Introduction to elastography

Hepatitis D virus (HDV) is a hepatotropic virus that causes acute and chronic liver disease [1]. HDV is often described as a “satellite virus” or an “incomplete virus”, as it can only coexist and complete its life cycle in the presence of the hepatitis B virus (HBV) [2]. Compared to the general population, HBV/HDV coinfections are more common in individuals with human immunodeficiency virus (HIV) on account of their shared routes of transmission. The main transmission routes of HDV are parenteral and sexual contact. Mother-to-child transmission is rare but can also occur [3]. Currently, HDV infection is considered a relatively uncommon disease as a result of the universal promotion of HBV vaccination [4]. However, due to the lack of widespread HDV screening, its prevalence may be underestimated [5]. HDV infection may be missed, particularly in people with HIV (PWH) receiving antiretroviral therapy (ART) with anti-HBV activity and coincidental HBV suppression [6]. Moreover, the majority of studies assessing the prevalence of HDV are based solely on anti-HDV antibodies, without the confirmation by HDV RNA testing. The prevalence of HDV is estimated to be between 0.16% and 1.00% in the global general population, and between 4.5% and 14.6% in HBsAg-positive patients, using anti-HDV antibodies [7]. Among PWH, the estimated prevalence of HDV infection based on anti-HDV screening is 1.7% [5].

In Poland in the general population among HBsAg positive patients in a single-centre study from Silesia, anti-HDV antibodies were detected in 3.4% of HBV-infected patients. However, HDV infection was assessed only serologically and was not confirmed by HDV RNA PCR [8]. More recent multicentre data from eastern Poland showed a lower prevalence of HDV. Among 398 patients with chronic HBV, only 6 (1.5%) were anti-HDV positive, whereas active HDV replication confirmed by RT-PCR HDV RNA was found in 1 (0.25%) of the cohort [9]. In people living with HIV/HBV coinfection, Polish data are very limited. In a study from the Łódź centre, anti-HDV antibodies were found in 6 of 28 HIV/HBV-coinfected patients, corresponding to 21.4%; however, HDV RNA PCR was not performed. Similar results were obtained in a European cohort with prevalence of 22.2%, also diagnosed only with anti-HDV antibody [10, 11].

HDV diagnosis is important for PWH, since, compared to HBV monoinfection, HDV coinfection increases the risk of cirrhosis, hepatic decompensation, hepatocellular carcinoma, the necessity for liver transplant, and mortality [12]. Moreover, it may impact HIV outcomes, because the combination of HIV with other viral hepatitis types can promote immune activation, causing dysfunction of CD4+ and CD8+ T lymphocytes and natural killer cells. These effects may result in poor immune recovery during antiretroviral therapy and progression to acquired immunodeficiency syndrome (AIDS) [13].

The aim of this study was to characterise patients with HDV infection among HIV/HBV-coinfected patients in HIV treatment centres in Warsaw, Poland, in 2012, and to analyse factors associated with the risk of acquiring HDV. A secondary outcome analysed was whether HDV infection in HIV/HBV-coinfected patients is a risk factor of mortality in a 12-year observation period. Additional outcomes analysed were whether other characteristics (age, sex, liver cirrhosis, HCV infection, or CD4+ T lymphocyte count) are risk factors of mortality in HIV/HBV-coinfected individuals.

Material and methods

Patients

In this observational cohort study, all patients attending two HIV treatment centres in Warsaw in 2012 were examined, and those with hepatitis B surface antigen (HBsAg) positivity were recruited. Inclusion criteria were age ≥ 18 years old, confirmed HIV-1 infection and HBV coinfection. Exclusion criteria were acute hepatitis and incomplete data at the point of recruitment.

Assessments

In all patients in 2012, anti-HDV antibody and HDV RNA analysis in human plasma were performed. The presence of HCV coinfection, the occurrence of cancer, opportunistic infections, lymphocyte T CD4+ cell count, HIV-1 viral load, HBsAg status, and HBV DNA were also analysed at the moment of recruitment, along with the prevalence of liver cirrhosis, the length of HIV treatment, applied antiretroviral therapy, and HIV transmission route.

HIV-1 infection was indicated by a positive western blot test for HIV-1 or detectable HIV-1 RNA. HBV infection was defined as HBsAg positivity at the moment of recruitment (in 2012) and detectable HBV DNA at the point of recruitment or in past medical history. HDV infection was indicated by a positive HDV RNA test at the point of recruitment (in 2012). Active HCV infection was defined by a positive HCV RNA, and past infection by positive anti-HCV antibodies, negative HCV RNA, and no history of HCV treatment.

The patients were followed for 12 years in terms of development of liver cirrhosis or death. All antiretroviral regimens applied during the observation period were also analysed, including whether the patients disrupted antiretroviral therapy (ART) at any point during the observation, whether the patients acquired HCV infection (indicated by a positive HCV RNA test in a patient with a previous negative result) and, if so, whether antiviral therapy was applied. All patients were evaluated in terms of developing cancer of opportunistic infection during the observation. HBV DNA, HBsAg status, lymphocyte T CD4+ cell count, and HIV-1 viral load were also analysed after 12 years.

HDV RNA was extracted from 150 µl of human plasma using a Ribo Virus spin column extraction kit (Sacace Biotechnologies, Italy), according to the manufacturer’s instructions. Quantitative detection of HDV RNA was performed using the HDV Real-TM Quant kit (Sacace Biotechnologies, Italy), a one-step real-time PCR assay designed for the detection and quantification of HDV RNA. The HDV Real-TM Quant assay has a linear quantification range from 30 to 100,000,000 copies/ml. The analytical sensitivity of the assay was no less than 30 copies/ml, and analytical specificity was reported to be 100% based on internal validation by the manufacturer.

Plasma samples from patients chronically infected with HBV were processed using an Abbott m2000sp automated sample preparation system with the mSample Preparation System DNA kit (Abbott Molecular, USA). HBV DNA was extracted via magnetic microparticle-based technology. The Abbott RealTime HBV assay is linear across a dynamic range from 10 IU/ml to 1 × 10⁹ IU/ml. The assay is validated for all major HBV genotypes (A-H) and is intended for the monitoring of HBV DNA levels in chronically infected patients, including assessment of response to antiviral therapy.

Quantification of HCV RNA was performed using the COBAS AmpliPrep/COBAS TaqMan HCV Test, version 1.0 (Roche Molecular Systems, Inc., Branchburg, NJ, USA), a fully automated real-time PCR assay designed for the detection and quantification of hepatitis C virus RNA in human plasma or serum. The assay provides quantitative results over a dynamic range of 43 to 69,000,000 IU/ml.

HIV-1 viral load was assessed by the Abbott RealTime HIV-1 assay using an in vitro reverse transcription-polymerase chain reaction (RT-PCR) assay with homogeneous real-time fluorescent detection on the automated m2000 System for the quantification of human immunodeficiency virus type 1 (HIV-1) in human plasma. The range of the performed test was 40 to 10,000,000 copies/ml. An HIV viral load of < 50 copies/ml was defined as undetectable, according to European AIDS Clinical Society 2023 guidelines [14].

Lymphocyte T CD4+ cell count was assessed via immunophenotyping analyses, including measurement of the absolute count of T lymphocyte (CD3+) subsets: CD4+ (helper/inducer), CD8+ (suppressor/cytotoxic), and CD4 : CD8 ratio. Counts were determined using the flow cytometry method with the application of the three-colour direct immunofluorescence reagent TriTEST (BD Biosciences, North Ryde, Australia).

The diagnosis of liver cirrhosis was based on the following criteria: laboratory test results typical for impaired liver function for at least 3 months (thrombocytopenia, hypoalbuminaemia, elevated liver enzymes, prolonged international normalised ratio [INR]) for at least 3 months with any of the following features: liver cirrhosis diagnosed in histopathological examination, F4 fibrosis in elastography examination, or radiological image of liver cirrhosis (abdominal ultrasound examination, computed tomography, or magnetic resonance imaging).

Statistical analysis

The Shapiro-Wilk test was performed to verify the normality of the distributions in the analysed variables. The two-sample Student t-test or Mann-Whitney U test was used to evaluate the difference in the mean values among quantitative variables, and χ² or Fisher exact tests were performed for categorical variables. Univariate and multivariable logistic regression analysis was used to evaluate the associated factors. Multivariable logistic regression was adjusted by age (< 35 years old vs. ≥ 35 years old) and sex (male vs. female). The p-value was set at 0.05. All statistical analyses were performed using Python 3.7 and Statistica 13.3 software (StatSoft, Kraków, Poland).

Results

Patients

The study included 65 patients diagnosed with both HIV and HBV infection in 2012. Among them, 64 (98.46%) had already been receiving antiretroviral therapy and 1 (1.54%) was newly diagnosed with HIV and died before ART implementation. The mean duration of ART was 5.44 years (standard deviation [SD] = 4.93 years). Among the 65 patients assessed in 2012, 15 (23.08%) tested positive for anti-HDV antibodies and 9 (13.85%) positive for HDV RNA. Eight patients with detectable HDV RNA also tested positive for anti-HDV antibodies, and 1 patient tested positive for HDV RNA and negative for anti-HDV antibodies. The baseline characteristics of the patients are presented in Table 1.

Among 6 patients with liver cirrhosis, 3 were also diagnosed with alcoholic liver disease. Decompensated cirrhosis was observed in 3 individuals. Thrombocytopenia was present in 3 patients with liver cirrhosis, oesophageal varices in 3 patients, ascites in 3 patients, encephalopathy in 3 patients. Ultrasound features of liver cirrhosis were observed in 6 patients, computed tomography was performed in 2 patients, magnetic resonance imaging in 4 patients, 1 patient had liver cirrhosis confirmed in liver biopsy. In 8 patients with a detectable HIV viral load (VL), there was a range of 51-165351 copies/ml, a mean of 25998.38 copies/ml, and an SD of 56962.79 copies/ml. In 15 patients with detectable HBV DNA, there was a range of 22-687,000 iU/ml, a mean of 44724.08 iU/ml, and an SD of 171349.9 iU/ml.

Risk factors for HDV infection in HIV/HBV-coinfected patients

Age, sex, HIV transmission route, and HCV infection were analysed as risk factors for HDV infection in HIV/HBV-coinfected patients. Univariate and multivariate analyses are presented in Table 2.

Evaluation after 12 years

In 2024, there were 42 patients in continued care (64.62%), 8 patients who had died (12.31%), and 15 who were lost to follow-up (23.08%). The median of long-term follow-up for all patients was 12 years (IQR 7 years). The median follow-up for patients who died or were lost to follow-up was 2 years (IQR 5 years). The causes of death were liver cirrhosis (2 cases), AIDS (1), pneumonia (1), anal cancer (1), injury (1), and unknown (2). The 6 patients lost to follow-up reported moving to another country. Among the 16 patients with anti-HDV antibodies or detectable HDV RNA in 2012, 8 were in continued care (50.00%), 3 had died (18.75%), and 5 were lost to follow-up (31.25%).

Among the 42 patients in continued care in 2024: 23 were HBsAg positive, 11 were HBsAg negative, and data were unavailable for 8; 1 had detectable HBV DNA (2 280 000 iU/ml), 23 had undetectable HBV DNA, and 18 patients had no data; and 4 had detectable HIV RNA (in the range 53-566 copies/ml, mean 190.50 copies/ml), and 38 had undetectable HIV RNA.

Among the patients with no liver cirrhosis in 2012, two had developed it by 2024, and none had developed hepatocellular carcinoma. Seven patients were newly infected with HCV between 2012 and 2024, and a total of 14 patients were treated for HCV infection. All treated patients reached a 48-week sustained virological response (SVR), 11 with direct-acting antiviral medications (DAA) and 3 with interferon (INF)-based therapy. Mean CD4+ T lymphocyte count after 12 years was 610.12 cells/mm³ (SD = 264.19 cells/mm³), and it increased significantly from 2012 (p = 0.004).

Antiretroviral therapy and treatment interruptions

The ART regimens administered to 42 patients were analysed during the 12-year observation. During this time, all patients underwent 3-6 switches of at least one antiretroviral drug (either within a class of antiretroviral drugs or between classes). There were 4 classes of applied antiretroviral agents: protease inhibitors (PI), nucleoside reverse transcriptase inhibitors (NRTI), integrase strand transfer inhibitors (InSTI), and non-nucleoside reverse transcriptase inhibitors (NNRTI).

Among the applied regimens there were: 24 patients (57.14%) receiving 1 PI + 2 NRTI followed by 1 InSTI + 2 NRTI; 8 patients (19.05%) receiving only 1 PI + 2 NRTI; 5 patients (11.90%) receiving 1 NNRTI + 2 NRTI followed by 1 InSTI + 2 NRTI; 4 patients (9.52%) receiving only 1 NNRTI + 2 NRTI; and 1 patient (2.38%) receiving 1 PI + 2 NRTI followed by 1 NNRTI + 2 NRTI followed by 1 InSTI + 2 NRTI.

Among the 42 patients, 7 (16.67%) interrupted antiretroviral therapy for at least 1 month but all resumed the treatment.

Risk factors of mortality in HIV/HBV-coinfected patients

Univariate and multivariate analysis of risk factors of mortality in HIV/HBV-coinfected patients are presented in Table 3. Patients lost to follow-up were excluded from the analysis.

Discussion

The global prevalence of HDV is estimated to be between 0.16% and 1.00%, and between 4.5% and 14.6% in HBsAg-positive patients [7]. However, many HDV prevalence studies are outdated, and prevalence estimates vary widely depending on the study. In particular, there are limited data concerning the occurrence of HDV in the PWH population. In a study by Béguelin et al. analysing 1556 PWH positive for HBsAg in the Swiss HIV Cohort Study and EuroSIDA between 1988 and 2019, the prevalence of HDV coinfection was 15.2%, and 66% of HDV-positive individuals had active HDV replication [15]. In 2024, Schinas et al. performed a cross-sectional study in Greece, showing that only 1 in 30 patients with HIV/HBV-coinfection tested positive for anti-HDV antibodies, and none for HDV RNA [16]. In Italy, 15.2% of HIV/HBV-coinfected patients were positive for anti-HDV antibodies, and 17% of anti-HDV positive PWH tested positive for HDV-RNA [17]. Considering studies from different centres assessing the Polish population, the prevalence of HDV infection ranges from 1.5% to 3.4% among patients with HBV infection, and from approximately 21.4% to 22.2% among people with HIV, when based exclusively on anti-HDV antibody testing. HDV replication confirmed by HDV RNA PCR was detected in 1 of 6 individuals, corresponding to 16.6% [8-11]. In our study, among patients attending two HIV centres in 2012, 65 tested positive for HBsAg, 23% of whom were positive for anti-HDV antibodies, and 53% of anti-HDV-positive individuals had detectable HDV RNA. Our findings from a population studied in 2012 differ substantially from those currently reported in other Polish studies, probably because contemporary ART regimens in people with HIV/HBV usually include agents active against HBV, particularly tenofovir. Although tenofovir does not act directly against HDV, long-term suppression of HBV may reduce HBsAg availability and indirectly limit the persistence of active HDV infection. Another possible explanation is the generally small sample size, in which even minor numerical differences can substantially affect the observed percentage.

The prevalence of HDV infection seems to decrease with the implementation of HBV vaccination programmes but, owing to immigrant influx, in some countries it has increased. In Spain, the highest prevalence of HDV infection (13%) was reached between 1983 and 1987, and the lowest between 2003 and 2012. However, among immigrants, the highest prevalence (10.7%) was reached in 2008-2012 [18]. Similar results were reported in another study conducted in Greece, which showed an overall stable prevalence of HDV infection in 2,137 HBsAg carriers, 4.1% in the 1997-2003 group, and 4.4% in 2004-2010, with a significant difference between Greek (2.8%) and immigrant patients (7.8%), most of whom were from Balkan countries [19].

The prevalence of HDV infection may be underestimated, since testing for HDV is not routinely performed and even unavailable in many centres. In our study, HDV testing was performed in all HIV/HBV-coinfected patients in 2012. We further aimed to evaluate factors associated with the risk of HDV infection in PWH. Significant differences were found in the characteristics of HIV/HBV- and HIV/HBV/HDV-coinfected patients. Firstly, differences were observed according to the route of HIV transmission. Although HDV infection may be transmitted both by sexual contact and injection drug use, the latter seems to be a much more common route. In our study population, 78% of the HDV-positive population was infected with HIV through injection drug use and, among the recruited patients who injected drugs, 41.2% were infected with HDV. Moreover, in multivariate analysis, injection drug use was a statistically significant risk factor for acquiring HDV infection (p < 0.05), which may suggest that this route of transmission is highly effective, and that this population is especially vulnerable to HDV infection. In the Swiss HIV Cohort Study and EuroSIDA, the prevalence of HDV infection among individuals who inject drugs was also high, reaching 50.5% [15]. Studies performed globally also indicate that HDV prevalence is higher among people who inject drugs [20, 21]. In our cohort of patients, there were no individuals with HDV infection among men having sex with men (MSM), and only one person who acquired HIV by heterosexual contact. Global studies also indicate that MSM represent a population at risk of HDV infection, although injection drug use is the predominant route of transmission [4].

As they share a route of transmission, HCV infection is also indicated as a risk factor for HDV infection [4, 22]. It is estimated that HCV is found in more than 10% of HBV or HBV/HDV-infected patients worldwide [23]. In our study, among all patients with HIV/HBV/HCV infection, 30.4% were also infected with HDV. In an Italian study by d’Arminio Monforte et al., HCV infection was an independent factor for HDV antibody positivity in HIV/HBV-coinfected individuals [24]. A study from Pakistan also observed a strong association of HBV and HDV coinfection with HCV infection [23]. Furthermore, in people with HIV/HBV/HDV/HCV coinfections, the incidence of severe liver-related events is higher than in people without HCV coinfection [24].

A meta-analysis by Chang et al. showed that HDV infection increases the risk of liver cirrhosis and hepatocellular carcinoma compared to HBV monoinfection [25]. Liver cirrhosis was found in 9% of our study cohort, but in 33% of patients with HDV infection. Our results seem to be lower than in other countries, since, in Italy, 71% of HIV/HBV patients who tested positive for HDV RNA had advanced liver disease [17]. Worldwide, nearly half of the patients with HDV infection have cirrhosis at the time of diagnosis [26]. This may result from the fact that approximately 10-15% of patients with hepatitis D progress to cirrhosis within 1 to 2 years, and 70-80% of patients within 5 to 10 years [27].

In our study, liver cirrhosis was not only more prevalent in the HDV-coinfected subgroup, but it was also the only statistically significant risk factor for mortality among HIV/HBV-coinfected patients. Another factor significantly associated with mortality was detectable HDV RNA. HDV infection is associated with an increased risk of progression to cirrhosis and the development of hepatocellular carcinoma compared to patients with HBV monoinfection. The risk of cirrhosis and cancer is even higher in PWH [7]. A systematic review and meta-analysis by Kamal et al. showed that infection with HDV is associated with a 2-fold higher risk of hepatocellular carcinoma development compared to HBV mono-infection [28]. Furthermore, HDV coinfection leads to earlier hepatic decompensation than HBV alone [27]. In our study, deaths occurred for various reasons, but liver cirrhosis was the cause in only two patients. HDV coinfection was not significantly associated with higher odds of mortality, in contrast to other studies that report a higher all-cause mortality rate per 100 person-years in HDV coinfection [27]. Moreover, significantly higher mortality was observed among HDV-coinfected individuals who were not treated for HBV [29]. Studies also point to female sex and older age as risk factors for HDV infection, although we did not observe these relationships in our cohort [22].

In our study, all patients except for one were receiving antiretroviral therapy consisting of 3 drugs, including 2 NRTIs, which also covers HBV treatment. According to current recommendations by the European Association for the Study of the Liver, nucleoside analogues should be given in patients with decompensated cirrhosis, in patients with compensated cirrhosis and detectable HBV DNA, and in patients without cirrhosis if HBV DNA levels are higher than 2,000 IU/ml. Unfortunately, nucleoside analogues have low efficacy in the control of HDV infection [30]. Another treatment option for chronic hepatitis D is pegylated interferon alpha, which should be considered in all patients, irrespective of whether they have cirrhosis [30]. A new alternative for HDV infection and compensated liver disease is bulevirtide, which inhibits the entry of HDV into hepatocytes, limiting the ability of the virus to replicate [31]. However, the optimal dose and duration of bulevirtide treatment have not yet been defined [30]. None of our patients was receiving specific anti-HDV treatment because, during the study period 2012-2024, there was no treatment programme for HDV in Poland.

Significance and limitations of the study

Our study is significant because of the long follow-up of patients with HIV/HBV coinfection (12 years) and the analysis of HDV prevalence in PWH in Poland. However, the study has several limitations that need to be addressed. Firstly, the study cohort was relatively small, limiting the statistical power to reliably evaluate the association between the analysed factors and PWH mortality. Secondly, we performed HDV RNA analysis only once during the observation period. This could result in underdiagnosis of HDV infection during the observation, and also did not allow the course of infection to be monitored. Moreover, due to the long-term observation, there was a significant number of patients who were lost to follow-up (15 individuals) or had missing laboratory data in 2024 (8 patients without HBsAg assessment and 18 patients without HBV DNA assessment). Nevertheless, these findings provide a basis for further studies involving larger populations.

Conclusions

Patients with HIV/HBV/HDV coinfection had significantly higher rates of liver cirrhosis and HCV infection and non-significantly higher mortality. Injection drug use was identified as a significant risk factor for HDV. Despite comorbidities, long-term ART was effective in achieving viral suppression and immune recovery. Routine HDV screening is recommended in HIV/HBV-coinfected individuals, especially among those who inject drugs.

Disclosures

This research received no external funding.

The study was approved by the Bioethics Committee at the Medical University of Warsaw (consent number: AKBE/165/2025).

The authors declare no conflict of interest.

References