UCSF

OBJECTIVE: Traumatic popliteal vascular injuries are associated with the highest risk of limb loss of all peripheral vascular injuries. A method to evaluate the predictors of amputation is needed because previous scores could not be validated. In the present study, we aimed to provide a simplified scoring system (POPSAVEIT [popliteal scoring assessment for vascular extremity injuries in trauma]) that could be used preoperatively to risk stratify patients with traumatic popliteal vascular injuries for amputation. METHODS: A review of patients sustaining traumatic popliteal artery injuries was performed. Patients requiring amputation were compared with those with limb salvage at the last follow-up. Of these patients, 80% were randomly assigned to a training group for score generation and 20% to a testing group for validation. Significant predictors of amputation (P < .1) on univariate analysis were included in a multivariable analysis. Those with P < .05 on multivariable analysis were assigned points according to the relative value of their odds ratios (ORs). Receiver operating characteristic curves were generated to determine low- vs high-risk scores. An area under the curve of >0.65 was considered adequate for validation. RESULTS: A total of 355 patients were included, with an overall amputation rate of 16%. On multivariate regression analysis, the risk factors independently associated with amputation in the final model were as follows: systolic blood pressure <90 mm Hg (OR, 3.2; P = .027; 1 point), associated orthopedic injury (OR, 4.9; P = .014; 2 points), and a lack of preoperative pedal Doppler signals (OR, 5.5; P = .002; 2 points [or 1 point for a lack of palpable pedal pulses if Doppler signal data were unavailable]). A score of ≥3 was found to maximize the sensitivity (85%) and specificity (49%) for a high risk of amputation. The receiver operating characteristic curve for the validation group had an area under the curve of 0.750, meeting the threshold for score validation. CONCLUSIONS: The POPSAVEIT score provides a simple and practical method to effectively stratify patients preoperatively into low- and high-risk major amputation categories.