Predictors of splenectomy after failure of non-operative management: An analysis of the nation trauma database from 2013 to 2014. (January 2021)

• Record Type: Journal Article
• Authors: Nguyen, Allan; Orlando, Alessandro; Yon, James R; Mentzer, Caleb J; Banton, Kaysie; Bar-Or, David
• Is Part Of: Trauma. Volume 23:Number 1(2021)
• Page Start: 13
• Page End: 23
• Publication Date: 2021-01
• Subjects: Blunt splenic injury -- non-operative management -- National Trauma Data Bank -- failure of non-operative management; Traumatology -- Periodicals; Disaster medicine -- Periodicals; Wounds and injuries -- Periodicals; Electronic journals; 617.1
• DOI: 10.1177/1460408620911489
• Languages: English
• ISSNs: 1460-4086

Abstract

Introduction: There is practice variability in non-operative management (NOM) of blunt splenic trauma. This is particularly true for management decisions following failure of NOM, i.e. splenectomy versus angioembolization (AE). The objective of this study was to identify predictors of splenectomy versus AE in patients who failed NOM. Methods: We included adult patients from the National Trauma Data Bank for 2013–2014, who had a splenic injury and who were admitted to a Level I Trauma Center (L1TC). Patients undergoing splenectomy after 2 h of emergency department arrival were deemed to have failed NOM. Multivariate logistic regression modeling was used to identify independent predictors of intervention after failed NOM. Results: There were 2284 patients admitted for splenic injury between 2013 and 2014 who failed NOM. A total of 1253 patients underwent AE and 1031 patients underwent splenectomy. Seven independent factors were identified that predicted failure of NOM: penetrating injury, community L1TC, hospital bed size, number of trauma surgeons on call, functional dependence, chronic steroid use, and cirrhosis. Conclusions: Seven independent variables were identified that predicted failure of NOM. These results contribute to the body of data regarding management of blunt splenic injury. Knowing predictive factors could help personalize management of patients, minimize delay of care, efficient resource allocation, and inform future studies.