Endotracheal aspirate as an alternative to bronchoalveolar lavage fluid for the diagnosis of Pneumocystis pneumonia by real-time polymerase chain reaction.

Objectives: The purpose of this study was to evaluate the diagnostic performance of an endotracheal aspirate (ETA) real-time polymerase chain reaction (RT-PCR) assay for the diagnosis of Pneumocystis jirovecii pneumonia in intensive care unit (ICU) patients with suspected pneumonia.

Methods: We conducted a retrospective analysis of a prospective cohort in a 28-bed medical ICU. Adult patients who underwent both ETA and bronchoalveolar lavage (BAL) sampling for P. jirovecii real-time PCR within 48 hours between October 2018 and December 2024 were included. Diagnostic performance metrics of the ETA RT-PCR assay were calculated using BAL RT-PCR results as the reference standard (cycle threshold [Ct] ≤40). The concordance of Ct values between ETA and BAL samples was assessed using Pearson correlation.

Results: Among 249 included patients, 69 (27.7%) were positive for P. jirovecii by BAL RT-PCR. The ETA RT-PCR showed a sensitivity of 85.5% (95% confidence interval [CI]: 75.0-92.8%), a specificity of 92.8% (95% CI: 88.0-96.1%), an accuracy of 90.7% (95% CI: 86.4-94.0%), a positive likelihood ratio of 11.84, and a negative likelihood ratio of 0.16. Ct values between ETA and BAL samples were strongly correlated (r = 0.78, p < 0.001). Discordant results were observed in 23 patients (9.2%), with false-positive ETA samples showing significantly higher Ct values (median: 34.6; interquartile range [IQR]: 33.1-35.4) than true positives (median: 26.4; IQR: 24.2-31.0; p < 0.01).

Discussion: ETA RT-PCR for P. jirovecii demonstrates high diagnostic concordance with BAL fluid RT-PCR in ICU patients and may serve as a minimally invasive alternative. However, elevated Ct values in ETA samples warrant further confirmatory testing due to possible colonization.