Comparative Performance of Pediatric Risk of Mortality IV and Pediatric Index of Mortality 3 in Critically Ill Children with Cancer: A Prospective Observational Study

Kushal R Kalvit; Shilpushp J Bhosale; Jacob G Pulinilkunnathi; Swapna C Vijayakumaran; Anjana M Shrivastava; Atul P Kulkarni

doi:10.5005/jp-journals-10071-24979

Comparative Performance of Pediatric Risk of Mortality IV and Pediatric Index of Mortality 3 in Critically Ill Children with Cancer: A Prospective Observational Study

Indian J Crit Care Med. 2025 Jun;29(6):486-491. doi: 10.5005/jp-journals-10071-24979. Epub 2025 Jun 5.

Authors

Kushal R Kalvit¹, Shilpushp J Bhosale², Jacob G Pulinilkunnathi³, Swapna C Vijayakumaran⁴, Anjana M Shrivastava², Atul P Kulkarni²

Affiliations

¹ Guys and St Thomas' NHS Foundation Trust, London, United Kingdom.
² Department of Anesthesiology, Critical Care and Pain, Division of Critical Care Medicine, Tata Memorial Centre, Homi Bhabha National Institute, Mumbai, Maharashtra, India.
³ Department of Critical Care, Caritas Hospital Thellakom, Kottayam, Kerala, India.
⁴ Department of Critical Care Unit, Apollo Hospital, Chennai, Tamil Nadu, India.

Abstract

Background and aims: In pediatric critical care, pediatric risk of mortality (PRISM) and pediatric index of mortality (PIM) scoring systems are the most widely used severity of illness scoring systems. The current versions of these scoring systems (PRISM IV and PIM 3) have not been validated in critically ill oncology patients. The aim of this study was to evaluate the comparative performance of PRISM IV and PIM 3 in critically ill children with cancer.

Patients and methods: About 415 critically ill pediatric patients were recruited at a single-center, mixed medical-surgical ICU in a prospective observational study. Patients with an expected ICU stay of greater than 24 h were included. Demographic, physiological, and laboratory parameters required to calculate the PRISM IV and PIM 3 scores were collected; laboratory data from 2 hours prior to 4 hours after admission, and physiological data within 4 hours of admission were collected. The worst value was considered for calculating the scores. Both scores and their predicted mortality risk were calculated using online calculators.

Results: The mortality at ICU discharge and at hospital discharge was 32.7% and 36.1%, respectively. For the prediction of hospital mortality, the AUROC for PRISM IV score was 0.71and the PIM 3 estimated mortality (%) was 0.78, respectively. There was a statistically significant positive correlation between the PRISM IV score and PIM 3 estimated mortality (%) with the hospital mortality. The logistic model for PRISM IV score also passed the goodness-of-fit test indicating a good model fit. The goodness-of-fit test for the PIM 3 model suggests that the model may not fit the data well and may require further investigation or improvement.

Conclusions: Both PRISM IV and PIM 3 scores showed acceptable discriminative ability between survivors and non-survivors in the critically ill pediatric oncology population. However, only the PRISM IV score showed good calibration in the prediction of survival.

How to cite this article: Kalvit KR, Bhosale SJ, Pulinilkunnathi JG, Vijayakumaran SC, Shrivastava AM, Kulkarni AP. Comparative Performance of Pediatric Risk of Mortality IV and Pediatric Index of Mortality 3 in Critically Ill Children with Cancer: A Prospective Observational Study. Indian J Crit Care Med 2025;29(6):486-491.

Keywords: Mortality; PIM 3; PRISM IV; Pediatric intensive care; Pediatric oncology; Prospective observational study; Scoring methods.