The Impact of Lower Respiratory Tract Sample Cultures on Outcomes in Community-Acquired Pneumonia: A Propensity Score-Matched Cohort Study.

Abstract

This study aimed to determine whether collecting a lower respiratory tract sample (LRTS) for bacterial microscopy, culture, and resistance (MCR) testing affects outcomes in patients with community-acquired pneumonia (CAP). This was a cohort study including adults admitted to hospital with CAP. The primary outcome was the duration of narrow-spectrum antibiotic treatment. Secondary outcomes included intravenous and total antibiotic durations, time to discharge, and 90-day mortality. Propensity score matching (PSM) balanced covariates between those who did and did not have an LRTS taken. Subgroup analyses focused on CAP due to Streptococcus pneumoniae or Haemophilus influenzae. After PSM, the cohort consisted of 1434 patients. LRTS collection did not impact narrow-spectrum antibiotic use but was associated with longer intravenous (0.6 days longer, P = .001) and total antibiotic treatments (10.4 vs 9.9 days, P = .036). Time-to-event analysis showed lower probability of early discharge for those with an LRTS (hazard ratio [HR], 0.88 [95% confidence interval {CI}:, .79-.98]) and higher probability for those with CAP due to S. pneumoniae or H. influenzae (HR, 1.44 [95% CI: 1.22-1.71]). Survival analysis showed lower 90-day mortality for patients with an LRTS (HR, 0.78 [95% CI: .61-.99]) and for those with CAP due to S. pneumoniae or H. influenzae (HR, 0.38 [95% CI: .24-.62]). LRTS collection did not directly affect the use of narrow-spectrum antibiotics but was associated with extended antibiotic treatments overall. Additionally, LRTS collection was linked to longer hospital stays and reduced mortality. Until more sensitive methods for determining CAP etiology become available, LRTS MCR testing should remain a standard investigation for patients admitted with CAP.