Crohn's disease (CD) and ulcerative colitis (UC) are the major forms of idiopathic inflammatory bowel disease (IBD) with a remitting and relapsing course. To induce and maintain remission in IBD, most patients receive immunosuppressive agents, including corticosteroids, thiopurine antimetabolites such as azathioprine and 6-mercaptopurine, and anti-tumor necrosis factor α (anti-TNF-α) agents such as infliximab.1 Reactivation of hepatitis B virus (HBV) replication is a well-known complication in patients with chronic HBV infection who receive cytotoxic or immunosuppressive therapy.2 Because HBV infection remains highly prevalent in Asia3 and the incidence of IBD is rapidly increasing in some Asian countries,4, 5 understanding the clinical features of HBV infection in IBD patients is very important. Little is known, however, about the clinical characteristics of IBD patients with chronic HBV infection. A multicenter study from Spain showed that treatment with ≥2 immunosuppressants was an independent predictor of liver dysfunction in IBD patients infected with HBV.6 A hospital-based study in Greece indicated that IBD patients infected with HBV can be safely treated with immunosuppressants if they are also treated with preventive antiviral treatment.7 Those two studies, however, included only 25 and 11 IBD patients who were positive for hepatitis B surface antigen (HBsAg), respectively, indicating the need for further investigations of the clinical features and courses of HBV infection in patients with IBD. We therefore assessed the influence of immunosuppressants on the course of chronic HBV infection in patients with IBD. We also compared the clinical course of IBD patients with HBV infection with matched IBD controls without HBV infection at a single tertiary center in Korea.
- Top of page
- PATIENTS AND METHODS
To our knowledge, this study is the largest to date to assess the clinical courses of chronic HBV infection and IBD in patients with both diseases, and the first to determine the outcome of chronic HBV infection in Asian patients with IBD who were treated with immunosuppressive regimens. In Korea, the prevalence of HBV was almost 8% of the general population until the 1980s.11 After the immunization for all neonates was begun since 1991 (National Immunization Program), the prevalence of HBV infection was markedly reduced, especially in children (0.6%), adolescents (1.6%), and in their early 20s (3%).12 However, the prevalence in adults still showed no significant differences to the prevalence rates in the past.13 This study differs from previous studies, in that we included HBsAg-positive IBD patients, but excluded IBD patients with antibodies to hepatitis B core antigen (anti-HBc) alone. Although HBV may be reactivated by immunosuppressive treatment in patients with anti-HBc alone, its incidence is very rare. Therefore, we sought to include only patients at high risk of HBV reactivation. This study also differs from previous studies in that we evaluated not only the clinical outcomes of chronic HBV infection but also those of IBD by comparing IBD patients with and without HBV infection. This was possible due to our inclusion of a relatively large number of HBsAg-positive patients. We found that prolonged immunosuppression (>3 months) was significantly related to liver dysfunction, and that IBD patients with HBV infection were less frequently treated with immunosuppressants than IBD patients without HBV infection. In addition, several clinical outcomes, including hospital admissions, surgery, and mortality, were worse in IBD patients with than without HBV infection.
Reactivation of HBV has been reported in 21% to 53% of patients with chronic HBV infection who receive cytotoxic therapy for hemato-oncological malignancies.14–16 In addition, the rates of HBV reactivation in patients with HBV and rheumatoid arthritis who receive anti-TNF-α therapy were 0% in those who received antiviral prophylaxis and 63% in those who did not.17 However, less is known about the rate of HBV reactivation in IBD patients administered immunosuppressive therapy, such as corticosteroids, thiopurines, and anti-TNF-α agents.18
In a recent multicenter study from Spain, HBV reactivation was observed in 9 of 25 (36%) HBsAg-positive patients with IBD who had received immunosuppressive agents.6 Of these, six patients developed hepatic failure, with three undergoing liver transplantation and one dying. Considering the higher prevalence of chronic HBV infection in our IBD patients (4.1%) than in IBD patients in Western countries (mean prevalence 1%),18 HBV reactivation in IBD patients is likely a more common problem in Asian than in Western patients. We observed liver dysfunction related to chronic HBV infection in 17.2% of 134 HBsAg-positive patients, and in 25.7% of 70 HBsAg-positive patients who had received immunosuppressive therapy. This rate of HBV reactivation was lower than the 36% reported in Spain. In addition, only one of our patients developed fulminant hepatic failure; this patient recovered following full supportive care without liver transplantation, a finding different from the results of the Spanish study. This discrepancy may be partially explained by relatively short periods of follow-up (median 56 months) and immunosuppression (median 6 months) in our patients, and by ethnic differences in the response of carriers to HBV infection.19
Of our 70 HBsAg-positive patients receiving immunosuppressive therapy, only 11 received antiviral prophylaxis, with three (27.3%) developing liver dysfunction. This rate of liver dysfunction despite antiviral prophylaxis was higher than in previous reports,6, 17 but we observed no difference in the rate of liver dysfunction between patients who did and did not receive antiviral prophylaxis. In contrast, other studies, including the study from Spain, observed a trend toward a lower rate of HBV reactivation in patients who did than did not receive antiviral prophylaxis, although the differences were not statistically significant due to the small number of patients in earlier studies.6 Our finding, of a high rate of HBV reactivation despite antiviral prophylaxis, may have been due to patient failure to follow guidelines for antiviral prophylaxis. The consensus of the European Crohn's and Colitis Organisation recommends that chronic HBsAg carriers receive prophylactic antiviral treatment with nucleotide/nucleoside analogs, starting 2 weeks prior to the initiation of treatment with steroids, azathioprine, or anti-TNF-α, and continuing for 6 months after their withdrawal.20 We found that three patients received inappropriate antiviral prophylaxis because of a strict Korean government health insurance reimbursement policy. Until now, the cost of prophylactic antiviral treatments in patients without liver dysfunction (i.e., an increase of 2-fold the upper normal limit of serum ALT) was not reimbursed by public health insurance in Korea. For the same reason, only 11 of 70 HBsAg-positive patients who had received immunosuppressive therapy also received antiviral prophylaxis. Among the eight patients who received adequate antiviral prophylaxis, one (12.5%) experienced liver dysfunction. Therefore, our results indicate the importance of strict antiviral prophylaxis, in accordance with treatment guidelines.
We found that immunosuppressants were less frequently used for HBsAg-positive than HBsAg-negative IBD patients, likely due to the fear of HBV reactivation and to the strict Korean insurance policy on antiviral prophylaxis. A lower rate of immunosuppressant use may have an adverse effect on the prognosis of patients with IBD. We found that clinical outcomes, including admission rate, mean number of admissions per person, surgery rate for UC, and mortality were worse in HBsAg-positive than HBsAg-negative IBD patients. Especially the higher rate of total proctocolectomy in HBsAg-positive than HBsAg-negative UC patients may have been due to the desire to avoid repeated and prolonged use of immunosuppressants. Thus, the greater use of immunosuppressants, together with appropriate antiviral prophylaxis, may have reduced the rate of total proctocolectomy in HBsAg-positive UC patients.
Our findings must, however, be interpreted against the background of potential limitations. First, this was a retrospective study. Therefore, the evaluation of HBV reactivation and the use of antiviral prophylaxis were not standardized in our patients, thus precluding an accurate estimation of the frequency of HBV reactivation and of the efficacy of antiviral prophylaxis. Second, this study included ethnically distinct Korean patients only, with vertical transmission of HBV being the major mode of chronic HBV infection in Korea. Therefore, our results may not be applicable to Western populations or patients with other modes of transmission. However, there is no evidence of ethnic differences of HBV reactivation in patients with rheumatic diseases, with studies performed in both Asian and Western populations.17, 21
In conclusion, liver dysfunction in HBV-infected IBD patients was more frequent in those receiving prolonged immunosuppression. We also found that immunosuppressants were less frequently used for IBD patients with than without HBV infection, and that IBD patients with HBV infection had worse clinical outcomes than those without HBV infection. These findings suggest that HBV-infected IBD patients should receive prophylactic antiviral agents and immunosuppressants without delay. Further prospective, long-term follow-up studies are needed to confirm these observations.