Extended-duration thromboprophylaxis after ventral hernia repair: a risk model to predict venous thrombotic events after hospital discharge

Abstract

Abstract Background Venous thromboembolism (VTE) is a major source of morbidity and mortality after ventral hernia surgery, but the risk of VTE after discharge has not been reported. Study design Data from the American College of Surgeons-National Surgical Quality Improvement Program (ACS-NSQIP) were used to investigate the risk of post-discharge VTE. Current procedural terminology (CPT) codes identified all reported patients who underwent ventral hernia repair from 2011 to 2017. We created a multivariable regression model for post-discharge VTE, using the 2011–2016 dataset to develop the model and 2017 as a validation set. The prediction model was used to create a risk calculator as a mobile application. Results The rate of VTE after surgery was 0.62% (878 of 141,065) with 48% occurring after discharge from the hospital. The final predictor model consisted of eight variables: age > 60 years, male sex, body mass index (BMI) ≥ 35 kg/m2), operative time > 2 h, concurrent panniculectomy, post-operative hospitalization > 1 day, presence of bleeding disorder, and emergency operation. The model had good calibration and discrimination (Hosmer–Lemeshow goodness-of-fit test, p = 0.71; c-statistic = 0.71). Threshold analysis showed a strategy of extended-duration thromboprophylaxis was optimized when the risk of post-discharge VTE was > 0.3%. Conclusion Forty-eight percent of VTEs after ventral hernia repair occur after discharge, particularly in older, male, obese patients undergoing longer and complex operations that require hospitalization > 1 day. Post-discharge thromboprophylaxis should be considered in these patients, particularly when risk of VTE exceeds 0.3%.