ABSTRACT
To determine the frequency of Ventilator-Associated Pneumonia [VAP] and to identify the associated factors, causative organisms and outcome of VAP in children admitted to ICU. Cross-sectional, observational study. Medical ICU [MICU] of the Children/'s Hospital and Institute of Child Health, Lahore, from August 2008 to March 2009. All children admitted to MICU and requiring ventilation during the study period were included and monitored for any features suggestive of VAP. Partial septic screen was done in all suspected cases. VAP was labelled when any patient on the ventilator for more than 48 hours had at least 2 of the following features of nosocomial infection - fever > 101'F, TLC < 4000 or > 15000 per mm[3], neutrophils > 85%, CRP > 48 mg/L or new findings on chest examination suggestive of pneumonia' and radiological evidence of new or progressive and persistent infiltrates. Percentages were compared using chi-square test with the significance at p-value less than 0.05. Of the 93 children requiring mechanical ventilation during the study period, 16 developed VAP [17%]. Almost half [46%] were younger than 1 year with male to female ratio of 1.2:1. Children developing VAP required ventilation for 13.5 [+ 10.1] days compared to 7.7 [+ 5.5] days in those who did not develop VAP. The common organisms isolated were Pseudomonas, Klebsiella and E. coli. Factors associated with increased frequency of VAP included age less than 1 year, unplanned emergency intubation and use of continuous intravenous sedation. Features that strongly suggested underlying VAP included purulent tracheal secretions compared to increased secretions alone, CRP > 48 mg/L, positive radiological findings and positive tracheal aspirate culture. Overall mortality was 23% among the ventilated cohort. Thirty two percent of them had VAP compared to only 13% among those who survived to discharge [p = 0.03]. The frequency of VAP was 17% in this series. Factors significantly associated with VAP were age less than 1 year, unplanned intubation and continuous sedation. The important predictors of VAP included purulent tracheal secretions, high CRP and persistent new radiological findings