Circulation of an atypical hepatitis C virus (HCV) strain at a dialysis unit in northeast India

Abstract Patients undergoing hemodialysis are at an increased risk of hepatitis C virus (HCV) infection. The implementation of standard infection control measures can substantially decrease the risk of infections and other nosocomial infections. To study the HCV infection rates and genotypes in maintenance hemodialysis subjects in a dialysis unit. A total of 196 maintenance hemodialysis subjects were tested for HCV RNA for one year at a tertiary care teaching hospital in northeast India continuously. Genotyping was performed using direct sequencing (Sanger sequencing) of the 5′UTR‐core region. The HCV infection rate was 26.0%. On phylogenetic analysis, 29 sequences clustered around genotype 3 and subtype 3f were observed. High sequence similarities (75–100% homology) were observed among the isolated sequences. High molecular similarities in the isolates from the same dialysis unit with a high infection rate (26.0%) over a relatively short period of study (10 months) indicated an ongoing nosocomial transmission. Nosocomial transmission by subtype 3f is rare, and it has not been reported in dialysis cases previously. The strain is most likely evolving from common strains such as 3b or 3i and may spread due to migration or movement of people. Urgent implementation of adequate infection control measures is required.

The prevalence of HCV in northeast India is higher than that in the entire nation (Phukan et al., 2003). However, the genotype distribution of HCV in northeast India has not been thoroughly investigated, and only a few studies have reported the predominant HCV genotypes 3 and 6 ( Barman et al., 2018;Christdas et al., 2013;Gupta et al., 2017;Shah et al., 2016;Win et al., 2016). The infection control activities are suboptimal even in tertiary care centers. Therefore, we conducted this hospital-based study to determine the infection rate of HCV, elucidate the co-infection pattern with other similar viral infections such as hepatitis B virus (HBV) and human immunodeficiency virus (HIV) infections, and understand the distribution of prevalent HCV genotypes.

| MATERIAL S AND ME THODS
This cross-sectional study was conducted between January 2016 and December 2016 (including two months of data analysis). As shown in the flow diagram (Figure 1), the inclusion criteria were as follows: 1) patients undergoing maintenance hemodialysis for chronic kidney disease (stage 5 or 4) or end-stage kidney disease (ESKD) as defined previously (Agar et al., 2007) and 2) adults (>18 years). The exclusion criteria were as follows: 1) acute renal failure patients, 2) other acute conditions requiring dialysis, and 3) HCV-positive or HIV-HBV patients on initial screening at admission or anytime during the study).
All samples were continuously collected. and epidemiological data of all participants were recorded in a standard pretested proforma. Approximately 6 mL of blood was collected from all participants, taking all necessary aseptic and antiseptic measures, and the samples were immediately dispatched to the laboratory.
Subsequently, the serum was separated, aliquoted into five parts (approx. 500 µL each), and stored at −80°C. One aliquot was withdrawn, and viral RNA was extracted using the manual High Pure Viral Nucleic Acid Kit (Roche Diagnostics, Rotkreuz, Switzerland) according to the manufacturer's instructions. Extracted RNA was further processed for quantitative detection of HCV RNA by performing real-time polymerase chain reaction (RT-PCR) using the Cobas TaqMan 48 analyzer (Roche Diagnostics, Rotkreuz, Switzerland).
For genotypic identification, we selected the first HCV RNApositive sample every week or four positive samples per month (as no positive cases were reported in some weeks), totaling 40 samples in 10 months. For initial amplification, direct sequencing was conducted to target a 405 bp region in the 5′UTR-core region using conventional PCR (Verma & Chakravarti, 2008). RNA extraction was performed using the QIAamp Viral RNA Mini Kit (Qiagen diagnostics, Hilden, Germany), and subsequently, cDNA transcription was conducted using an outer antisense primer (IR below) using the RevertAid First Strand cDNA It is worth mentioning that for genotype determination of HCV direct sequencing of a conserved region (e.g., 5'UTR) followed by phylogenetic analysis (or similar) is considered a reference method (Gold standard) and it is very reproducible, hence commonly resorted to.
Sequence data obtained from the above process underwent multiple sequence alignment using BioEdit software package v.7.0.1 (freeware under open source certification). According to the Tajima-Nei model, phylogenetic analyses were conducted using the rate heterogeneity parameter and the neighbor-joining method. The obtained sequences were compared with reference sequences, which were downloaded from the GenBank database as previously described (Verma & Chakravarti, 2008). Confidence values were calculated using bootstrap analysis, and a consensus tree was prepared using MEGA v 3.0 (Kumar et al., 2004). To determine the percentage similarities of sequences, homology analysis was performed using the DNA Starmega Align v 5.00 software (DNA Star Inc, Madison Wisconsin, United State).

| RE SULTS
A total of 207 patients were enrolled; of which, 11 were rejected (baseline data is included in Table 1 while Figure 1 depicts the flow).
Hence, subject enrollment was on a continuous basis, though some samples were rejected later, which was unavoidable.
No HIV-positive case was detected while HBV (HBsAg) and HCV-HBV dual infections were detected in 4.8% and 1.8% of the participants, respectively (Table 2). Conventional PCR resulted in successful amplification of 33 samples, whereas no amplification was observed in seven samples (the gel diagram shown in Figure 2).
On direct sequencing, 29 samples yielded useful sequence data (four samples failed or did not possess any useful sequence) that were deposited in the GenBank database (see Data Availability Statement ).
Phylogenetic analyses (Figure 3) of sequence data along with downloaded sequences revealed that all 29 samples were clustered in the 3f region. Furthermore, a homology analysis demonstrated that the percent identity of sequences varied from 85.6% to 100% (https:// doi.org/10.5281/zenodo.4294357).
Sixteen machines were present in the dialysis unit; of these, three were reserved for infectious patients (i.e., HIV-, HCV-, and HBVinfected patients). No isolation policy existed for infected patients.
Due to resource constraints, the dialyzers were routinely reused a maximum of 12 times. Each reused dialyzer contained the patient's name, and after each session, dialyzers were placed in separate containers labeled as clean, HCV, HBV, or HIV. A common sink was present, where dialyzers used by seronegative patients were rinsed first followed by those of HBV-and HCV-infected patients. Blood samples were collected in the treatment room next to the nurses' sta- On inquiry in adjoining and other units/wards/clinical departments in the hospital, no clustering or localization of HCV cases was found. Additionally, apart from receiving treatment in the same dialysis unit, none of the positive cases was epidemiologically or family or otherwise linked. As no study on HCV infections (with genotype and subtype detection) from the locality was available, baseline data of HCV genotypes in the locality could not be found.

| DISCUSS ION
We observed a high infection rate and an ongoing nosocomial transmission of HCV in a busy hemodialysis unit with low priority for infection control measures.
Male: Female ratio of 7:1 overwhelmingly tilted toward males despite gender not being a criterion for admission into the dialysis unit.
It may be a general pattern of CRF or common predisposing factors (Diabetes etc.) here. However, no specific reason could be found for this overwhelming male preponderance. As all risk factors were equally distributed among both genders, the high number of males in the study did not affect the results. (Data not shown).
Overall, 26.0% (51/196) of the dialysis patients were HCVinfected, and more than 90% of them were over 40 years of age ( 0.050 detection system such as enzyme-linked immunosorbent assay. Antibody detection leads to low sensitivity, that is, the antibody levels in immunosuppressed patients, such as hemodialysis subjects, may not be up to the detectable levels (Sypsa et al., 2005).
The HBsAg positivity rate was 4.8% while HCV-HBsAg co-infection was 1.8%. No HIV-positive subjects were observed.
HIV infection in hemodialysis subjects is rare, which is attributed to the low infection rates of HIV as compared to those of HCV or HBV and aggressive HIV screening before dialysis. HCV/HBV co-infection is a commonly detected dual infection in hemodialysis patients, and most studies have reported its prevalence to be in the range of 1-5%, similar to our findings (Mittal et al., 2013;Schiller et al., 2015).
Phylogenetic analysis (Figure 3) revealed that all 29 sequences were clustered in the area belonging to genotype 3 and subtype 3f, suggesting a common source of infection. All samples were collected from the same hospital, and they underwent dialysis at the same unit. Homology analysis for the percent identity varied from 85.6% to 100%, which is an indication of common source transmission. A 2019 HCV outbreak in a Dutch hospital dialysis unit used the 5′UTR & NS5A region sequencing techniques to successfully determine the infection (Heikens et al., 2019). An Italian study also reported an HCV outbreak in a dialysis unit using the same method as ours, that is, direct/capillary sequencing, to determine infections due to the same genotypes (Senatore et al., 2016).
To the best of our knowledge, this is the first study reporting an HCV 3f strain-related outbreak in our region. Reports on other HCV 3 subtypes causing chronic hepatitis or dialysis-related infections exist; however, HCV 3f-related outbreaks could not be retrieved from published literature. Multiple reports on the dialysis-related transmission of HCV genotypes such as 1b, 1a, and 3a, exist (Hmaied et al., 2006;Izopet et al., 1999;Ko et al., 2020). Recently, a Pune (India)-based study reported an HCV outbreak in a dialysis unit, where 84.6% of the isolates belonged to HCV 1b with 99.3% nucleotide sequence similarity, indicating a nosocomial transmission (Roy et al., 2019).
The genotype distribution patterns in such cases usually reflect the distribution in the general population (Ko et al., 2020).
The emergence of the uncommon 3f strain as a potential outbreak agent can have other inferences. As previously mentioned, the increased incidence of HCV due to an uncommon subtype may indicate the spread and transmission due to human migration-related infections (Gupta et al., 2017). Some studies have reported the transmission of novel HCV genotypes/subtypes due to migration (Borgia et al., 2018) According to the phylogenetic tree (Figure 3), it is obvious that the current strain is genetically related to subtype 3i and perhaps to the more commonly found 3b. Its associations with other prevalent strains such as 3a and 1b seem distant. Hence, it can be assumed that this particular strain is derived from the 3 g (or 3b) strain during transmission (Figure 3). A broader study with more subjects can provide additional clarifications.
Our study also indicates strong evidence of nosocomial patient-to-patient HCV transmission. Given these circumstances, the most likely cause of transmission is a breach of infection control procedures (Duong & McLaws, 2016;Jadoul, 2018). The high nucleotide similarity of the HCV strains isolated from patients indicates a common source, possibly a chronic HCV infection case.

| CON CLUS IONS
The infection rate of HCV in chronic dialysis or maintenance hemodialysis subjects in a tertiary care hospital in northeast India was high (26.0%), and the co-infection rate with HBV was 1.8%. An uncommon subtype of HCV (3f subtype) seemed to cause nosocomial transmission in the dialysis unit. Phylogenetic analyses indicated that this uncommon subtype could emerge from other subtypes such as 3 g or more common subtypes such as 3b. This new subtype may be linked to the spread of infection due to migration. Strict implementation and regular audits of standard infection control measures are urgently required (was advised to the concerned dialysis unit).

E TH I C S S TATEM ENT
Protocols and procedures were reviewed and approved by the Gauhati Medical College Institutional Ethics (approval no. MC/2/2015/82 dt13.5.2015). Written informed consent was obtained from each participant. Medical College, Guwahati, for allowing us the sample and data collection.

CO N FLI C T O F I NTE R E S T
None declared.
The raw sequence data generated during the current study are avail-