Human respiratory syncytial virus diversity and epidemiology among patients hospitalized with severe respiratory illness in South Africa, 2012–2015

Abstract Background We aimed to describe the prevalence of human respiratory syncytial virus (HRSV) and evaluate associations between HRSV subgroups and/or genotypes and epidemiologic characteristics and clinical outcomes in patients hospitalized with severe respiratory illness (SRI). Methods Between January 2012 and December 2015, we enrolled patients of all ages admitted to two South African hospitals with SRI in prospective hospital‐based syndromic surveillance. We collected respiratory specimens and clinical and epidemiological data. Unconditional random effect multivariable logistic regression was used to assess factors associated with HRSV infection. Results HRSV was detected in 11.2% (772/6908) of enrolled patients of which 47.0% (363/772) were under the age of 6 months. There were no differences in clinical outcomes of HRSV subgroup A‐infected patients compared with HRSV subgroup B‐infected patients but among patients aged <5 years, children with HRSV subgroup A were more likely be coinfected with Streptococcus pneumoniae (23/208, 11.0% vs. 2/90, 2.0%; adjusted odds ratio 5.7). No significant associations of HRSV A genotypes NA1 and ON1 with specific clinical outcomes were observed. Conclusions While HRSV subgroup and genotype dominance shifted between seasons, we showed similar genotype diversity as noted worldwide. We found no association between clinical outcomes and HRSV subgroups or genotypes.

Conclusions: While HRSV subgroup and genotype dominance shifted between seasons, we showed similar genotype diversity as noted worldwide. We found no association between clinical outcomes and HRSV subgroups or genotypes. coinfection was found to negatively affect disease outcomes including increased odds of hospitalization. 2 HRSV is divided into two phylogenetically distinct subgroups (HRSV A and B), and these are further classified into multiple distinct genotypes based on sequence variability in the C-terminus of the envelope glycoprotein G (G protein). [4][5][6][7] Since 2002, novel genotype BA, and since 2009 novel genotype ON1, displaying 60-and 72-nucleotide sequence duplications (HRSV subgroups B and A, respectively) within the second variable domain of the C-terminus have emerged. 4,[7][8][9][10][11][12][13] To date, these novel genotypes have disseminated globally, becoming the dominant genotypes in circulation. 4,[7][8][9][10]14 Contrasting findings have found no clear consensus between HRSV subgroups/genotypes and clinical outcomes. 4,[7][8][9][10][11][12]15 Inconsistencies observed between studies may be due to small sample sizes and geographic variances in population diversity and herd immunity, which could predispose to or protect individuals from adverse clinical outcomes.
While no HRSV vaccines are currently available, global HRSV surveillance aided the development of multiple vaccine candidates that are in various stages of evaluation. 16 The high degree of sequence variance and complex circulation patterns of HRSV resulted in challenges for development of an effective HRSV vaccine. 13,[16][17][18][19] These challenges include global and regional gaps in knowledge of HRSV epidemiology, subgroup and genotype geographic circulation patterns, target patient demographics, and outcomes for HRSVassociated clinical illness. 16 Regular updates of these knowledge gaps provide an information baseline that can be used to evaluate regional vaccine effectiveness when implemented and can inform vaccine formulation suitability and timing.
In this study, we aimed to describe the seasonal patterns and prevalence of HRSV subgroups and genotypes among hospitalized patients with severe respiratory illness (SRI) in South Africa, from 2012 through 2015. We also aimed to compare the associations of HRSV subgroups and genotypes with epidemiologic characteristics and clinical outcomes.

| Study design and population
We enrolled participants in a prospective hospital-based SRI surveillance program from 01 January 2012 through 31 December 2015 at Edendale and Klerksdorp-Tshepong Hospitals situated in periurban areas in the subtropical KwaZulu-Natal and temperate North West Provinces of South Africa, respectively. The SRI case definition included hospitalized individuals of any age with illness onset of any duration prior to admission meeting age-specific inclusion criteria.
Children aged 2 days to <3 months included any hospitalized patient with diagnosis of suspected sepsis or physician-diagnosed lower respiratory tract infection. 20 Children aged 3 months to <5 years included any hospitalized patient with physician-diagnosed lower respiratory tract infection, including bronchitis, bronchiolitis, pneumonia, and pleural effusion. 20 Individuals aged ≥5 years included any hospitalized patient presenting with lower respiratory tract infection with temperature ≥38 C or history of fever and cough. 20

| Sample collection
Respiratory samples (nasopharyngeal aspirates for children <5 years of age and combined nasopharyngeal and oropharyngeal swabs from individuals ≥5 years of age) were collected and placed in universal transport medium (Copan, California, USA), stored at 4 C-8 C and transported within 72 h of collection to the National Institute for Communicable Diseases (NICD), Johannesburg, South Africa, for testing. 20 Whole blood samples (EDTA) were collected from consenting patients for the detection of Streptococcus pneumoniae (S. pneumoniae) by lytA polymerase chain reaction (PCR).

| Determination of HIV status
HIV infection status was obtained from a combination of two sources: (i) patient clinical records and (ii) for consenting patients, dried blood spots that were tested by enzyme-linked immunosorbent assay (ELISA) for patients aged ≥18 months and PCR for children aged <18 months if the ELISA was reactive or exposure status was unknown.  and F1 reverse primers as described. 25 HRSV-positive samples with insufficient volume or positives with cycle threshold (Ct)-values of >35 were not PCR amplified for sequencing as they were unlikely to successfully amplify. PCR products were analyzed and viewed following gel electrophoresis on a 1% agarose gel and purified using the

| CONCLUSION
This study builds upon our understanding of the clinical and epidemiological relevance of HRSV, its subgroups, and genotypes among patients hospitalized with SRI in South Africa. Changes in HRSV diversity over the study period demonstrated the presence and dominance of similar subgroups and genotypes found elsewhere in the world. This may suggest that no region-specific considerations will be required when selecting appropriate vaccine target strains.
Furthermore, we showed that the youngest age group (<6 months), which demonstrated the highest HRSV infection prevalence (47%) would benefit most from HRSV vaccine implementation, similar to findings from previous studies from South Africa (55%). 2,50 While no significant differences were found between the clinical outcomes and underlying illnesses associated with HRSV A and B infection or between HRSV A genotypes NA1 and ON1, a statistically significant association between HRSV A and S. pneumoniae coinfection in young children warrants further investigation of disease outcomes.

DISCLAIMER
The findings and conclusions in this report are those of the author(s) and do not necessarily represent the official position of the funding agencies.

ACKNOWLEDGMENTS
We would like to thank the SRI surveillance teams in Edendale and Klerksdorp-Tshepong Hospitals for their contributions to this study.
This work was supported by funds from the United States Centers for