Microbiological sampling in lower respiratory tract infections: comparing nasopharyngeal and sputum specimens

To provide targeted antimicrobial treatment in lower respiratory tract infections (LRTIs), obtaining a microbiological diagnosis is a prerequisite. Despite guideline recommendations, diagnostic sputum sampling is rarely performed in the emergency room (ER). There is a lack of studies comparing microbiological findings from upper and lower respiratory tract specimens, especially for culture.

We conducted a prospective cohort quality study in the ER including 101 adult patients admitted with suspected LRTI between June 2023 and March 2024. Participants provided nasopharyngeal swabs and spontaneous or induced sputum for bacterial culture, and polymerase chain reaction (PCR) detecting viruses as well as bacteria causing atypical pneumonia. Sputum quality was assessed by light field microscopy.

PCR detected pathogens in 34.7% and culture in 31.7% of the participants. We observed no difference in the microbiological yield between the sampling sites. Negative and positive percent agreement between sample sites by PCR was 97.1% and 94.1%, respectively, whereas it was 84.1% and 47.8% by culture. Performing sputum induction in participants not able to produce spontaneous sputum significantly increased the proportion of participants with a microbiological diagnosis. Only one mild adverse event occurred in the 62 participants who underwent sputum induction.

In this study, sampling site for PCR analyses was of less importance in LRTI. For culture, agreement between sampling sites was poor, and sputum probably provides more representative results than nasopharyngeal specimens in LRTI. Sputum induction is feasible and safe and should be encouraged when patients cannot produce spontaneous sputum.