Dr E. B. Keeffe, Stanford University Medical Center, 750 Welch Road, Suite #210, Palo Alto, CA 94304-1509, USA. E-mail: email@example.com
Evidence regarding the outcomes of viral super-infection in patients with chronic liver disease and practical strategies for hepatitis A and B vaccination of these individuals are reviewed. Patients with acute hepatitis A and chronic hepatitis B have a more severe clinical course and a higher death rate compared with otherwise healthy individuals with hepatitis A, and these differences are most pronounced in older patients and those with histological evidence of chronic hepatitis or cirrhosis, rather than in asymptomatic hepatitis B carriers. Patients with acute hepatitis A super-infection and chronic hepatitis C have an increased risk of fulminant hepatitis and death. In addition, patients with other chronic liver diseases also appear to be at increased risk for more severe disease with superimposed hepatitis A. Patients with chronic hepatitis B and hepatitis C virus co-infection have more severe laboratory abnormalities, more severe histological disease, a greater frequency of cirrhosis and complications of cirrhosis, and a higher incidence of hepatocellular carcinoma. Vaccines for both hepatitis A and B are safe and effective if used early in the course of chronic liver disease. Hepatitis A and B vaccination should be part of the routine management of patients with chronic liver disease, preferably as early as possible in the natural course of their disease.
Since at least 400 bc, when Hippocrates described epidemic jaundice, viral hepatitis has been a relentless public health problem. Despite a declining incidence of acute viral hepatitis in recent years, it remains the most common cause of chronic liver disease world-wide.1 Globally, chronic hepatitis B virus (HBV) infection affects over 350 million people, and up to 40% of these individuals may progress to cirrhosis, liver failure or hepatocellular carcinoma.2 Chronic hepatitis C virus (HCV) infection affects roughly half as many people. Many of the estimated 170 million people with chronic HCV infection may also progress to cirrhosis, end-stage liver disease and hepatocellular carcinoma.3 Industrialized countries, whilst benefiting from sanitation and vaccination programmes, are not spared from endemic viral hepatitis or its long-term morbidity and mortality. Chronic liver disease and cirrhosis account for approximately 400 000 hospitalizations and nearly 30 000 deaths annually.4,5 Of the total number of patients with chronic liver disease, over 17 000 were listed for liver transplantation in 2003, and more than 10% of these individuals were expected to die before undergoing transplantation.6
As the pool of patients with chronic liver disease grows, and acute viral hepatitis continues to occur with only a slightly reduced incidence, it is inevitable that a greater number of individuals with chronic liver disease will be at risk for superimposed acute and chronic hepatitis. To minimize the occurrence of acute hepatitis in patients with chronic liver disease, a variety of organizations have recommended hepatitis A and B vaccination of these patients. The Centers for Disease Control and Prevention, the National Institutes of Health, the Veteran's Health Administration and the American Liver Foundation are among the organizations endorsing vaccination of patients with chronic liver disease.7,8 The purpose of this article is to review vaccination in patients with chronic liver disease, specifically the evidence for worse outcomes associated with the occurrence of viral super-infection, and practical strategies for the vaccination of these individuals.
Hepatitis a super-infection in chronic liver disease
Hepatitis A virus (HAV) infection is common, accounting for up to half of the reported cases of acute viral hepatitis in the USA.9 In general, hepatitis A is a self-limited illness with a recovery time measured in months. Young children are often asymptomatic, whereas adults are more likely to be symptomatic and may present with jaundice. Although most patients recover relatively quickly and without long-term sequelae, about 15% may have a prolonged cholestatic syndrome or a relapsing course over 6–9 months. A smaller subset will progress to fulminant hepatic failure and death or transplantation, with estimated case fatality rates ranging from 0.01–0.03% to 0.2% of cases amongst hospitalized patients.10,11 Two major risk factors have been identified in patients who develop fulminant hepatic failure: age over 40 years and the presence of underlying chronic liver disease.10–12
Acute hepatitis A and chronic hepatitis B
A number of epidemiological studies have assessed the outcome of acute HAV infection superimposed on chronic HBV infection. Two large studies are commonly cited to support the contention that acute HAV super-infection results in higher morbidity and mortality than isolated acute HAV infection.12–14 The first study analysed a large outbreak of hepatitis A that occurred in Shanghai in the late 1980s.13 Over 300 000 cases of acute hepatitis A were reported, and nearly 30 000 of these cases probably occurred in HBV carriers, based on the estimated regional seroprevalence of hepatitis B surface antigen (HBsAg). The fatality rate was subsequently analysed and shown to be 5.6 times higher in HBsAg carriers than in HBsAg-negative patients, although the absolute fatality rate was low: 0.05% for HBsAg-positive compared with 0.009% for HBsAg-negative patients.12 Persuasive though this evidence is, differences exist between HBV carriers in the USA and China. In Shanghai, much of the population, particularly the older population, has serological evidence of past HAV infection, which is associated with life-long immunity against hepatitis A. Indeed, the mean age of affected patients in the Shanghai epidemic was 28 years.13 Furthermore, the age of HBV infection probably differs between the two populations, with most Chinese being infected at birth or early in life. The age of hepatitis B e-antigen (HBeAg) seroconversion is also known to affect the clinical course of chronic hepatitis B, with infection at a young age often leading to quiescent disease by early adult life.15 As such, it is not unreasonable to suggest that HAV super-infection may follow a different course in the USA when compared with outcomes in China.
In addition to the Shanghai data, epidemiological data from the Centers for Disease Control and Prevention are often cited to quantify the risks of HAV super-infection of patients with chronic HBV infection in the USA, as well as HAV infection superimposed on other forms of chronic liver disease.12,14 Between 1983 and 1988, a total of 115 551 cases of hepatitis A were reported to state branches of the Centers for Disease Control and Prevention.14 Amongst these patients, there were 381 deaths, resulting in a case fatality rate of 0.33%. Within this group of 381 deaths, there were 27 patients with chronic HBV infection and 107 patients with other chronic liver disease. Based on the HBsAg carrier rate in the USA, the fatality rate amongst chronic HBsAg carriers was subsequently analysed and calculated to be 11.7%, compared with 0.2% for patients with no liver disease, a greater than 50-fold increased risk of death.12
Other studies paint a more complex picture. A study from Japan suggests that chronic HBV infection with histological disease is the predisposing factor accounting for a more adverse outcome from superimposed acute hepatitis A.16 Six patients with HAV super-infection of chronic HBV infection were compared with 80 cases of isolated HAV infection. No significant differences in laboratory markers of hepatic inflammation were found. The authors then reviewed an additional 23 patients with HAV super-infection of chronic hepatitis B reported in the Japanese literature. Based on liver biopsy data available from 17 of these 23 patients, the authors concluded that histological disease, rather than HBsAg seropositivity, might predispose patients to a worse outcome associated with HAV super-infection.16 Asymptomatic HBsAg-positive carriers might well have a clinical course similar to patients with acute hepatitis A alone. Although the Japanese study did not reveal a difference in hepatic markers of inflammation, a Greek study did.17 Fourteen of 90 patients with acute hepatitis A had serological evidence of HBV carrier status, whilst another nine had evidence of previous infection. The mean serum alanine aminotransferase levels were significantly higher in patients with evidence of current or previous HBV infection, compared with those with acute hepatitis A without the presence of HBV markers. However, there were no deaths, and the mean bilirubin levels did not differ. Another case series in Thailand found worse outcomes associated with HAV super-infection.18 Twenty asymptomatic HBsAg-positive patients, eight HBsAg-positive patients with cirrhosis and four anti-HCV-positive patients with cirrhosis were compared with 100 patients with HAV infection without underlying liver disease. None of the patients with isolated HAV infection developed severe hepatitis. However, of the patients with pre-existing liver disease, nearly half (47%) developed fulminant or submassive hepatitis and 28% died. This raw difference in mortality rates does not, unfortunately, account for age differences. Patients with existing liver disease were decades older than patients with isolated HAV infections, and all deaths occurred in patients over 50 years of age.18 It can thus be hypothesized that the patients who died (already more ill than an average group referred to a tertiary care centre) were more likely to do so as a result of either their age or their underlying liver disease, as the relative risk of these intertwined factors cannot be separated.
Other case series have not found underlying chronic HBV infection to predict a worse outcome of acute hepatitis A. A case series in Taiwan analysed 143 patients with hepatitis A, 28 of whom were HBsAg carriers, and failed to find any clinical or biochemical difference between the two groups.19 It is noteworthy that the patients were young, ranging from 5 to 30 years of age, lending credence to the argument that it is age and/or underlying histological chronic hepatitis or cirrhosis that leads to a worse outcome, rather than merely the presence of HBV infection. A recent review of 50 patients with acute hepatitis A, 19 of whom developed fulminant hepatitis (10 eventually requiring transplantation), failed to reveal serological evidence of chronic viral hepatitis.20 Furthermore, in a multivariate analysis, age did not emerge as an independent risk factor for the development of a fulminant course. In addition to these studies, a number of other small case series have failed to reveal a difference in outcomes of isolated HAV infection compared with HAV/HBV super-infection.21–24
Acute hepatitis A and chronic hepatitis C
Acute hepatitis A, when occurring in patients with underlying chronic hepatitis C, may also cause significantly increased morbidity and mortality. The most dramatic observation of HAV super-infection in HCV patients was seen in a well-cited Italian study.25 Over a 7-year period, 432 patients with chronic hepatitis C and no serological evidence of previous HAV infection were followed. Seventeen (3.9%) developed acute hepatitis A. Out of this subset, seven patients (41%) developed fulminant hepatitis, six of whom died. Interestingly, patients with chronic HBV infection did not have worse outcomes with HAV super-infection, in contrast with the data from the studies in Shanghai and by the Centers for Disease Control and Prevention. Ten patients with chronic HBV infection developed acute hepatitis A, and all recovered. In addition to this seeming refutation of previous large epidemiological studies in Shanghai and by the Centers for Disease Control and Prevention, questions have been raised about the unexpectedly high rate of fulminant hepatitis A. This high rate was not seen in data from the Italian specific surveillance system of acute viral hepatitis.26 Furthermore, other trials and national surveillance systems, such as in Finland, Switzerland and Norway, have failed to confirm this high rate of fulminant HAV infection in patients with chronic HCV infection.27,28 Case series in Brazil and Switzerland reporting hundreds of patients with hepatitis A and hepatitis C have not revealed an increased rate of fulminant hepatitis.29,30 In addition, two large reviews of fulminant hepatitis A cases in tertiary care centres have failed to reveal a large number of cases with underlying chronic HCV infection.20,31 The Italian study, although provocative, requires confirmation.
Acute hepatitis A and miscellaneous chronic liver diseases
There is only limited published evidence suggesting that other chronic liver diseases may predispose to a worse outcome associated with chronic hepatitis A. An analysis of data from the Centers for Disease Control and Prevention revealed that patients with forms of chronic liver diseases other than HBV were at a 23-fold increased risk of death.12,14 Poor outcomes from hepatitis A superimposed on chronic liver disease other than chronic HBV or HCV infection have been seen in other small case series. A report of four occurrences of fatal fulminant hepatitis A in patients with alcoholic liver disease (two cases) and cryptogenic cirrhosis (two cases) supports the contention that, in addition to chronic hepatitis B, other chronic liver disease can increase the morbidity and mortality of acute hepatitis A.32 A review of all patients admitted with severe hepatitis A to a tertiary care hospital in London between 1974 and 1999 revealed 97 cases, but only four had underlying chronic HCV or HBV infection.31 However, three of these patients, all with cirrhosis, were amongst the 29 who died, and an additional six patients required transplantation. Interestingly, post-mortem examination revealed that 20% of patients had evidence of underlying chronic liver disease, supporting the assertion that chronic liver disease, regardless of the aetiology, predisposes to worse outcomes.31
By reviewing this growing experience with acute hepatitis A in patients with chronic liver diseases, four points can be made. First, patients with acute hepatitis A and chronic hepatitis B have a more severe clinical course and a higher death rate compared with otherwise healthy individuals with hepatitis A. Second, these differences are most pronounced in older patients with histological evidence of chronic hepatitis or frank cirrhosis, rather than younger, asymptomatic HBsAg carriers. The natural experiments that provide much of the data have not allowed the clear separation and examination of confounding factors of age and histology. Third, patients with acute hepatitis A super-infection and chronic hepatitis C may have an increased risk of fulminant hepatitis and death, although this still needs corroboration. Finally, patients with other chronic liver diseases, apart from chronic HBV and HCV infection, also appear to be at increased risk for more severe disease.
Hepatitis b super-infection in chronic liver disease
Acute hepatitis B, like hepatitis A, can be associated with super-infection in patients with pre-existing chronic liver disease. However, HBV infection has an age-associated risk of chronicity not present with HAV infection. Acute hepatitis B results in chronic infection in up to 95% of infected neonates. By contrast, 95–99% of adults with intact immune systems will recover from acute HBV infection without long-term sequelae.15 As such, the forms of HBV super-infection are more varied. The literature examining HBV infection superimposed on pre-existing liver disease generally falls into one of three categories: acute HBV super-infection, chronic HBV and HCV co-infection, and occult HBV infection. In addition, the interaction between HBV and hepatitis D (delta) virus (HDV), as well as HBV infection and alcoholic liver disease, is well described, and will not be discussed in this review. The implications of acute and chronic co-infection with HBV and HCV have been reviewed recently.33,34
Acute hepatitis B and chronic liver disease
Acute hepatitis B superimposed on chronic hepatitis C has been implicated as a cause of fulminant hepatitis. A survey of 40 patients with fulminant hepatitis found a significant role for co-infection.35 Seventeen of these patients were found to be HBsAg-positive. Of this group, eight (47%) were also HCV RNA-positive. Other case series have not shown as dramatic a correlation between co-infection and clinical outcomes. One case series reviewed patients enrolled in the Copenhagen Hepatitis Acuta Programme over a 2-year period.36 Sixteen patients with acute HBV infection and detectable anti-HCV were identified. All 16 recovered from the acute illness. Other small case reports have shown discordant results, with either rapid clearance of serum HBsAg (two patients) or progression to hepatic failure and death (one patient).37,38 Indeed, in two reported cases, it was noted that HBsAg and anti-HCV (as well as HCV RNA) became undetectable, suggesting that acute HBV may suppress pre-existing HCV and its antibody, resulting in the occasional resolution of both infections.38,39 This sort of viral interference has been well documented in chronic HBV/HCV co-infection.40–43 In addition to a complex interrelationship between co-habitating viruses, it has been suggested for some time that hepatitis vaccination itself, as opposed to HBV infection, may invoke a similar response.44 A recent trial of 91 patients designed to evaluate the safety of hepatitis B vaccine in patients with chronic HCV infection found evidence of a possible therapeutic benefit, based on a significant decrease in serum alanine aminotransferase levels when compared with HCV patients with evidence of immunity to HBV who were not immunized.34 In summary, acute hepatitis B super-infection remains a complicated subject. Relatively few cases are available for study, and existing case series provide conflicting results. Some patients progress to rapid hepatic failure and death. Other rare patients, perhaps from a complex interference from dual viral replication or an enhanced host response, clear both the superimposed acute infection and the underlying chronic infection.
Chronic hepatitis B and chronic liver disease
Regardless of the final verdict regarding the severity and outcome of acute HBV infection in patients with chronic hepatitis C, patients with chronic HBV and HCV co-infection appear to have worse outcomes than patients with either infection alone. A case-controlled retrospective study in Saudi Arabia compared 20 patients with combined HCV and HBV infection with 33 controls with only chronic HCV infection.45 The dual infection group was found to have significantly more cirrhosis (95% vs. 48.5%) and hepatocellular carcinoma (63% vs. 15%). Other similar studies in Italian and French patients have also shown increased risks of both cirrhosis and hepatocellular carcinoma.40,46–48 In addition to a greater rate of progression to cirrhosis and hepatocellular carcinoma, other evidence of worsened disease has been noted. In a case-controlled study, the experience of 19 patients with chronic HBV and HCV co-infection and 17 patients with combined HBV, HCV and HDV infection was examined.49 In addition to more severe histological disease and greater elevations of hepatic markers of inflammation, co-infected patients, compared with patients with isolated HCV infection (matched for age, duration of disease and hepatitis risk factors), were less likely to respond to anti-viral therapy. An Israeli study at the Hadassah Medical Center liver unit reviewed 92 consecutive patients with chronic HCV infection.50 Markers of HBV infection, either past or present, were observed in 66% of patients. Although some differences in markers of hepatic injury were observed, with higher bilirubin levels in co-infected patients, the initial clinical and physical symptoms were similar between patients with isolated HCV infection and co-infected patients. However, HBV co-infection was associated with more complications, including bleeding oesophageal varices, hepatic encephalopathy, spontaneous bacterial peritonitis and hepatocellular carcinoma. In summary, chronic HBV and HCV co-infection probably results in: (i) more severe laboratory abnormalities; (ii) more severe histological disease; (iii) more frank cirrhosis and complications of cirrhosis; and (iv) a higher incidence of hepatocellular carcinoma.
Occult hepatitis B virus infection and chronic liver disease
The above studies have generally defined chronic HBV infection by the persistent detection of HBsAg. However, patients can also manifest HBV infection by detection of serum HBV DNA without the presence of HBsAg, a situation known as occult hepatitis B infection. Cacciola et al. examined liver and serum samples from 200 HBsAg-negative patients with chronic HCV infection.51 Sixty-six patients were found to have serum HBV DNA by testing with a polymerase chain reaction assay. Twenty-two of these patients (33%) had cirrhosis, compared with 26 of the 134 patients (19%) with HCV infection but no detectable serum HBV DNA. Cacciola et al. also reviewed the published epidemiological and molecular studies, and found that a preponderance of data indicated that persistent HBV infection in HBsAg-negative patients may have a crucial role in the development of hepatocellular carcinoma.51 However, a number of recent studies have cast doubt on this association. A study of 210 Taiwanese HBsAg-negative patients with chronic hepatitis C did not reveal an increased risk of cirrhosis or hepatocellular carcinoma in the 15% of patients with detectable serum HBV DNA.52 Furthermore, serum HBV DNA was detectable in an equal percentage of healthy HCV-negative controls. Another study analysed 24 Japanese patients with hepatocellular carcinoma associated with chronic hepatitis C.53 Although HBV DNA was found in 12 patients, other techniques, such as Southern hybridization and end-point dilution, revealed an extremely low copy number of HBV DNA. Indeed, most samples contained significantly less than one copy of HBV DNA per cell. If HBV DNA incorporation plays a critical role in carcinogenesis, monoclonal tumour cells might be expected to have at least one copy per cell. Recent work looking at the distribution of HBV DNA using in situ hybridization techniques found a significantly lower percentage of hepatocyte involvement in occult HBV infection, compared with HBsAg-positive HBV infection, and confirmed the low viraemia levels found in other studies.54 These studies combined cast doubt on the significance of occult HBV infection in patients with chronic HCV infection.
Hepatitis a and b vaccination in chronic liver disease
Given the growing evidence that hepatitis A super-infection is probably worse in at least some, and probably most, patients with chronic liver disease, and that hepatitis B may be worse in at least some varieties of super-infection, vaccination seems reasonable. Furthermore, as risk factors for the various forms of viral hepatitis overlap, patients with chronic liver disease are often at a greater risk of viral super-infection.33,34 A greater risk of infection, coupled with a likely worse outcome, warrants immunization. Currently, a variety of vaccines are available for hepatitis A and B (Table 1). In evaluating HAV and HBV vaccination in patients with chronic liver disease, three questions need to be answered: (i) who needs vaccination?; (ii) is it safe?; and (iii) is it effective? Answering these questions will provide an effective strategy for applying hepatitis vaccines in chronic liver disease.
Table 1. Common hepatitis vaccine formulations, doses and schedules (adults)
EL.U., enzyme-linked immunosorbent assay units; HAV, hepatitis A virus; HBV, hepatitis B virus; HBsAg, hepatitis B surface antigen; i.m., intramuscularly.
1440 EL.U. in 1 mL i.m.
Month 0, and 6–12 months later
∼ 50 U in 1 mL i.m.
Month 0, and 6–12 months later
20 µg recombinant HBsAg in 1 mL i.m.
Month 0, 1 and 6 or month 0, 1, 2 and 12
10 µg recombinant HBsAg in 1 mL i.m.
Month 0, 1 and 6
Twinrix (Engerix-B and Havrix)
720 EL.U. inactivated HAV, 20 µg recombinant HBsAg in 1 mL i.m.
Month 0, 1 and 6
Vaccination against hepatitis A in patients with chronic liver disease
HAV vaccine is currently recommended for three groups: communities with endemic HAV infection (> 20/100 000), individuals at increased risk of HAV exposure, and individuals with increased risk of severe disease.7 Patients with chronic liver disease are considered to be at risk for severe disease, and also have an increased risk of HAV exposure. Patients with serological evidence of previous hepatitis A are considered to have probable lifelong immunity. As such, the seroprevalence of HAV markers plays a central role in determining a vaccination strategy. If it was assumed that no potential vaccine recipients had previous exposure to HAV, it would make sense to provide vaccination without pre-screening for anti-HAV. Alternatively, if there was a sufficiently high endemic rate of HAV exposure, there would be little point in either pre-screening or providing vaccination, as patients would probably be immune and, in any case, the vaccine would provide close to no additional protection at a population level. Indeed, in many endemic areas, immunity to hepatitis A approaches 100%.55,56 Based on this reasoning, many have argued that routine vaccination or screening for anti-HAV does not need to be provided to patients with chronic liver disease who are adult immigrants from endemic areas.57–59
In industrialized countries, anti-HAV seroprevalence rates in patients with chronic liver disease are higher than those in the general population, but do not reach the levels where immunity can be assumed. On average, about 30% of Americans have detectable anti-HAV.60 Antibody prevalence is age related, with younger cohorts having a lower level of immunity, reflecting the reduced prevalence of the disease over time. The rate of anti-HAV prevalence amongst patients with chronic liver disease is assumed to be higher, but is not definitively known. In one study from Spain, the prevalence of anti-HAV in patients with chronic hepatitis B and C was 75%.61 This varied from 20% at 19 years of age to 93% over the age of 40 years. The rate of anti-HAV in the USA is lower. In a review of 473 patients with chronic hepatitis C, hepatitis B or alcoholic liver disease, the average anti-HAV seroprevalence rate was found to be 55%, with an age-related range from 44% to 65%.62 This corresponds with other surveys in patients with chronic HCV in the USA.63 In one recent large review of 1092 patients with chronic hepatitis C, confirmed by detectable HCV RNA, however, the anti-HAV seroprevalence was found to be similar to that in the general population at 33%.64
Based on this range of seroprevalence, a variety of authors have analysed vaccination strategies, and have attempted to answer the question of whether to pre-screen candidates for the presence of anti-HAV antibodies. Three general vaccination strategies have been evaluated: (i) universal vaccination, in which serologies are not checked; (ii) selective or deferred vaccination, in which serologies are checked prior to vaccine administration; and (iii) semi-selective vaccination, in which the first vaccine injection in a series is given at the same time as a test for IgG anti-HAV is drawn.65 In general, targeted vaccination strategies have been found to be the most cost-effective.66 Even at the lowest estimated anti-HAV seroprevalence rates, selective or deferred vaccination is the most cost-effective strategy in patients with chronic liver disease, if the recommended HAV vaccination schedule is to be used.64 In a study of 100 relatively healthy military patients with chronic liver disease, the anti-HAV seroprevalence was found to be 53%.67 Selective or deferred vaccination was found to be the most cost-effective strategy. Sensitivity analysis of the prevalence rate revealed this to be a robust strategy; the prevalence rate of anti-HAV had to decrease to below 12% before universal vaccination became a more cost-effective strategy.
Patients can receive HAV vaccination with little worry about vaccine-related adverse events. Since the late 1980s, hepatitis A vaccine has been studied extensively in individuals of all ages, and has been shown to be safe.68–70 In a large multi-centre study of 475 patients (including 188 healthy controls), no significant adverse events were found to be associated with the use of hepatitis A vaccine in patients with chronic liver disease.71 Mild general adverse effects, such as fatigue and headache, were higher in chronic liver disease, although these symptoms were difficult to differentiate from patients' baseline illness. More importantly, this trial also answered questions regarding the efficacy of HAV vaccine in patients with chronic liver disease. Mean geometric titres of anti-HAV were significantly lower in chronic liver disease than in healthy patients. However, approximately 95% of patients with chronic liver disease, compared with 98% of healthy controls, seroconverted to an immune-protected state (defined as previously seronegative patients who achieved anti-HAV titres of ≥ 30 mIU/mL).71 A smaller trial in Chinese patients also noted similar seroconversion rates in 65 chronic HBV patients vs. healthy controls given HAV vaccine, although anti-HAV titres were lower in patients with chronic liver disease.72
Although studies, such as those cited above, demonstrate the immunogenicity of HAV vaccine in mild and moderate chronic liver disease, the efficacy of HAV vaccines in advanced or decompensated liver disease has not been thoroughly investigated. The trend towards lower, although still effective, anti-HAV titres in patients with chronic liver disease raises the concern that severe liver disease will result in an ineffective antibody response. Evidence exists that this is indeed the case. An evaluation of 72 patients with end-stage liver disease awaiting liver transplantation found that only 35 patients (48.6%) seroconverted following a normal HAV vaccination schedule.73 Although age, gender and race did not influence seroconversion, a worse stage of liver failure (defined by the United Network for Organ Sharing criteria at the time of listing) and the presence of an alcoholic component of liver disease were predictive of lower seroconversion rates. Another trial analysed 24 patients with severe liver disease, 16 with end-stage liver disease and eight liver transplant recipients.74 After HAV vaccine was given according to the routine schedule, none of the liver transplant recipients converted, compared with seven of the 14 liver failure patients. Additional trials to evaluate the safety and efficacy of HAV vaccine in liver transplant recipients and patients with decompensated liver failure have shown equally dismal results.75, 76 These results are summarized in Table 2.
Table 2. Hepatitis A vaccine trial results — immunogenicity
Number of patients (n)
Seroconversion at month 7 (%)
CLD, chronic liver disease; GMC, geometric mean concentration of anti-HAV; HAV, hepatitis A virus; HBV, chronic hepatitis B virus infection; HCV, chronic hepatitis C virus infection.
Based on this growing pool of evidence, a reasonable approach to hepatitis A vaccination is to test for anti-HAV, and then to give HAV vaccine to those without serological markers of previous infection who have mild to moderate chronic liver disease. Patients with chronic liver disease from HAV endemic areas, such as India, are the only defined group that can be excluded from an HAV ‘test and treat’ strategy. Given the high seroconversion rate for patients with mild to moderate disease, post-vaccination titres do not need to be drawn. As it appears that HAV vaccine has poor immunogenicity in advanced liver disease, vaccination should be provided as soon as possible in the natural history of chronic liver disease. The current evidence does not support an effective vaccination strategy in patients with advanced liver disease, even though these individuals might benefit the most from protection against viral super-infection. As this group has extremely low seroconversion rates, post-immunization anti-HAV screening may be appropriate for risk assessment and for the selection of patients appropriate for yet-to-be recommended or proven repeat vaccinations or immunological vaccine adjuncts. A practical vaccination strategy is outlined in Figure 1.
Vaccination against hepatitis B in patients with chronic liver disease
In addition to the universal use in children and adolescents, HBV vaccination, like HAV vaccination, is recommended for those at high risk for infection or at high risk for poor outcomes if infected. In practice, this recommendation applies to patients with non-HBV chronic liver disease.77 For the majority of patients without chronic liver disease, pre-screening serologies are not recommended prior to immunization; universal, not selective, immunization strategies are used. However, in groups with a prevalence of previous infection expected to be greater than 30%, pre-screening has been found to be cost-effective.78 HCV-infected patients are reported to have serological evidence of HBV infection in roughly two-thirds of cases, warranting pre-vaccination screening,64 a common practice in patients with chronic liver disease.33,34
Serological pre-vaccination screening is not straightforward in patients with chronic liver disease. More than anti-HBs testing is necessary. Occasionally, patients with chronic liver disease, particularly with chronic hepatitis C, will have a serological pattern with isolated anti-HBc as the only marker of HBV infection.79 This serological pattern is thought to be due to inhibition of HBsAg synthesis by pre-existing HCV infection.80,81 Furthermore, low-level occult HBV infections will, by definition, evade serological testing. Patients with previous HBV infection lacking HBsAg and anti-HBs have immunological memory, and will mount a brisk anamnestic response to vaccine stimulation.82 Patients without previous infection, who represent false positive results, will not have this brisk response. A diagnostic algorithm can be used for patients with isolated anti-HBc.83,84 Anyone with isolated anti-HBc should be re-tested for HBV markers to reduce the likelihood of laboratory error. If patients remain persistently anti-HBc positive, the challenge is to identify those who are false positives from those with previous immunity who have lost HBsAg or those who have low-level occult HBV infection. A single dose of hepatitis vaccine should be administered, and follow-up quantitative anti-HBs serology should be determined in 1 month. A high titre of anti-HBs at this time indicates immunity; no further vaccination is necessary. However, if the titre is low (less than 10 mIU/mL), a full three-dose course of vaccination should be given. If post-vaccination titres are still low or undetectable, occult HBV infection may be present, and these patients are not expected to respond to vaccination. At this point, serum HBV DNA testing is appropriate. Also important and not to be forgotten is the acute presentation of HBV infection, where, during the core window, only anti-HBc is present. This presentation is thought to account for only a small number of cases in this setting.82
Hepatitis B vaccines, similar to hepatitis A vaccines, are quite safe. A recent review of studies addressing the safety and tolerability of HBV vaccine in patients with chronic HCV found few side-effects.85 Of the three trials reviewed, mild local reactions occurred in 7–20% of vaccinees, and general reactions, such as headache, fever and nausea, occurred in 2–19%, with no serious adverse reactions being reported.71,86,87 Hepatitis B vaccine has also been found to be safe in patients with alcoholic liver disease. 88 The efficacy of HBV vaccination varies depending on the stage of chronic liver disease. In patients with chronic hepatitis C, seroprotection rates have ranged from 69% (using 10 µg doses) to 100% (with 20 µg doses) after a full course of HBV vaccine.71,86,87,89 Patients with alcoholic liver disease convert at a lower rate, with reported seroprotection ranging from 18% to 75%. 88,90,91 Rosman et al. demonstrated that an accelerated, high-dose regimen (40 µg at 0, 1, 2 and 6 months) achieved higher seroconversion rates when compared with the standard regimen (20 µg at 0, 1 and 6 months), i.e. 75% vs. 46%. 88 These studies are summarized in Table 3. In liver transplant candidates and recipients, low seroconversion rates are well described. Less than one-third of patients seroconvert after receiving standard dosing; providing a double dose (40 µg) and an accelerated schedule (0, 1 and 2 months) increases the seroconversion rate to between 44% and 62%. 92,93 A prospective trial at the Mayo Clinic of 356 liver transplant recipients, using an even more accelerated, double-dose schedule of HBV vaccine (40 µg at 0, 2 weeks, 4 weeks and 6 months), found a seroconversion rate of 36%.94 Two years post-transplantation, the seroconversion rate had dropped to 8%. The authors concluded that seroconversion before or after liver transplantation is low, that the antibody titre decreases rapidly after transplantation, and that a variety of factors, such as age, genetic predisposition and severity of liver disease, may negatively influence responsiveness to vaccination.94 Given the higher rate of non-responders compared with HAV vaccines, post-vaccination serological testing may be appropriate, particularly in the sickest patients. Some patients, particularly those with some minimal rise in titres, may fully seroconvert with repeat vaccination.95
Table 3. Hepatitis B vaccine trial results — immunogenicity
Hepatitis B vaccination requires a strategy similar to hepatitis A vaccination. Pre-vaccination serology should be checked in all patients. This panel needs to include anti-HBc to identify present or past HBV-infected patients with isolated core antibody as the only serological manifestation of HBV infection. Depending on the serological test results, a vaccination strategy is provided in Figure 2. Despite declining efficacy as disease progresses, the excellent safety profile allows for a trial of accelerated, double-dose HBV vaccine for patients with advanced chronic liver disease. Post-immunization serological testing in patients, particularly with moderate or advanced disease, can then identify hyporesponders who may benefit from additional doses of vaccine.
Compared with patients without liver disease, patients with chronic liver disease are more likely to experience severe complications from acute viral hepatitis A or B. Historical data suggest that acute hepatitis A can result in significantly increased mortality in patients with underlying hepatitis B and C and probably other chronic liver diseases as well. Hepatitis B, whether manifest by acute, chronic or occult infection, has the potential to significantly worsen the course of underlying liver disease. Strong epidemiological evidence suggests an increased occurrence of fulminant liver failure, cirrhosis and hepatocellular carcinoma in patients with HBV super-infection. Vaccines for both hepatitis A and B are available, safe and of proven efficacy if used early in the course of chronic liver disease. As available vaccines are both safe and immunogenic, it is not surprising that a variety of government and non-government health organizations endorse the use of these vaccines in patients with chronic liver disease. Hepatitis A and B vaccination should be part of the routine management of patients with chronic liver disease, preferably as early as possible in the natural course of their illness. A combined HAV and HBV vaccine provides a convenient approach for patients requiring both vaccines.