Comparison of intensive care of injured children between pediatric-based and non-pediatric-based intensive care units in a university hospital in Taiwan.

Abstract

Intensive care management of the injured child requires a multidisciplinary approach and meticulous attention to detail. However, the overwhelming majority of injured children are cared for by surgeons in surgical intensive care units (ICU) that see both adult and pediatric patients. There have been no previous reports of studies comparing the outcome in surgical ICUs (SICU) dealing with patients of all ages versus the outcome in pediatric ICUs (PICU). This study sought to determine differences in the outcome of pediatric intensive care between the SICU and PICU of our hospital. From Jan. 1989 to Aug. 1995, 118 children (68 boys and 50 girls), with an average age of 9.19 years (range: 3 months to 16 years), were admitted to our SICU. During the same period, 65 children (42 boys and 23 girls), with an average age of 5.04 years (range: 2 months to 16 years), were admitted to our PICU. Most of these patients received surgical intervention and were exclusively under surgeons' management. The decision to admit patient to the SICU or PICU was made by surgeon based on the availability of ICU beds. Pediatric risk of mortality (PRISM) score was used as a scoring system to assess disease severity in children. The most common cause for admission in both ICUs was traffic accidents. The average hospitalization duration in the SICU was 4.06 days (range 1 day to 23 days) and 3.34 days (range 1 day to 17 days) in the PICU. The average PRISM score was 7.87 (range 0-41) in the SICU and 6.48 (range 0-35) in the PICU. The overall mortality rate in the SICU was 12.7% (15/118) and 7.7% (5/65) in the PICU. There was a significant difference in patients' age and operative status but no significant difference in admission duration, PRISM score, and mortality rate between the SICU and PICU groups. The regression coefficients of the selected predictor variables and the impact on outcome showed one more score of PRISM would increase 1.5 fold of risk to become poor outcome while operation had lower risk (0.1 fold) to develop poor outcome. In conclusion, disease diversity and severity were similar among PICU and pediatric SICU patients in this study. The outcome was better in PICU patients although the difference was not statistically significant. The PRISM score is a useful measure to predict poor outcome in ICU patients after adjustment with confounders.