Neonatal Modified Pediatric Risk of Mortality Scoring System III and Perfusion Index in Neonatal Intensive Care Units

Abstract Background Pediatric Risk of Mortality (PRISM) score III presents an excellent discriminatory performance and prediction, being used in many PICUs as a prognostic score to assess gravity of disease(Kaur et al., 2020). In our study we modified the PRISM score III variables to produce the neonatal modified PRISM score III (NM PRISM score III) using validated cut offs which were more specific to neonatal age to be able to predict risk of mortality in both preterm and full term neonates admitted to NICU. The perfusion index (PI), derived from the pulse oximeter signal, has been reported to reflect inadequate peripheral perfusion in critically ill newborns (Montiero et al., 2017). The combined use of NM PRISM score III and PI can be a helpful method in identifying mortality and morbidity in NICUs accurately. Aim to test the use of the NM PRISM score III in predicting risk of mortality in NICUs. Methods In this prospective observational study, we adjusted the PRISM score III to neonatal based parameters score; neonatal modified PRISM score III (NM PRISM score III) and applied it to 112 neonates admitted to NICU, Ain Shams university hospitals on day1 and day3 of admission using the most abnormal values of physiologic and laboratory data to calculate the score for patients who survived and patients who died. Perfusion index was also used as an adjuvant to the NM PRISM score III (combined use of NM PRISM score III and PI) as an indicator of peripheral circulation. Results There was a significant difference between patients who survived and patients who died for NM PRISM score III with median 4 and IQR 2.00-4.00 at day1 and median 3 and IQR 0.00-7.00 at day3, with p value < 0.001 both at day1 and day3 of admission. By using ROC curve NM PRISM score III can be used to predict mortality at day3 as the area under the ROC curve at day3 was 0.975, with sensitivity 93.1% and Specificity 91.6%. The combined use of NM PRISM score III and PI at day3 of admission can also be used as it represented the best discriminative tool for predicting mortality as the area under the ROC curve at day3 was 0.980 with sensitivity 96.6 % and specificity 91.6%. Conclusion Combined use of NM PRISM score III and Perfusion index at day III can predict mortality accurately in critically ill neonates admitted to NICU.