Fulminant hepatitis A: Disappearing, but not soon enough


  • See Article on Page 1589.

  • Potential conflict of interest: Nothing to report.

One of the great achievements in hepatology during the past 40 years has been the development of highly effective and safe vaccines for the prevention of hepatitis B and A. While the impact of the former on improved global health has been identified, the impact of the latter will take longer to assess, because universal immunization has not been implemented. Nonetheless, it seems likely that the long-term benefits of vaccination against hepatitis A virus (HAV) will be significant. Since the introduction of hepatitis A vaccines in 1995, reported cases of hepatitis A in the United States have declined by more than 85% (Fig. 1).1 While part of the decline may reflect the cyclical pattern of HAV infection, it seems reasonable to assume that HAV vaccine use also has contributed. Eradication of hepatitis A in the United States is possible, but until now published federal recommendations only supported vaccination of individuals within high-risk groups, such as international travelers, homosexual men, injecting drug users, and children in communities with attack rates of hepatitis A above the national average. Because nearly 50% of reported cases have no identified specific risk factor, clearly these guidelines were imperfect. The recent but belated recommendation of the Advisory Committee on Immunization Practices (ACIP) to begin immunization of all children at age 12 to 23 months is extremely promising,2 because in the future, risk factor assessment as a basis for vaccination will become anachronistic. For the present, however, this vaccine-preventable disease continues to be a cause of both morbidity and mortality, as reflected by the occurrence of hepatitis A-associated acute liver failure. Although most common in adult and late adult life, fulminant hepatitis A does not respect sex or ethnicity, and even children, in whom the disease is usually clinically mild, also may be affected.

Figure 1.

Incidence of reported hepatitis A in the United States from 1966–2005.1


HAV, hepatitis A virus; ACIP, Advisory Committee on Immunization Practices; ALFSG, Acute Liver Failure Study Group; UNOS, United Network for Organ Sharing.

No specific pharmacologic treatment or immunotherapy for typical hepatitis A and its clinical variants, such as relapsing hepatitis or acute liver failure, have been developed. Experimental therapies designed to inhibit HAV replication, e.g., via small interfering RNAs3 or cell transplantation4 have not reached clinical trials in fulminant hepatitis A, and the ultimate therapy for hepatitis A-associated acute liver failure—liver transplantation—is life-saving in many instances but results in long-term immunosuppression with its own potential consequences. In addition, although only a handful of cases have been reported, recurrence of clinical hepatitis A has followed transplantation for fulminant hepatitis A.5

Estimates of the incidence of acute liver failure in the United States are generally given as 2000–2500 cases annually, yet there is no national registry of cases, and neither death certificate nor discharge coding are sufficiently reliable to confirm these numbers. Hepatitis A has long been considered a minor cause of acute liver failure and death.6 Case-fatality rates have been estimated to be under 1% for children and young adults, rising to about 2.5% in individuals over age 50 years. Fortunately, the incidence of fulminant hepatitis A, based on the observations of the liver transplant centers comprising the Acute Liver Failure Study Group (ALFSG) and data from the United Network for Organ Sharing (UNOS), reported in this issue,7 has declined in recent years. Because there are no data on the incidence of hepatitis A-associated acute liver failure of lesser severity, a condition which is not referred to a transplant center, it is possible that data from transplant centers may underestimate the true incidence of this disorder. To resolve this issue, population-based studies, not yet available, would be helpful. However, it seems likely that the decline of fulminant hepatitis A is real and not simply a reflection of a change in the severity of the disease, a changing population at risk, or better methods of management. Most likely, the decline in fulminant hepatitis A reflects the parallel decline in reported cases of hepatitis A in the United States.

The definition of acute liver failure used operationally by the ALFSG, namely the presence of encephalopathy and coagulopathy occurring up to 26 weeks from the onset of illness in the absence of pre-existing liver disease, has been time-honored since its introduction in the 1960s by Trey and Davidson. However, the universality of the defining features may be open for question. Recent studies indicate that in children with acute liver failure, encephalopathy is not an invariable finding.8 Additionally, patients with late-onset liver failure generally have a course suggesting chronic disease with a higher frequency of ascites, renal insufficiency, and a poorer survival rate than those with earlier onset,9 although this has not been extensively studied nor has it been fully discussed in the AASLD position paper on management of acute liver failure.10 Recent studies in children confirm that prolonged duration of illness before the onset of acute liver failure is associated with higher death rates and the need for transplantation.11 This entity has been termed late-onset liver failure. Perhaps such patients should be analyzed as a distinct group and not included in the acute liver failure category. The relatively small numbers of patients in the present study of fulminant hepatitis A preclude separate analyses, but such analyses should be considered in future presentations of larger acute liver failure databases, perhaps in developing countries.

The observations reported here regarding prognostic factors in fulminant hepatitis A are provocative. The presence of 3 or more of these criteria—elevation of serum creatinine to ≥ 2.0 mg/dL, ALT ≤ 2600 IU/L, intubation status, and need for pressors, all assessed on study day 1—predicted death or the need for transplantation in 89%–100% of subjects. The statistical model developed for fulminant hepatitis A performed better than other prognostic models. Although this model is based on what may well be one of the largest series of fulminant hepatitis A cases reported, it should be emphasized that only 29 patients were collected between 1998 and 2005, or roughly 5 cases annually. Validation may be impossible to achieve as the U.S. incidence continues to decline. Because the model was based solely on fulminant hepatitis A cases, it may be irrelevant for other causes of acute liver failure.

Existing data suggest that the seroprevalence of hepatitis A in patients with chronic liver disease in the United States is significantly higher than that found in the general population.12 Although an association of acute liver failure due to hepatitis A superimposed on chronic liver disease has served as the basis for the current recommendation that all susceptible patients with chronic liver disease be vaccinated,13 it should be noted that the association remains somewhat controversial.14 Nonetheless, the risk-benefit ratio and available cost-effectiveness estimates15–17 strongly favor the current recommendation to vaccinate patients with chronic liver disease and to do so as early as possible, before the development of advanced disease. The exclusion of patients with known preexisting liver disease and the small number of patients in the ALFSG fulminant hepatitis A database also impeded assessment of this subject.

In summary, the observations reported by Taylor et al.7 indicate that fulminant hepatitis A is disappearing. Unfortunately, it is not disappearing rapidly enough to eliminate the need for liver transplantation in a subset of acutely infected patients. Given the excellent immunogenicity and safety of the currently available inactivated HAV vaccines for more than a decade, the continuing occurrence of fulminant hepatitis A must be viewed as a failure of our health care system. The ACIP's recent recommendation for routine early childhood immunization beginning at age 12 to 23 months is a first step that needs immediate implementation throughout the country. Ideally, it should have been accompanied by a recommendation for catch-up vaccination for children, adolescents, and susceptible adults. This proposed recommendation accordingly may simply be too great an opportunity to miss.