Introduction

Rheumatoid arthritis (RA) is the most common form of chronic inflammatory joint disease associated with extra-articular and systemic effects [1, 2]. The extra-articular manifestations of RA can occur at any age after onset and more common in seropositive RA patients [3, 4]. Many studies have indicated that infections may contribute to the pathogenesis of the disease [5,6,7,8,9]. In addition, the coexistence of RA and various infections is speculated to facilitate the production of a wide range of autoantibodies as well [10,11,12].

In a series of studies, we described the coexistence of rheumatologic and autoimmune diseases with several infectious diseases [13, 14]. Patients with rheumatic diseases are at a higher risk of infection due to several factors including dysregulated immune-pathways in the host, and immunosuppressant therapy [15]. We previously showed that hepatitis C virus (HCV) infection is more common among systemic lupus erythematosus (SLE) patients compared with a large group of age- and sex-matched controls [16]. This was similarly illustrated regarding hepatitis B virus (HBV) among SLE patients [13]. Viruses, such as HBV and HCV as mentioned above, seem to be associated with many systemic rheumatic diseases [14, 17, 18]. A nationwide study from Taiwan based on extracted data between the years 1999 and 2009 showed a higher prevalence and risk for HBV infection in RA patients compared with non-RA cohort [19]. Hsu et al. [20] took this further and showed, in a recent study, HBV core antigen is present in the synovium of 64% of patients with RA and chronic HBV infection demonstrating an important role of HBV in the pathogenesis and progression of RA. HBV and hepatitis D virus (HDV) are DNA viruses and the former represents the leading cause of chronic hepatitis, cirrhosis, and hepatocellular carcinoma (HCC). According to the WHO, HBV infection is prevalent in the Western Pacific and African regions where around 6–6.5% of the population are affected [21]. Serologic testing for HBV is mandatory in RA patients especially those planned to receive potentially hepatotoxic medications such as methotrexate or immunosuppressant therapy, namely biologic agents [22]. Vaccination to nonimmune patients is warranted and treatment should be offered to those chronically ill. Though considered safe, Kanduc and Shoenfeld suggested a safer approach to HBV vaccine production [23].

In the current study, we aimed to assess whether RA and chronic HBV infection coexist more commonly, by conducting a cross-sectional study utilizing an extensive database of Clalit Health Services (CHS).

Methods

Measures

A cross-sectional population study has been conducted utilizing the CHS database. The latter is the largest healthcare maintenance organization in Israel, providing public and semi-private health services for approximately 4,400,000 members representing half of the Israeli population. The CHS chronic disease registry receives input data from a variety of sources, namely pharmaceutical, medical, and administrative computerized operating systems.

RA diagnosis was defined as such when there was a documented diagnosis of this entity at least twice in the medical records registered by a physician in the community or when they were listed in the diagnoses of discharge letters from hospitals (specialist). All RA patients detected in the CHS database were considered eligible and, as such, enrolled in this study. Controls were randomly selected from the CHS database, with the exclusion of RA patients.

Similar to RA, HBV infection was determined according to the subjects’ medical records. These diagnoses were validated by a systematic methodology, performing regular checks of these diagnoses logistically by comparing different sources of information and directly by physician-patient individual level. The validity of the diagnoses in the registry has been shown to be high and was used for research purposes in the past [24,25,26,27].

Statistical analyses

Using a χ2 test for gender and socioeconomic status and a t test for age; the distribution of sociodemographic and clinical factors was compared between patients with and without RA. Crude ORs and 95% confidence intervals (CIs) are presented. Logistic regression models were used to estimate the association between RA and chronic hepatitis B in a multivariate analysis. Statistical analysis was performed using R Statistical Software (version 3.2.2; R Foundation for Statistical Computing, Vienna, Austria).

Results

The study included 11,782 RA patients and 57,973 age- and sex-matched controls. In both study groups, the median age was around 61 years old, with a predominance of the female gender (77%). Pertinent to SES, no significant difference was noted between RA patients and controls. The smoking rate was significantly higher in RA patients than in controls (32.8% vs 28.8%, respectively, < 0.001). Regarding alcohol abuse habit, no significant difference was found among the two study groups (p = 0.474). Interestingly, there was a significantly higher proportion of chronic HBV infection in RA patients as compared with controls (1.19% and 0.63%, respectively; p < 0.001). Further details regarding the basic characteristics of both study groups are reported in Table 1.

Table 1 RA patients and matched controls basic characteristics
Full size table

The interaction between RA and various factors in terms of association with chronic HBV infection reported the highest OR of 2.32 (95%CI 1.33–3.90) among those of high SES (Table 2). The association was consistent in both sexes although it was more prominent in females. Moreover, the association was significant through all SES and age groups.

Table 2 Interaction between RA and various factors in terms of association with chronic HBV infection
Full size table

In a multivariate logistic regression analysis and after adjusting for confounders including age, socioeconomic class, alcohol abuse habit, and smoking, RA was significantly associated with chronic HBV infection (OR = 1.89, 95%CI 1.46–2.90). Moreover, alcohol abuse habit (OR 4.13, 95%CI 2.211–7.072) and smoking (OR 1.25, 95%CI 1.029–1.514) were significantly associated with chronic HBV infection (Table 3). By contrast, high SES was inversely associated with chronic HBV infection (OR 0.44, 95%CI 0.334–0.580) (Table 3).

Table 3 Logistic regression—covariates associated with chronic HBV infection
Full size table

Discussion

In this large cross-sectional study, a higher proportion of chronic HBV infection in RA patients compared with controls was found.

The association between rheumatic/autoimmune diseases and infection has long been described in medical literature [13, 28, 29]. The link between infection and RA is bidirectional. On the one hand, several infections such as EBV, parvoviruses, and retroviruses have been reported to trigger RA disease although establishing a causal relationship between the two conditions is extremely difficult [30, 31]. On the other hand, RA patients are at a higher risk of infection and this can be attributed to various factors such as a dysregulated immunity related to the disease itself or anti-disease medications [32, 33].

HBV infection can manifest with an array of rheumatic manifestations and autoimmune laboratory findings. These manifestations seem to occur in around 20% of patients infected with HBV whether acute or chronic and thought to be mediated mainly by circulating immune complexes [34, 35]. These include, among others, arthralgias, arthritis, myalgia, fatigue, and fibromyalgia [36].

Regarding HBV and RA, Lin et al. [37] compared HBV testing in the USA and Taiwan and showed low testing rates in both countries, though this rate is increasing in the last decade. In fact, the association between RA and HBV is described in the medical literature mainly in the context of the reactivation of HBV infection in RA patients receiving immunosuppressive therapy including glucocorticoids as well as biologic agents [38, 39]. This phenomenon is known since the introduction of biologic agents in RA [40]. For this reason, antiviral therapy is recommended for such patients while on biological drugs [22]. Our report supports this recommendation, given the excessive cooccurrence of RA and the diagnosis of chronic HBV infection.

Jeong et al. [41] in a study from South Korea compared RA population with matched non-RA population and illustrated a higher prevalence of comorbidities in RA patients. Significant associations were found between RA and myocardial infection, angina pectoris, thyroid disease, and depression, but only a nonsignificant trend was observed concerning HBV infection in RA patients. Although there are some similarities between the Korean and our study, our study was designed to primarily investigate the prevalence of chronic HBV infection in RA, demonstrating a statistically significant association between the two conditions. It is pertinent to mention that HBV incidence and prevalence differ worldwide; this seems to be one of the main reasons for the difference between the findings of the two studies.

Our study has several strengths, mainly being based on a very large database. However, there are several limitations that should be acknowledged. The main limitation is the lack of data concerning the medical treatment of RA patients, serology, and their disease activity.

In conclusion, we found a higher proportion of chronic HBV infection among RA patients in a large control case study. Physicians in general and those dealing with RA patients in particular should be aware of these findings and probably consider HBV infection screening in all RA patients regardless of their planned or current medical therapy.