Background: During the past 40 years, management of solid organ injury in pediatric trauma patients has shifted to highly successful nonoperative management. Our purpose was to characterize children requiring operative intervention. We hypothesized that older children would be more likely to require operative intervention. In particular, we wanted to examine potential outcome disparities between children who were operated upon immediately and those in whom attempted nonoperative management failed. Additionally, we asked whether attempted nonoperative management, when failed, put children at higher risk for mortality or morbidities such as increased blood product transfusions or lengths of stays.
Methods: Retrospective cohorts from seven Level I pediatric trauma centers were identified. Blunt splenic, hepatic, renal, or pancreatic injuries were documented in 2,944 children <1 to 19 years of age from January 1993 to December 2002. Data collected included demographics, hemodynamics, blood transfusions, Glasgow Coma Scale score, Injury Severity Score, hospital length of stay (LOS), intensive care unit (ICU) LOS, and mortality. Analysis involved 140 (4.8%) of 2,944 patients requiring operation. Two cohorts were characterized: (1) immediate operation (IO), defined as laparotomy </=3 hours after arrival (n = 81; 58%) and (2) failed nonoperative management (F-NOM), defined as laparotomy >3 hours after arrival (n = 59; 42%).
Results: Comparing the two cohorts, no age differences were found. Compared with F-NOM, IO had significantly worse hemodynamics, Injury Severity Score, and Glasgow Coma Scale score and was associated with liver injuries. Pancreatic injuries were significantly associated with F-NOM. While controlling for injury severity to compare IO versus F-NOM, linear regression revealed equivalent blood transfusions, ICU LOS, hospital LOS, and mortality rates.
Conclusion: IO and F-NOM are rare events and independent of age. When operated upon for appropriate physiology, the timing of operation in pediatric solid organ injury is irrelevant and not detrimental with respect to blood transfusion, mortality, ICU and hospital LOS, and resource utilization.