Predicting ventilator-associated pneumonia in elderly patients requiring mechanical ventilation through the detection in tracheal aspirates.

Objective: In this study, we evaluated the clinical utility of tracheal aspirates α-amylase (AM), pepsin, and lipid-laden macrophage index (LLMI) in the early diagnosis of ventilator-associated pneumonia (VAP) in elderly patients on mechanical ventilation.

Methods: Within 96 hours of tracheal intubation, tracheal aspirate specimens were collected from elderly patients on mechanical ventilation; AM, pepsin, and LLMI were detected, and we analyzed the potential of each index individually and in combination in diagnosing VAP.

Results: Patients with VAP had significantly higher levels of AM, pepsin, and LLMI compared to those without VAP (P < 0.001), and there was a positive correlation between the number of pre-intubation risk factors of aspiration and the detection value of each index in patients with VAP (P < 0.001). The area under a receiver operating characteristic (ROC) curve (AUC) of AM, pepsin, and LLMI in diagnosis of VAP were 0.821 (95% CI:0.713-0.904), 0.802 (95% CI:0.693-0.892), and 0.621 (95% CI:0.583-0.824), the sensitivities were 0.8815, 0.7632, and 0.6973, the specificities were 0.8495, 0.8602, and 0.6291, and the cutoff values were 4,321.5 U/L, 126.61 ng/ml, and 173.5, respectively. The AUC for the combination of indexes in diagnosing VAP was 0.905 (95% CI:0.812-0.934), and the sensitivity and specificity were 0.9211 and 0.9332, respectively. In the tracheal aspirate specimens, the detection rate of AM ≥ cutoff was the highest, while it was the lowest for LLMI (P < 0.001). The detection rates of AM ≥ cutoff and pepsin ≥ cutoff were higher within 48 hours after intubation than within 48-96 hours after intubation (P < 0.001). In contrast, the detection rate of LLMI ≥ cutoff was higher within 48-96 hours after intubation than within 48 hours after intubation (P < 0.001). The risk factors for VAP identified using logistic multivariate analysis included pre-intubation aspiration risk factors (≥3), MDR bacteria growth in tracheal aspirates, and tracheal aspirate AM ≥ 4,321.5 U/L, pepsin ≥ 126.61 ng/ml, and LLMI ≥ 173.5.

Conclusion: The detection of AM, pepsin, and LLMI in tracheal aspirates has promising clinical utility as an early warning biomarker of VAP in elderly patients undergoing mechanical ventilation.