Prevalence of occult HBV infection in Western countries

2.1 OBI in patients with immunosuppression

In patients with OBI, the cccDNA can start actively replicating on immunosuppression. Thus, enhanced viral replication during the administration of immunosuppressive regimens results in the virological reactivation of OBI and in possible hepatitis reactivation, especially after the discontinuation of immunosuppressive drugs, when immune reconstitution occurs with the emergence of HBV-specific T cells and infected hepatocytes are targeted by the immune system.^17-21 The risk of OBI reactivation is estimated as high when there is marked immunosuppression, particularly in onco-hematological patients (from 21% to 67%), in those receiving hemopoietic stem cell transplantation and in those treated with monoclonal antibodies.^22-33

Current evidence suggests that the risk of hepatitis B virus reactivation (HBVr), both in current and resolved HBV infection is proportionally linked to the serological status of the patients and to the level, intensity, and duration of immunosuppression achieved (Table 1).^22-36

Under these conditions, HBVr is associated with a mortality rate close to 20%, due to hepatic failure or to the progression of the underlying disease after the discontinuation of specific treatment.^{17, 35, 36}

Recently there has been a growing number of cases of reactivation of OBI reported in patients undergoing immunosuppressive therapy for non-oncological diseases, such as inflammatory intestinal diseases, dermatological diseases, and rheumatology diseases. However, in this setting, the studies are few and with a small number of patients enrolled.³⁷

2.2 OBI and liver diseases

Another interesting issue is the correlation between OBI and liver disease. The patients with OBI typically have suppressed HBV replication and a low viral load. Thus, most of the OBI patients have normal liver histology or minimal fibrosis.^{38, 39}

However, in the presence of an underlying chronic liver disease due to other etiologies, OBI may be associated to liver cirrhosis and/or HCC.^40-45

A long-term observational cohort study showed that in hepatitis C virus (HCV) chronic patients OBI was associated with the more severe complications of chronic liver disease, more frequent development of HCC and increased risk of liver-related death compared to OBI-negative patients.¹⁶

It must be emphasized that histological data is rare in patients without chronic liver disease because of the ethical problems in performing liver biopsy. Thus, there are more histological data in patients with OBI and HCV chronic infection.

2.3 OBI and blood transfusion

Another important issue concerns the impact of OBI in the safety of blood transfusion. The risk of HBV transmission through blood transfusion from an OBI carrier has been defined as a very rare occurrence, although some cases have been reported.^46-54

In fact, OBI frequently shows phases of low levels of viremia alternating with periods of absence of HBV DNA in the serum,^{48, 52-54} so the potential blood infectivity of an OBI patient may fluctuate over time, it is always very low and depends on the transfused viral dose (viral load × plasma volume transfused), the presence of anti-HBs in donor and/or recipients and the general immune status of the recipient.^{49, 54-56}

A recent study by Candotti et al⁵⁷ investigated 3 repeat HBsAg-negative donors from Slovenia who had undetectable HBV DNA by highly sensitive nucleic acid amplification test (NAT) and who transmitted HBV to 9 recipients following transfusion of blood components. This study has enabled a revised estimation of the minimal HBV infectious dose from the previous 20 IU/mL to approximately 3.0 IU/mL of HBV DNA. The NAT sensitivity required to prevent HBV transmission by transfusion would need to be lowered from the current 3.4 IU/mL.⁵⁸ The clinical outcome of HBV infection by OBI transmission may depend also on the immune status of the recipient. For example, Tonziello et al described the case of HBV transmission to an immunosuppressed patient by plasma donation from an HBsAg-negative subject, but with very low serum HBV DNA (about 50 IU/mL) and five mutations in the major hydrophilic region of HBsAg.⁴

4.1 Prevalence of OBI defined as HBV DNA positivity in liver tissue

The recent Taormina statement identified the detection of HBV DNA in the liver as the diagnostic gold standard for OBI.⁹ Although it is defined as the gold standard, for epidemiological purposes, this definition has a significant limitation: it requires a liver biopsy, an invasive procedure usually reserved for subjects with chronic liver diseases, who are clearly not representative of the general population.

Therefore, not surprisingly, only a few studies have investigated the prevalence of OBI in subjects without liver diseases.

Table 2 summarized studies evaluating OBI prevalence in healthy subjects. In the late 1990s, two studies carried out in the United Kingdom⁷⁰ and United States⁷¹ investigated the prevalence of OBI in patients with nonviral fulminant hepatitis in the absence of previous chronic liver disease, and reported positivity in 0 out of 22 and 3 of 10 subjects enrolled, respectively. The latter study reported also 5.1% prevalence among 59 liver transplantation recipients. We should consider, however, that the molecular assays used in these studies might have had a suboptimal diagnostic accuracy. A different approach was tried by Raimondo et al⁷²: a liver biopsy was performed in 98 subjects during abdominal surgery, and an OBI prevalence of 16.3% was found. More recently, three studies conducted in Italy have investigated the presence of OBI in liver donors, which represents an interesting setting for this purpose.^73-75 In 2009 Toniutto et al⁷³ found a prevalence of 44.8% in 29 liver donors, 3 of whom were anti-HBc-positive, with a prevalence among the recipients of 26.7%. 5 years later, Tandoi et al⁷⁴ demonstrated the presence of OBI in 30 out of 50 anti-HBc-positive and in none of the 20 anti-HBc-negative age-matched donors. In 2018, the same group tested the liver tissue of 100 anti-HBc-positive donors for HBV DNA with a high-sensitivity PCR assay and found positivity in 52 cases.⁷⁵

However, as already discussed, most of the studies on OBI prevalence enrolled patients with chronic liver disease. In Table 3 we summarize the characteristics of these studies.^{5, 42, 70-101} In 1999, Cacciola et al⁷⁶ analyzed the liver tissue of 200 HCV-positive subjects and 50 patients with other liver diseases and demonstrated the presence of OBI in 73 (29.2%) of them. The same group tested for HBV DNA in the liver tissue of 107 patients with HCC and of 192 patients with chronic hepatitis or cirrhosis and reported a prevalence of OBI of 63.5% among subjects with HCC and of 32.8% in the control group.⁴² Two years later, they published a cohort study enrolling 280 patients (66 with cirrhosis and 214 with chronic hepatitis), demonstrating an OBI prevalence of 35.5%.⁹⁰ In 2000, a cross-sectional study conducted in the United States on 285 HBsAg-negative HCV-RNA-positive subjects treated with interferon found that only 7 patients (2.5%) had positive HBV DNA in the liver.⁸³

In 2008, Sagnelli et al⁹⁵ investigated the presence of HBV DNA in the blood, peripheral blood mononuclear cells and liver of 89 patients with CHC and found an overall positivity rate in the liver of 39.3%, but with significant variability according to the anti-HBc and anti-HBs serostatus (from 11.1% in anti-HBc-negative to 73% in anti-HBc-positive/anti-HBs-negative patients). More recently, our group⁵ sought for HBV DNA in neoplastic and nonneoplastic tissue of 68 patients with HCC and found positivity in 19.1% in cancer tissue and 5.9% in nontumorous tissue. In 2010 Levast et al⁹⁷ analyzed the prevalence of OBI in 113 French patients with CHC who underwent a liver biopsy and found positivity in 5 (4.4%) of them. The following year, a case-control study⁹⁸ was conducted in the United States on 28 patients with HCC and 55 controls who participated in the Hepatitis C Antiviral Long-term Treatment against Cirrhosis (HALT-C) trial and showed an OBI prevalence of 10.7% in the HCC cases and of 23.6% in the controls. Cardoso et al¹⁰¹ prospectively enrolled 100 Portuguese patients with CHC from 2008 to 2011 and showed a presence of HBV DNA in the liver samples of 57 of them. Regarding the prevalence of OBI in the HIV-infected population, Raffa et al⁹² investigated the presence of OBI in 101 HIV/HCV coinfected patients and showed a prevalence of 41%; no difference was found in the anti-HBc serostatus among OBI positive and negative subjects. The following year, a cross-sectional study⁹⁴ enrolling 52 HIV/HCV-positive patients demonstrated the presence of OBI in 7 (13.5%) subjects. Finally, in 2013 Cassini et al⁹⁹ found a 29.2% prevalence of OBI in 24 anti-HBc-positive HIV-infected subjects; most of them had CHC.

The results of a recent meta-analysis are interesting, reporting OBI prevalence considering the serostatus for anti-HBc: the prevalence of OBI was 51% (95% CI, 40-62%) in the 823 anti-HBc-positive subjects, and 19% (95% CI, 10-30%) in the 1041 anti-HBc-negative subjects.⁵⁹ Moreover, in the same meta-analysis the authors also evaluated the role of the presence of anti-HBs in anti-HBc-positive subjects: the pooled overall prevalence of OBI was 43% (95% CI, 27-60%) in the 176 anti-HBs-negative/anti-HBc-positive subjects and 51% (95% CI, 31-72%) in the 257 anti-HBs/anti-HBc-positive subjects.⁵⁹ Then, evaluating the data from studies comparing anti-HBc-positive and negative subjects, the authors found that the prevalence of OBI was higher in the 641 anti-HBc-positive subjects than in the 1041 anti-HBc-negative (prevalence ratio [PR], 2.29, 95% CI, 1.61-3.26; P < .001).⁵⁹

Finally, Table 4 summarizes studies evaluating OBI prevalence in a particular subgroup of patients, the anti-HBs positive and anti-HBc negative subjects.

4.2 Prevalence of OBI defined as HBV DNA positivity in serum

The characteristics of studies evaluating the prevalence of OBI defined as HBV DNA positivity in plasma are shown in Table 5.^101-121 The authors describe the results of the different studies according to the type of population enrolled.

The studies with a larger number of patients enrolled were those evaluating the prevalence of OBI in blood donors.^102-106 Specifically, Candotti et al¹⁰² in 2008 performed a multicenter cohort study (five blood centers in Europe and two in South Africa) on blood donors of all European origins. Of the 2,003,974 subjects tested, 74 were identified as harbouring OBI, 72 were anti-HBc-positive and 40 anti-HBs-positive subjects. The authors evaluated the virological characteristics of the 74 subjects with OBI: HBV DNA load ranged between unquantifiable (13 cases) and 5640 IU/mL, 14 had genotype A2, 43 genotype D, and 17 were not genotyped.¹⁰² Kiley et al¹⁰⁴ in 2014 performed a similar study in Australia: during the 2 years of the study period, 42 OBI carriers out of 2,673,521 blood donors were detected.

In studies enrolling subjects with chronic liver diseases or risk factors, the prevalence of OBI was on average higher.^{120, 121}

Considering patients with chronic kidney disease, Mina et al¹¹⁵ in 2010 performed a study on patients with end-stage renal failure (ESRF) in Greece: OBI was identified in 3 (0.9%) of the 346 HBsAg-negative patients. In the United Kingdom, Sowole et al¹¹⁷ showed OBI in 138 (17.4%) out of 793 HBsAg-negative ESRF patients. Stratta et al¹¹⁸ in 2009 evaluated the OBI prevalence in 300 HBsAg-negative Italian patients on a waiting list for kidney transplant, with an identification in 10 cases (3.3%).

Tramuto et al¹¹⁴ in 2013 performed a cross-sectional study in Italy on 339 HBsAg-negative immigrant subjects from North Africa (43), Eastern Europe (54), South-East Asia (60), and sub-Saharan Africa (181). Overall, 11 (3.2%) had HBV DNA in their plasma and 4 were also anti-HIV-positive; in fact, the presence of anti-HIV was the only factor independently associated with a probability of observing OBI (P < .001).¹¹⁴ Chadwick et al¹⁰⁸ performed a multicenter study in the United Kingdom to study the prevalence of OBI in HBsAg-negative African immigrants with HIV infection: OBI was identified in 15 (4.4%) of the 335 patients enrolled, more frequently in the 216 antiretroviral therapy (ART)-naïve than in 114 on ART (6.5% vs 0.9%); the only factor significantly associated with OBI was anti-HBc positivity (odds ratio [OR], 7.41, 95% CI, 2.0-27.6).

In patients with HIV infection, the prevalence of OBI ranged from 0% (95% CI, 95%: 0-3.75) to 16.67% (95% CI, 10.37-25.69).^108-113 In a cohort study performed in the Netherlands, 191 ART-naïve HIV-positive, HBsAg-negative, and anti-HBc-positive patients were included in the study: OBI was identified in 26 (13.6%), more frequently in patients with a low CD4 cell count.¹⁰⁸ A similar study performed in Italy involving 405 HBsAg-negative/HIV-positive subjects identified OBI in 24 (5.9%), more frequently if positive for anti-HBc alone (8/71) or for anti-HBc and anti-HBs (6/88), and less frequently in those without HBV markers (9/183).¹¹³ In 2008, Palacios et al¹¹¹ performed a cross-sectional, multicenter study in Spain on 202 anti-HIV-positive/anti-HBc-positive/anti-HBs-negative patients, with HBV DNA positivity in plasma in 5 (2.4%).

Finally, considering the subjects with chronic liver diseases due to another etiology, Georgiadou et al¹²⁰ analyzed the prevalence of OBI in 196 Greek patients with nonviral chronic liver disease: 54 with autoimmune hepatitis-type 1, 2 with autoimmune hepatitis-type 2, 7 with autoimmune hepatitis and primary biliary cirrhosis, 3 with overlap syndrome, 93 with primary biliary cirrhosis, and 37 with primary sclerosing cholangitis: HBV DNA was detected in the serum of 24 (12.2%) patients.