Effect of length of surgery on the incidence of venous thromboembolism after benign hysterectomy

Abstract

Venous thromboembolism is a leading cause of morbidity and mortality postoperatively. The current venous thromboembolism risk assessment tools have not been validated in gynecologic patients. Most patients undergoing hysterectomy for benign indications will receive mechanical or pharmacologic prophylaxis based on preoperative risk assessment. However, current guidelines do not incorporate newer data that indicate additional risk of venous thromboembolism with prolonged surgery times or mode of hysterectomy. This study aimed to determine the effect of length of surgery, or operative time, on the risk of venous thromboembolism within 30 days after hysterectomy and determine whether differences in the effect of operative time exist across age, body mass index, and surgical approach. We performed a secondary analysis of prospectively collected surgical quality improvement data using the American College of Surgeons National Surgical Quality Improvement Program database, which contains demographic and perioperative information and 30-day postoperative outcomes from >500 hospitals, and targeted data files including procedure-specific risk factors and outcomes for a subset of hospitals. We analyzed patients undergoing abdominal, vaginal, or laparoscopic hysterectomy for benign conditions from 2014 to 2017, identified by the Current Procedural Terminology codes. We excluded patients with cancer, patients whose surgery was not performed by a gynecologist, patients who were not in the targeted files, and patients with missing operative time or with an operative time of <30 minutes. Patients were compared with respect to the incidence of venous thromboembolism and operative time, stratified by age, body mass index, and surgical approach. Multivariable logistic regression was performed; operative time was treated as a continuous, linear variable. A total of 70,606 patients were included. The 30-day venous thromboembolism incidence was 0.4% (n=259). Patients with venous thromboembolism were more likely to be obese, have inpatient procedures, and had, on average, greater uterine weight. Hysterectomy approach was vaginal in 11,641 patients, laparoscopic in 41,557 patients, and abdominal in 17,408 patients. After adjustment, for each 60-minute increase in operative time, there was a 35% increase in the odds of venous thromboembolism (adjusted odds ratio, 1.35; 95% confidence interval, 1.25-1.45). Stratified by surgical approach, the odds of venous thromboembolism per 60-minute increase in operative time was greatest among abdominal hysterectomy (adjusted odds ratio, 1.49; 95% confidence interval, 1.35-1.65) compared with laparoscopic hysterectomy (adjusted odds ratio, 1.20; 95% confidence interval, 1.05-1.38) and vaginal hysterectomy (adjusted odds ratio, 1.27; 95% confidence interval, 0.97-1.66) (P=.01). Increasing body mass index and increasing age did not modify the impact of operative time on venous thromboembolism incidence (P=.66 and P=.58, respectively). Every 60-minute increase in operative time was independently associated with a 35% increased odds of venous thromboembolism within 30 days of hysterectomy, and this risk was cumulative. Minimally invasive hysterectomy had lower odds of venous thromboembolism than abdominal hysterectomy across all time points.