Aman Banerjee, MD; Jeffrey A. Claridge, MD, MS, FACS
Background: Non-operative management of hemodynamically stable patients with blunt splenic injury (BSI) has become the standard of care. However, practice patterns can vary by institution and occasionally by provider. Regionalization of trauma systems and implementation of evidence-based protocols can be used to reduce this variability and improve rates of success of non-operative management. The purpose of this multi-institutional study was to characterize patients with BSI within the system, determine regional practice patterns and outcomes, specifically, the spleen salvage rate within the trauma system prior to implementation of a system wide protocol.
Methods: The trauma registry for the Northern Ohio Trauma System, a regional trauma system made up of 2 healthcare systems was queried for patients older than 14 years diagnosed with BSI from 2008 through 2010. The primary outcome measure was spleen salvage rate, defined as a patient being discharged from the hospital with the spleen in situ. Secondary outcomes included mortality, ICU length of stay (LOS), overall LOS, ventilator days and rate of splenic artery embolization (SAE). The Level I center was compared to regional trauma centers, additionally, individual hospital analysis was also performed.
Results: 328 patients were identified. The Level I center treated more BSI patients during the study period 261 vs. 67. Patients treated at the Level I center were more severely injured with ISS 21.7 ±0.8 vs. 15.6 ±1.4, p <0.001. The Level I center patients tended to treat a higher percentage of patients with AIS head, chest, abdomen and lower extremity scores of 3 or greater p = 0.031, 0.001, 0.035 and 0.001, respectively. The Level I center treated a higher percentage of grade 3 and 4 spleen injuries 46% vs. 28.3, p = 0.003. SAE rates were higher at the Level I center 19.5% vs. 7.5%, p = 0.01. This was most pronounced for grade 3 and 4 spleen injuries 39.2 % vs. 15.8%, p = 0.05. Splenectomy rates and spleen salvage rates were identical at 14.9% (NS) and 85.1% (NS), respectively. Total LOS and ICU LOS was longer at the Level I center 8.8 vs. 5.3 (0.006) and 5.3 vs. 3.1 (0.05), respectively. Across the system splenectomy rates ranged from 8.7% to 100% (0.01). SAE rates ranged from 0% to 19.5% (NS).
Conclusion: The study established the splenectomy rate across the trauma system at 14.9% with a spleen salvage rate of 85.1%. There exists significant regional practice variability as seen in utilization of SAE, specifically, in grade 3 and 4 injured spleens. Use of SAE may account for the observation that although the Level I center treated a significantly higher proportion of grade 3 and 4 injuries the rate of splenectomy was similar to that of the regional hospitals. Centers that incorporated splenic artery embolization into their practice had higher rates of spleen salvage across the system; however this did not reach statistical significance. The data generated will serve as a point of reference with which comparison can be made with future studies within the trauma system.