Thrombophylaxis in Patients Undergoing Bariatric Surgery

Background:  Patients undergoing bariatric surgery are at increased risk for symptomatic venous thromboembolism (VTE). Although the incidence of clinical VTE is lower than in other general and orthopedic surgical procedures, pulmonary embolism is an independent predictor of death after gastric bypass surgery. The optimal strategy for postoperative thrombophylaxis in bariatric surgery has yet to be elucidated. In 2010 our institution proposed a guideline for postoperative thrombophylaxis for bariatric surgery based on risk stratification. Patients were considered high risk if they had history of VTE or a BMI >/= 60 kg/m² or if they had two or more of the following: age > 50, BMI >/= 50 kg/m², male sex, recent history of smoking, obstructive sleep apnea, varicose veins, or hormone therapy within 30 days. All patients undergoing bariatric surgery received enoxaparin 40 mg SC twice daily in addition to mechanical thrombophylaxis including venodynes and compression stockings during the hospital stay. Prophylactic IVC filter insertion was not recommended. High-risk patients received an extended course of anticoagulation for 10 days after discharge.   

Methods:  We retrospectively reviewed the medical charts of 692 patients who underwent bariatric surgery at Dartmouth-Hitchcock Medical Center from 2009 to 2014. We analyzed the incidence of VTE and bleeding complications under two different institutional thrombophylaxis protocols: an earlier protocol of VTE prophylaxis based on surgeon’s preference (n = 62) and our subsequent protocol based on risk stratification (n = 630). 

Results:  Prior to implementing our protocol the incidence of VTE and bleeding complications in patients who underwent bariatric surgery was 1.6% (1 in 62) for each. After 2010, with the implementation of extended VTE prophylaxis for high-risk patients, the incidence of postoperative VTE was 0% (0 in 630) and the incidence of postoperative bleeding was 2.7% (17 in 630). Of 17 patients who developed postoperative bleeding, 14 (82%) developed bleeding before postoperative day 3 during the hospitalization and 3 (18%) experienced delayed bleeding after hospital discharge. Of 3 patients with delayed bleeding, only 1 was on the extended thrombophylaxis protocol. Severe bleeding, defined as the need for packed red blood cell transfusions or re-operation, occurred in 1.6% (10 in 630).

Conclusion:  Our study demonstrates that a protocol based on risk stratification is effective at reducing the risk of symptomatic VTE in patients undergoing bariatric surgery. Although the incidence of all postoperative bleeding appears to have increased slightly after implementation of the protocol, the incidence of severe bleeding appears unchanged. As most of the post-operative bleeding events occurred during the hospitalization period, extending the length of pharmacologic thromboprophylaxis was not associated with an increase in bleeding. This study is limited by its retrospective design and the small number of patients studied prior to implementing the extended VTE prophylaxis protocol. Nevertheless, the findings are promising with respect to VTE risk reduction after bariatric surgery and suggest that prospective evaluation of the thromboprophylaxis protocol is in order.