Splenic artery embolization versus splenectomy: Analysis for early in-hospital infectious complications and outcomes.

BACKGROUND: Splenic artery embolization (SAE) has gained increasing acceptance as an important adjunct in the treatment of splenic injuries. Residual immunologic function of the spleen after embolization and its consequences on early infectious complications still remain intensely debated. The purpose of this study was to compare SAE and splenectomy (SP) in terms of early in-hospital infectious complications and outcomes.

METHODS: Two-year retrospective Trauma Quality Improvement Program database prognostic study. Patients with grade IV to V splenic injury requiring SAE or SP were included in the final analysis. Examined variables were demographics, mechanism of injury, Abbreviated Injury Scale (AIS), Injury Severity Score, Organ Injury Scale, admission vital signs, blood transfusion in the first 24 hours, early infectious complications, and outcomes. Multivariate analysis adjusted for patient and injury-related variables was used to identify independent predictors for infectious complication and mortality.

RESULTS: During the study period, 4,063 patients with a grade IV to V splenic injury managed with SAE or SP were included in the study. SAE was performed in 461 (11.3%) patients. The early infectious complication rate was 23.1% in the SP group and 11.7% in the SAE group (p < 0.001). Stepwise logistic regression analysis identified age 65 years or older, Glasgow Coma Scale (GCS) score less than 9, Head AIS score of 3 or greater, SP, and blood transfusion in the first 24 hours as independent predictors for early infectious complications. The unadjusted overall mortality was 12.7% in the SP group and 5.4% in the SAE group (p < 0.001). Age 65 years or older, GCS score less than 9, hypotension, head AIS score of 3 or greater, and blood transfusion in the first 24 hours were independent risk factor for mortality. SP was not an independent risk factor in terms of mortality. Subgroup analysis in patients with isolated splenic injury showed age 65 years or older, GCS score less than 9, and blood transfusion in the first 24 hours as independent factors associated with early infection.

CONCLUSION: Our study supports the effectiveness of SAE in hemodynamically stable patients with a grade IV to V splenic injury. SP is associated with an increased risk of early infectious complications but is not an independent risk for mortality.

LEVEL OF EVIDENCE: Therapeutic study, level IV.