The pediatric index of mortality 3 score to predict mortality in a pediatric intensive care unit in Palembang, South Sumatera, Indonesia
Journal Title: Paediatrica Indonesiana - Year 2017, Vol 57, Issue 3
Abstract
Background For critically ill patients in the pediatric intensive care unit (PICU), a scoring system is helpful for assessing the severity of morbidity and predicting the risk of mortality. The Pediatric Index of Mortality (PIM) 3 score consists of ten easy simple variables, so that the probability of death can be assessed prior to undergoing advanced therapies. The PIM 3 score in inexpensive and comprised of routine laboratory variables performed in PICU patients. In Indonesia, studies to validate the PIM 3 score have been limited.Objective To evaluate the PIM 3 score for predicting the probability of death in the PICU, Dr. Mohammad Hoesin Hospital (MHH), Palembang.Methods A prospective, cohort study was performed in the PICU, MHH, Palembang, from February to April 2016. The PIM 3 score was calculated within 2 hours of patients admission to the PICU by an android calculator application. PIM3 score and mortality were analyzed by Mann-Whitney test; calibration was performed by Hosmer-Lameshow goodness of fit test, discrimination was done by receiver operating characteristic (ROC) curve analysis; and standardized mortality ratio (SMR) was calculated.Results During the study period there were 81 PICU patients, 69 children were included, ranging in age from 1,5 to 187 months. The overall mortality rate was 40,58%. The most common illnesses in our subjects were malignancy (17,4%), post non-thoracic surgery (14,5%), dengue shock syndrome (14,5%), respiratory disease (13%), and neurological disease (11,6%). Subjects’ PIM3 scores ranged from 1,02% to 58,84%, with means of 26,08% in non-survivors and 13,05% in survivors. The SMR was 2,24, indicating that death was underpredicted. The AUC of 0,771 (95% CI of 0,651 to 0,891) indicated that the PIM3 score had good discrimination.Conclusion In Mohammad Hoesin Hospital, Palembang, South Sumatera, the PIM 3 can be used to predict mortality in PICU patients, but the score should be multiplied by a factor of 2.24. This recalibration is needed due to the presumed lower standard of care at this hospital compared to that of the originating PIM 3 institutions in developed countries.
Authors and Affiliations
Destiana Sari, Indra Saputra, Silvia Triratna, Mgs. Saleh
Keywords
Related Articles
- EP ID EP438219
- DOI 10.14238/pi57.3.2017.164-70
- Views 93
- Downloads 0
Correlation of heart failure severity and N-terminal pro-brain natriuretic peptide level in children
Background Heart failure affects morbidity and mortality in children with heart disease. There is no single, specific test to diagnose heart failure. The modified Ross Reithmann scoring system has been used to classify...
Using family atopy scores to identify the risk of atopic dermatitis in infants
Background Atopic dermatitis is the first manifestation ofallergic disease in early life. Early interventions may prevent thedevelopment of allergy disease. Allergy trace cards have beenused to identify the level of alle...
Developmental performance in small for gestational age children with and without catch-up growth
Background Infants born small for gestational age (SGA) have an increased risk of developmental delay. The influence of catch-up growth on developmental function remains unknown.Objective To compare the development of SG...
Lung function test in children with left-to-right shunt congenital heart disease
Background Increased pulmonary blood flow may lead to abnormal lung function in children with left-to-right (L to R) shunt congenital heart disease. This condition has been linked to considerable mortality and morbidity,...
Soluble transferrin receptor as an indicator of iron deficiency and febrile seizures
Background Iron deficiency (ID) has a high incidence inIndonesia, and is a risk factor for febrile seizures. The most suitableassay to detect iron deficiency in the presence of inflammationhas not yet been defined. An in...