Ventilator-Associated Pneumonia: How Do the Different Criteria for Diagnosis Match Up?

Abstract

Ventilator-associated pneumonia (VAP) affects up to 30 per cent of ICU patients and has been associated with increased morbidity and mortality. We identified factors associated with prolonged latency of VAP and evaluated its effects on survival and additional outcomes. We also determined the sensitivity of various clinical definitions of VAP, including the Centers for Disease Control and Prevention (CDC) 2013 criteria. We hypothesized that the CDC 2013 criteria would have poor sensitivity. We collected data on 102 subjects who developed VAP between 2012 and 2017. We conducted a Kaplan-Meier survival analysis with Cox proportional hazards regression and generalized linear models/ANOVA to look at predictor variables along with multivariate models for each outcome. White patients, nonsurgical patients, patients with renal failure, altered mental status, increased FiO₂, and increased positive end-expiratory pressure had worse survival. Trauma patients, patients with positive sputum cultures, and patients with suspected pneumonia had better survival. Sensitivity of the CDC 2013 criteria was only 44.1 per cent. Our results emphasize the importance of having a high index of suspicion for VAP in ventilator-dependent patients. The 2013 CDC criteria failed to detect 55.9 per cent of confirmed VAP cases. These results are concerning because undetected VAP can have devastating consequences for patients.