OBJECTIVE: To determine risk factors for hysterectomy, pelvic pain, and continued menorrhagia after rollerball endometrial ablation.
METHODS: All women having rollerball endometrial ablations between 1990 and 2000 were sent standardized questionnaires on pre‐ and postablation symptoms, satisfaction with the ablation, and subsequent gynecologic surgery. Pathology reports from surgery after the ablation were reviewed when available.
RESULTS: Two hundred forty women had a rollerball ablation during this period and 174 (72.5%) responded to the questionnaire. The average age of women at the time of the ablation was 43.1 years and the mean follow‐up time since the ablation was 49 months. Seventy‐four percent of women were satisfied with the ablation and 92% reported decreased or absent menstrual bleeding since the ablation. However, 13% of women reported new or worsening pelvic pain symptoms since the ablation and 21 women (12%) had a hysterectomy after the ablation for continued symptoms. A previous tubal ligation was a risk factor for having a hysterectomy (hazard ratio of 3.3, P = .03) and for having worsened pelvic pain (hazard ratio of 3.2, P = .05) after an ablation. Women who had a previous tubal ligation were more likely to use pain medications for pelvic pain after an ablation. Age over 35 at the time of the ablation was predictive of less bleeding after the ablation. Pathologic findings consistent with the postablation tubal sterilization syndrome were observed in five surgical specimens for an incidence of 6%.
CONCLUSION: Having a tubal ligation is a risk factor for the development of pelvic pain and for having a hysterectomy after rollerball endometrial ablation. The incidence of pathologically confirmed postablation tubal sterilization syndrome is 6%, but clinical manifestations of this syndrome may be higher. Relatively older age at the time of the ablation is associated with a higher rate of improved bleeding symptoms after ablation.
Previous tubal ligation is a risk factor for new or worse pelvic pain and hysterectomy after rollerball endometrial ablation.
Department of Obstetrics and Gynecology, University of Iowa Hospitals and Clinics, Iowa City, Iowa; and Ob‐Gyn Associates PC, Cedar Rapids, Iowa.
Address reprint requests to: Bradley J. Van Voorhis, MD, University of Iowa Hospitals and Clinics, Department of Obstetrics and Gynecology, 200 Hawkins Drive, Iowa City, IA 52242; E‐mail: brad‐van‐firstname.lastname@example.org.
Received January 16, 2002. Received in revised form April 18, 2002. Accepted May 13, 2002.
Endometrial ablation is becoming an increasingly popular means of treating patients with menorrhagia. Most women experience reduced menstrual bleeding, reduced pelvic pain, and satisfactory outcomes from ablation.1 However, some women require hysterectomy after the ablation for menstrual problems. The hysterectomy rate after ablation is generally around 10%,2–4 but as high as 34%5 in the published literature. Although continued menorrhagia is the predominant reason for dissatisfaction with ablations, some women are dissatisfied due to the development of new or increased pelvic pain. One cause of pelvic pain is the postablation tubal sterilization syndrome; a syndrome that has been described in several case reports of women who have had a previous tubal sterilization and an endometrial ablation or resection.6–8 These women have presented with cyclic, severe cramping pain due to the presence of trapped endometrial tissue leading to hematosalpinges and endometriosis in the segment of the fallopian tube proximal to the site of ligation. Thus far the syndrome has been defined by pathologic findings after additional surgery. The clinical manifestations of this disorder may be more common than has been appreciated. We hypothesized that having a tubal ligation procedure before an endometrial ablation may be a risk factor for development of pain after an ablation, dissatisfaction with the procedure, and may predispose women for additional surgical procedures, including hysterectomy, after an ablation.
MATERIALS AND METHODS
Approval for this study was granted by the University of Iowa Human Subjects Committee. All patients having an endometrial ablation using a hysteroscopic rollerball from January 1990 to January 2000 at the University of Iowa or at Ob‐Gyn Associates, a private practice in Cedar Rapids, Iowa, were identified using Current Procedural Terminology codes. All patients were sent a standardized questionnaire (available on request) that detailed gravidity, parity, and tubal ligation history. Satisfaction with the ablation was rated as very satisfied, satisfied, neutral, unsatisfied, or very unsatisfied. Women were also asked if they had menstrual periods since the ablation and, if so, to characterize the bleeding in a typical cycle. Menstrual bleeding after the ablation was compared with menstrual bleeding before the procedure as less, the same, or heavier. The presence of both pre‐ and postablation pelvic pain was determined, and, if present, pain was characterized as mild, moderate, or severe. Any gynecologic surgery that occurred since the ablation was recorded and women were asked about the symptoms and or diagnoses leading to the additional surgery. Finally, current use of hormonal medications including hormone replacement therapy and oral contraceptive pills, and use of pain medications for control of pelvic pain, was detailed.
Women who did not respond to the first questionnaire were sent a second questionnaire. Those not responding a second time were contacted by phone and asked to return the questionnaire. The pathology reports of all patients having pelvic surgery of any type were also reviewed when available.
Only women who had complete endometrial ablations using hysteroscopy and rollerball electrocautery were included. All patients had been pretreated with either monthly gonadotropin‐releasing hormone (GnRH) agonist, Lupron Depot (TAP Pharmaceuticals, Deerfield, IL), or 600 mg of Danocrine per day for at least 1 month preoperatively. Intraoperatively, a standard operative hysteroscope equipped with a 5‐mm rollerball attachment connected to a Force 2 electrosurgical generator (Valleylab, Boulder, CO) was used at settings of 30–80 W of coagulating current. A 3% sorbitol solution was used as the distension medium with continuous monitoring of fluid balance. Attempts were made to completely ablate the uterine cavity, including the region of the tubal ostia. Greater than 90% of the procedures were directly performed and supervised by the authors (BJV and GS).
All questionnaire responses were entered into a database and the results were then analyzed. Comparisons between groups of women who had tubal ligations and those who did not were made by t tests for continuous variables and by χ2 analysis for categorical variables. Multivariable Cox proportional hazards models were used to test the hypothesis that age, gravidity, previous tubal ligation, and pelvic pain before the ablation would affect the outcomes and the occurrence of hysterectomy after endometrial ablation when controlling for the time of follow‐up from ablation. Survival analysis was performed using a Kaplan‐Meier plot to evaluate the outcome of hysterectomy over time in patients with and without a previous tubal ligation.
We identified a total of 240 women (95 from the University of Iowa, 145 from Ob‐Gyn Associates) who had a rollerball ablation during the study period. A total of 174 women responded to our survey for a response rate of 72.5%. There was no significant difference in the response rate when comparing the two populations of women (76% for the University of Iowa, 70% for Ob‐Gyn Associates). There was no difference in age or time since the ablation when comparing women who responded to the survey with nonresponders.
The mean age of women at the time of ablation was 43.1 years, with a range of 16–76 years. The mean follow‐up time (time from ablation to survey completion) was 49 months, with a range of 13–132 months. The median gravidity was 2 and median parity was 2. Greater than 90% of the women surveyed were white, consistent with the population in Iowa.
Table 1 demonstrates the outcomes of endometrial ablation in the whole population of women studied. A large majority of women (74%) considered themselves to be either very satisfied or satisfied with the results of the ablation, whereas 15% were unsatisfied or very unsatisfied with the results. Likewise, a large majority of women (92%) reported that menstrual bleeding was either absent or reduced after the ablation, as compared with before the ablation. Many women noted pelvic pain, including dysmenorrhea, before the ablation, with fewer women having this complaint after the ablation. However, 13% of women reported either the new onset of pelvic pain or worse pelvic pain symptoms after the ablation. Eleven women (6%) required a second ablation for continued menorrhagia, 21 women (12%) had a subsequent hysterectomy, and 5 women had additional laparoscopic pelvic surgery.
Comparisons between women who had an endometrial ablation and either did or did not also have a tubal ligation are listed in Table 2. Of women who had a tubal ligation, a large majority (80 of 90 women) had their tubal ligation in a separate procedure before their endometrial ablation, while 10 had the procedures concurrently. Women who had a tubal ligation were younger and had a higher mean gravidity and parity than women with no history of a tubal ligation. Following ablation, women with a tubal ligation had a similar satisfaction rate, had a lower amenorrhea rate but a higher percentage of reduced menstrual bleeding than women who did not have a tubal ligation. Despite having similar rates of preablation pelvic pain, women who had a tubal ligation were more likely to report postablation pelvic pain, new onset or worse pain postablation, and use of pain medications for pelvic pain. There was no difference in postablation contraceptive medication use, including oral contraceptive pills or progestins. Hysterectomies were performed more often in women who had a previous tubal ligation.
Multivariable Cox proportional hazard testing, performed to control simultaneously for multiple variables and to account for differing lengths of follow‐up time, revealed that having a tubal ligation was a risk factor for having a hysterectomy after endometrial ablation (Table 3). Women with a previous tubal ligation had a significantly greater chance of having a hysterectomy after ablation, by survival curve and analysis (Figure 1). Tubal ligation was also a risk factor for the development of new or worsened pain after ablation. Tubal ligation status had no effect on the percentage of women reporting reduced bleeding (Table 3), the amenorrhea rate (data not shown), or on the satisfaction rate after ablation (data not shown).
The age of the woman at the time of the ablation was a significant factor influencing outcomes after ablation. The effect of age on bleeding outcomes was not linear, so age was made a categorical variable in 5‐year increments (Table 3). Women in each age category over the age of 35 were more likely to report reduced bleeding after ablation as compared with women under the age of 35 (Table 3). Women who had their ablation performed when they were over the age of 40 were more likely to report amenorrhea after ablation as compared with women who had ablations performed when they were under the age of 35 (data not shown). In some cases, amenorrhea may have been due to the onset of menopause, as some ablations were performed to control bleeding in women on hormone replacement therapy. Forty‐three percent of women over the age of 50 at the time of the survey reported current use of hormone replacement medications. Women over the age of 45 were significantly less likely to report new onset pelvic pain after the ablation (Table 3) and were more likely to report satisfaction with the procedure than younger women (results not shown). Age at the time of the ablation did not predict postablation hysterectomy. The woman's gravidity, parity, and the presence of preablation pelvic pain had no effect on any outcomes after endometrial ablation.
A total of 21 hysterectomies were performed in women after the ablation was performed. Of 15 hysterectomies performed in women with a previous tubal ligation, 4 (27%) were performed for menorrhagia alone, 9 (60%) were performed for both menorrhagia and pelvic pain, and 2 (13%) were performed for pelvic pain only. Of six hysterectomies in women with no history of a tubal ligation, three (50%) were performed for menorrhagia alone, two (33%) were for menorrhagia and pelvic pain, and one (17%) was done for pain alone. These differences were not statistically significant.
Of the 21 hysterectomies performed, 15 had pathology reports that we could review. Six had hysterectomies performed at outside hospitals and we did not have permission to obtain these reports. Three of 15 specimens had pathologically confirmed postablation tubal sterilization syndrome. Four women had leiomyomas and two had endometriosis. Eight specimens had no reported pathological abnormalities and thin endometrial linings were noted. No cases of adenomyosis were found. Of 11 women who had a repeat ablation, two (18%) eventually had a hysterectomy for continued menorrhagia.
There were five additional pelvic surgeries in women after an endometrial ablation, four in women with a previous tubal ligation and one with no tubal ligation. Three women had benign ovarian cysts removed laparoscopically and two had dilated proximal tubal segments causing cyclic pelvic pain. Pathology from these two procedures was consistent with the postablation tubal sterilization syndrome. Thus, a total of five women had pathologically confirmed postablation tubal sterilization syndrome (three removed by hysterectomy and two by laparoscopy) for an incidence of 6% (5 of 90 women who had a tubal ligation and endometrial ablation). All pathologically confirmed cases occurred in women who had their tubal ligation as a separate procedure before the endometrial ablation.
The postablation tubal sterilization syndrome is a known complication of endometrial ablation.6–8 The syndrome has been reported after both rollerball ablation and endometrial resection, and after a variety of tubal ligation methods including cautery, Falope rings, and Pomeroy tubal ligation. In all previously reported cases and in all pathologically confirmed cases that we report, the tubal ligation was performed at least 1 year before the endometrial ablation. These women present clinically with severe, cyclic, unilateral, or bilateral pain usually beginning between 5 and 10 months after the ablation that is sometimes associated with light menses. The pathophysiology is thought to be related either to regeneration of endometrium in the transitional zone between tubal epithelium and the endometrium, or to reflux of endometrium into the proximal fallopian tube during the ablation. In either case, uterine scarring after the ablation leads to trapping of the tissue and the development of hematosalpinges and endometriosis within the tube proximal to the site of the tubal ligation. Resection of the dilated fallopian tubes has been reported to be curative.6
Having a previous tubal ligation is a risk factor for the development of new onset or worse pelvic pain after complete endometrial ablation. Furthermore, women who have had a tubal ligation report a higher rate of using pain medications for relief of pelvic pain after an ablation. There was no difference in use of hormonal medications, including birth control pills, ruling this out as a cause for reduced pain among women not having a tubal ligation for contraception. We found a rate of pathologically proven postablation tubal sterilization syndrome of 6%, a rate consistent with the 8.4% incidence reported by Bae et al.8 However, this may be an underestimate of the true incidence, as dilated proximal tubal segments may not be noticed unless the surgeon and pathologist are aware of this disorder. In addition, our findings suggest that more women may be symptomatic with the syndrome yet not require surgery.
Having a tubal ligation before an ablation was a risk factor for subsequently having a hysterectomy with a hazard ratio of 3.3 when controlling for age, length of follow‐up, gravidity, and the prevalence of preablation pelvic pain. Our findings confirm a recent study that found a hazard ratio of 2.2 for the risk of hysterectomy after ablation in women with a tubal ligation.9 Among women who had a history of a tubal ligation and subsequently had a hysterectomy, 73% reported pelvic pain as one of the indications for the hysterectomy. Three of six (50%) women having a hysterectomy but no history of tubal ligation listed pelvic pain as one of the indications for the hysterectomy. This difference was not statistically significant, probably due to small numbers. Increasing pain may be part of the explanation for the higher hysterectomy rate found in women who have had a tubal ligation before endometrial ablation.
Another possible explanation for the higher hysterectomy rate in women having both ablation and tubal sterilization is that these women may, in general, be predisposed to having a hysterectomy. In a large cohort study, Hillis et al observed that women who have tubal sterilization are four to five times as likely as women whose partners undergo vasectomy to have a hysterectomy.10 The reason for this is not clear as, outside of the setting of endometrial ablation, women who have tubal sterilization are no more likely than other women to have menstrual abnormalities including dysmenorrhea.11 Women who are willing to have a surgical procedure for birth control may be more likely to resort to a surgical procedure like hysterectomy for control of menstrual abnormalities.
Hysterectomy rates after rollerball endometrial ablation procedures have varied widely from less than 1%12 to 34%.5 Some of the variability can be explained by differing lengths of follow‐up and different methods for calculating the hysterectomy rate. There is no question that short‐term follow‐up studies underestimate the hysterectomy rate, as the occurrence of hysterectomies seems to be steady for the first 3 years after ablation with a possible decline after that.2,13 Studies reporting a projected hysterectomy rate by life‐table analyses to account for differing lengths of follow‐up will have a higher hysterectomy rate than studies reporting actual percentages of hysterectomies performed in the population studied.
Besides having a previous tubal ligation, other patient‐related factors may influence the rate of hysterectomy after rollerball ablation. Younger women (less than 35 years) have been reported to have an increased risk of hysterectomy after rollerball ablation,9 a finding also noted after laser ablation.14 This finding has not been confirmed in other studies of women having either endometrial resection or laser ablation, although in both studies there was a trend towards improved outcomes in relatively older women.13,15 We found that older age at the time of the ablation significantly predicted satisfaction and reduced bleeding after the ablation but did not predict the occurrence of hysterectomy.
The need for a repeat ablation has been reported to be a risk factor for hysterectomy after rollerball ablation,9 endometrial resection,13 and laser ablation.15 The effect of intrauterine pathology (polyps or fibroids) on the outcomes of endometrial ablations is not clear. Phillips et al reported decreased hysterectomy rates after laser ablation when intrauterine pathology was present.15 However, O'Connor and Magos reported improved outcomes after endometrial resections when the endometrial cavity was free of pathology.13 We could not study these variables as relatively few women had repeat ablations and endometrial ablations were largely restricted to women with a normal uterine cavity.
Although we and others9 have found that a previous tubal ligation is a risk factor for pelvic pain and hysterectomy after rollerball ablation, it is important to stress that a majority of women in this circumstance are satisfied with the results of the ablation and are able to avoid hysterectomy. We therefore feel that ablations should still be offered as an option for women with a previous tubal ligation, although women should be apprised of the increased risks. Methods to avoid the postablation tubal sterilization syndrome are not known, although Mc‐Causland and McCausland have described performing a partial endometrial ablation where only one wall of the endometrial cavity is ablated and the cornual regions of the cavity are avoided to reduce scarring.16 Preliminary satisfaction rates with this procedure are encouraging but long‐term outcomes are needed before this can be recommended. Whether or not newer global endometrial ablation procedures, including “balloon” procedures and heated‐saline ablations, cause the postablation tubal sterilization syndrome is not known. Should a patient with a previous tubal ligation develop pelvic pain after ablation, the gynecologist must be aware of the syndrome, as laparoscopic resection of dilated proximal fallopian tubes may be curative and hysterectomy might be avoided in some of these women.
1. Ke, RE. Endometrial ablation: An alternative to hysterectomy. Clin Obstet Gynecol 1997;40:914–27.
2. Chullapram T, Song JY, Fraser IS. Medium-term follow-up of women with menorrhagia treated by rollerball endometrial ablation. Obstet Gynecol 1996;88:71–6.
3. Molloy D, Taylor PT. Gynaecological surgery after endometrial ablation. Med J Aust 1994;161:604–6.
4. Tsaltas J, Taylor N, Healey M. A 6-year review of the outcome of endometrial ablation. Aust NZ J Obstet Gynaecol 1998;38:69–72.
5. Unger JB, Meeks GR. Hysterectomy after endometrial ablation. Am J Obstet Gynecol 1996;175:1432–7.
6. Townsend DE, McCausland V, McCausland A, Fields G, Kauffman K. Postablation-tubal sterilization syndrome. Obstet Gynecol 1993;82:422–4.
7. Webb JC, Bush MR, Wood MD, Park GS. Hematosalpinx with pelvic pain after endometrial ablation confirms the postablation-tubal sterilization syndrome. J Am Assoc Gynecol Laparosc 1996;3:419–21.
8. Bae IH, Pagedas AC, Perkins HE, Bae DS. Postablation-tubal sterilization syndrome. J Am Assoc Gynecol Laparosc 1996;3:435–8.
9. Dutton C, Ackerson L, Phelps-Sandall B. Outcomes after rollerball endometrial ablation for menorrhagia. Obstet Gynecol 2001;98:35–9.
10. Hillis SD, Marchbanks PA, Tylor LR, Peterson HB. Higher hysterectomy risk for sterilized than nonsterilized women: Findings from the US collaborative review of sterilization. Obstet Gynecol 1998;91:241–6.
11. Peterson HB, Jeng G, Folger SG, Hillis SA, Marchbanks PA, Wilcox LS. The risk of menstrual abnormalities after tubal sterilization. N Engl J Med 2000;343:1681–7.
12. Baggish MS, Sze EHM. Endometrial ablation: A series of 568 patients treated over an 11-year period. Am J Obstet Gynecol 1996;174:908–13.
13. O'Connor J, Magos A. Endometrial resection for the treatment of menorrhagia. N Engl J Med 1996;335:151–6.
14. Garry R, Shelley-Jones D, Mooney P, Phillips G. Six hundred endometrial laser ablations. Obstet Gynecol 1995;85:24–9.
15. Phillips G, Chien PRW, Garry R. Risk of hysterectomy after 1000 consecutive endometrial ablations. Br J Obstet Gynaecol 1998;105:897–903.
© 2002 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
16. McCausland AM, McCausland VM. Partial rollerball endometrial ablation: A modification of total ablation to treat menorrhagia without causing complications from intrauterine adhesions. Am J Obstet Gynecol 1999;180:1512–21.