No evidence for patient-to-patient transmission of hepatitis C virus during upper gastrointestinal endoscopy: molecular studies on three acute hepatitis C patients.

BACKGROUND: The risk of patient-to-patient transmission of hepatitis C virus (HCV) during endoscopy remains controversial. Using molecular approaches, we examined the possibility of patient-to-patient transmission of HCV in three patients who developed acute hepatitis C 1-6 months after examination by upper gastrointestinal endoscopy (UGIE) in a hospital endoscopy unit in Japan. METHODS: For the source of HCV infection, we used frozen sera obtained from potential candidates who underwent UGIE earlier than three index patients on the same days in the same unit. HCV genotype was determined by multiplex polymerase chain reaction (PCR) with genotype-specific primers. The 1087-nucleotide (nt) sequence of the NS5B region of the HCV genome was compared between index patients and their HCV-viremic candidates. RESULTS: The three index patients were exclusively infected with HCV of genotype 1b. Among a total of 60 candidate patients who underwent UGIE earlier than the index patients, 14 were positive for anti-HCV, of whom 12 had detectable HCV-RNA (1b, n = 9; 2a, n = 1; 2b, n = 2) on sera collected during each UGIE. Shared identity within the 1087-nt NS5B sequence was less than 95.0% between index patients and HCV/1b-infected candidates (n = 3, 1 and 5, respectively). None of the remaining 46 candidates who were negative for anti-HCV at UGIE examination tested positive for HCV-RNA, nor seroconverted to anti-HCV on their sera, which most likely excludes the possibility of HCV viremia despite the anti-HCV-negative serology at UGIE examination. CONCLUSION: The present study suggests that patient-to-patient transmission of HCV during UGIE is infrequent.

Molecular analysis of transmission of hepatitis C virus in a nurse who acquired acute hepatitis C after caring for a viremic patient with epistaxis.

A 23-year-old nurse (HC-IP) developed acute hepatitis C. Intrafamilial transmission of hepatitis C virus (HCV) was suspected initially because her parents were carriers of HCV of the same genotype (1b) as that of Patient HC-IP. However, the HCV isolate from Patient HC-IP and those from her parents shared identities of only 92.4-92.7% in the 1,087-nucleotide (nt) sequence within the NS5B region. It was then suspected that she contracted HCV infection during medical practice. Sixteen patients with antibodies to HCV (anti-HCV) were hospitalized 1-3 months before she became positive for anti-HCV. Upon analysis of stored serum samples, 14 of the 16 patients were found to be positive for HCV RNA, and 9 of the 14 viremic patients had genotype 1b HCV. Although the shared identities between the HCV isolate from Patient HC-IP and those from eight of the nine patients were merely 90.6-93.9% within the 1,087-nt NS5B sequence, the HCV isolate from the remaining one patient (HC-P12) was 99.7% identical to that from Patient HC-IP. Upon analysis of the E1 and E2 junctional region including hypervariable region 1 (283 nt), there was a close relationship (99.3-100%) between clones obtained from Patients HC-IP and HC-P12. Although the nurse HC-IP had a finger injury, she took care of Patient HC-P12, a 70-year-old man with HCV-related cirrhosis and recurrent epistaxis, occasionally without wearing protective gloves. This study indicates the occurrence of HCV transmission by exposure of nonintact skin to blood in health care settings.

Analysis of the full-length genome of a subgenotype IIIB hepatitis A virus isolate: primers for broadly reactive PCR and genotypic analysis.

Among six known subgenotypes (IA, IB, IIA, IIB, IIIA, and IIIB) of human hepatitis A virus (HAV), the complete genomic sequence has not been determined for IIIB. In this study, the full-length genomic sequence of a IIIB HAV isolate (HA-JNG06-90F) recovered from a Japanese patient who contracted sporadic hepatitis A in 1990, was determined. The HA-JNG06-90F genome, which comprised 7462 nt excluding the poly(A) tail, was related most closely to NOR-21 of subgenotype IIIA with an identity of 89.1%, and was only 82.6-83.4% similar to human HAV isolates of genotypes I and II over the entire genome. Comparison of full-length genomic sequences of 20 reported isolates and HA-JNG06-90F generated optimal results for separation of different levels: the nucleotide identities were 80.7-86.6% at the genotype level, 89.1-91.9% at the subgenotype level, and 94.6-99.7% at the isolate level. Similar ranges of nucleotide identity were observed when comparing partial nucleotide sequences of the VP1-2B (481 nt; primer sequences at both ends excluded) and 3C/3D (590 nt) regions, which were amplifiable by PCR with primers designed from well-conserved areas of the HAV genome. All 66 samples with IgM-class HAV antibodies tested positive for HAV RNA by both VP1-2B (481 nt)-PCR and 3C/3D (590 nt)-PCR: subgenotype assignment was concordant in all samples tested (IA [n = 61], IB [n = 1], IIIA [n = 2] and IIIB [n = 2]). These results suggest that two broadly reactive PCRs using primers derived from the VP1-2B and 3C/3D regions, respectively, may be applicable to universal detection and phylogenetic analysis of various HAV strains.

Distinct changing profiles of hepatitis A and E virus infection among patients with acute hepatitis, patients on maintenance hemodialysis and healthy individuals in Japan.

To compare the epidemiologic profiles of hepatitis A virus (HAV) and hepatitis E virus (HEV) infections in Japan, the prevalence of clinical or subclinical HAV and HEV infections was investigated serologically and molecularly among 128 consecutive patients (age, mean +/- standard deviation, 37.5 +/- 14.7 years) who contracted acute hepatitis between 1989 and 2005 in a city hospital, and among 416 hemodialysis patients (60.1 +/- 12.6 years) and 266 medical staff members (34.6 +/- 11.4 years) at the same hospital, using stored periodic serum samples collected since the start of hemodialysis or employment, respectively. Between 1989 and 1995, among 93 patients with acute hepatitis, 51 (54.8%) were diagnosed with hepatitis A and only one patient with hepatitis E. Between 1996 and 2005, however, among 35 patients, only 3 (8.6%) were diagnosed with hepatitis A and 2 (5.7%) with hepatitis E. Although subclinical HEV infection was recognized in four hemodialysis patients (one each in 1979, 1980, 1988, and 2003) and two medical staff members (1978 and 2003) in previous studies, none of the 191 hemodialysis patients who had been negative for anti-HAV at the start of hemodialysis contracted HAV infection during the observation period of 7.6 +/- 6.4 years. Only one (0.4%) of the 246 medical staff members who had been negative for anti-HAV at the start of employment acquired hepatitis A during the observation period of 7.9 +/- 8.0 years: none had subclinical HAV infection. Clinical or subclinical HEV infection has occurred rarely during the last three decades, while HAV infection has markedly decreased at least since 1996.

Serological and molecular studies on subclinical hepatitis E virus infection using periodic serum samples obtained from healthy individuals.

Subclinical hepatitis E virus (HEV) infection among healthy individuals was studied serologically and molecularly. Serum samples collected at screening between March and April 2004 (or just before retirement) from 266 medical staff members (35 males, 231 females) who had been working for 8.8 +/- 8.5 (mean +/- standard deviation, range, 0.3-35.1) years in a city hospital in Japan and serum samples that had been collected from these staff members at the start of employment were tested for IgA, IgM, and IgG antibodies to HEV (anti-HEV) by in-house enzyme-linked immunosorbent assays. Overall, six subjects (2.3%) tested positive for anti-HEV IgG at the screening; among them, four subjects (1.5%) had already been positive for anti-HEV IgG at the start of employment and two subjects (0.8%) seroconverted after initiation of employment. Periodic serum samples that had been collected from the two seroconverted subjects were tested for HEV antibodies and HEV RNA. The two subjects became positive for anti-HEV IgG in 1978 or 2003, respectively, with no discernible elevation in alanine aminotransferase (ALT) level, and continued to be seropositive up through the screening date. Although anti-HEV IgM was not detectable in the two subjects, one was infected transiently with Japan-indigenous HEV strain of genotype 3 and the other was positive transiently for anti-HEV IgA. The present study indicates that even an individual with subclinical HEV infection had evidence of transient viremia in the absence of ALT elevation and that anti-HEV IgA detection may be useful for serological diagnosis of recent subclinical HEV infection.

Prevalence of hepatitis E virus infection among hemodialysis patients in Japan: evidence for infection with a genotype 3 HEV by blood transfusion.

To investigate the prevalence of hepatitis E virus (HEV) infection among patients on maintenance hemodialysis, serum samples collected in January 2003 from 416 patients who had been undergoing hemodialysis for 7.6 +/- 6.3 (mean +/- standard deviation) (range, 0.3-26.0) years in a dialysis unit in Japan and serum samples that had been collected from these patients at the start of hemodialysis were tested for IgG antibodies to HEV (anti-HEV IgG) by an "in-house" enzyme-linked immunosorbent assay (ELISA). Overall, 39 patients (9.4%) had anti-HEV IgG in January 2003, and included 35 patients (8.4%) who had already been positive for anti-HEV IgG at the start of hemodialysis and 4 patients (1%) who seroconverted after initiation of hemodialysis. Periodic serum samples that had been collected from the four seroconverted patients were tested for HEV antibodies and HEV RNA. The four patients became positive for anti-HEV IgG in 1979, 1980, 1988, or 2003, and continued to be seropositive until the end of the observation period. Although anti-HEV IgM was not detectable in the four patients, three were infected transiently with apparently Japanese indigenous HEV strains of genotype 3. The patient who contracted HEV infection in 1979 had been transfused with 2 U of blood 21 days before the transient viremia: one of the two stored pilot serum samples had detectable HEV RNA with 100% identity to that recovered from the patient. Our study provides evidence of transfusion-transmitted HEV infection in Japan in 1979, and that the prevalence of de novo HEV infection during hemodialysis was low (1.1% or 4/374).