- Email: [email protected]
Pathologic Characteristics of Hysterectomy Specimens in Women Undergoing Hysterectomy after Global Endometrial Ablation
Original Article
Pathologic Characteristics of Hysterectomy Specimens in Women Undergoing Hysterectomy after Global Endometrial Ablation Erin T. Carey, MD, Sherif A. El-Nashar, MBBCh, MS, Matthew R. Hopkins, MD, Douglas J. Creedon, MD, PhD, William A. Cliby, MD, and Abimbola O. Famuyide, MD* From the Department of Obstetrics and Gynecology, University of North Carolina (Dr. Carey), and the Department of Obstetrics and Gynecology, Mayo Clinic, Rochester, Minnesota (Drs. El-Nashar, Hopkins, Creedon, Cliby, and Famuyide).
ABSTRACT Study Objective: To describe uterine pathologic features in women who underwent hysterectomy because of failed global endometrial ablation (GEA). Design: Retrospective cohort study from 1998 through 2005 (Canadian Task Force classification III). Setting: Tertiary referral center. Patients: Sixty-nine women who underwent hysterectomy because of GEA failure. Interventions: Pathology reports were available for 67 patients. Descriptions of hysterectomy specimens after GEA were reviewed. Measurements and Main Results: Rates of pathologic findings in hysterectomy specimens after failed GEA were determined. Reasons for hysterectomy in the 67 patients with available pathology reports were bleeding in 34 (51%), pain in 19 (28%), and bleeding and pain in 14 (21%). The pathology reports of these specimens showed leiomyomas in 33 specimens (49%); intramural myomas were present in 15 women (44%) who underwent hysterectomy because of bleeding and 8 women (42%) who underwent hysterectomy because of pain. Hematometra was identified in 7 pathologic specimens (10%). Specifically, hematometra was identified in specimens from 5 of 19 women who underwent hysterectomy because of pain (26%). Conclusion: Hematometra was a significant finding in women who underwent hysterectomy because of persistent pain after GEA. A possible pathologic predictor of GEA failure may be intramural leiomyomas. Journal of Minimally Invasive Gynecology (2011) 18, 96–99 Ó 2011 AAGL. All rights reserved. Keywords:
Global endometrial ablation; Hysterectomy; Menorrhagia
Endometrial ablation was initially developed as an alternative to hysterectomy in women with excessive menses in whom medical management failed. The first ablation procedures were hysteroscopy-dependent and time-consuming, and success was directly proportionate to the provider’s surgical skill [1]. The US Food and Drug Administration approves several methods of global endometrial ablation (GEA) that have comparable clinical efficacy and are largely independent of hysteroscopic skill. These methods include thermal balloon ablation, circulated hot fluid ablation, The authors have no commercial, proprietary, or financial interest in the products or companies described in this article. Presented as an oral plenary presentation at the 39th Annual Meeting of the AAGL, Las Vegas, NV, November 8–12, 2010. Corresponding author: Abimbola O. Famuyide, MD, Department of Obstetrics and Gynecology, Mayo Clinic, 200 First St. SW, Rochester, MN 55905. E-mail: [email protected] Submitted April 27, 2010. Accepted for publication October 7, 2010. Available at www.sciencedirect.com and www.jmig.org 1553-4650/$ - see front matter Ó 2011 AAGL. All rights reserved. doi:10.1016/j.jmig.2010.10.004
bipolar radiofrequency ablation, cryotherapy, and microwave endometrial ablation [2]. Global endometrial ablation is now a well-established and effective treatment for refractory menorrhagia [2]. Despite the efficacy of GEA, as many as 30% of women will require hysterectomy after treatment [3,4]. Demographic and clinical factors predictive of GEA failure include younger age (,45 years), history of tubal ligation, preoperative dysmenorrhea, and presence of submucosal myomas [5–7]. However, few studies have characterized the uterine pathologic findings in women in whom GEA failed. For example, persistent bleeding after failed GEA has been associated with thermal necrosis or regenerating endometrium at histopathologic examination after hysterectomy [8]. Considerable uterine scarring also has been noted after ablative therapies, but its clinical significance in women in whom ablation failed is unclear [9,10]. Some evidence indicates that endometrial regrowth after GEA can lead to obstructive pain syndromes such as
Carey et al.
Hysterectomy after Endometrial Ablation
97
hematometra and postablation tubal sterilization syndrome, resulting in GEA failure secondary to pain [10]. In an effort to determine common pathologic findings and complications after GEA, a descriptive pathologic review was conducted of specimens from hysterectomies performed after failed GEA. Materials and Methods This retrospective population-based cohort study was approved by the Mayo Clinic Institutional Review Board. Only patients with a consent form on file allowing use of their medical records were included. Women who underwent GEA from January 1998 through December 2005 at Mayo Clinic (Rochester, MN), were identified from our patient database. From these patients, the subgroup of women who underwent hysterectomy after GEA was identified. The indication for hysterectomy was determined from clinical notes and operative reports; the pathology reports from the hysterectomy specimens were reviewed in detail by 2 independent practitioners (E.T.C. and S.A.E.-N.). Pathologic findings in women who underwent hysterectomy after GEA failure were evaluated. Patients were grouped according to the indication for hysterectomy: persistent bleeding, persistent pain, or persistent bleeding and pain. Within each of these categories, the relative percentages of various pathologic diagnoses were determined. Differences in pathologic findings between groups were analyzed using the independent t test, Wilcoxon rank sum test, c2 test, and Fisher exact test by using commercially available statistical software (JMP version 7.0.0; SAS Institute, Inc., Cary, NC). Survival analysis using a life table was used to calculate the cumulative hysterectomy rate (Table 1).
Fig. 1. Flow of patients through the study. GEA 5 global endometrial ablation.
for 1 patient. Although this group was not used as a comparison group, the pathologic findings in these specimens were included in the present descriptive study. In the other 69 patients, hysterectomy was performed because of GEA failure, defined as hysterectomy because of persistent bleeding, pain, or both, after GEA (rate of hysterectomy due to GEA failure, 9.7% [69 of 711]). Two hysterectomies were performed at outside institutions, and the operative reports were not available for review; thus, these patients were excluded. The remaining 67 patients (GEA Table 2
Pathologic characteristics of hysterectomy specimens according to outcome of GEAa Characteristic
Results During the 8-year study, 711 women (mean [SD] age, 43.3 [5.6] years) underwent GEA, either bipolar radiofrequency ablation or thermal balloon ablation, to treat menorrhagia. Of these, 77 women (10.8%) underwent hysterectomy after GEA (Fig. 1), at age 42.1 (5.9) years. In 8 patients, hysterectomy was not performed because of GEA failure: 2 patients had complex hyperplasia with atypia, 1 had cervical intraepithelial neoplasia, 2 had symptomatic myomas, 1 had pelvic organ prolapse, 1 had a complex adnexal mass, and no pathologic evaluation was available Table 1
Cumulative hysterectomy rate after endometrial ablation in the study duration Time to event (year)
At risk
Failure
1 2 3 4 5 6 7
629 596 560 559 522 404 263
4% 6% 8% 8% 10% 11% 12.5%
Confidence interval 1 2.5% 4% 6% 7% 8% 9% 10.%
Confidence interval 2 5% 7% 11% 12% 13% 15% 17.5%
Uterine length, cm Uterine weight, g Endometrial finding Proliferative Secretory Atrophic Ablative necrosis Other Adenomyosis Endometriosis Endosalpingiosis Hematosalpinx/hydrosalpinx Hematrometra Leiomyoma Submucosal Intramural Subserosal Intramural and submucosal Intramural and subserosal Submucosal, intramural, and subserosal Largest leiomyoma, mm Myoma size, mm R10 R50
GEA failure (n 5 67)
Non-GEA failure (n 5 7)
9.3 (1.2) 151.7 (92.5)
9.6 (1.1) 213.0 (109.5)
32 (48) 18 (27) 8 (12) 8 (12) 1 (1) 19 (28) 11 (16) 4 (6) 5 (7) 7 (10) 33 (49) 1 (3) 19 (58) 4 (12) 5 (15) 2 (6) 2 (6)
4 (57) 0 1 (145) 2 (29) 0 2 (29) 2 (29) 0 0 0 2 (29) 0 1 (50) 0 0 1 (50) 0
18.9 (13.1)
27.5 (31.8)
21 (64) 1 (3)
1 (50) 1 (50)
GEA 5 global endometrial ablation. a Values are given as mean (SD) or No. (%).
p Value .58 .23 .59
..99 .60 ..99 ..99 ..99 .44
.94 ..99 .12
98
Journal of Minimally Invasive Gynecology, Vol 18, No 1, January/February 2011
failure group) were categorized into 3 subgroups on the basis of the clinical reason for hysterectomy: persistent bleeding (n 5 34), persistent pain (n 5 19), or persistent bleeding and pain (n 5 14) (Fig. 1). In the GEA failure group, the pathology reports identified myomas in 33 patients (49%), adenomyosis in 19 (28%), ablative necrosis in 8 (12%), and hematometra in 7 (10%) (Table 2). Subgroups of the GEA failure group were compared according to indication for hysterectomy (bleeding vs pain) (Table 3). In the 34 women who underwent hysterectomy only because of bleeding, 18 (53%) had myomas, which were intramural in 16 (89%). These percentages were similar to those in the subgroup who underwent hysterectomy because of pain: 10 of 19 (53%) had myomas, which were intramural in 9 (90%). In contrast, in women in whom the indication for hysterectomy was pain and bleeding, myomas were present in only 5 of 14 specimens (36%), 3 of which were intramural. Endometrial findings were described as active (proliferative or secretory), atrophic, or ablative necrosis. A total of 9 patient had ablative necrosis; 5 out of 22 (18.5%) in patients who failed before 1 year compared to 4 out of 55 (7.3%) in patients who failed after 1 year (Chi square-Exact Fisher test, P 5 0.147). Ablative necrosis was identified in 2 specimens (6%) in the bleeding-only group and 4 specimens (21%) in the pain-only group (Table 3); ablative necrosis was not identified in the bleeding and pain group. Other endometrial findings included regrowth of active endometrium. Active endometrium was identified in 50 patients in the entire GEA failure group (75%) and in 28 patients (82%) in the bleeding-only group. Only 12 of 50 patients (24%) with endometrial regeneration had adenomyosis. Table 3
Pathologic characteristics of hysterectomy specimens according to indication for surgerya Indication for surgery Characteristic
Bleeding (n 5 34)
Pain (n 5 19)
Uterine weight, g Endometrial finding Proliferative Secretory Atrophic Ablative necrosis Adenomyosis Endometriosis Endosalpingiosis Hematrometra Leiomyoma Submucosal Intramural Subserosal Intramural and submucosal Intramural and subserosal Submucosal, intramural, and subserosal
145 (65.6)
173.2 (139.6)
17 (50) 11 (32) 4 (12) 2 (6) 10 (29) 6 (18) 4 (12) 1 (3) 18 (53) 0 11 (61) 2 (11) 3 (17)
7 (37) 3 (16) 4 (21) 4 (21) 6 (32) 2 (11) 0 5 (26) 10 (53) 0 6 (60) 1 (10) 1 (10)
1 (6)
1 (10)
1 (6)
1 (10)
a
Values are given as mean (SD) or No. (%).
p Value
.87 .49 .12 .03 .98
Fig. 2. Pelvic sonogram in a patient who underwent hysterectomy because of pain shows anechoic fluid in the left cornu suggestive of hematometra (arrowheads). TRANS UTE FUND 5 transverse view of the uterine fundus region.
Hematometra was identified in 7 patients (10%) in the GEA failure group. Only 4 of the 7 patients underwent imaging preoperatively (ultrasonography in 3 and computed tomography in 1); however, hematometra was identified in all 4 patients (Fig. 2). Of the 7 patients with hematometra, 5 (71%) were in the pain-only group, 1 was in the bleedingonly group, and 1 was in the bleeding and pain group. Six of the 7 women (85%) with hematometra had undergone previous bilateral tubal ligation, and GEA was performed using thermal balloon ablation in 4 and bipolar radiofrequency ablation in 3. Of the 5 patients with hematometra in the pain-only group, only 1 had a diagnosis of adenomyosis at hysterectomy. Discussion In the present study, histopathologic findings of uterine specimens after GEA failure were described. The literature is limited insofar as descriptions of uterine pathology in women in whom GEA fails. Factors that have been suggested as causes of GEA failure include intramural leiomyomas [11], necrotic myometrium, inflammatory response, and substantial scarring [9]. In a retrospective pathologic analysis of 20 uteri removed because of persistent bleeding after ablation, the 2 most common findings were ablative necrosis (a known endometrial response to thermal damage) and regenerating endometrium, which suggested that these factors are causes of persistent bleeding [8]. The ablative necrosis was identified within three months of the endometrial ablation. In the present study, the majority of the hysterectomy specimens with ablative necrosis were performed within the first year following the ablation procedure, however this was not statistically significant. Regenerating endometrium was also increased in the group of women with GEA failure due to persistent bleeding. In the present study, regenerating endometrium was increased in the group of women with GEA failure due to persistent bleeding. Deep adenomyosis also has been described as a risk factor for failure of GEA [12]; however, the complete role of adenomyosis in GEA failure or how it influences GEA
Carey et al.
Hysterectomy after Endometrial Ablation
outcomes is unknown. Adenomyosis has also been associated with endometrial regeneration, which has been previously associated with failure of endometrial ablation secondary to postablation bleeding [13]. These findings were not supported in the present study. A high rate of endometrial regrowth was noted in women in whom GEA failed; however, the rates of endometrial regeneration and adenomyosis were not significantly different between patient groups. Other reports have related postablation pain to the presence of adenomyosis [10]; however, adenomyosis was present only in a minority of patients who underwent hysterectomy because of pain alone. The rate of adenomyosis was similar in all groups including those with hysterectomy unrelated to GEA failure. This information suggests that factors other than adenomyosis may contribute to regrowth of the endometrium and that the presence of adenomyosis cannot reliably predict GEA failure. Submucosal leiomyomas are another known risk factor for GEA failure [14]. In the present study, most of the myomas identified in the specimens were intramural. The role of intramural myomas in GEA outcome is unknown. Loffer [15] suggested that the presence of submucosal and intramural myomas does not change the efficacy of ablation for control of bleeding; others believe that intramural myomas increase the failure rate of GEA, although no studies have shown this with statistical significance [11]. In the present study, the percentage of intramural myomas was higher in the GEA failure group; however, this result was not statistically significant. Therefore, further studies are warranted. Additional research is needed to specifically evaluate intramural myomas in the context of endometrial ablative therapy and their function in GEA failure. The strongest relationship found in the present study was between pain and hematometra. The cause of hematometra after ablation is thought to be from preservation or regeneration of the endometrium in the setting of substantial scarring, which may block the outlet and result in obstructive symptoms [10]. Women with hematometra often have cyclic pelvic pain or chronic pelvic discomfort [16]. Those who have undergone tubal ligation are at increased risk of GEA failure as a result of pain, known as postablation tubal sterilization syndrome [17]. The pain is thought to be caused by distention of the proximal end of the fallopian tube during endometrial growth with no outlet into the peritoneum. Pain secondary to hematometra seems to be an important cause of failure; patient selection and appropriate counseling should be initiated before ablation therapy. A limitation of the present study is that it was a retrospective review of pathology reports. Although a pathologist evaluated the specimen at surgery and generated a routine report, no additional pathologic review of the specimens was obtained in the context of postablation therapy. This could be considered in the future to further evaluate specific factors such as superficial vs deep adenomyosis as it relates to endometrial regrowth and GEA failure.
99
In conclusion, pathologic findings of hysterectomy specimens after GEA were reviewed. Further delineation of these factors may facilitate determination of consistent pathologic predictors of GEA failure. Additional histopathologic information will benefit preoperative assessment in these patients, thus improving GEA failure rates and patient satisfaction. Postablation monitoring of the endometrium is a central area of research focus in the future. References 1. Overton C, Hargreaves J, Maresh M. A national survey of the complications of endometrial destruction for menstrual disorders: the MISTLETOE study. Br J Obstet Gynaecol. 1997;104:1351–1359. 2. Lethaby A, Hickey M, Garry R. Endometrial destruction techniques for heavy menstrual bleeding. Cochrane Database Syst Rev. 2005. CD001501. 3. Abbott J, Hawe J, Hunter D, Garry R. A double-blind randomized trial comparing the Cavaterm and the NovaSure endometrial ablation systems for the treatment of dysfunctional uterine bleeding. Fertil Steril. 2003;80:203–208. 4. Dickersin K, Munro MG, Clark M, et al., Surgical Treatments Outcomes Project for Dysfunctional Uterine Bleeding (STOP-DUB) Research Group. Hysterectomy compared with endometrial ablation for dysfunctional uterine bleeding: a randomized controlled trial [published correction appears in Obstet Gynecol. 2008;112(2 pt 1):381]. Obstet Gynecol. 2007;110:1279–1289. 5. Dutton C, Ackerson L, Phelps-Sandall B. Outcomes after rollerball endometrial ablation for menorrhagia. Obstet Gynecol. 2001;98:35–39. 6. Bongers MY, Mol BW, Brolmann HA. Prognostic factors for the success of thermal balloon ablation in the treatment of menorrhagia. Obstet Gynecol. 2002;99:1060–1066. 7. El-Nashar SA, Hopkins MR, Creedon DJ, et al. Prediction of treatment outcomes after global endometrial ablation. Obstet Gynecol. 2009;113: 97–106. 8. Coad JE. Endometrial ablation for dysfunctional uterine bleeding: histologic insights into treatment failures [abstract]. J Minim Invasive Gynecol. 2007;14:S25. 9. Colgan TJ, Shah R, Leyland N. Post-hysteroscopic ablation reaction: a histopathologic study of the effects of electrosurgical ablation. Int J Gynecol Pathol. 1999;18:325–331. 10. McCausland AM, McCausland VM. Long-term complications of endometrial ablation: cause, diagnosis, treatment, and prevention. J Minim Invasive Gynecol. 2007;14:399–406. 11. Ziegler WF, Sites C, Badger G, Shamonki M. Prediction of endometrial ablation success by preoperative findings. Primary Care Update Ob/ Gyns. 1998;5:204. 12. Quemere MP, Cravello L, Roger V, d’Ercole C, Blanc B. Impact of adenomyosis on results of endometrial ablations [in French]. Contracept Fertil Sex. 1999;27:357–363. 13. Tresserra F, Grases P, Ubeda A, Pascual MA, Grases PJ, Labastida R. Morphological changes in hysterectomies after endometrial ablation. Hum Reprod. 1999;14:1473–1477. 14. Gemer O, Kruchkovich J, Huerta J, Kapustian V, Kroll D, Anteby E. Perioperative predictors of successful hysteroscopic endometrial ablation [published online ahead of print December 7, 2006]. Gynecol Obstet Invest. 2007;63:205–208. 15. Loffer FD. Improving results of hysteroscopic submucosal myomectomy for menorrhagia by concomitant endometrial ablation. J Minim Invasive Gynecol. 2005;12:254–260. 16. Cooper JM, Brady RM. Intraoperative and early postoperative complications of operative hysteroscopy. Obstet Gynecol Clin North Am. 2000;27:347–366. 17. Townsend DE, McCausland V, McCausland A, Fields G, Kauffman K. Postablation tubal sterilization syndrome. Obstet Gynecol. 1993;82:422–424.
Pathologic Characteristics of Hysterectomy Specimens in Women Undergoing Hysterectomy after Global Endometrial Ablation Erin T. Carey, MD, Sherif A. El-Nashar, MBBCh, MS, Matthew R. Hopkins, MD, Douglas J. Creedon, MD, PhD, William A. Cliby, MD, and Abimbola O. Famuyide, MD* From the Department of Obstetrics and Gynecology, University of North Carolina (Dr. Carey), and the Department of Obstetrics and Gynecology, Mayo Clinic, Rochester, Minnesota (Drs. El-Nashar, Hopkins, Creedon, Cliby, and Famuyide).
ABSTRACT Study Objective: To describe uterine pathologic features in women who underwent hysterectomy because of failed global endometrial ablation (GEA). Design: Retrospective cohort study from 1998 through 2005 (Canadian Task Force classification III). Setting: Tertiary referral center. Patients: Sixty-nine women who underwent hysterectomy because of GEA failure. Interventions: Pathology reports were available for 67 patients. Descriptions of hysterectomy specimens after GEA were reviewed. Measurements and Main Results: Rates of pathologic findings in hysterectomy specimens after failed GEA were determined. Reasons for hysterectomy in the 67 patients with available pathology reports were bleeding in 34 (51%), pain in 19 (28%), and bleeding and pain in 14 (21%). The pathology reports of these specimens showed leiomyomas in 33 specimens (49%); intramural myomas were present in 15 women (44%) who underwent hysterectomy because of bleeding and 8 women (42%) who underwent hysterectomy because of pain. Hematometra was identified in 7 pathologic specimens (10%). Specifically, hematometra was identified in specimens from 5 of 19 women who underwent hysterectomy because of pain (26%). Conclusion: Hematometra was a significant finding in women who underwent hysterectomy because of persistent pain after GEA. A possible pathologic predictor of GEA failure may be intramural leiomyomas. Journal of Minimally Invasive Gynecology (2011) 18, 96–99 Ó 2011 AAGL. All rights reserved. Keywords:
Global endometrial ablation; Hysterectomy; Menorrhagia
Endometrial ablation was initially developed as an alternative to hysterectomy in women with excessive menses in whom medical management failed. The first ablation procedures were hysteroscopy-dependent and time-consuming, and success was directly proportionate to the provider’s surgical skill [1]. The US Food and Drug Administration approves several methods of global endometrial ablation (GEA) that have comparable clinical efficacy and are largely independent of hysteroscopic skill. These methods include thermal balloon ablation, circulated hot fluid ablation, The authors have no commercial, proprietary, or financial interest in the products or companies described in this article. Presented as an oral plenary presentation at the 39th Annual Meeting of the AAGL, Las Vegas, NV, November 8–12, 2010. Corresponding author: Abimbola O. Famuyide, MD, Department of Obstetrics and Gynecology, Mayo Clinic, 200 First St. SW, Rochester, MN 55905. E-mail: [email protected] Submitted April 27, 2010. Accepted for publication October 7, 2010. Available at www.sciencedirect.com and www.jmig.org 1553-4650/$ - see front matter Ó 2011 AAGL. All rights reserved. doi:10.1016/j.jmig.2010.10.004
bipolar radiofrequency ablation, cryotherapy, and microwave endometrial ablation [2]. Global endometrial ablation is now a well-established and effective treatment for refractory menorrhagia [2]. Despite the efficacy of GEA, as many as 30% of women will require hysterectomy after treatment [3,4]. Demographic and clinical factors predictive of GEA failure include younger age (,45 years), history of tubal ligation, preoperative dysmenorrhea, and presence of submucosal myomas [5–7]. However, few studies have characterized the uterine pathologic findings in women in whom GEA failed. For example, persistent bleeding after failed GEA has been associated with thermal necrosis or regenerating endometrium at histopathologic examination after hysterectomy [8]. Considerable uterine scarring also has been noted after ablative therapies, but its clinical significance in women in whom ablation failed is unclear [9,10]. Some evidence indicates that endometrial regrowth after GEA can lead to obstructive pain syndromes such as
Carey et al.
Hysterectomy after Endometrial Ablation
97
hematometra and postablation tubal sterilization syndrome, resulting in GEA failure secondary to pain [10]. In an effort to determine common pathologic findings and complications after GEA, a descriptive pathologic review was conducted of specimens from hysterectomies performed after failed GEA. Materials and Methods This retrospective population-based cohort study was approved by the Mayo Clinic Institutional Review Board. Only patients with a consent form on file allowing use of their medical records were included. Women who underwent GEA from January 1998 through December 2005 at Mayo Clinic (Rochester, MN), were identified from our patient database. From these patients, the subgroup of women who underwent hysterectomy after GEA was identified. The indication for hysterectomy was determined from clinical notes and operative reports; the pathology reports from the hysterectomy specimens were reviewed in detail by 2 independent practitioners (E.T.C. and S.A.E.-N.). Pathologic findings in women who underwent hysterectomy after GEA failure were evaluated. Patients were grouped according to the indication for hysterectomy: persistent bleeding, persistent pain, or persistent bleeding and pain. Within each of these categories, the relative percentages of various pathologic diagnoses were determined. Differences in pathologic findings between groups were analyzed using the independent t test, Wilcoxon rank sum test, c2 test, and Fisher exact test by using commercially available statistical software (JMP version 7.0.0; SAS Institute, Inc., Cary, NC). Survival analysis using a life table was used to calculate the cumulative hysterectomy rate (Table 1).
Fig. 1. Flow of patients through the study. GEA 5 global endometrial ablation.
for 1 patient. Although this group was not used as a comparison group, the pathologic findings in these specimens were included in the present descriptive study. In the other 69 patients, hysterectomy was performed because of GEA failure, defined as hysterectomy because of persistent bleeding, pain, or both, after GEA (rate of hysterectomy due to GEA failure, 9.7% [69 of 711]). Two hysterectomies were performed at outside institutions, and the operative reports were not available for review; thus, these patients were excluded. The remaining 67 patients (GEA Table 2
Pathologic characteristics of hysterectomy specimens according to outcome of GEAa Characteristic
Results During the 8-year study, 711 women (mean [SD] age, 43.3 [5.6] years) underwent GEA, either bipolar radiofrequency ablation or thermal balloon ablation, to treat menorrhagia. Of these, 77 women (10.8%) underwent hysterectomy after GEA (Fig. 1), at age 42.1 (5.9) years. In 8 patients, hysterectomy was not performed because of GEA failure: 2 patients had complex hyperplasia with atypia, 1 had cervical intraepithelial neoplasia, 2 had symptomatic myomas, 1 had pelvic organ prolapse, 1 had a complex adnexal mass, and no pathologic evaluation was available Table 1
Cumulative hysterectomy rate after endometrial ablation in the study duration Time to event (year)
At risk
Failure
1 2 3 4 5 6 7
629 596 560 559 522 404 263
4% 6% 8% 8% 10% 11% 12.5%
Confidence interval 1 2.5% 4% 6% 7% 8% 9% 10.%
Confidence interval 2 5% 7% 11% 12% 13% 15% 17.5%
Uterine length, cm Uterine weight, g Endometrial finding Proliferative Secretory Atrophic Ablative necrosis Other Adenomyosis Endometriosis Endosalpingiosis Hematosalpinx/hydrosalpinx Hematrometra Leiomyoma Submucosal Intramural Subserosal Intramural and submucosal Intramural and subserosal Submucosal, intramural, and subserosal Largest leiomyoma, mm Myoma size, mm R10 R50
GEA failure (n 5 67)
Non-GEA failure (n 5 7)
9.3 (1.2) 151.7 (92.5)
9.6 (1.1) 213.0 (109.5)
32 (48) 18 (27) 8 (12) 8 (12) 1 (1) 19 (28) 11 (16) 4 (6) 5 (7) 7 (10) 33 (49) 1 (3) 19 (58) 4 (12) 5 (15) 2 (6) 2 (6)
4 (57) 0 1 (145) 2 (29) 0 2 (29) 2 (29) 0 0 0 2 (29) 0 1 (50) 0 0 1 (50) 0
18.9 (13.1)
27.5 (31.8)
21 (64) 1 (3)
1 (50) 1 (50)
GEA 5 global endometrial ablation. a Values are given as mean (SD) or No. (%).
p Value .58 .23 .59
..99 .60 ..99 ..99 ..99 .44
.94 ..99 .12
98
Journal of Minimally Invasive Gynecology, Vol 18, No 1, January/February 2011
failure group) were categorized into 3 subgroups on the basis of the clinical reason for hysterectomy: persistent bleeding (n 5 34), persistent pain (n 5 19), or persistent bleeding and pain (n 5 14) (Fig. 1). In the GEA failure group, the pathology reports identified myomas in 33 patients (49%), adenomyosis in 19 (28%), ablative necrosis in 8 (12%), and hematometra in 7 (10%) (Table 2). Subgroups of the GEA failure group were compared according to indication for hysterectomy (bleeding vs pain) (Table 3). In the 34 women who underwent hysterectomy only because of bleeding, 18 (53%) had myomas, which were intramural in 16 (89%). These percentages were similar to those in the subgroup who underwent hysterectomy because of pain: 10 of 19 (53%) had myomas, which were intramural in 9 (90%). In contrast, in women in whom the indication for hysterectomy was pain and bleeding, myomas were present in only 5 of 14 specimens (36%), 3 of which were intramural. Endometrial findings were described as active (proliferative or secretory), atrophic, or ablative necrosis. A total of 9 patient had ablative necrosis; 5 out of 22 (18.5%) in patients who failed before 1 year compared to 4 out of 55 (7.3%) in patients who failed after 1 year (Chi square-Exact Fisher test, P 5 0.147). Ablative necrosis was identified in 2 specimens (6%) in the bleeding-only group and 4 specimens (21%) in the pain-only group (Table 3); ablative necrosis was not identified in the bleeding and pain group. Other endometrial findings included regrowth of active endometrium. Active endometrium was identified in 50 patients in the entire GEA failure group (75%) and in 28 patients (82%) in the bleeding-only group. Only 12 of 50 patients (24%) with endometrial regeneration had adenomyosis. Table 3
Pathologic characteristics of hysterectomy specimens according to indication for surgerya Indication for surgery Characteristic
Bleeding (n 5 34)
Pain (n 5 19)
Uterine weight, g Endometrial finding Proliferative Secretory Atrophic Ablative necrosis Adenomyosis Endometriosis Endosalpingiosis Hematrometra Leiomyoma Submucosal Intramural Subserosal Intramural and submucosal Intramural and subserosal Submucosal, intramural, and subserosal
145 (65.6)
173.2 (139.6)
17 (50) 11 (32) 4 (12) 2 (6) 10 (29) 6 (18) 4 (12) 1 (3) 18 (53) 0 11 (61) 2 (11) 3 (17)
7 (37) 3 (16) 4 (21) 4 (21) 6 (32) 2 (11) 0 5 (26) 10 (53) 0 6 (60) 1 (10) 1 (10)
1 (6)
1 (10)
1 (6)
1 (10)
a
Values are given as mean (SD) or No. (%).
p Value
.87 .49 .12 .03 .98
Fig. 2. Pelvic sonogram in a patient who underwent hysterectomy because of pain shows anechoic fluid in the left cornu suggestive of hematometra (arrowheads). TRANS UTE FUND 5 transverse view of the uterine fundus region.
Hematometra was identified in 7 patients (10%) in the GEA failure group. Only 4 of the 7 patients underwent imaging preoperatively (ultrasonography in 3 and computed tomography in 1); however, hematometra was identified in all 4 patients (Fig. 2). Of the 7 patients with hematometra, 5 (71%) were in the pain-only group, 1 was in the bleedingonly group, and 1 was in the bleeding and pain group. Six of the 7 women (85%) with hematometra had undergone previous bilateral tubal ligation, and GEA was performed using thermal balloon ablation in 4 and bipolar radiofrequency ablation in 3. Of the 5 patients with hematometra in the pain-only group, only 1 had a diagnosis of adenomyosis at hysterectomy. Discussion In the present study, histopathologic findings of uterine specimens after GEA failure were described. The literature is limited insofar as descriptions of uterine pathology in women in whom GEA fails. Factors that have been suggested as causes of GEA failure include intramural leiomyomas [11], necrotic myometrium, inflammatory response, and substantial scarring [9]. In a retrospective pathologic analysis of 20 uteri removed because of persistent bleeding after ablation, the 2 most common findings were ablative necrosis (a known endometrial response to thermal damage) and regenerating endometrium, which suggested that these factors are causes of persistent bleeding [8]. The ablative necrosis was identified within three months of the endometrial ablation. In the present study, the majority of the hysterectomy specimens with ablative necrosis were performed within the first year following the ablation procedure, however this was not statistically significant. Regenerating endometrium was also increased in the group of women with GEA failure due to persistent bleeding. In the present study, regenerating endometrium was increased in the group of women with GEA failure due to persistent bleeding. Deep adenomyosis also has been described as a risk factor for failure of GEA [12]; however, the complete role of adenomyosis in GEA failure or how it influences GEA
Carey et al.
Hysterectomy after Endometrial Ablation
outcomes is unknown. Adenomyosis has also been associated with endometrial regeneration, which has been previously associated with failure of endometrial ablation secondary to postablation bleeding [13]. These findings were not supported in the present study. A high rate of endometrial regrowth was noted in women in whom GEA failed; however, the rates of endometrial regeneration and adenomyosis were not significantly different between patient groups. Other reports have related postablation pain to the presence of adenomyosis [10]; however, adenomyosis was present only in a minority of patients who underwent hysterectomy because of pain alone. The rate of adenomyosis was similar in all groups including those with hysterectomy unrelated to GEA failure. This information suggests that factors other than adenomyosis may contribute to regrowth of the endometrium and that the presence of adenomyosis cannot reliably predict GEA failure. Submucosal leiomyomas are another known risk factor for GEA failure [14]. In the present study, most of the myomas identified in the specimens were intramural. The role of intramural myomas in GEA outcome is unknown. Loffer [15] suggested that the presence of submucosal and intramural myomas does not change the efficacy of ablation for control of bleeding; others believe that intramural myomas increase the failure rate of GEA, although no studies have shown this with statistical significance [11]. In the present study, the percentage of intramural myomas was higher in the GEA failure group; however, this result was not statistically significant. Therefore, further studies are warranted. Additional research is needed to specifically evaluate intramural myomas in the context of endometrial ablative therapy and their function in GEA failure. The strongest relationship found in the present study was between pain and hematometra. The cause of hematometra after ablation is thought to be from preservation or regeneration of the endometrium in the setting of substantial scarring, which may block the outlet and result in obstructive symptoms [10]. Women with hematometra often have cyclic pelvic pain or chronic pelvic discomfort [16]. Those who have undergone tubal ligation are at increased risk of GEA failure as a result of pain, known as postablation tubal sterilization syndrome [17]. The pain is thought to be caused by distention of the proximal end of the fallopian tube during endometrial growth with no outlet into the peritoneum. Pain secondary to hematometra seems to be an important cause of failure; patient selection and appropriate counseling should be initiated before ablation therapy. A limitation of the present study is that it was a retrospective review of pathology reports. Although a pathologist evaluated the specimen at surgery and generated a routine report, no additional pathologic review of the specimens was obtained in the context of postablation therapy. This could be considered in the future to further evaluate specific factors such as superficial vs deep adenomyosis as it relates to endometrial regrowth and GEA failure.
99
In conclusion, pathologic findings of hysterectomy specimens after GEA were reviewed. Further delineation of these factors may facilitate determination of consistent pathologic predictors of GEA failure. Additional histopathologic information will benefit preoperative assessment in these patients, thus improving GEA failure rates and patient satisfaction. Postablation monitoring of the endometrium is a central area of research focus in the future. References 1. Overton C, Hargreaves J, Maresh M. A national survey of the complications of endometrial destruction for menstrual disorders: the MISTLETOE study. Br J Obstet Gynaecol. 1997;104:1351–1359. 2. Lethaby A, Hickey M, Garry R. Endometrial destruction techniques for heavy menstrual bleeding. Cochrane Database Syst Rev. 2005. CD001501. 3. Abbott J, Hawe J, Hunter D, Garry R. A double-blind randomized trial comparing the Cavaterm and the NovaSure endometrial ablation systems for the treatment of dysfunctional uterine bleeding. Fertil Steril. 2003;80:203–208. 4. Dickersin K, Munro MG, Clark M, et al., Surgical Treatments Outcomes Project for Dysfunctional Uterine Bleeding (STOP-DUB) Research Group. Hysterectomy compared with endometrial ablation for dysfunctional uterine bleeding: a randomized controlled trial [published correction appears in Obstet Gynecol. 2008;112(2 pt 1):381]. Obstet Gynecol. 2007;110:1279–1289. 5. Dutton C, Ackerson L, Phelps-Sandall B. Outcomes after rollerball endometrial ablation for menorrhagia. Obstet Gynecol. 2001;98:35–39. 6. Bongers MY, Mol BW, Brolmann HA. Prognostic factors for the success of thermal balloon ablation in the treatment of menorrhagia. Obstet Gynecol. 2002;99:1060–1066. 7. El-Nashar SA, Hopkins MR, Creedon DJ, et al. Prediction of treatment outcomes after global endometrial ablation. Obstet Gynecol. 2009;113: 97–106. 8. Coad JE. Endometrial ablation for dysfunctional uterine bleeding: histologic insights into treatment failures [abstract]. J Minim Invasive Gynecol. 2007;14:S25. 9. Colgan TJ, Shah R, Leyland N. Post-hysteroscopic ablation reaction: a histopathologic study of the effects of electrosurgical ablation. Int J Gynecol Pathol. 1999;18:325–331. 10. McCausland AM, McCausland VM. Long-term complications of endometrial ablation: cause, diagnosis, treatment, and prevention. J Minim Invasive Gynecol. 2007;14:399–406. 11. Ziegler WF, Sites C, Badger G, Shamonki M. Prediction of endometrial ablation success by preoperative findings. Primary Care Update Ob/ Gyns. 1998;5:204. 12. Quemere MP, Cravello L, Roger V, d’Ercole C, Blanc B. Impact of adenomyosis on results of endometrial ablations [in French]. Contracept Fertil Sex. 1999;27:357–363. 13. Tresserra F, Grases P, Ubeda A, Pascual MA, Grases PJ, Labastida R. Morphological changes in hysterectomies after endometrial ablation. Hum Reprod. 1999;14:1473–1477. 14. Gemer O, Kruchkovich J, Huerta J, Kapustian V, Kroll D, Anteby E. Perioperative predictors of successful hysteroscopic endometrial ablation [published online ahead of print December 7, 2006]. Gynecol Obstet Invest. 2007;63:205–208. 15. Loffer FD. Improving results of hysteroscopic submucosal myomectomy for menorrhagia by concomitant endometrial ablation. J Minim Invasive Gynecol. 2005;12:254–260. 16. Cooper JM, Brady RM. Intraoperative and early postoperative complications of operative hysteroscopy. Obstet Gynecol Clin North Am. 2000;27:347–366. 17. Townsend DE, McCausland V, McCausland A, Fields G, Kauffman K. Postablation tubal sterilization syndrome. Obstet Gynecol. 1993;82:422–424.