Occult hepatitis B virus infection in a North American adult hemodialysis patient population
Article first published online: 14 OCT 2004
Copyright © 2004 American Association for the Study of Liver Diseases
Volume 40, Issue 5, pages 1072–1077, November 2004
How to Cite
Minuk, G. Y., Sun, D. F., Greenberg, R., Zhang, M., Hawkins, K., Uhanova, J., Gutkin, A., Bernstein, K., Giulivi, A. and Osiowy, C. (2004), Occult hepatitis B virus infection in a North American adult hemodialysis patient population. Hepatology, 40: 1072–1077. doi: 10.1002/hep.20435
- Issue published online: 14 OCT 2004
- Article first published online: 14 OCT 2004
- Manuscript Accepted: 23 JUL 2004
- Manuscript Received: 29 APR 2004
- Health Canada, Blood Borne Pathogens Division
Hepatitis B virus (HBV) infections continue to occur in adult hemodialysis units. A possible contributing factor is the presence of occult HBV (serum hepatitis B surface antigen [HBsAg] negative but HBV DNA positive). Two hundred forty-one adult hemodialysis patients were screened for occult HBV. HBV DNA testing was performed by real-time polymerase chain reaction (PCR) with 2 independent primer sets (core promoter and surface). Two (0.8%) of the 241 patients were HBsAg positive. Of the remaining 239 HBsAg-negative patients, 9 (3.8%) were HBV DNA positive. Viral loads in these individuals were low (102-104 viral copies/mL). Seven of the 9 (78%) were nt 587 mutation (sG145R mutant) positive. Demographic, biochemical, and HBV serological testing did not help to identify those with occult HBV. In conclusion, the prevalence of occult HBV in adult hemodialysis patients in this North American urban center is approximately 4 to 5 times higher than standard HBsAg testing would suggest. The majority of these infections are associated with low viral loads and a high prevalence of the sG145R mutant. Finally, the demographic, biochemical, and/or serological features of HBV DNA–positive subjects do not distinguish these individuals from the remainder of the dialysis patient population. (HEPATOLOGY 2004; 40:1072–1077.)
Despite the development of an effective hepatitis B virus (HBV) vaccine and extensive infection control guidelines, HBV infections continue to occur in dialysis units throughout North America and Europe.1–3 Based on the results of hepatitis B surface antigen (HBsAg) testing, the incidence of such infections is thought to be 0.05%-1% per year.4 However, this likely represents an underestimate of the true incidence, as the use of more sensitive monoclonal-based assays for HBsAg detection increase positive findings in dialysis patients by 120%.5
The relatively low acceptance and response rates to the HBV vaccine among dialysis patients likely contributes to ongoing transmission, as does the need for vaccine boosts to maintain antibody to HBsAg (anti-HBs) at protective levels.4, 6–8 Additional contributing factors include “breakdowns” in the application of universal and/or dialysis-specific infection control measures.4, 7, 9 With the development of specific and sensitive polymerase chain reaction (PCR)–based testing for HBV DNA, the presence of occult HBV infection (HBV DNA positivity in the setting of negative serum HBsAg) represents yet another possible explanation for ongoing transmission.
To date, few studies have documented the prevalence of occult HBV infection in renal dialysis patients. In 3 small studies of 33, 5, and 67 HBsAg-negative dialysis patients, 50%, 40%, and 0%, respectively, were HBV DNA positive.10–12 Of note, the majority of HBV DNA–positive individuals in these studies had serological evidence of previous HBV infection (HBV seropositive), but as many as 39% were HBV seronegative.
The present study documents the prevalence of HBV DNA positivity in a large, North American renal dialysis patient population and determines whether specific demographic, biochemical, and/or serological features could serve to identify these individuals for additional testing and follow-up.
Patients and Methods
The study was performed from May 2003 to September 2003 in Winnipeg, Manitoba, an urban center in central Canada with a population of approximately 680,000. Subjects for the study were recruited from 2 of the main hemodialysis units in the city (Health Sciences Centre and Seven Oaks General Hospital). All patients, regardless of age, sex, or ethnic background, were invited to participate. After written, informed consent was obtained, blood was drawn for liver biochemistry (alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase, gamma-glutamyltransferase, albumin, bilirubin, and international normalized ratio for prothrombin times), HBV serology (HBsAg, anti-HBs, and antibody to hepatitis B core antigen [anti-HBc]), and HBV DNA testing. Liver biochemistry testing was performed at the Health Sciences Centre's Clinical Chemistry Department, while HBV serology was performed at the Cadham Provincial Laboratory. Testing for HBsAg was performed by a third-generation monoclonal enzyme immunoassay, according to the manufacturer's instructions (AUSZYME Monoclonal Diagnostic Kits; Abbott Laboratories, North Chicago, IL). The sensitivity of the assay is 0.3 to 0.7 ng/mL. Testing for anti-HBc was by the CORE IMx system and for anti-HBs by the AUSAB IMx system (Abbott Laboratories).
HBV DNA testing was performed using real-time PCR with 2 independent sets of primers (core promoter and surface), as described previously.13 Briefly, HBV DNA was extracted from sera using a High Pure Viral Nuclei Acid Isolation Kit (Roche Diagnostics, Laval, Quebec, Canada). The sequences of the primers, anchor and sensor probes used in this study are summarized in Table 1. Wild-type HBV adw1 subtype was purchased from ATCC (Manassas, VA). The HBV fragment was cut from a pBR322 plasmid at the BamH I site, separated on 1% agarose gel, and reamplified. The purified DNA concentration was determined with a spectrophotometer at 260 nm, and the corresponding copy number of HBV virus was calculated. Serial dilutions ranging from 1 × 105 to 1 copy per 1 μL volume of H2O were prepared. Quantitative PCR of serum HBV copy number based on the 6 serial dilutions (external standards) was performed in the presence of 2 μL of genomic DNA of patient isolate. The lowest detection limit of this assay (10 viral copies/mL) was analyzed in parallel with standards and calculated by genome copies/mL.
|Target||Sequences (5′-3′)||Position (nt)||Product Size (bp)||Annealing Temperature (°C)|
|Sensor probe||5′-LC-Red 640-TACAAAACCTATGGATGGAAACTGC-3′-P||571–595|
|Mutation site||G A, nt 587|
|Basal core promoter mutation|
|Sensor probe||5′-LC-Red 640-GGGAGGAGATTAGGTTAAAGGTCTTTG-3′-P||1744–1770|
PCR reactions were performed in a total volume of 20 μL containing 2 μL DNA template, 2 μL LightCycler DNA Master Hybridization Mixture (Roche Diagnostics) (Taq DNA polymerase, reaction buffer, dNTP mixture, and 10 mmol/L MgCl2), 1.6 μL 25 mmol/L MgCl2, 0.2 μmol/L each of the probes, and 0.5 μmol/L each of the primers. Samples were loaded into composite plastic/glass disposable capillaries, centrifuged, and placed in the LightCycler sample carousel (Roche Diagnostics).
After amplification was complete, a melting curve was generated. The melting curves were converted to melting peaks by plotting the negative derivative of the fluorescence with respect to temperature (−dF/dT) against temperature. Melting curves and quantitative analysis of the data were performed using LightCycler analysis software 3.5.
To determine the sensitivity of the detection system for mutant HBV DNA among wild-type HBV DNA, mutant and wild-type DNA were mixed at mutant:wild-type ratios of 100:0, 50:50, 25:75, 10:90, 5:95, 2:98, and 0:100 prior to processing for melting analysis.
To prevent carryover contamination during PCR, each step of the procedure was performed in a separate room with dedicated equipment and directional flow from the beginning of the procedure to the end. Negative controls containing serum or water were also included in each extraction run, and an extra negative control containing water was included during each PCR run.
Hospital charts were reviewed to determine which patients were known to be HBsAg positive prior to the study (HBsAg testing is performed monthly in this patient population) and those who had received and/or completed HBV vaccination.
Student t tests were performed for parametric and Mann-Whitney tests for nonparametric data. The Bonferoni correction factor was employed for multiple comparisons. A chi-square test of association (F test when warranted) was used to examine differences in proportions. All statistical analyses were performed using the Number Cruncher Statistical Systems 2001 (Kaysville, UT).
This study was approved by the University of Manitoba Conjoint Ethics Committee for Human Experimentation.
Of a possible 277 dialysis subjects, 252 (90%) agreed to participate in the study. None of the 25 individuals who refused to participate were known to be HBsAg positive. Age information was available for 250 of the 252 participants. The mean age (±SD) was 59.0 ± 15.5 years (range, 18 to 88 years).
Sex information was available for 249 individuals. There were 135 males (54%) and 114 females (46%). There was no difference in age/sex distribution of the sample population. Thus, the mean age (±SD) for males was 58.6 ± 15.1 years; for females, 59.4 ± 16.1 years (P = .78). Approximately 60% of patients were Caucasian, 30% were Aboriginal, and 10% were Southeast Asian.
The results of liver biochemistry tests for the study population are provided in Table 2. Thirty-two patients (14%) had elevated alanine aminotransferase, 38 (16%) had elevated aspartate aminotransferase, 84 (36%) had elevated alkaline phosphatase, and 143 (61%) had elevated gamma-glutamyltransferase values. Serum albumin levels were low in 62 (26%), bilirubin levels elevated in 4 (2%), and international normalized ratio values increased in 39 (18%) subjects.
|Test (Normal Range)||N||Mean ± SD||Median||Range||95% CI|
|ALT (0–30 U/L)||236||21.1 ± 35.6||15||4–391||16.6–25.7|
|AST (10–32 U/L)||236||26.8 ± 37.5||20||7–411||22.0–31.6|
|ALP (30–120 U/L)||235||121.6 ± 70.3||105||38–478||112.6–130.6|
|GGT (4–26 U/L)||236||58.2 ± 78.8||34||7–612||48.2–68.3|
|Albumin (35–50 G/L)||235||37.5 ± 6.0||37||19–54||36.7–38.3|
|Bilirubin (3–18 μmol/L)||234||7.7 ± 4.1||7||1–39||7.2–8.3|
|INR (0.9–1.1)||217||1.2 ± 0.9||1||0.9–12||1.1–1.4|
Sufficient sera were available for HBsAg, anti-HBs, and anti-HBc testing in 241 (96%) individuals. There were only 2 positive HBsAg results (0.8%). Both were males and known to be HBsAg positive prior to the onset of the study. One hundred fifty-two (63%) were anti-HBs positive and 21 (8.7%) were anti-HBc positive.
Of the 228 individuals with information available regarding their HBV vaccination status, 211 (92.5%) had been vaccinated on at least 1 occasion. The majority of these individuals (202/228, 88.6%) had completed 4 vaccinations. Only 1 (0.4%) had received hepatitis B immunoglobulin in the past.
As indicated in Tables 3 and 4, there were 9 HBsAg-negative, HBV DNA–positive individuals, representing 3.8% of the 239 subjects tested. Their mean (±SD) age was 61.4 ± 7.3 years, and 5 (56%) were male. Seven of the 9 (78%) were Caucasian, and 2 (22%) were Southeast Asian. All had normal liver enzyme and function tests.
|Patient No.||Age||Sex||Ethnicity (C, A)||ALT (<30 IU/L)||HBsAg||Anti-HBc||Anti-HBs||HBV DNA (×102 copies/mL)||sG145R-mutant/wild type||Vaccinated X1/Complete|
|Test (Normal Range)||Group||N||Mean ± SD||Median||Range||P||95% CI|
|Age (years)||HBV DNA pos.||9||61.4 ± 7.3||59||54–73||.636||55.8–67.1|
|HBV DNA neg.||241||58.9 ± 15.7||59||18–101||56.9–61.0|
|Sex (% male)||HBV DNA pos.||9||5 (56%)|
|HBV DNA neg.||240||135 (54%)|
|ALT (0–30 U/L)||HBV DNA pos.||9||12.9 ± 6.6||13.5||4–24||.390||7.4–18.4|
|HBV DNA neg.||223||21.5 ± 36.6||15||4–391||16.7–26.3|
|AST (10–32 U/L)||HBV DNA pos.||9||17.4 ± 5.7||18.5||10–25||.151||12.6–22.2|
|HBV DNA neg.||223||27.3 ± 38.5||20||7–411||22.2–32.3|
|ALP (30–120 U/L)||HBV DNA pos.||9||88.4 ± 32.9||72||51–135||.147||60.8–115.9|
|HBV DNA neg.||222||122.9 ± 71.6||105||38–478||113.5–132.3|
|GGT (4–26 U/L)||HBV DNA pos.||9||23.9 ± 16.4||20||7–53||.031||10.1–37.6|
|HBV DNA neg.||223||59.1 ± 79.9||36||8–612||48.6–69.6|
|Albumin (35–50 G/L)||HBV DNA pos.||9||37.3 ± 8.8||37.5||19–47||.782||29.9–44.6|
|HBV DNA neg.||222||37.5 ± 6.0||38||19–54||36.7–38.3|
|Bilirubin (3–18 μmol/L)||HBV DNA pos.||9||7.9 ± 2.7||8||5–13||.668||5.6–10.1|
|HBV DNA neg.||221||7.8 ± 4.2||7||1–39||7.2–8.3|
|INR (0.9–1.1)||HBV DNA pos.||9||1.1 ± 0.2||1.05||1–1.5||.725||0.9–1.2|
|HBV DNA neg.||204||1.2 ± 0.9||1||0.9–12||1.1–1.4|
Two of the 9 (22%) HBV DNA–positive individuals were anti-HBc positive, and 5/8 (63%) were anti-HBs positive. One of the anti-HBc–positive individuals was also anti-HBs positive. Thus, 3 (33%) HBV DNA–positive individuals were HBV seronegative. Viral loads were low, ranging from 6.5 × 102 to 2.5 × 104 viral copies/mL.
Seven of the 9 (78%) HBV DNA–positive subjects were sG145R mutant–positive. In 4 cases, a mixed sG145R-mutant/wild-type infection was present; in 3, the sG145R-mutant was the only form identified; and in 2, only wild-type virus was detected. For the 6 sG145R mutant–positive individuals for whom information was available, 6 had received at least 1 HBV vaccination, and 5 completed the series of 4 injections. None of the subjects had received hepatitis B immunoglobulin.
The results of this study indicate that in this North American urban center, the prevalence of HBV viremia in adult hemodialysis patients is 3.8% or 4 to 5 times higher than what standard monoclonal antibody–based HBsAg testing would have suggested. The results also indicate that the majority of these infections are associated with low viral loads (<105 copies/mL) and a high prevalence of the surface nt 587 (sG145R) mutation. Finally, the demographic, biochemical, and/or serological features of HBV DNA–positive subjects do not help to distinguish these individuals from those who are HBV DNA negative.
An important issue related to these findings is whether occult HBV infection can be transmitted to others. Previous data indicate that infection can occur in susceptible chimpanzees, infants, and transfusion or organ recipients following exposure to HBsAg-negative, HBV DNA–positive blood.14–23 That none of the occult carriers in this study were dialyzed by the same dialysis equipment or staff argues against but does not necessarily exclude nosocomial transmission.
Another important issue relates to what impact (if any) occult HBV infection has on the health of these individuals. Previous reports describe associations between occult HBV and fulminant hepatic failure, chronic hepatitis, cirrhosis, and hepatocellular carcinoma.24 If so, antiviral treatment and/or screening for hepatocellular carcinoma may be warranted in certain cases. In favor of treatment are the low viral loads described, which portend a good response to antiviral therapy (at least with interferon-based treatment).25 Against are the normal liver enzymes, which tend to be associated with poor responses to treatment.26
Although the HBV DNA prevalence rate in this study (3.8%) was higher than the HBsAg-positive rate of 0.8%, it was significantly lower than the 58% and 40% figures reported by Spanish and Swiss investigators, respectively, in their HBsAg-negative dialysis patients.10, 11 A number of possible explanations exist for this finding. First, the prevalence of HBV infection in the general populations of Spain and Switzerland are higher than in Canada (1.7% and 1.0% vs. 0.5%, respectively) and prevalence rates of occult HBV tend to correlate with “background” prevalence rates of HBV infection.24, 27–29 Second, patients in the Spanish and Swiss studies were selected on the basis of either a past history of HBV infection or a positive anti-HBc result. These selection criteria would favor the enrollment of more high-risk patients. Third, investigators in the other studies employed 1 rather than the suggested 2 independent sets of HBV DNA primers, which raises concerns regarding false-positive results in the European studies.24 Finally, the relatively small number of subjects enrolled in the other studies must also be considered. Of note, all 3 studies employed commercially available, monoclonal anti-HBs based assays for HBsAg detection, which although more sensitive have a greater possibility of false-negative results in the setting of S mutants than polyclonal antibody–based assays.30
Vaccination of susceptible patients and staff, avoidance of dialyzer reuse and use of dedicated dialysis rooms, machines, and staff for infected patients have all been advocated as means of limiting HBV transmission within dialysis units.1, 7 Of these, only vaccination of susceptible patients and staff would have had an effect on limiting the risk of transmission of occult HBV. Unfortunately, as the 63% prevalence of anti-HBs in our dialysis patients suggests, vaccine acceptance and/or response rates are suboptimal in this patient population.
A somewhat surprising finding was the high prevalence (7/9) of the nt 587 surface (sG145R) mutation in our HBV DNA–positive patients. To date, this mutation has most commonly been described in a small percentage (<5%) of high-risk newborns and HBsAg-positive transplant recipients provided with active and passive HBV immunoprophylaxis following birth and transplantation, respectively.31, 32 In the present study, all but one of the HBV DNA–positive patients had been vaccinated against HBV, but none had received hepatitis B immunoglobulin. Although sG145R mutants/variants have been described in association with occult HBV infection, the highest rate reported to date (30%) is still significantly lower than the 78% documented in the present study.33, 34
The failure of demographic, biochemical, and/or serological findings to suggest the presence of occult HBV is unfortunate but not unexpected. Occult infections have been described in both high and low prevalence regions and various ethnic populations with no age or gender predilections.24 Liver enzyme abnormalities, which tend to be attenuated in dialysis patients regardless of the cause of the liver disease, are uncommon in patients with occult HBV.35 Finally, while occult HBV is more common in the setting of HBV seropositivity, it has also been described in those who are HBV seronegative, as was the case in our study.24
In conclusion, HBV viremia is approximately 4.5 times more common in dialysis patients than the results of standard HBsAg testing would suggest. Until data exist indicating whether nosocomial transmission of occult HBV can occur in susceptible dialysis patients and/or staff, screening with sensitive PCR-based assays of all dialysis patients for HBV DNA regardless of demographic, biochemical, or serological findings seems prudent. Additional studies are required to determine how frequently such screening should be performed and what policy should be in place for those who test positive.
The authors thank the Dialysis staff at the Health Sciences Centre and Seven Oaks Hospital for their assistance and Mrs. S. Zdanuk for her prompt and accurate typing of the manuscript.
- 1Centers for Disease Control and Prevention (CDC). Hepatitis-control measures for hepatitis B in dialysis centers. In: Hepatitis surveillance report no. 41. Atlanta, GA: U.S. Department of Health and Human Services, Public Health Service, CDC, 1977: 12–17.
- 2CDC. Perspectives in disease prevention and health promotion update: universal precautions for prevention of transmission of human immunodeficiency virus, hepatitis B virus and other bloodborne pathogens in health-care settings. MMWR Morb Mortal Wkly Rep 1988; 37: 377–388.
- 22Transmission of hepatitis B by transplantation of livers from donors positive for antibody to hepatitis B core antigen. The National Institute of Diabetes and Digestive and Kidney Diseases Liver Transplantation Database. Gastroenterology 1997; 113: 1668–1674., , , , , , et al.