and E. Hodaj et al., Rate of natural disease progression in patients with chronic hepatitis C, J Hepatol 38 (2003), pp. 307–314. SummaryPlus | Full Text + Links | PDF (211 K) | Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (21)

[11] S.T. Khouri, R.M. , P.M. de Oliveira, V.M. Figueiredo, D.V. Barbosa and V.P. Lanzoni et al., Rebiopsy in patients with untreated hepatitis C: a useful procedure, J Clin Gastroenterol 40 (2006), pp. 347–352. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (0)

[12] P. Pradat, N. Voirin, H.L. Tillmann, M. Chevallier and C. Trepo, Progression to cirrhosis in hepatitis C patients: an age-dependent process, Liver Int 27 (2007), pp. 335–339. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (0)

[13] M. Mizokami and Y. Tanaka, Tracing the evolution of hepatitis C virus in the United States, Japan, and Egypt by using the molecular clock, Clin Gastroenterol Hepatol 3 (2005), pp. S82–S85. SummaryPlus | Full Text + Links | PDF (139 K) | View Record in Scopus | Cited By in Scopus (4)[14] Y. Tanaka, F. Kurbanov, S. Mano, E. Orito, V. Vargas and J.I. Esteban et al., Molecular tracing of the global hepatitis C virus epidemic predicts regional patterns of hepatocellular carcinoma mortality, Gastroenterology 130 (2006), pp. 703–714. SummaryPlus | Full Text + Links | PDF (397 K) | View Record in Scopus | Cited By in Scopus (10)