ABSTRACT: Endometrial ablation is a well-established minimally invasive gynecologic procedure for the treatment of abnormal uterine bleeding in premenopausal women. However, 16% of women have persistent abnormal bleeding or pain after ablation and fail therapy; repeat ablation or hysterectomy is required. Most postablation studies that investigated failure grouped bleeding and pain together; few examined pain alone or attempted to identify causes of pain. No previous studies have focused on identifying preablation predictors for the development of pelvic pain alone after ablation.
The aim of this retrospective study was to identify preablation patient characteristics associated with an increased risk of postablation pelvic pain. Data were obtained from a cohort of 437 women who underwent endometrial ablation between 2006 and 2010 at a large academic medical center. Patients were identified through use of Current Procedural Terminology codes (58563, 58353, and 58356) for any type of endometrial ablation (rollerball or global). Individual patient data were abstracted for multiple conditions and comorbidities. Postablation outcomes examined were bleeding; pain; and treatment with hysterectomy, repeat ablation, hormone-based interventions, and/or analgesics. Bivariate analysis of patient demographics (type of ablation, age at ablation, parity, smoking status, body mass index, uterine size, and history of prior tubal ligation or uterine surgery) and the incidence of pain after endometrial ablation was performed using the χ2, Fisher exact, and/or independent t tests. A final multivariate analysis with logistic regression was conducted to control for confounders, effect modifiers, and significant bivariate variables and to determine which specific patient characteristics were associated with pelvic pain after endometrial ablation.
Of the 437 women who underwent ablation, 91 (20.8%) reported pain after ablation. Patients were followed for up to 6.5 years after ablation. Median follow-up was 794 days, and the median number of days for pain onset was 301 days. Seventy-five percent of patients who developed pain reported it within about 2 years of the procedure. Postablation treatment in women reporting pain was primarily hysterectomy (15.1%) or hormonal treatment (9.4%). The median time to hysterectomy for patients with pain was 570 days. Of the 20.8% of patients reporting postablation pelvic pain, only 6.3% underwent subsequent hysterectomy for this indication. Before ablation, the following patient characteristics were significantly associated with the development of postablation pain: dysmenorrhea (adjusted odds ratio [aOR] = 1.73), smoking status (aOR = 2.31), prior tubal ligation (aOR = 1.68), and age younger than 40 years (aOR = 1.90). A personal history of endometriosis was not a statistically significant predictor of postablation pain. Neither an ultrasound finding of adenomyosis nor body mass index was a significant risk factor for postablation pain. Patients with all 4 proven risk factors for postablation pain (dysmenorrhea, smoking, prior tubal ligation, and age <40 years) had a 53% (95% confidence interval, 0.40–0.66) likelihood of developing pain. These preexisting conditions should be considered when counseling patients on expected outcomes after an endometrial ablation procedure.
Division of Minimally Invasive Gynecologic Surgery, Vanderbilt University Medical Center, Nashville, TN