Skip Navigation

Clinical and Virological Improvement of Hepatitis B Virus—Related or Hepatitis C Virus—Related Chronic Hepatitis with Concomitant Hepatitis A Virus Infection

  1. Evangelista Sagnelli1,2,
  2. Nicola Coppola1,2,
  3. Mariantonietta Pisaturo1,2,
  4. Raffaella Pisapia1,2,
  5. Mirella Onofrio1,2,
  6. Caterina Sagnelli2,
  7. Antonio Catuogno1,
  8. Carlo Scolastico2,
  9. Felice Piccinino2, and
  10. Pietro Filippini2
  1. 1Division of Infectious Diseases, Azienda Ospedaliera San Sebastiano, Caserta
  2. 2Department of Public Medicine, Section of Infectious Diseases, Second University of Naples, Naples, Italy
  1. Reprints or correspondence: Prof. Evangelista Sagnelli, Dept. of Public Health, Section of Infectious Diseases, Second University of Naples, c/o Ospedale Gesù e Maria, Via D. Cotugno 1, 80135 Naples, Italy (evangelista.sagnelli{at}unina2.it).

Abstract

Background. We evaluated the clinical and virological characteristics of hepatitis A virus infection in persons concomitantly infected with hepatitis B virus (HBV) or hepatitis C virus (HCV).

Methods. We enrolled 21 patients with acute hepatitis A and chronic hepatitis with no sign of liver cirrhosis, 13 patients who were positive for hepatitis B surface antigen (case B group), 8 patients who were anti-HCV positive (case C group), and 21 patients with acute hepatitis A without a preexisting liver disease (control A group). Two control groups of patients with chronic hepatitis B (control B group) or C (control C group) were also chosen. All control groups were pair-matched by age and sex with the corresponding case group.

Results. Fulminant hepatitis A was never observed, and hepatitis A had a severe course in 1 patient in the case B group and in 1 patient in the control A group. Both patients recovered.

On admission, HBV DNA was detected in 1 patient in the case B group (7.7%) and in 13 patients (50%) in the control B group; HCV RNA was found in no patient in the case C group and in 16 patients (81.2%) in the control C group. of 9 patients in the case B group who were followed up for 6 months, 3 became negative for hepatitis B surface antigen and positive for hepatitis B surface antibody, 2 remained positive for hepatitis A surface antigen and negative for HBV DNA, and 4 became positive for HBV DNA with a low viral load. of 6 patients in the case C group who were followed up for 6 months, 3 remained negative for HCV RNA, and 3 had persistently low viral loads.

Conclusion. Concomitant hepatitis A was always self-limited, associated with a marked inhibition of HBV and HCV genomes, and possibly had a good prognosis for the underlying chronic hepatitis.

Acute hepatitis A is a self-limiting disease that is almost always asymptomatic in babies and children [1] and is frequently symptomatic in adults; fulminant hepatitis is rare in adults and unusual in children [26]. A high rate of fulminant hepatitis in patients with preexisting chronic liver disease without cirrhosis has been described in Italy [7], a finding not confirmed by other authors [812].

The existence of interference between hepatitis A virus (HAV) and hepatitis B virus (HBV) or between HAV and hepatitis C virus (HCV) has been poorly investigated. In fact, an inhibitory effect of acute HAV infection on HBV replication in chronic carriers of hepatitis B surface antigen (HBsAg) has been reported anecdotally [5, 13], but no information regarding HAV-HCV interaction is available.

Recent Italian epidemiological studies have shown an increasing prevalence of chronic hepatitis among adults who lack a naturally acquired immunity to HAV [12, 14, 15] and have demonstrated an increased incidence of hepatitis A associated with the ingestion of raw or partially cooked shellfish [16]. Consequently, from January 2001 to December 2003, we observed 143 consecutive patients with HAV infection, 13 of whom had preexisting chronic hepatitis B and 8 of whom had preexisting chronic hepatitis C. In these patients, we studied the clinical and virological impact of concomitant HAV infection using recently developed technology. Data were compared with those obtained from control groups, and the results are reported here.

Patients and Methods

Patients. One hundred forty-three consecutive patients with symptomatic hepatitis A hospitalized in 2 infectious diseases wards in southern Italy (1 in Caserta and 1 in Naples) from January 2001 to December 2003 were included in the study. These 2 wards had been using the same clinical and laboratory approach for years and had cooperated in several clinical investigations. In each case, HAV etiology was determined on the basis of the detection of anti-HAV IgM in serum.

Of these 143 patients, 13 had a preexisting chronic HBV infection (case B group), and 8 had a preexisting chronic HCV infection (case C group). The 13 patients in the case B group had been asymptomatic, chronic carriers of HBsAg for 18–36 months, 10 of whom had steadily abnormal serum alanine aminotransferase (ALT) levels (chronic hepatitis) and 3 of whom had persistently normal ALT levels (asymptomatic carriers). The 8 patients in the case C group had been HCV antibody carriers for at least 16 months, 7 of whom had chronic hepatitis and 1 of whom was an asymptomatic carrier. None of the 21 patients in the case B group or the case C group had undergone liver biopsy or had been studied for HBV or HCV replication, respectively.

To compare acute hepatitis A in the HBV and HCV carriers with that occurring in patients without HBsAg and HCV infection, we chose for each patient in the case B and case C groups an HBsAg-negative patient with acute hepatitis A who was negative for HCV antibody from among the 122 patients hospitalized over the same period, and we pair-matched them by age (±5 years), sex, and number of days between hospitalization and onset of symptoms (±2 days). This new group of patients was named the “control A group.”

On admission, no patient in the 3 above-mentioned groups showed circulating hepatitis B core IgM antibody or hepatitis D virus antibody, nor had they been vaccinated against HAV.

Acute hepatitis A was considered to be severe when prothrombin activity dropped to ⩽25% during observation or when patients developed portosystemic encephalopathy or ascites; for the diagnosis of fulminant hepatitis, accepted criteria were applied [17].

Nine of the 13 patients in the case B group and 6 of the 8 patients in the case C group were followed up for 6 months, 3 patients (2 in the case B group and 1 in the case C group) were observed for only 2–4 months, and 3 patients were lost to follow-up because of lack of compliance.

To evaluate whether concomitant HAV infection may induce clinical and/or virological variations in chronic HBV or HCV infection, we formed corresponding control groups for both the case B and case C groups. For each patient in the case B group, we chose 2 patients among those first observed at our wards in the same period with asymptomatic HBV chronic infection without concomitant HAV infection, and we pair-matched them by age (±5 years) and sex; before deciding on a treatment, we observed these patients for 6 months to define the disease as chronic [18]. This group was named the “control B group” and comprised 26 patients (16 males and 10 females; median age, 19 years [range, 16–36 years]). Similarly, for each patient in the case C group, 2 patients with asymptomatic chronic HCV infection without concomitant HAV infection were chosen using the same criteria, to form a control group for the case C group. This new group, named the “control C group,” comprised 16 patients (12 males and 4 females; median age, 24 years [range, 16–28 years]).

All patients in the case B and case C groups and the corresponding control groups were evaluated for the presence of symptoms or signs characterizing liver cirrhosis (i.e., clinical aspects of cirrhosis on physical examination) [19]; abnormalities in the liver function tests, including a blood platelet count <100,000 platelets/mm3; and ultrasound evidence characterizing liver cirrhosis [20]. This evaluation was performed for all patients with hepatitis A at the time of remission of the disease and in all patients without hepatitis A at the time of the first observation.

For each patient in the case B, case C, and control A groups, plasma samples were obtained at admission and during the initial phase of hepatitis A at an interval of 4–6 days; the samples were stored at -80°C and were not thawed until used.

For patients in case B and C groups and control B and C groups with a 6-month duration of follow-up, plasma samples were obtained throughout the observation period and stored at -80°C.

Routine methods. HAV, HBV, and hepatitis D virus serum markers were determined by a commercial enzymatic immunoassay (HAVAB-M-EIA, HAVAB-EIA, AUSZYME MONOCLONALE, AUSAB-EIA, CORZYME, and CORZYME-M [Abbott Laboratories] for anti-HAV IgM, anti-HAV IgG, HBsAg, hepatitis B surface antibody, hepatitis B core antibody, and anti—hepatitis B core IgM, respectively; and ETI-EBK-PLUS, ETI-AB-DELTAK-2, and ETI-DELTA-IgGMK-2 [DiaSorin] for hepatitis B envelope antigen, hepatitis B envelope antibody, anti—hepatitis D virus IgG, and anti—hepatitis D virus IgM, respectively). The HCV antibody was determined by a third generation commercial enzymatic immunoassay (Ortho Diagnostic Systems). Liver function tests were performed by routine methods.

HBV and HCV molecular biology techniques. HCV RNA qualitative assay was performed by RT-PCR in the 5′ noncoding region of the viral genome using a commercial kit (HEPA-Check-C; Nuclear Laser Medicine). The lowest limit of detection for this method, as stated by the manufacturer, is 200 copies/mL. We determined HCV viral load using a quantitative RT-PCR (HCV Amplicor Monitor; Roche), with the lowest limit of detection, as stated by the manufacturer, of ∼600 IU/mL, on all HCV RNA—positive plasma samples.

Plasma HBV load was determined using PCR on all plasma samples from 13 patients in the case B group. Viral DNA was first extracted from 200 µL of each plasma sample using microspin columns (QIAamp Blood Kit; Qiagen GmbH) and then amplified using primers located in the core region of HBV, as described elsewhere [21]. The lowest limit of detection by this method is 200 copies/mL [22]. In all HBV DNA—positive plasma samples, we determined the HBV load by quantitative PCR using a HBV Amplicor Monitor (Roche). The lowest limit of detection by this method is 400 copies/mL, as stated by the manufacturer.

HBV and HCV genotyping were performed using the Line-Probe Assay (INNO-LIPA HBV Genotyping and INNO-LIPA HCV II, respectively; Innogenetics).

HAV molecular biology techniques. Viral RNA was extracted from plasma samples using microspin columns (QIAamp Viral RNA Mini Kit; Qiagen GmbH) and amplified by nested RT-PCR using primers made at 5′—nontranslated region of HAV, as described elsewhere [23].

Statistical analysis. The mean values were compared using Student's t test; the differences in the proportions were assessed using the χ2 test with the Yates correction or by Fisher's exact test. A P value <.05 was considered to be statistically significant.

Results

At admission to our ward, all patients had jaundice and had been symptomatic for 2–11 days. Fulminant hepatitis was observed neither in the 21 patients with chronic hepatitis B or C and concomitant HAV infection nor in the 21 patients in the control A group. A severe course of illness was observed in only 2 patients, 1 patient (7.7%) in the case B group and 1 patient (5%) in the control A group, who recovered in 36 and 39 days, respectively. A normal, self-limited course of hepatitis A was observed in the remaining 40 patients (95.2%), with a time to recovery of 25–38 days. In no patient did the clinical, laboratory, and ultrasonographic data suggest the presence of liver cirrhosis, most likely because the patients are young adults who probably became infected with HBV or HCV at a young age, when disease severity tends to be mild and when the progression to cirrhosis is slow. Other anamnestic, demographic, biochemical, and virological data obtained on presentation are shown in table 1. In addition, in the case C group, 1 patient was an injection drug user and another was a sexual partner of a chronic carrier of HCV; in the case B group, 3 patients had at least 1 family contact who was HBsAg positive, and 2 patients were sexual partners of chronic carriers of HBsAg.

Figure 1
Figure 1

Prevalence of plasma hepatitis A virus RNA among patients with acute-phase hepatitis A in the case groups and in the control A group.

Table 1
Table 1

Demographic characteristics and risk factors for the acquisition of hepatitis A virus infection and clinical and biochemical data obtained on admission from patients with acute hepatitis A.

On admission, 10 patients (76.9%) in the case B group, 7 patients (87.5%) in the case C group, and 18 patients (85.7%) in the control A group were positive for HAV RNA; no significant difference was observed in the prevalence of HAV RNA across these 3 groups at each subsequent screening point (figure 1). All patients who were positive for HAV RNA became negative within 22 days after the onset of symptoms (figure 1).

On admission, plasma HBV DNA was less frequently detected in patients in the case B group (1 [7.7%] of 13 patients) than in patients in the control B group (13 [50%] of 26 patients); the only patient in the case B group in whom HBV DNA was detected had a low HBV load (4.8 × 103 copies/mL) at admission and became negative for HBV DNA 5 days later. All patients in the case B group were negative for hepatitis B envelope antigen and positive for hepatitis B envelope antibody throughout the course of HAV infection. In the control B group, 6 patients (23%) were positive for hepatitis B envelope antigen, 1 patient (4.4%) was negative for hepatitis B envelope antigen and hepatitis B envelope antibody, and 19 patients (72.6%) were positive for hepatitis B envelope antibody.

Three of the 9 patients in the case B group who were followed up for 6 months showed persistently normal serum ALT levels, which were associated with serum HBsAg clearance and seroconversion to hepatitis B surface antibody; seroconversion was persistent in 2 patients and transient in 1 (table 2). Two patients in the same group remained HBsAg-positive with persistently normal serum ALT levels and undetectable levels of plasma HBV DNA; the remaining 4 patients were found to be HBsAg positive and had low levels of HBV DNA and persistently abnormal ALT serum levels during the follow-up period (table 2).

Table 2
Table 2

Biochemical, serological, and virological test results and outcome by day of follow-up for 13 chronic carriers of hepatitis B surface antigen (HBsAg) with acute hepatitis A.

Throughout the observation period, no change in hepatitis B envelope antigen and hepatitis B envelope antibody status was observed in the 9 patients in the case B group or in the 26 patients in the control B group. HBV genotyping was performed only on plasma samples with measurable levels of HBV DNA; genotype D was identified in all 3 samples that were genotyped.

None of the 8 patients in the case C group had detectable plasma HCV RNA at admission or during the acute phase of illness, whereas the corresponding prevalence in the control C group was 81.2%. of the 6 patients of the case C group who were followed up for 6 months, 3 remained negative for HCV RNA throughout the observation period, and 3 patients became positive for HCV RNA with a low level of viral replication (<5 × 103) (table 3). HCV genotyping, which was performed only on plasma samples with a measurable level of HCV RNA, identified genotype 1b in patient 14 and genotype 2a/2c in patient 20 (table 3).

Table 3
Table 3

Biochemical, serological, and virological test results and outcome for 8 chronic carriers of hepatitis C virus antibody with acute hepatitis A.

Patients in the case B group and the case C group having a follow-up period of 6 months were evaluated again with biochemical and virological tests 1–3 years later. None of them had been treated by their physicians with antiviral drugs, and all of their conditions had remained substantially unchanged. Patients with high ALT serum levels refused to undergo liver biopsy.

Discussion

Fulminant hepatitis never occurred, and a severe clinical form was rare in patients with chronic hepatitis B or C with concomitant HAV infection. In addition, no difference was found in the clinical presentation and course of acute hepatitis A between patients with preexisting chronic HBV or HCV infection and patients without it. The favorable outcome of acute hepatitis A in chronic carriers of HBV or HCV is probably because of the frequent mildness of the chronic disease in young patients, most of whom have been infected early in life, when the progression of the disease is slow. Accordingly, a recent multicenter prospective Italian survey [12] showed an overall HAV antibody seroconversion rate of 1.2 per 100 persons per year in 76 susceptible individuals, 74 of whom had chronic hepatitis and 2 of whom had cirrhosis; none of the seroconverters had liver cirrhosis, nor had they shown overt acute hepatitis A or worsening of the preexisting chronic liver disease. Moreover, the Italian Surveillance System of Acute Viral Hepatitis reported that none of the 179 persons carrying HBsAg and/or HCV antibody with acute hepatitis A observed from 1992 to 1996 showed fulminant hepatitis [8].

Instead, a report from the Centers for Disease Control and Prevention (Atlanta) estimated a case fatality rate of 11.7% among patients with underlying chronic hepatitis B, a rate that is 58-fold higher than the rate among patients with no preexisting liver disease [5]. In addition, during an epidemic of acute hepatitis A in Shanghai in 1988, 15 patients with chronic HBV infection died of fulminant hepatitis A [4].

The different rates of patients with cirrhosis reported in these published studies may be a reason for the discrepancy observed in the incidence of fulminant hepatitis A [4, 5, 24]. It seems reasonable to assume that concomitant HAV infection usually shows a normal, self-limited clinical course in patients with mild or moderate chronic liver disease, whereas severe or life-threatening acute hepatitis A may occur more frequently in patients with preexisting liver cirrhosis, as was observed in patients with HBV-related liver cirrhosis in the Shangai epidemic in 1988 [4] and in a study from Japan published 2 years later [25]. Other studies, however, showed high rates of fulminant hepatitis A, even in HBsAg-positive patients with chronic hepatitis without cirrhosis [2628], suggesting that some other mechanism of liver failure might have been implicated in these cases. Cooksley hypothesized that, in chronic carriers of HBsAg, hepatic failure may be more common when HAV infection occurs in the immuno-eliminative phase of HBV infection, which is characterized by active, specific cell-mediated cytolysis [29]. It has not been established whether HAV infection may enhance the HBV immune response, but it has been demonstrated that an unrelated viral superinfection with lymphocytic choriomeningitis virus in a hepatitis B—infected transgenic mouse causes a down-regulation of HBV in the hepatocytes and an activation of virus-specific cytotoxic T lymphocytes, with enhancement of the liver injury [30]. The hypothesis of an unrelated immunological mechanism of liver injury favoring the development of fulminant hepatitis A in persons with chronic HCV infection without cirrhosis was advanced by Vento [7], who suggested a role of a predisposing HLA haplotype for autoimmune hepatitis.

Positivity for HBV DNA and HCV RNA was not determined before the onset of acute hepatitis A in patients in case B and case C groups, respectively. We may assume, however, that the prevalences of positivity in these 2 groups would be similar to the prevalences found in the control B group and the control C group and those reported in several published papers [3135], ∼50% for HBV DNA and 80%–85% for HCV RNA. On the basis of this evidence, we can infer that concomitant HAV infection in these chronic carriers of HBV or HCV was associated with a marked inhibition of HBV or HCV genomes, respectively.

Analysis of data of the 6-month follow-up of 9 patients in the case B group and 6 patients in the case C group shows that one-half of the patients in both groups experienced a clear improvement with respect to their preexisting chronic hepatitis. The suppression of the HBV genome was particularly marked in 3 patients who showed a persistent absence of HBsAg, persistently normal serum ALT levels, and no detectable HBV DNA in plasma during the follow-up period; clinical improvement was also observed in another patient who remained HBsAg-positive but progressed to an inactive phase of the illness with no sign of HBV replication. The mechanism of this suppression of HBV replication remains unclear, but an action exerted on HBV by a cytokine production induced by HAV may be hypothesized. A sharp increase in IFN-γ production was demonstrated during HAV superinfection in 1 HBsAg-positive patient with chronic hepatitis who was treated with lamivudine [36].

Because 3 of the 6 patients in the case C group also showed a clear improvement, and because they remained HCV-RNA negative with a steady normalization of their serum ALT levels throughout the follow-up period, it seems that HAV may inhibit HCV replication and induce, at least in some cases, a favorable outcome of chronic hepatitis C; this, however, needs to be confirmed with larger studies.

Similar viral interactions have been observed in patients with concurrent acute HBV-HCV infections [34, 3740], in chronic carriers of HCV with concomitant acute HBV infection [4144], and in chronic carriers of HBV with concomitant acute HCV infection [4547]. These studies and the present investigation suggest that acute hepatitis infections may strongly suppress the chronic replication of other hepatitis viruses. Therefore, the temporal sequence and nature of the infections (whether acute or chronic) may be important factors modulating the interaction between viruses and hosts in patients with multiple hepatitis viral infections, with substantial clinical consequences. However, it cannot be ruled out that extensive hepatocellular necrosis due to acute viral hepatitis—either A, B, or C—may limit chronic replication of HBV or HCV infection.

In conclusion, concomitant HAV infection in chronic carriers of HBsAg or HCV antibody with no sign or symptom of liver cirrhosis seems to be characterized by a self-limited clinical and virological course. The consequent inhibition on HBV and HCV genomes may induce, in some cases, a clearance of chronic viral replication and remission from chronic hepatitis.

Acknowledgments

We thank Michele D'Errico, Michela Picozzi, and Lorenza Scialla, as well as all other nurses for their invaluable technical assistance.

Financial support. Progetto di Ricerca di Interesse Nazionale, 2003.

Potential conflicts of interest. All authors: no conflicts.

  • Received November 3, 2005.
  • Accepted January 24, 2006.

References

| Table of Contents