RESUMO
BACKGROUND: Postoperative groin complication is a common cause of morbidity in vascular surgery. Prophylactic wound adjuncts addressing this issue have been shown to reduce complications in high-risk patients, but their widespread implementation is limited by their high cost. This study introduces a risk prediction model for patients at a high risk for groin complication which can be accessed through the iPhone application, Vasculink. METHODS: A literature search identified risk prediction models for groin complication in vascular surgery. Odds ratios of risk factors that were present in at least 2 published models were calculated with a pooled effect size. The weighted risk for each factor was used to create our model and a cutoff point defining high risk patients was chosen. The initial model was assessed and validated using a split-sample methodology on a cohort identified via a retrospective chart review of all patients undergoing open vascular surgery at our institution between 2017 and 2020. Model performance was assessed using the C-statistic. RESULTS: Risk factors included in our model were female gender, body mass index ≥28 kg/m2, ever-smoker, reoperation, use of prosthetic, emergency, and end-stage renal disease. Of 216 patients, 131 were at a high risk. The overall groin complication rate was 43%, and specific complication rates were 27% infection, 14.8% seroma, and 6.9% hematoma. Our model's sensitivity and specificity were 92.47% and 60.98%, respectively. The C-statistic is 0.768. CONCLUSIONS: By using risk factors identified in the literature we have been able to establish a highly sensitive risk prediction model for groin complication following open vascular surgery. By incorporating our model into an iPhone application, Vasculink, we hope to facilitate preoperative decision making regarding the use of prophylactic wound adjuncts.
