Incidence, risk factors, clinico-microbiological profile, change in ventilator settings needed and outcome of 135 ventilator associated pneumonia cases in pediatric intensive care unit (PICU) of a tertiary care centre in Eastern India

Abstract

Introduction: Ventilator associated pneumonia (VAP) is the second most common nosocomial infection diagnosed in mechanically ventilated patients with incidence of 20-36%, mainly caused by Gram-negative organisms in our country. Decrease in PaO₂/FIO₂ (arterial oxygen tension/fractional inspired oxygen) is an early marker of VAP. Impaired consciousness, re-intubation and continuous sedation are the most important risk factors of VAP. We aimed to study the incidence, risk factors, clinico-microbiological profile, change in ventilator settings needed and outcome of VAP in the pediatric intensive care unit (PICU) of a tertiary care centre in Eastern India.

Methods: This retrospective, observational study was conducted from January 2015 to June 2017, including 300 patients. We diagnosed VAP using Centre for Disease Control and Prevention (CDC) criteria and analysed the data using the Statistical Package for Social Sciences (SPSS®) version 20.0.

Results: Incidence of VAP was 45%, with higher incidence in infants with prolonged ventilation, use of continuous sedation and H2 blockers, re-intubation, presence of genetic syndromes and impaired consciousness. Gram-negative organisms (94%) (P. aeruginosa [45.93%], K. pneumoniae [25.18%], E. coli [14.81%], Acinetobacter spp. [8.14%]) outnumbered Gram-positive organisms (6%) (S. aureus [2.96%], Enterococcus spp. [2.22%] and S. pneumoniae [0.7%]). Resistance to common antibiotics was found in many cases. Multivariate analysis identified nasogastric tube feeding (adjusted odds ratio [OR] = 1.88; 95% CI = 0.8-2.3), use of H2-blockers (adjusted OR = 2.04; 95% CI = 0.51-4.5), use of continuous sedation (adjusted OR = 2.779; 95% CI = 0.7-4.9), re-intubation (adjusted OR = 4.861; 95% CI = 1.6-17.8) and duration of ventilation > 1 week (adjusted OR = 5; 95% CI = 0.7-6.3) as the risk factors for VAP. Purulent tracheal secretions (p < 0.0001), positive tracheal aspirate culture (p < 0.0001) and a suggestive chest radiograph (p < 0.0001) were the strongest predictors of development of VAP. The PaO₂/FIO₂ ratio was lower in the VAP group in all the three points of comparison but was not significant. The tidal volumes, peak and mean pressures, positive end-expiratory pressures (PEEP) and FIO₂ were higher in VAP patients both on days 3 and 5 of ventilation as compared to non-VAP patients but the differences were not statistically significant. Duration of PICU stay (16.5 ± 10.1 days) and mortality (53.3%) was higher in VAP patients compared to non-VAP patients (11.5 ± 9.2 days and 40.6%).

Conclusion: Identifying and minimizing the risk factors and proper choice of antibiotics as per sensitivity would improve outcome. Characteristics and parameters of mechanical ventilation were not influenced by the development of VAP. The variables of ventilation would not be sensitive for diagnosing VAP and clinical, radiological and microbiological criteria remain the tools for diagnosing VAP.