Background

Viral hepatitis is a global health issue that claims the lives of 1.34 million people each year. This is more than the human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS) related mortality of 1 million and upcoming to 1.67 million tuberculosis-related mortality in a single year [1,2,3]. Chronic liver disease associated with hepatitis B virus (HBV) and hepatitis C virus (HCV) is becoming a significant cause of morbidity and mortality among people living with HIV [4]. All three of these viruses share a similar route of transmission [5], such as unsafe sexual contact, blood and blood products, vertical transmission (mainly HBV), horizontal (child-to-child) and injections (mainly HCV) [6] and are preventable; HCV is curable [7].

An estimated 37.7 million people are living with HIV, more than two-thirds of whom dwell in sub-Saharan Africa (SSA) [8]. Among these, 2.7 million are co-infected with HBV and 2.3 million with HCV [9]. According to Barth et al., the rate of HBV infection among people living with HIV (PLWHIV) in Africa is 15%, while HCV infection is 7% [10]. Hepatitis B and C viruses are the common causes of mortality from chronic liver disorders such as Hepatocellular Carcinoma (HCC) and cirrhosis. In SSA, HBV is the common cause of HCC, with 36,700 HCC-related deaths in 2020, and this death rate is expected to double by 2040 [11].

Before the implementation of HAART, coinfection of HIV with HBV and HCV increased mortality [12, 13]. Moreover, HBV infection increases the hepatotoxicity of HAART [14]. Even among HAART-initiated HIV patients, chronic HBV and HCV infections are increasing health challenges [15]. Untreated HIV–hepatitis coinfection, according to the World Health Organization (WHO), hastens progression of HBV and/or HCV–related liver disease, HCC impeding effective HIV treatment [9].

Access to HBV immunization may be lower in HIV-infected infants, or a full dose vaccination may not be protective, and they may be infected by HBV, as a result of a weaker immunologic response and declining titer of antibodies against hepatitis B surface antigens (HBsAg). Besides, the risk of developing chronic hepatitis is about 90% if HBV is acquired during birth or soon after birth, compared to a lower (5%) risk when the infection is acquired in adulthood [16]. Children in SSA are at increased risk of HBV and HCV because home birth is common (> 40%), which increases the risk of perinatal transmission, limits access to HBV vaccination, and needs special attention [17].

In SSA, resources for establishing laboratories with reliable diagnostic tools are limited, which sets back the WHO’s end viral hepatitis target [1]. The prevalence of HBV and HCV infection in HIV-infected children varies significantly with geographic areas and remains to be determined [18]. Those few studies conducted in SSA confirm this, revealing 5.3% in Nigeria and 15% in Tanzania [19, 20]. The Ethiopian Federal Ministry of Health (FMOH) has included HBV vaccination in its Expanded Program of Immunization (EPI) and set a goal of using aggregated national data for evidence-based decisions towards ending viral hepatitis [21]. Published reports on HIV, HBV, and HCV coinfection in children are scarce in the country, even in SSA [22] and lacking in the study area, so determining the local prevalence of these viruses in this segment of population will be the basis for establishing pooled figures in the country. The result will guide health managers, including the Ethiopian FMOH, to give due attention and design effective treatment and control policies. Therefore, this study aimed to provide figures of the seroprevalence and associated factors of HBV and HCV among HAART-experienced children at the University of Gondar Comprehensive Specialized Hospital (UoGCSH).

Materials and methods

Study design and area and period

We conducted a cross-sectional study at the UoGCSH ART clinic from January to May 2020 among HAART-experienced HIV-infected children. The hospital is one of the oldest teaching hospitals and is found in the historic town of Gondar, 740 km away from Addis Ababa, the capital city of Ethiopia. The hospital has been providing different health services such as medical, surgical and many other services in both inpatient and outpatient settings for more than 7 million people from Gondar province and neighboring regions. It launched a free-of-charge HIV/AIDS intervention services and has been providing free diagnosis, treatment, and monitoring in its ART clinic for both pediatric and adult patients since 2003.

Study population

We included 241 HAART-experienced children in this study by using a convenient, non-probabilistic sampling technique. We systematically included all the participants in the study consecutively until the end of the study period (May-2020). This study included all under 15-year-old HIV-infected children attending their follow-up at the UoGCSH, taking HAART and did not know their HBV and HCV status and had no known history HBV vaccination during the study period (Fig. 1).

Fig. 1
figure 1

The flow diagram of the inclusion process of study participants

Full size image

Data collection and laboratory procedures

We collected the socio-demographic characteristics of children, including age, gender, residence, educational status, family income, family size, and family occupation by using pretested and structured questioners via a face-to-face interview. Moreover, we gathered the detailed clinical data of the children such as WHO HIV disease stage at the time of sample collection, type of HAART, and duration of HAART they had been taking at the time of sample collection, by reviewing the medical records of HIV-infected children. Based on the WHO guidelines, the most commonly prescribed HAART regimen consists of nucleoside reverse transcriptase inhibitors (NRTIs: lamivudine (3TC), abacavir (ABC), and zidovudine (AZT), tenofovir disoproxil fumarate (TDF)) together with a non-nucleoside reverse transcription inhibitor (NNRTI, nevirapine or efavirenz) or a protease inhibitor (lopinavir). Currently, the preferred HAART regimen for those children is the release of a new, dispersible tablet formulation of dolutegravir (DTG) plus two nucleoside reverse transcriptase inhibitors (NRTIs) [23]. As the study participants were HIV-infected children, WHO anthro (for children aged ≤ 5 years) and anthro-plus (for children aged > 5 years) software were applied for calculation of their anthropometric measures such as Z-scores of Weight-for-Age (WAZ), Height-for-Age (HAZ) and Body Mass Index for Age (BAZ) based on WHO nutritional assessment guideline. Underweight children are those whose z-score of weight-for-age (WAZ) is less than − 2, while stunted means children who have lower than height-for-age of less than − 2 Z-scores [24].

Regarding the laboratory method, 5 ml of venous blood was collected following standard operating procedures (SOPs) for HIV-viral load determination, centrifuged to extract serum for the analysis of HBsAg and anti-HCV antibodies using enzyme-linked immunosorbent assay (ELISA). The serum specimen was stored at – 80 °C until processed and brought to room temperature and analyzed by thawing once before analysis. Then, the HBsAg and anti-HCV antibody (IgG and IgM) were detected on sera using ELISA according to the manufacturer’s instructions using AiD™ HBsAg ELISA and Wantai Aid™ anti-HCV ELISA Plus, ELISA kits (Beijing Wantai Biological Pharmacy Enterprise Co., Ltd, China). Furthermore, HIV Ribonucleic acid (RNA) viral load was determined directly by an advanced molecular technique using TAQMAN® AMPLICOR HIV-1 MONITOR (Roche Molecular Systems) according to the manufacturer's instruction by well-trained laboratory technologists. To ensure the quality of data, pre-analytical, analytical, and post-analytical phases were strictly maintained. Safety procedures and specimen handling procedures were strictly followed for the tests. (ELIZA, and HIV viral load). The manufacturer procedures and SOPs were strictly followed.

Data analysis and interpretation

Data were coded and entered into EPI-info version 4.4 and were transferred to Statistical Package for Social Science (SPSS) version 20 for analysis. Descriptive statics like frequencies and percentages were used to summarize the results. The distribution of data was assessed through Shapiro–Wilk test and a p-value > 0.05 in the Shapiro–Wilk test was considered as normally distributed data. Both crude odds ratio (COR) and adjusted odds ratio (AOR) with 95% confidence interval (CI) were calculated to determine the extent to which the risk factors are associated with hepatitis infection. Variables such as sex, age, WAZ grouped, BAZ grouped, HAZ grouped, educational status, family marital status, family educational status, family size, lymphopenia, anemia, WHO HIV stage, HIV viral load were subjected to univariable analysis for calculating COR. To identify the independent factors, variables with p < 0.2 at univariable analysis were included in multivariable analysis. The model was built by dropping the most insignificant factors one at a time in a stepwise manner and finally, in the multivariable analysis, a variable with a p-value < 0.05 was considered as statistically significant.

Results

Socio-demographic and clinical characteristics of HAART-experienced children

The total number of study participants in the study was 241, of whom 121 (50.2%) were boys. Among these study participants, 86.7% had been taking the HAART for a year or longer. The median age of participants was 13 years, with an interquartile range (IQR)-of 11–14. In addition, 232 (96.4%), 52 (21.6%), and 204 (84.6%) of the participants were in the WHO clinical stage of HIV-I, had an HIV viral load of above 1000 (both at the time of data collection), and had been living with HIV-infected parents, respectively. All 241 of the study participants were on ART regimens that contain 3TC in its backbone. The overall seroprevalence of HBsAg and anti-HCV Abs among HAART-experienced children who did not know their HBV and/or HCV infection status before, was 12.4% (95% CI = 8.7–17). The seroprevalence of HBsAg was 9.5% (N = 23, 95%CI: 6.6–13.2), and anti-HCV Abs were 2.9% (N = 7, 95% CI = 0.8–5.8). There was no coinfection of HBV and HCV in this study (Table 1).

Table 1 Socio-demographic and clinical characteristics of HAART experienced children at the UoGCSH, 2020
Full size table

The association of socio-demographic features of HAART-experienced children with HBV infection

Among the variables computed in univariable logistic regression to determine the association with HBsAg seropositivity, weight (WAZ) (COR = 3.96: 95% CI; 1.15–13.65 and P = 0.029) and HIV-viral load (COR = 3.14: 95% CI; 1.10–8.97 and P = 0.033) were entered to and analyzed in multivariable analysis. The result of multivariable analysis revealed that WAZ was significantly associated with HBsAg seropositivity (AOR = 3.87: 95% CI; 1.04–14.46, and P = 0.044) (Table 2).

Table 2 Factors associated with HBV infection among HAART experienced children at the UoGCSH, 2020
Full size table

The association of socio-demographic and clinical features of HAART-experienced children with HCV infection

Regarding the association variables with anti-HCV seropositivity, the variable WAZ showed a statistically significant association with HCV infection (AOR = 4.54: 95% CI; 1.21–17.04 and P = 0.025) (Table 3).

Table 3 Factors associated with HCV infection among HAART-experienced children at the UoGCSH, 2020
Full size table

Discussion

The World Health Organization designed a strategy for eliminating viral hepatitis as part of the Global Health Sector Strategy (GHSS) agenda 2030 for sustainable development. It aimed to succeed in diagnosing 90% of infected individuals, treating 80% of diagnosed cases, reducing death by 65%, and vaccinating 90%. Unfortunately, less than 5% of chronic viral hepatitis patients know their status. Knowledge of the burden of HBV and HCV and incorporation of the target population who needs the services in each country are two of the five strategic directions for the GHSS goal [1]. Therefore, this study reveals the seroprevalence of HBV and HCV among HIV-infected children (the population who need the services), which supports determining the national prevalence of viral hepatitis in this population segment.

All the study participants received ART containing 3TC in the backbone as the Ethiopian HIV treatment guidelines recommended. Lamivudine (3TC) is also active against HBV and is commonly used in the treatment of HBV or combined with other drugs for the treatment of HIV or HIV–HBV [25]. The 3TC is accessible at a lower cost in resource-limited settings like Ethiopia [26]. The national HIV treatment guideline requests screening for baseline HBV serology (HBsAg) [25] But HBV diagnosis before initiation of HIV treatment and close monitoring of patients on for HBV viremia and biochemical status are not implemented in Ethiopia. Mutation and rapid progression of drug resistance, exacerbation of HBV disease and liver enzyme flare are the potential risks in this high HIV/HBV burden area [26, 27]. This study did not assess the effect of 3TC because all the study participants were exposed to the drug, tested once and not followed for HBV response.

The overall seropositivity rate of hepatitis among HAART-experienced children was 12.4% (95% CI = 8.7–17). From these, 9.5% were seropositive for HBsAg and the rest 2.9% were seropositive for anti-HCV antibodies. The overall result was comparable to the seropositivity rate of both HBsAg and anti-HCV reported in Nigeria (14.0%) [28]. The result was also analogous to results reported in Guinea, (HBsAg = 8.16%) [29], Sierra Leone (HBsAg = 7.4% and anti-HCV = 1.5%) among children receiving ART[30];Tanzania (HBsAg = 7%) [31], Rwanda (HBsAg = 7%) [32], and Nigeria (HBsAg = 10% and anti-HCV = 1.7%) among ART-naïve HIV-infected children [33]. The SSA is a region of high HBV endemicity (8% and above) [34]. The result of our study indicates that the seroprevalence of HBV and HCV is high among children who are taking HAART.

The result of this study reveals a higher positivity rates of HBsAg compared to results reported by investigators from West Africa (2.2–5.3%) [20, 35,36,37,38] and comparable positivity rates of anti-HCV (0.5–3.9%) results from the same region [20, 37, 39].

The variation might be because of population and geographic differences. Some authors included infants, while others included those adolescents who were aged above 15 years old. But, in this study, children aged 2 to 15 years old were the study population. Efforts to ensure health care, expand HBV vaccination programs, and the goal of ending viral hepatitis might also be responsible for the difference. The result of this study indicates that Ethiopia is far behind the goal set by the WHO to end viral hepatitis by 2030 [40]. The WHO recommendations for routine screening of HIV patients for HBV and HCV infections were not implemented in Ethiopia, in which resources are limited, and the result of our study reflects this weak move towards the target [41].

Unlike the result of this study, a study involving ART-naïve HIV-infected children in Bahir Dar, Ethiopia, reported a reduced seroprevalence of HBV (HBsAg = 2%) and a borderline higher HCV (anti-HCV = 5.5%) [18]. In addition, a study done in Nigeria reported a lower HBV (HBsAg = 3%) and a higher HCV (anti-HCV = 11%) infection among HAART-experienced children [28]. The difference in the length of exposure to HAART and anti-HBV activity of the regimens could have reduced HBV infection. In contrast, the higher HCV infection rate might be attributed to weak immune reconstitution in these children. Hepatitis C virus coinfection in HAART-experienced HIV patients has been shown to have a low immune/CD4 reconstitution [18, 41].

Among the variables analyzed in this study, weight was the variable significantly associated with HBV and HCV infections. The likelihood of HBV seropositivity in HAART-experienced underweight children was nearly fourfold higher than the likelihood of HBV seropositivity in normal-weight children (AOR = 3.87). Similarly, the odds of HCV seropositivity among underweight (WAZ) children was at least 4.5 times higher than the odds of HCV among normal-weight children. The higher likelihood of hepatitis seropositivity among underweight children might be associated with poor nutritional status and associated poor immune status or immunosuppression by HIV infection, and the resulting lower clearance capacity for the viruses. Strengths of this research: this research focused on the high-risk population segment. Hence, it can be an example of the local prevalence of HBV and HCV among HAART experienced children living with HIV in the high burdened SSA region. We included large sample size and ELISA rather than rapid diagnostic techniques for detection. It can also be used as baseline data for further study and control strategy. This study was limited by the lack of data on the history of vaccination, absence of HBV-core antibody tests and molecular tests for detection HBV and HCV, lack of longitudinal data and absence of liver function tests. There were two children who had history of vaccination and were excluded.

Conclusions

The findings of this study showed that the seroprevalence of hepatitis B and C virus infection was high among HAART-experienced children with unknown previous hepatitis status. The variable being underweight was associated with both hepatitis B and C viruses. Therefore, screening HIV-infected children and implementing appropriate remedial measures are crucial to reducing these infections. Large-scale population-based studies on are needed to determine the national prevalence to help reach the WHO end viral hepatitis goal by 2030.