MORTALITY RISK ASSESSMENT IN PEDIATRIC INTENSIVE CARE UNIT OF A DEVELOPING COUNTRY USING PRISM SCORE

Abstract

Objective: To assess the application and efficacy of PRISM score in predicting mortality rate in a tertiary care PICU of a developing country.

Study Design: Prospective cohort study.

Place and Duration of Study: This study was carried out at Pediatric Department, Pak Emirates Military Hospital Rawalpindi, from Mar 2017 to Aug 2017.

Material and Methods: In this study 370 consecutive admissions as per the inclusion criteria were enrolled. PRISM score was calculated as recommended by original authors and outcome for each patient was recorded as expired or discharged. Hosmer-Lemeshow goodness-of-fit test and Area under the ROC Curve were used to ascertain the power of calibration and discrimination of the score.

Results: Out of the 370 patients with a mean age of 19 months, 282 (76.2%) survived and 88 cases expired (23.8%). Male to Female ratio was 1.56:1 and the majority of patients were infants (225/370). Infants recorded highest mortality percentage (24%). Infectious diseases were the most common etiology. There was no association of gender with outcome however length of stay had a positive correlation with survival. Mean PRISM score was 14.0 with scores significantly lower in survivors (p<0.001). Pearson Chi-square revealed significant association between PRISM score categories and child outcome. Hosmer & Lemeshow statistic was non-significant (p=0.244) at 10.30, revealing an acceptable goodness of fit of the regression model. Area under ROC curve was .885 indicating good predicting power of the model.

Conclusion: PRISM score showed good discriminatory capacity and was found to be a useful tool for predicting prognosis of pediatric patients admitted to a tertiary pediatric intensive care unit in resource constrained developing countries.