Objective
We sought to determine the overall effectiveness and risk factors for failure of hydrothermal ablation in the management of abnormal uterine bleeding.
Study Design
We performed a retrospective cohort analysis of patients who underwent hydrothermal ablation for abnormal uterine bleeding at our institution from July 2005 through February 2008. Variables analyzed included patient demographics, insurance status, body mass index, bleeding pattern, obstetric history, prior medical therapy and duration, uterine characteristics, and tobacco use history.
Results
In all, 159 patients were identified and 142 charts were eligible for evaluation. A total of 45 patients (31.6%) had return of preablation vaginal bleeding. Menometrorrhagia was a significant predictor for failure ( P = .027) and subsequent hysterectomy ( P = .0025). Younger age ( P = .044), tobacco use ( P = .042), and Medicaid/Medicare insurance status ( P = .039) were also associated with a higher risk of failure.
Conclusion
Women who are younger, use tobacco products, and have menometrorrhagia are more likely to fail hydrothermal ablation.
Abnormal uterine bleeding (AUB), defined as prolonged, excessive, or irregular vaginal bleeding, occurs in up to 30% of reproductive-age women and is a common reason for referral to a gynecologic practice. Medical management with combination oral contraceptives, progestins, or the progesterone-containing intrauterine device remains first-line therapy for women with AUB. Various surgical options exist for patients who fail medical therapy. Hysteroscopic endometrial resection or ablation is an effective alternative to hysterectomy in the treatment of AUB. Compared to hysterectomy, benefits of endometrial ablation include a shorter operative time, a shorter postoperative recovery period, lower cost, and the ability to perform the procedure in an outpatient setting with only regional anesthesia. However, the lower initial cost of endometrial ablation is offset by the potential need for additional treatment or procedures to achieve the desired result.
Multiple endometrial ablation techniques have been developed with only modest differences in efficacy and risk. Existing data suggest that the newer endometrial ablation techniques compare favorably to transcervical endometrial resection with regard to success rates and complication profiles. At our institution, we have used the Hydro ThermAblator (HTA; Boston Scientific, Natick, MA) since July 2005. A previously published prospective, randomized trial with 36-month follow-up demonstrated that 53% of women developed amenorrhea after undergoing ablation with the HTA, an additional 41% reported normalization of menses, and 6% had no improvement in their bleeding and ultimately required hysterectomy for definitive therapy. The results obtained with this device compare favorably to results obtained with the gold standard rollerball ablation technique while requiring less operator experience.
Multiple studies have sought to identify perioperative risk factors predictive of outcomes of endometrial ablation. However, the results of these studies are conflicting. The best perioperative predictor of success is increasing age. The presence of intracavitary lesions such as leiomyomas and endometrial polyps have varying impact on ablation success with some studies indicating no effect on outcome, some showing increased success rates, and others reporting a higher incidence of failure. The HTA has been demonstrated to be effective in women with submucous leiomyomas up to 4 cm in diameter.
While indicated for menorrhagia, Florida Medicaid currently requires all women at our institution with either menorrhagia or menometrorrhagia to undergo endometrial ablation prior to approval for a hysterectomy. There are limited data on the effectiveness of endometrial ablation for women with menometrorrhagia.
We performed this study to determine the effectiveness of the HTA in the management of women with AUB. Specifically, we sought to determine if there is a subset of women in whom HTA reliably can be predicted to be ineffective.
Materials and Methods
We identified and reviewed the medical records of all patients who underwent hydrothermal ablation for premenopausal AUB at our institution from July 2005 through February 2008. The study protocol was reviewed and approved by the institutional review board.
All patients were required to have an endometrial biopsy to exclude hyperplasia or carcinoma and cervical cytologic screening to exclude dysplasia within 1 year prior to undergoing endometrial ablation. Transvaginal ultrasonography or saline-infusion sonohysterography was performed preoperatively to measure the endometrial stripe and to evaluate the uterus for anatomic abnormalities. Endometrial ablation was not offered to any patients with submucous myomas ≥4 cm. Patients were pretreated with either a combination oral contraceptive or a progestin for at least 30 days preceding the ablation procedure. All procedures were performed at a tertiary academic medical center or its affiliated ambulatory surgical center. Patients were placed under general anesthesia and their endometrial cavity was explored using the HTA hysteroscope to evaluate for intrauterine pathology. After examination of the uterine cavity and confirmation of an adequate cervical seal, hydrothermal ablation was performed in the routine manner. Patients were categorized as having endometrial polyps when the diagnosis was made by direct hysteroscopic visualization at the time of HTA. In cases where endometrial polyps were identified, they were subsequently resected after the endometrial ablation was completed.
Cases were excluded if the endometrial ablation procedure was incomplete or aborted or if subsequent follow-up data were unavailable. Following treatment with the HTA we searched clinical documents and medical records for differences in reported bleeding patterns such as frequency or number of daily pad changes. We also searched for the number of days or weeks in a month during which the patient reported abnormal vaginal bleeding. Symptoms were then compared to what was documented prior to the ablation procedure. A determination of endometrial ablation success was made if a patient reported amenorrhea or a reduction in the number or frequency of pad changes and/or the number of days or weeks in a month during which they experienced bleeding and they did not require any additional medical or surgical therapy at their most recent follow-up visit. For patients in whom the procedure was considered a success, the time interval from the ablation procedure to the most recent follow-up visit was recorded. A determination of endometrial ablation failure was made if the patient reported either the same or an increase in the number or frequency of pad changes and/or the number of days or weeks in a month during which they experienced bleeding and/or they required additional medical or surgical therapy. For this group of patients, the interval of time from the ablation procedure to the reinstitution of medical therapy or subsequent surgery was recorded. Women who reported a decrease in bleeding but were not satisfied and required additional medical therapy or hysterectomy were categorized as treatment failure.
Additional variables recorded were patient age, race, insurance status, body mass index, parity, bleeding pattern (menorrhagia vs menometrorrhagia), duration of bleeding symptoms prior to the ablation, uterine size, endometrial stripe thickness, presence of polyps, presence of leiomyomas, presence of uterine malformations, and tobacco use history. Patients were divided into those who failed ablation and those who did not, and variables were analyzed to determine if any significant differences existed between the 2 groups. Patients were also compared based on their pretreatment bleeding pattern to determine ablation success rates in those patients with menorrhagia compared to those with menometrorrhagia. In patients who failed endometrial ablation we searched for a subgroup of patients that required hysterectomy. In this subgroup, the indications for the hysterectomy and the presence adenomyosis on final pathology were recorded.
Categorical variables were analyzed using χ 2 tests or Fisher’s exact tests. Continuous variables were analyzed using the Wilcoxon rank sum test. A P value of < .05 was considered statistically significant.
Results
We identified 160 patients who underwent endometrial ablation for AUB using the HTA from July 2005 through February 2008. One patient was excluded because her procedure was aborted before completion due to a loss of system pressure. The total ablation time was not recorded for this patient. Seventeen patients were lost to follow-up, leaving 142 charts eligible for evaluation. Patient ages ranged from 22-53 years with a median age of 40 years. Ninety-seven patients (68.3%) had amenorrhea or a significant reduction in vaginal bleeding with a mean follow-up of 12.4 months (range, 1–41 months). In contrast, 45 patients (31.6%) had return of preablation vaginal bleeding requiring additional treatment with a mean interval between ablation and failure of 10.1 months (range, 2–28 months). Hysterectomy was ultimately performed on 26 of the 142 patients (18.3%). The principal indications for hysterectomy in this subgroup of patients were either menorrhagia or menometrorrhagia. Five patients reported pelvic pain or dysmenorrhea as a secondary indication. Uterine sizes ranged from 56-244 g (mean weight of 134 g). Adenomyosis was identified on final pathology in 12 of the 26 patients (46.2%) requiring hysterectomy. Other histologic findings on hysterectomy specimens included leiomyomas in 9 of 26 patients, with the largest submucosal leiomyoma measuring 2.6 cm. Two of 26 patients were found to have endometrial polyps, the largest measuring 2.2 cm. There were no hysterectomy specimens with pathologic diagnoses of endometrial hyperplasia. When analyzing continuous variables, there was no difference in outcomes based on body mass index, parity, duration of symptoms prior to ablation procedure, uterine size, or endometrial stripe thickness ( Table 1 ). Also, when analyzing categorical variables, there was no difference in race and diagnosis of obesity (body mass index ≥30) ( Table 2 ).
| Perioperative variables | Success (n = 97) | Failure (n = 45) | P value | OR | 95% CI | ||
|---|---|---|---|---|---|---|---|
| Median | IQR a | Median | IQR a | ||||
| Age | 42.0 | 36.0–46.0 | 40.0 | 34.0–42.0 | .044 | 0.95 | 0.90–1.00 |
| Body mass index | 28.9 | 24.2–36.0 | 32.4 | 26.0–38.6 | .15 | 1.03 | 0.99–1.07 |
| Parity | 2.0 | 1.0–3.0 | 2.0 | 1.5–3.0 | .50 | 1.03 | 0.78–1.36 |
| Preprocedure duration of symptoms, mo | 24.0 | 12.0–48.0 | 24.0 | 12.0–60.0 | .48 | 1.00 | 0.99–1.01 |
| Uterine volume, cm 3 | 248 | 189–363 | 283 | 198–411 | .36 | 1.00 | 0.99–1.03 |
| Endometrial thickness, mm | 6.0 | 4.3–9.0 | 6.8 | 4.0–10.0 | .72 | 1.04 | 0.96–1.14 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
