Respiratory viral and bacterial co-infections in adults in acute care: A multiplex PCR-based study.

backgroundData on the clinical significance of viral-bacterial co-detection by molecular methods in adults is limited. We aimed to determine how often respiratory viruses and bacteria are co-detected in acutely ill adults and whether co-detection is associated with worse clinical outcomes than detection of a virus or bacterium alone.

methodWe prospectively enrolled 998 adult patients presenting to the emergency department with symptoms of respiratory infection, fever, chest pain, or poor general condition. Nasal swab samples were taken at entry and analysed by multiplex PCR for 16 respiratory viruses. PCR for 7 bacterial pathogens was performed later on 912 frozen-thawed samples. Medical records were reviewed for the clinical outcomes. Analysis of variance (ANOVA) was used to compare the differences in CRP and age between detection groups. The risk of major outcomes was evaluated using multivariate logistic regression.

resultsBoth viral and bacterial PCR were positive in 36 of 912 patients (4%). The most common virus in the Co-detection group was rhinovirus (21/36, 58%) and the most common bacterium Streptococcus pneumoniae (20/36, 56%). The patients in the Co-detection group were younger than those in the Sole bacteria group, who in turn were younger than those in the Sole virus group. Relative to the Negative group, the risk of pneumonia was highest in the Sole bacteria group (aOR 3.53, p < 0.01) and lower, though still elevated, in the Co-detection group (aOR 2.36, p = 0.03). The risk for antibiotic treatment during 30-day follow-up was greatest in the Co-detection group (aOR 4.42), followed by the Sole bacteria (aOR 2.40) and Sole virus (aOR 2.15) groups (all p < 0.01). There was no significant difference between the groups in the risk of death or ICU care.

conclusionViral and bacterial co-detection was rare in acutely ill adults. Although co-detection was associated with the highest risk of antibiotic treatment, it did not carry the highest risk of pneumonia and was not associated with death or ICU admission. Routine use of multiplex PCR for detecting respiratory viral and bacterial pathogens is therefore unlikely to add clinical value in this population. Larger cohorts powered for severe outcomes and real-time result reporting, are needed to determine whether respiratory viral bacterial testing can meaningfully guide clinical management.