A Comparison of PRISM-IV and PIM-III Scores in Predicting Mortality in the Paediatric Intensive Care UnitABSTRACT OBJECTIVES: To determine the sensitivity, specificity and diagnostic accuracy of PRISM-IV score and PIM-III score in predicting mortality in cr

Abstract

ABSTRACT

OBJECTIVES:        To determine the sensitivity, specificity and diagnostic accuracy of PRISM-IV score and PIM-III score in predicting mortality in critically ill children receiving critical care.

STUDY DESIGN:    Cross-sectional validation study.

SETTING/DURATION OF STUDY:         Department of Paediatrics, Combined Military Hospital, Lahore, Aug 2022 to Aug 2024.

METHODOLOGY:   We studied 300 children admitted to the paediatric intensive care unit aged between the ages of 1 to 12 years. Patients who were diagnosed as suffering from severe malnutrition, or those suffering from chronic liver disease, nephrotic syndrome or had received blood product transfusions prior to admission to the PICU were excluded. All patients underwent scoring with PRISM-IV and PIM-III systems at the time of admission and were followed up till completion of twenty-eight days, or till the occurrence of mortality. A PRISM-IV score of >10 and a PIM-III score of >4 were considered to be high risk for the occurrence of mortality.

RESULTS:    Patient age upon enrollment was 4.0 (IQR: 6.0) years, with a slight female majority of 158 (52.7%). Common indications for admission were pulmonary infections (n=93, 31.0%) acute exacerbations of bronchial asthma (n=73, 24.3%)  and non-respiratory infections (n=57, 19.0%).  PRISM-IV had a sensitivity of 92.86%, specificity 86.11% and diagnostic accuracy 88.00% in predicting mortality, while PIM-III had a sensitivity, specificity and diagnostic accuracy of 92.86%, 96.76% and 95.67%, respectively, for the same.

CONCLUSION:       PIM-III and PRISM-IV have good diagnostic accuracy in predicting the occurrence of death in paediatric critical care.