Investigation of viral respiratory co-infections and associated risk factors among SARS-COV-2 suspects in Uganda

Abstract

Respiratory coinfections are usually understudied during pandemics like the ongoing COVID-19 pandemic yet preceding outbreaks of respiratory infections like influenza have implicated coinfections as a major cause of various symptoms that lead to severe morbidity and mortality. This study aimed at determining the respiratory co-infections existing among SARS-CoV-2 suspects, characterization of these respiratory co-infections and determining the predisposing factors associated with SARS-CoV-2 confirmed cases. A total of 400 surveillance samples from SARS-COV-2 suspects were tested using a multiplex PCR panel into which SARS-COV-2 was incorporated. Next-generation Sequencing (NGS) was done for four of the 400 samples for the genomic characterisation of the coinfections detected by the multiplex PCR. Bivariate and multivariate analyses were done by logistic regression using STATA to determine potential predisposing factors to SARS-COV-2. All Fastq files were analyzed using EDGE Bioinformatics.Viral coinfections of Rhinovirus A/B/C and Enterovirus A/B/C (2/400) were detected among the negative SARS-COV-2 samples, but none among the positives. There were also single infections of Coronavirus OC43/HK (1/400), Enterovirus A/B/C (1/400) and SARSrelated COV (2/400) among the negative SARS-CoV-2 samples. Genomic characterization revealed coinfection of Enterovirus C, Enterovirus D68 and Masteradenovirus C in one of the four samples which had been confirmed as an Enterovirus A/B/C and Rhinovirus A/B/C coinfection by mRT-PCR. Furthermore, sequencing revealed various bacterial respiratory co-infections in all four samples. There were no statistically significant predisposing factors associated with testing positive for SARS-CoV-2 and this could be attributed to the presence of many asymptomatic cases whereas the absence of coinfections among the SARS-COV-2 positive cases could either be due to viral interference, resource competition or impact of COVID-19 control measures. However, symptoms of cough, sore throat, muscle ache and body temperature were found to excellently predict testing positive for SARS-CoV-2. Interestingly, 75% of the samples that were positive for viral infections by mRT-PCR only revealed co-infection with an array of bacteria by NGS and this could be attributed to sample degradation, low viral concentration or even misdiagnosis by the multiplex PCR assay because the rtPCR assay also failed to detect an Adenovirus which was identified as Masteradenovirus C by NGS. Therefore, the use of PCR techniques alongside NGS for routine detection of respiratory pathogens gives more insight where one assay is limited.