Abstract: Comprehensive care for critically ill children in the pediatric intensive care unit (PICU) is done with the aim of achieving good outcomes. Severe disease in children is characterized by disruption of homeostatic processes, and can be evaluated by mortality scoring methods. There are several mortality scoring methods which can be used to predict mortality in children, the pediatric risk of mortality (PRISM) and pediatric index of mortality (PIM) are the most preferably used among all. The pediatric index of mortality 2(PIM2) is a key mortality prediction model for children receiving treatment in intensive care units, but its use has not been well validated in Indonesia.
Objective To evaluate the performance of PIM2 model in PICU patients.
Methods This cross-sectional study was conducted on PICU patients at Sanglah Hospital from November 2012 to April 2013. Patients underwent PIM2 scoring during their admission. The predictive ability of PIM2 scoring for patient mortality was analyzed using ROC curve.
Results A total of 54 patients were included in this study, of whom 8 (14.8%) died. Discrimination between survival and death was assessed by the area under the receiver operating characteristic curve and found to be 0.81 (95% CI 0.59 to 1.03). Sensitivity was 75 (95%CI 36 to 96)% and specificity was 98 (95%CI 87 to 99)%. The PIM2 cut off value was ≥ -0.99.
Conclusion The PIM 2 model has a good discriminatory power and calibration for predicting the mortality of children admitted to PICU and therefore is recommended for routine use in clinical practice.
Author: Monica Sampurna, Ida Bagus Suparyatha, I Gede Raka Widiana
Journal Code: jpkedokterangg160024