Development of quantitative multiplex real-time RT-PCR assays for detection of 13 conventional and newly discovered viruses associated with lower respiratory tract infections in children in South Africa

Abstract

Acute lower respiratory tract infection (ALRI) is a major cause of paediatric morbidity and mortality, annually accounting for approximately 2 million childhood deaths worldwide of which up to 90% resides in the developing world. In 12-39% of ALRI cases no aetiological agent is identified, despite comprehensive investigations, thus suggesting that additional unknown agents may be involved. Since 2001 a number of new viruses have been identified that may account for some of these cases including human metapneumovirus (hMPV), human bocavirus (hBoV), and two new coronaviruses (hCoV) NL63 and HKU1. The contribution of the recently identified respiratory viruses to annual seasonal lower respiratory tract disease in Sub-Saharan Africa where human immunodeficiency virus infections may exacerbate respiratory infections is not fully understood. In addition, the role and disease association of many of these viruses as primary or co-infecting pathogens, as well as the underlying factors that may determine the pathogenesis of these viruses, is not yet well defined. Quantitative multiplex real-time RT-PCR assays were developed and validated for the detection of 13 well recognized and newly identified viral causes of ALRI, including respiratory syncytial virus (RSV), influenza viruses A and B, parainfluenza virus (PIV) types 1, 2, and 3, adenovirus, hMPV, hBoV and hCoV-NL63, -HKU1, -229E, and -OC43. The newly designed assays were subsequently used to facilitate the investigation of the contribution of respiratory viruses in patients requiring hospitalisation or attending outpatient visits in public sector hospitals serving the Pretoria area, South Africa. During 2006, the prevalence of the aforementioned respiratory viruses was determined by investigating the well recognized viruses previously diagnosed by routine immunofluorescence assays (IFA) in 737 respiratory specimens as well as viruses retrospectively detected by multiplex real-time RT-PCR in a sample group of 319 specimens. The epidemiology and disease association of these respiratory viruses in children who were predominantly less than 5 years of age with acute respiratory tract infections was investigated. Specimens were received from 2 public sector hospitals in Pretoria, South Africa. In addition, the disease association of each virus as a single or co-infection in human immunodeficiency virus (HIV) infected/exposed and HIV-uninfected children as well as the role of viral load was investigated. The multiplex assays could detect 2.5-25 recombinant plasmid DNA/RNA (in vitro transcribed) copies/μl, with a co-efficient of variation of less than 3.1%. Validation on 91 known positive respiratory specimens indicated similar specificity to IFA or single-round PCRs used in the initial identification of the viruses. Application of the multiplex assays to IFA negative specimens improved the detection of respiratory viruses by up to 44%. In children less than 5 years of age RSV was identified in 35.1%, followed by PIV 3 (8.3%); adenovirus (5.6%); influenza A (4.2%); hMPV (4.2%); hBoV (3.8%); hCoV-NL63 (1.6%); influenza B (1.0%); and PIV1, PIV2, hCoV-OC43, hCoV-229E, hCoV-HKU1 in less than 1% of cases. Co-infections were more common for the new viruses ranging from 58% of hMPV cases to 84% for hCoV-NL63 relative to 27% of RSV cases. Viruses were most frequently identified in children <1-year. RSV activity peaked in autumn and winter, PIV 3 in spring, while influenza A and B were mostly detected in winter. The observed seasonal distribution of hBoV and hMPV was less defined compared to traditional viruses, with both viruses showing variability over the two years. Comparable hospitalisation rates were observed for RSV, hMPV, PIV 3 and adenovirus, where approximately 60% of infected children were hospitalised. In addition, a high frequency of hospitalisations was observed in patients for both hMPV and hBoV in HIV-infected/exposed children. Co-infections occurred at higher frequencies with the new viruses, were more frequently associated with severe disease and were frequent in HIV-infected/exposed patients. Viral load was associated with severe RSV disease (p=0.014) however no significant association was observed for the new viruses as single infections. However, where hMPV occurred as a co-infection, higher viral loads of either hMPV or co-infecting agents occurred in severe cases. This association was also observed for hBoV. Most cases of hCoV-NL63 and hCoV-OC43 were co-infections in hospitalised patients. The newly developed multiplex assays demonstrates an improved sensitivity and scope of detecting respiratory viruses relative to routine antigen detection assays while the quantitative utility may facilitate investigation of the role of co-infections and viral load in respiratory virus pathogenesis. RSV remains the most significant viral cause of paediatric ALRI in South Africa. Viruses not currently included in routine diagnostic assays collectively contributed to 11% of ALRI cases in children <2-years in South African hospitals.