Background: The extent to which systemic perioperative thromboembolic prophylaxis affects peroperative and postoperative bleeding during cholecystectomy is not known. This article reports on risk of bleeding in a national cohort of cholecystectomies.
Methods: All cholecystectomies registered in the Swedish Register of Gallstone Surgery and Endoscopic Retrograde Cholangiopancreatography (GallRiks) between 2005 and 2010 were reviewed. Peroperative bleeding was defined as bleeding that could not be controlled by standard surgical techniques, necessitated conversion to an open procedure or required peroperative blood transfusion. Postoperative bleeding was defined as bleeding that necessitated reoperation, transfusion or a prolonged hospital stay. Risk estimates were performed using univariable and multiple logistic regression, and reported as odds ratios (ORs).
Results: A total of 51 621 procedures were registered in GallRiks. Some 48 010 patients were included in the analyses, of whom 21 259 (44.3 per cent) received thromboembolic prophylaxis. Peroperative bleeding complications occurred in 400 (1.9 per cent) and postoperative bleeding in 296 (1.4 per cent) given thromboembolic prophylaxis, compared with 189 (0.7 per cent) and 195 (0.7 per cent) respectively without thromboprophylaxis. After adjusting for age, sex, indication for surgery, American Society of Anesthesiologists grade, mode of admission, operative approach, duration of surgery and hospital volume, the OR for peroperative or postoperative bleeding complications in the group receiving prophylaxis was 1.35 (95 per cent confidence interval 1.17 to 1.55). However, in a subgroup analysis the risk was increased in laparoscopic surgery only. At 30-day follow-up, a total of 74 patients (0.2 per cent) had developed postoperative thromboembolism, 43 (0.2 per cent) of those who received thromboembolic prophylaxis compared with 31 (0.1 per cent) of those who did not.
Conclusion: Thromboprophylaxis in patients undergoing laparoscopic cholecystectomy increased the risk of bleeding, but the occurrence of thromboembolic events was not significantly reduced. Identification of high- and low-risk patients is needed to guide clinical decisions regarding medical thromboprophylaxis.
2012. Vol. 99, no 7, 979-986 p.