Molecular bacterial detection in nasopharyngeal swabs does not predict the clinical course of lower respiratory tract infections in primary care: an ancillary study of a randomized controlled trial

Background: Point-of-care molecular tests on easily collected samples could support antimicrobial stewardship in primary care. Aim: To determine whether molecular bacterial detection in nasopharyngeal swabs is associated with a worse disease course in patients with lower respiratory tract infections (LRTIs). Design and Setting: Secondary analysis of a randomized controlled trial including 469 consecutive patients with LRTI, from 60 primary care practices in Switzerland. Methods: Polymerase chain reaction assays for respiratory viruses and bacteria were retrospectively performed on nasopharyngeal swabs. Kaplan-Meier analyses and a multivariate Cox regression were used to evaluate time to recovery according to bacterial detection. Results: A total of 345 patients were included, of which 142 (41%) had bacteria detected. The detection of bacteria did not influence median time to recovery (7 days in both groups, 95% confidence interval [CI] 6–8 days, p=0.29). Bacterial detection was not associated with time to recovery (adjusted hazard ratio 1.13, 95% CI 0.84–1.50, p=0.42), nor was the interaction term between bacterial detection and antibiotic prescription (adjusted hazard ratio 1.10, 95% CI 0.64–1.90, p=0.74), indicating independence from antibiotic treatment. The detection of bacteria was not associated with a different median duration of cough, sputum, or dyspnoea. The median duration of cough of any severity was 16 days (95% CI 15-18 days). Conclusion: Among patients with LRTI in primary care, molecular detection of bacteria in nasopharyngeal swabs did not predict a worse disease course. These findings cast doubts on the utility of syndromic molecular testing for managing LRTIs in this setting.