Update on epidemiology of hepatitis B in a low‐endemic European country: There is still much to do

The latest epidemiological data in Spain were obtained a decade ago and revealed a prevalence of hepatitis B surface antigen (HBsAg) of 0.7%; hence, updated epidemiological data are necessary. Our aim was to determine the prevalence of hepatitis B virus (HBV) infection, and to analyse associated factors and characterize chronic infection. A population‐based, cross‐sectional study was performed in Spain between July 2015 and April 2017. Participants from three regions were selected using two‐stage conglomerate sampling and stratified by age. Anthropometric and demographic data were collected, and blood samples were taken to detect serological markers of HBV infection and to quantify HBV‐DNA. The characterization of chronic HBV infection was based on ALT (alanine aminotransferase) values, HBV‐DNA levels, and results of transient elastography. The overall prevalence rates of HBsAg and antibody to hepatitis B core antigen (anti‐HBc) among 12 246 participants aged 20‐74 years (58.4% females) were 0.6% (95% CI [0.4‐0.7]) and 8.2% (7.7‐8.7), respectively. The risk factors for HBV infection identified in the multivariate analysis were age, nosocomial risk, and non‐Spanish nationality. Moreover, most patients HBsAg positive (76.6%) presented as hepatitis B e antigen (HBeAg)–negative chronic infection (formerly ‘inactive carriers’) and only 6 (9.4%) HBsAg carriers fulfilled current criteria for treatment. The current HBV burden in Spain remains low but virtually unchanged over the past 15 years. Increased efforts are still needed to reach the goal set forth by the World Health Organization (WHO) for HBV elimination by 2030.


Chronic hepatitis B virus (HBV) infection remains a major health
threat with approximately 250 million carriers worldwide, many of them being at risk of developing cirrhosis and even hepatocellular carcinoma. 1 There is great variability on a global scale regarding HBV chronic infection being Spain a low-endemic southern European country. 1 The implementation of vaccination programmes against HBV, along with other primary prevention measures, including healthcare infection control and antenatal screening, has led to a decrease in the incidence of acute and chronic hepatitis B (CHB) in many European Union (EU) countries. 1,2 However, migration is currently changing the prevalence of HBV in several low-endemic countries due to higher HBsAg prevalence rates in migrants and refugees from outside Europe. 1 In 2016, the World Health Assembly passed a resolution to eliminate viral hepatitis as a public health threat by 2030. 3 The targets include 90% global coverage of three-dose infant vaccination by 2020, diagnosis of 90% of people infected with HBV, and antiviral treatment of 80% of those diagnosed with HBV and eligible for treatment. Timely, reliable prevalence data are needed to identify the populations who are most affected in order to improve screening and treatment programmes and to monitor the performance and impact of these activities at a strategic level. There are some recent epidemiological reports in different regions worldwide, including countries in southern and western Europe with low prevalence, that indicate a decline in average HBsAg prevalence rates to 0.6%, or even 0.3% in Spain. 1,2 However, these reports are based on estimates from systematic reviews of peer-reviewed literature on HBV prevalence in the general population. Population-based studies are necessary to confirm these estimates and update the latest epidemiological data in Spain that come from studies carried out in the first decade of this century, which showed HBsAg prevalence figures of 0.7%. 4,5 The aim of this study was to update the current burden of HBV infection in Spain, to analyse the associated risk factors, and to characterize chronic infection. according to the different expected prevalence. An uptake of 9%-15% via telephone invitation was estimated. Recruitment was made through a structured phone call that allowed collecting anonymous data from those who declined. Participants underwent a physical examination and were invited to complete a self-reporting questionnaire that was supervised to decrease the missing values (Table S1).

| MATERIAL S AND ME THODS
Blood samples were collected for complete blood count, biochemistry, and serological markers, and they were stored in the Valdecilla Biobank. The presence of HBsAg, and antibody to hepatitis B core antigen (anti-HBc) was assessed by immunoassay (ARCHITECT-i2000SR, Abbott Laboratories). In HBsAg-positive patients, the HBV-DNA blood levels were measured using real-time polymerase chain reaction (PCR) and quantified using the automatic COBAS TaqMan HBV Test 2.0 equipment (15 IU/mL sensitivity). Abdominal ultrasound and transient elastography (FibroScan; Echosens) were performed to determine liver stiffness in HbsAgpositive patients.
A descriptive analysis was performed. The adjusted association with the anti-HBV seroprevalence was investigated with a logistic regression analysis by introducing variables that were related to anti-HBV seroprevalence in a univariate analysis (P < .1) or that were considered clinically relevant regardless of statistical significance.
The strength of association was estimated using the odds ratio (OR) with 95% CI. Analyses were performed using SPSS Statistics for

| RE SULTS
12 246 participants were recruited, which represents 22.8% of recruitment success (Figure 1). 83.5% of those who declined to participate answered a questionnaire. Significant differences between these two groups were observed according to the respondent's age, sex, and level of education (Table S2). However, only differences in the educational level seemed to discriminate between both groups, with 1.5 times more subjects presenting lower education levels in the group that declined to participate (51.6% vs 35.0% [P < .001]).
Factors that were significantly associated with HBsAg detection in the multivariate analysis were age, non-Spanish nationality, and nosocomial risk (Table S6). Eleven of 67 (16.4%) HBsAg-positive patients were not born in Spain, which equates to a prevalence of infection of 1.4% (95% CI 0.8-2.5) in this population, whereas the anti-HBc seroprevalence was 12.0 (93/772; CI 95% 9.9-14.5) (Figures S3 and   S4). Table S7 shows the variables associated with the risk of having

| D ISCUSS I ON
Updated HBV seroprevalence rates were 0.6% and 8.2% for HBsAg and anti-HBc, respectively, in Spain. The risk factors for HBV infection were age, nosocomial risk, and non-Spanish nationality.
The most recent data regarding HBV prevalence in Spain based on systematic reviews or modelling studies reveal discrepant rates F I G U R E 1 Flow chart of the study and characterization of HBsAg-infected patients. FSU, first-stage units; IC, informed consent; and SSU, second-stage units. † HBeAg status could not be assessed; ‡ the patient presented an acute co-infection HIV and HBV at the time of recruitment. He showed HBsAg seroconversion on HAART treatment that included tenofovir. § HBeAg patients could not be fully characterized due to the unavailability of either HBV-DNA or ET (FibroScan ® ). ¶ 4 in five patients who were HBeAg-negative with chronic hepatitis B were already undergoing treatment at the time of recruitment, whereas the other patient started the treatment after being diagnosed during the study. Another patient with HBeAg-negative chronic infection was being treated due to the risk of HBV transmission as a health professional worker (0.34%-0.60%) that may impact the planning of health and financial resources. 1,2 Our results are closer to the latest estimation from the recent modelling study of the Polaris Observatory, which accounts for approximately 190 000 chronically infected patients in the age range studied. 2 Interestingly, data from our study are practically identical to the data obtained 10-15 years ago in two large-scale population-based studies carried out in Spain. 4 HBV-mono-infected HBsAg-positive patients in 1999 revealed that 67% of these were 'inactive carriers', whereas 33% presented as chronic hepatitis. 8 The study also showed that most CHB infections were anti-HBe-positive (86%), and only 16% of these fulfilled criteria to initiate treatment. Our study confirms the predominance of anti-HBe-positive patients and also that only a very small percentage of patients require treatment based on current guidelines. The study has some limitations that could involve selection and participation bias that have been previously discussed. 6 We cannot rule out some misclassification in the stage of infection due to the study design. The sample size calculation was based on estimated slightly higher values of the prevalence figures for each age stratum 6 ; therefore, a larger number of subjects were studied, which helps provide a more accurate estimation of the parameters.
In conclusion, the current HBV burden in Spain remains relatively low but virtually unchanged in the last 15 years. Many efforts are still needed (educational, preventive, screening in at-risk populations, treatment, and drug research) to reach the WHO commitment of HBV elimination by 2030.

ACK N OWLED G EM ENTS
We thank Mr Angel Estébanez for assistance with database management and laboratory support. We want to particularly acknowledge the patients and the BioBank Valdecilla (PT17/0015/0019), integrated in the Spanish National Biobank Network for their collaboration.