98 The association between bariatric surgery and need for fibroid procedural intervention amongst women with fibroids

Abstract

To compare the prevalence of fibroid procedural interventions between women with fibroids who did and did not undergo bariatric surgery. Based on a previously validated algorithm, a large database of de-identified medical records housed within our institution was queried to identify women with a fibroid diagnosis based on ICD codes and pelvic imaging findings from 2008 through 2015. Amongst this cohort, women who underwent bariatric surgery (exposed cases) were identified using CPT codes. Women who did not have bariatric surgery (non-exposed cases) were identified using optimal matching in a 2:1 (non-exposed:exposed) ratio. Exposed and non-exposed cases were matched on age at fibroid diagnosis, race, body mass index (BMI) at the time of fibroid diagnosis or bariatric surgery, fibroid type (subserosal, intramural, submucosal) and diagnoses of hypertension, diabetes, and obstructive sleep apnea (OSA). The outcome was need for a fibroid procedural intervention (laparoscopic or abdominal myomectomy, hysterectomy, uterine artery embolization, or endometrial ablation) by the end of 2019. Baseline characteristics were compared using the Wilcoxon test and Pearson’s test for continuous and categorical variables, respectively. The crude rates of fibroid procedural intervention between exposed and non-exposed cases were compared using Pearson’s test with Yates’ continuity correction. A mixed-effects model was used to control for correlations amongst matched exposed and non-exposed cases sets, as well as the residual differences in BMI between exposed and non-exposed cases. Twenty-seven exposed cases and 54 matched non-exposed cases were analyzed. Summary statistics on age, race, fibroid type, hypertension, diabetes, and OSA diagnoses are presented in Table 1. Non-exposed cases had a significantly lower BMI than exposed (median 37 vs. 45, P < 0.0001). Nine out of 27 exposed (33.3%) and 20/54 non-exposed (37.0%) required a fibroid procedural intervention (P = 0.73). In mixed-effects modeling, women who underwent bariatric surgery had lower odds of requiring a procedural intervention although this difference did not reach statistical significance (OR = 0.94, 95% CI = 0.75-1.18, P = 0.61). We did not find evidence that weight-loss surgery was associated with decreased risk of fibroid procedural intervention amongst women with fibroids.