ABSTRACT
This is a prospective observational study, comparing between three general physiology based prognostic scoring systems, which have been primarily developed to measure severity of illness and predict patient's outcome in pediatric intensive care unit [PICU]: Pediatric index of Mortality score [PIM], Pediatric Risk of Mortality score [PRISM] using both the original and current coefficients; in addition to evaluating the type and number of Organ System Failure [OSF]. The main goal was to assess the performance of each score. We prospectively collected data of 750 patients [447 males and 303 females] consecutively admitted to PICU with crude mortality rate of 5.6%. The study demonstrated that PIM score had good performance as evidenced by good discrimination and calibration in both the overall population as well as in two different subgroups [neonates and nonsurgical], but not in surgical, in addition to it is free and simple to use. It has also been demonstrated that PRISM score [original coefficients] had a poor performance in the overall population, as well as in nonsurgical and neonates but not in surgical subgroup. Meanwhile, PRISM [current coefficients] score showed good performance in the overall population, as well as in two subgroups [surgical and non surgical] but not in neonates. This is indicating the need to update and recalibrate prognostic scoring systems to maintain good performance over time. The new coefficients used in PRISM [current coefficients] in this study to compute mortality risk estimates were just developed to show the change in mortality risk over time and do not constitute a new pediatric mortality prediction scoring system. This study found also that organ system failure [OSF] is an important independent risk factor of death and patients with multiple organ system failure [two or more organ system failures] are more likely to die than those with one or less organ system failure