19: The incidence of transfusion and associated factors in pelvic reconstructive surgery: An analysis of the national surgery quality improvement program database


      To assess the incidence of blood transfusion following pelvic reconstructive surgery in a large national surgical quality database and to identify transfusion-associated factors.

      Materials and Methods

      This retrospective cohort study was performed using the National Surgery Quality Improvement Program (NSQIP) database from the years 2009-2014. All women undergoing surgery for pelvic floor disorders were identified by CPT code. The incidence of blood transfusion was calculated. A multivariate logistic regression analysis was performed to estimate the odds of blood transfusion for various factors while controlling for other covariates. Data were analyzed using Stata 13.0.


      In the NSQIP database, there were 54,408 women who underwent pelvic reconstructive surgery from 2009-2014. The median age of patients was 57 years (range = 28-89). The majority of patients were overweight or obese, 35.6% (BMI = 25-29.9) and 36.7% (BMI ≥30), respectively. Most patients were white (91.4%), with 4.6% being African Americans, and the remainder being Asian, American Indian/Alaska Native, or Native Hawaiian/Pacific Islander. Hispanic ethnicity was reported by 9.3% of patients. American Society of Anesthesiologists (ASA) Class 1 and 2 represented a majority of the sample (76.5%). Concomitant hysterectomy was performed in 38.1% of the population. The incidence of blood transfusion in this population was 1.28% (n = 696). In the multivariate analysis, preoperative hematocrit less than 30 (odds ratio [OR] = 18.9), history of coagulopathy (OR = 3.4), concomitant hysterectomy (OR = 1.7), and BMI <18.5 (OR = 1.9) were statistically significantly associated (p < 0.05) with receiving a postoperative blood transfusion. When compared to ASA Class I, patients who were Class 3 (OR = 2.2), or Class 4 (OR = 5.2) had a statistically significantly increased odds of blood transfusion. When compared to white patients, African Americans (OR = 1.8) and Hispanics (OR = 1.9) had a statistically significant increased odds of postoperative blood transfusion. In this cohort, overweight and obese subjects had a decreased odds of transfusion (OR = 0.7). When compared to women ages 30-59, women less than 30 (OR = 2.9) and over 65 (OR = 1.09) had a statistically significant increased odds of transfusion while women ages 60-64 (OR = 0.7) had decreased odds of transfusion.


      Transfusion after pelvic reconstructive surgery is rare. The variables that are associated with transfusion are preoperative hematocrit less than 30, ASA class, bleeding disorders, concomitant hysterectomy, underweight status, and age less than 30 and over 65. Recognition of these factors can help guide preoperative counseling regarding transfusion risk after pelvic reconstructive surgery and individualize preoperative preparation.