LAUSR

PURPOSE There is significant discussion and uncertainty about the optimal management of symptomatic uterine leiomyomas (SULs). Nonsurgical procedures such as uterine artery embolization (UAE) have been developed. The goal of this study was to conduct a meta-analysis and an indirect treatment comparison to examine the comparative efficacy and safety of the surgical procedures to treat SULs compared with UAE. METHODS MEDLINE, EMBASE, Lilacs, and the Cochrane Central Register of Controlled Trials databases were searched from inception to February 2016. Ten randomized controlled trials comparing UAE versus hysterectomy, myomectomy, and laparoscopic occlusion of the uterine arteries in patients with SUL published in a peer-reviewed journal were included. Two reviewers independently selected studies, assessed quality, and extracted data. Discrepancies were resolved through consensus. FINDINGS Data from 986 patients submitted to UEA (n = 527) or surgery (n = 459) were analyzed. UAE had a lower risk of major complications (risk ratio [RR], 0.45 [95% CI, 0.22-0.95]; P = 0.04)and a higher risk of minor complications (RR, 1.65 [95% CI, 1.32-2.06]; P < 0.00001); UAE had a higher risk of re-intervention up to 2 years (RR, 3.74 [95% CI, 1.76-7.96]; P = 0.0006) and up to 5 years (RR, 5.01 [95% CI, 1.37-18.39]; P = 0.02); UAE had a similar risk of follicle-stimulating hormone levels >40 IU/L after 6 months (RR, 1.76 [95% CI, 0.24-12.95]; P = 0.58)and of recommending the procedure to another patient up to 5 years after treatment (RR, 1.00 [95% CI, 0.87-1.14]; P = 0.94). The indirect comparison between myomectomy and hysterectomy found that the 2 procedures were similar in the studied outcomes. IMPLICATIONS Compared with surgery, UAE had lower rates of major complications with an increased risk of re-intervention up to 2 and 5 years after the first procedure. UAE compared with surgery had a similar risk of ovarian failure and similar recommendation of the procedure to another patient. However, the number of trials was limited, and there was a high risk of bias in at least 2 domains. None of the trials blinded the participants and personnel or the outcome assessment. PROSPERO identifier: CRD42015026319.