A quasi-experimental study of a bundled diagnostic stewardship intervention for ventilator-associated pneumonia

OBJECTIVES:Diagnostic error in the use of respiratory cultures for ventilator-associated pneumonia (VAP) fuels misdiagnosis and antibiotic overuse within intensive care units (ICUs). In this prospective quasi-experimental study (NCT05176353), we aimed to evaluate the safety, feasibility, and efficacy of a novel VAP-specific bundled diagnostic stewardship intervention (VAP-DSI) to mitigate VAP overdiagnosis/overtreatment. METHODS:We developed and implemented a VAP-DSI using an interruptive clinical decision support tool and modifications to clinical laboratory workflows. Interventions included: gatekeeping access to respiratory culture ordering, preferential use of non-bronchoscopic bronchoalveolar lavage for culture collection, and suppression of culture results for samples with minimal alveolar neutrophilia. Rates of adverse safety outcomes, positive respiratory cultures, and antimicrobial utilization were compared between mechanically-ventilated patients (MVPs) in the 1-year post-intervention study cohort (2022-23) and 5-year pre-intervention MVP controls (2017-22). RESULTS:VAP-DSI implementation did not associate with increases in adverse safety outcomes but did associate with a 20% rate reduction in positive respiratory cultures per 1,000 MVP days (preintervention rate 127 [95% CI 122-131], postintervention rate 102 [95% CI 92-112], p<0.01). Significant reductions in broad-spectrum antibiotic days of therapy per 1,000 MVP days were noted following VAP-DSI implementation (preintervention rate 1199 [95% CI 1177-1205], postintervention rate 1149 [95% CI 1116-1184], p=0.03). CONCLUSIONS:Implementation of a VAP-DSI was safe and associated with significant reductions in rates of positive respiratory cultures and broad-spectrum antimicrobial use. This innovative trial of a VAP-DSI represents a novel avenue for ICU antimicrobial stewardship. Multicenter trials of VAP-DSIs are warranted.