Blunt splenic injury: use of a multidetector CT-based splenic injury grading system and clinical parameters for triage of patients at admission.

PURPOSE: To assess the use of a dual-phase multidetector computed tomography (CT)-based grading system alone and in combination with assessment of clinical parameters at triage of patients with blunt splenic injury for determination of appropriate treatment (observation, splenic artery embolization [SAE], or splenic surgery).
MATERIALS AND METHODS: This HIPAA-compliant retrospective study was approved by the institutional review board, and the requirement for informed consent was waived. Between January 2009 and July 2011, 171 hemodynamically stable patients with blunt splenic injury underwent multidetector CT at admission to the hospital. Images were reviewed by applying a multidetector CT-based grading system, and the amount of hemoperitoneum was quantified. Demographic data, vital signs, laboratory values, injury severity score, abbreviated injury severity, final treatment decision, and success of nonsurgical treatment were reviewed. Receiver operating characteristic curves and stepwise logistic regression analyses were performed to determine the optimal parameters for effective triage of patients.
RESULTS: One hundred seventy one patients with splenic injury underwent multidetector CT. At triage, clinical treatment decisions were made, and patients received either observation (85 of 171 [50%]) or splenic intervention (surgery, 19 of 171 [11%] or splenic angiography, 67 of 171 [39%]). Four patients underwent SAE after unsuccessful observation. Six of 171 (3.5%) other patients received unsuccessful nonsurgical treatment with SAE. No patients who received observation required splenectomy. Areas under the receiver operating characteristic curve (AUCs) showed that the CT grading system was the best individual predictor of successful observation (AUC, 0.95), and stepwise logistic regression analysis results showed that multidetector CT grade and the abbreviated injury scale score (AUC, 0.97; P = .02) were the best combination of variables for selection of patients for observation versus splenic intervention. The combination of abbreviated injury scale score, systolic blood pressure reading, and serum glucose level was the best triage model for decision making between splenectomy and SAE (AUC, 0.84).
CONCLUSION: The best individual predictor of successful observation in patients with blunt splenic injury was the CT-based grading system. Multidetector CT grade and abbreviated injury scale score were the best combination of variables for selection of patients for observation versus splenic intervention.