Design: Retrospective cohort study

Setting: The American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database from 2012-2020.

Patients or Participants: Women who underwent laparoscopic (conventional or robotic) myomectomy.

Interventions: N/A

Measurements and Main Results: Women who received ≥1 blood transfusions within 72 hours after laparoscopic myomectomy start time were compared with those who did not require blood transfusion. Multivariable analysis was to identify risk factors independently associated with the risk of transfusion. Two prediction models for the need for blood transfusion were developed based on multivariable results.

11,498 women underwent laparoscopic myomectomy during the study period, of which 331(2.9%) needed blood transfusion. In multivariable regression analysis of factors available before surgery, White race [adjusted Odds Ratio (aOR) 95% Confidence Interval (CI) 0.72 (0.53-0.98), p=0.034), bleeding disorders [aOR 95%CI 3.23 (1.63-6.45), p=0.001], ASA III/IV [aOR 95%CI 1.42 (1.02-1.98), p=0.037] and preoperative Hematocrit value <35.0% [aOR 95%CI 3.43 (2.73-4.31), p<0.001] were independently associated with risk of transfusion.

In the developed calculator, the risk of transfusion was 1.2% without risk factors, 2.4% with one, 6.1% with two and 10.8% with three risk factors. In the combined pre and intraoperative risk model, specimen weight >250 gram/≥5 intramural myomas and operation time >170 minutes were identified as additional risk factors for transfusion. The risk of transfusion ranged between 0.1% when no risk factors are present to 18.2% with five risk factors.

Conclusion: We identified preoperative and intraoperative independent risk factors for bleeding requiring blood transfusion among women undergoing laparoscopic myomectomy. A prediction model for risk of blood transfusion was developed based on the identified risk factors. Further studies are needed to validate this model.