• Martin A Croce, Karen J Brasel, Raul Coimbra, Charles A Adams, Preston R Miller, Michael D Pasquale, Chanchai S McDonald, Somchan Vuthipadadon, Timothy C Fabian, and Elizabeth A Tolley.
• Department of Surgery, University of Tennessee Health Science Center, Memphis, Tennessee 38163, USA. mcroce@uthsc.edu
• J Trauma Acute Care Surg. 2013 Feb 1;74(2):354-60; discussion 360-2.

BackgroundSince its introduction by the Institute for Healthcare Improvement, the ventilator bundle (VB) has been credited with a reduction in ventilator-associated pneumonia (VAP). The VB consists of stress ulcer prophylaxis, deep venous thrombosis prophylaxis, head-of-bed elevation, and daily sedation vacation with weaning assessment. While there is little compelling evidence that the VB is effective, it has been widely accepted. The Centers for Medical and Medicaid Services has suggested that VAP should be a "never event" and may reduce payment to providers. To provide evidence of its efficacy, the National Trauma Institute organized a prospective multi-institutional trial to evaluate the utility of the VB.MethodsThis prospective observational multi-institutional study included six Level I trauma centers. Entry criteria required at least 2 days of mechanical ventilation of trauma patients in an intensive care unit (ICU). Patients were followed up daily in the ICU until the development of VAP, ICU discharge, or death. Compliance for each VB component was recorded daily, along with patient risk factors and injury specifics. Primary outcomes were VAP and death. VB compliance was analyzed as a time-dependent covariate using Cox regression as it relates to outcomes.ResultsA total 630 patients were enrolled; 72% were male, predominately with blunt injury; and mean age, Injury Severity Score (ISS), and 24-hour Glasgow Coma Scale (GCS) score were 47, 24, and 8.7, respectively. VAP occurred in 36%; mortality was 15%. Logistic regression identified male sex and pulmonary contusion as independent predictors of VAP and age, ISS, and 24-hour Acute Physiology and Chronic Health Evaluation as independent predictors of death. Cox regression analysis demonstrated that the VB, as a time-dependent covariate, was not associated with VAP prevention.ConclusionIn trauma patients, VAP is independently associated with male sex and chest injury severity and not the VB. While quality improvement activities should continue efforts toward VAP prevention, the Institute for Healthcare Improvement VB is not the answer. Financial penalties for VAP and VB noncompliance are not warranted.Level Of EvidenceTherapeutic study, level III.

29 keywords...

hide…