Viral hepatitis is a serious public health problem affecting billions of people globally. Caused mainly by hepatitis viruses A, B, C, D and E, and rarely by cytomegalovirus (CMV), Epstein-Barr virus (EBV) and fungal infections, the spectrum of hepatitis range from sub-clinical to milder and life threatening illness including hepatocellular carcinoma [1, 2]. Globally two billion people are infected with HBV, and 350 millions of them have chronic (lifelong) infections, who are at high risk of death from liver cirrhosis and liver cancer that kill more than one million people globally each year [3]. In the Middle East and Indian sub-continent, HBV infection is of intermediate endemicity with chronic HBV carriage rate of 2-5% among general population [3]. In Bangladesh, there is paucity of information on the prevalence of HBV infections among general population and majority of the previous studies were conducted in selected group of people with higher risk factors such as blood donors, drug addicts, commercial sex workers (CSWs) or hospitalised patients [4–8]. However, a recent report showed 5.5% HBsAg positivity among the general population living in Savar, a semi-urban area on the outskirts of Dhaka [9]. Although HBsAg is the most reliable biological biomarker of HBV infection, and the anti-HBc antibody is an important marker for surveying the burden of HBV infection as it persists even after resolution of infection, and thus identifies both past and current HBV infection [10]. As majority of the previous studies in Bangladesh examined only the prevalence of HBsAg [2, 4–7, 11–16] and most of them were conducted in selected group of people with higher risk factors, we decided to estimate the prevalence of both HBsAg and anti-HBc among the general population of Dhaka, Bangladesh.

HCV infections is also a major global health problem with an estimated 170 million people chronically infected and 3-4 million people get new infections each year [17]. HCV infections lead to acute hepatitis in 20% cases, and chronic hepatitis in 50-80% cases, of whom 10-20% develop liver cirrhosis [17–20] and 1-5% develop liver cancer in 20-30 years [17]. The prevalence of HCV infection varies throughout the world with significant regional and ethnic differences. The highest number reported from Egypt following parenteral antischistosomal therapy with a prevalence of anti-HCV of 6-28% (mean 22%) depending on the regions [21]. Although some risk factors for acquiring HCV infections are present in 50% cases, no recognizable transmission factor could be identified in the remaining 50% [17, 22]. As there is also lack of information on the prevalence of HCV infection among the general population and nearly all previous studies were conducted in selected group of people [8, 23–25], we also decided to estimate the prevalence of anti-HCV in the same population.

It is evident that HBV and HCV infections are both major global health problems, and they are rapidly spreading in developing countries due to the lack of health education, poverty, illiteracy and lack of hepatitis B vaccination. As many chronically infected individuals remain asymptomatic, and thus undetected for many years, we planned this population-based serological study to determine the prevalence of HBsAg, anti-HBc, and anti-HCV among all age groups and to identify the possible risk factors for acquiring the infections. We hoped that the findings might guide eventually the development, adaptation, and evaluation of prevention strategies.

Although prevalence studies are not always easily undertaken in the developing countries such as Bangladesh due to high cost, we made efforts to prospectively estimate the prevalence of HBV and HCV infections among a population living at Kamalapur, a densely populated community in Dhaka, the capital city of Bangladesh.

The results of our study suggest a high HBV exposure among our study population. The HBsAg prevalence of 6.5% among our study population is within the range of 2-7%, reported by previous studies from selective population of Dhaka: 3% among healthy adults and children [33] and 3.5% among pregnant women [5]; however, much lower rate (0.8%) was observed among school children [16]. The prevalence was also within the same range as reported from some high-risk groups from Dhaka: 4.4% among non-injectable drug users (non-IDUs) and 6.2% in injectable drug users (IDUs) [24]; and 5.9% among truck drivers and helpers [34]. The higher rates among our study population could be attributed to the general lack of proper health care because of deprived socio-economic status (monthly household income of US $50) and less public health awareness about the transmission of HBV infection as well as the lack of hepatitis B vaccination in the community. Our prevalence was similar (6.5% vs. 6.4%) to that reported from the Rangpur district of Bangladesh [13]. However, higher prevalence has also been reported among the high-risk groups of Dhaka: 7.6% among women at a STD clinic [23]; 8% among IDUs [6]; 8.6% among surgically operated patients [7]; 9.7% among CSWs [8]; 19% among hepatitis patients [2]; and 47% among hepatocellular carcinoma patients [14].

The prevalence of anti-HBc in our study population was similar (22.6% vs. 21.1%) to that reported from healthy adults and children [33] but lower than that reported from the high-risk groups of Dhaka: 24.1% in non-IDUs and 31.8% in IDUs [24]; 35.2% among women at a STD clinic [23]; 48.1% among truck drivers and helpers [34]; 49.3% among women living near a truck stand [25]; and 73% among CSWs [8].

The 0.2% prevalence of anti-HCV observed in our study population is lower than that reported from high-risk groups of Dhaka: 0.8% among truck drivers and helpers [34]; 0.9% among women at a STD clinic [23]; 1.6% among women living near a truck stand [25]; 5.8% in non-IDUs and 24.8% in IDUs [24]; and 13% among hepatitis patients [2].

The intermediate rate of chronic HBV carriage of around 3% was observed in most general populations (clinics, villagers), suggesting that this population would benefit from universal hepatitis B vaccination [11]. In 2004, the Government of Bangladesh and UNICEF have introduced the hepatitis B vaccine into the Expanded Programme on Immunization (EPI) against six infectious diseases. The successful continuation of the programme is expected to reduce chronic HBV infections in the next generations. Since 90% of the HBV infected older children and adults successfully clear the infection and do not become chronic carriers, the prevalence of HBsAg alone might not describe the total burden of HBV infections. Therefore, estimation of the prevalence of anti-HBc, in addition to the estimation of the prevalence of HBsAg which is the most reliable biological biomarker of HBV infection, is much more informative about indicator of HBV disease burden among the population. This probably accounts for our higher anti-HBc (22.6%) than HBsAg (6.5%) seropositivity rates (Figure 1).

There are some limitations of our study. First, we did not perform some diagnostic tests for HBV, e.g. anti-HBc IgM, the presence of which indicates acute infection; and anti-HBs that differentiates susceptible persons from those immune persons, which can be due either to natural infection or hepatitis B vaccination. Second, we did not perform some diagnostic tests for HCV, e. g. recombinant immunoblot assay (RIBA) to confirm HCV exposure, or polymerase chain reaction (PCR) to detect HCV infected individuals. All the above limitations are mainly due to study cost constraints, mostly related to laboratory tests. The third limitation is that the study was conducted in a single population in Dhaka, and may not reflect all of Bangladesh, although the literature we have cited suggests that it should. The fourth limitation is the selection of Kamalapur as the study site, as the activities of ICDDR,B might influence the prevalence of HBV and HCV, which might be a biased location because the people of this area might have altered KAP (knowledge, attitude and performance) regarding infectious diseases due to prolonged presence of ICDDR,B team and their health education. A final limitation is the relatively short observation window, which may have missed important secular trends in the background prevalence of both the hepatitis B and C viruses.

The results of our study indicate intermediate level of endemicity of HBV infection such as overall intermediate prevalence of HBsAg; but high anti-HBc, indicating that the members of this urban community are highly exposed to a high prevalence of the hepatitis B virus, which may be acute, chronic or resolved, in an urban community in Dhaka, Bangladesh. We also noted higher prevalence among the family members residing at the same household of that of the HBsAg positive individuals, suggesting likely intra-household transmission. However, we observed a much lower prevalence of HCV infections in the same community. We also identified some independent risk factors, which could help health care providers and policy makers in designing and initiating effective preventive programmes. The findings also highlight the need for prevention and control of HBV infection in Bangladesh by implementing universal hepatitis B vaccination and creating public awareness to promote the use of disposable needles for ear-nose-body piercing.

Future follow-up studies are required to confirm the family clustering effect of HBV infection as observed in Korea [35], and to better define the transmission dynamics, and to identify their common risk factors for acquiring HBV infection by comparing family members of HBsAg positive participants with those who are negative for the tests. Such a study is planned to be conducted for the same community in Dhaka. Further, more long-term population-based surveillance studies, with extended serology of HBV infection, are needed to more accurately assess the hepatitis B true disease burden in Bangladesh, the impact of vaccination, and to guide prioritization of scarce health care resources.