Clinical Predictive Factors of Failure of Hysteroscopic Endometrial Ablation: Retrospective Cohort Study at a Tertiary University Hospital

Background: Endometrial ablation is a safe and effective minimally invasive surgical procedure. Despite the high success rate of endometrial ablation for heavy uterine bleeding management, some patients experience persistent symptoms after the procedure, ne- cessitating a hysterectomy. The aim of this study is to determine the pre-operative clinical predictive factors of failure of endometrial ablation in the management of uterine bleeding. Methods: Retrospective cohort study of endometrial ablation procedures performed for treating heavy uterine bleeding. Results: Ninety five patients were included in the study. The failure rate was 24.2%. There was a statistically significant association between ablation failure and fluid deficit ( p = 0.002) and intra-operative blood loss ( p = 0.047). There was a statistically significant moderate association between adenomyosis (p = 0.003, φ = 0.37) and failed endometrial ablation. However, the age, body mass index (BMI), parity, number of miscarriages, number of cesarean sections, uterine length, endometrial thickness and procedure duration had no significant association with endometrial ablation. There was no significant relationship between the uterine abnormalities in general and the outcome of the endometrial ablation ( p = 0.637). However, patients with combined adenomyosis and dysmenorrhea had a statistically significant association with outcome of the endometrial ablation ( p = 0.016, φ = 0.28) and were more likely to have a failed endometrial ablation (crude odds ratio (COR) = 4.67, 95% confidence interval (CI), 1.35–16.09). Logistic regres- sion to adjust for related factors revealed that the adenomyosis (adjusted odds ratio (AOR) = 50.83, 95% CI, 3.64–706.75, p = 0.003) and fluid deficit (AOR = 1.003, 95% CI, 1.000–1.006, p = 0.044) had a higher likelihood of an unsuccessful outcome. Hysterectomy was performed in 47.8% of patients who had failure of the ablation. Conclusions: Among pre-operative factors, adenomyosis, fluid deficit and combined adenomyosis and dysmenorrhea were found to predict failure of hysteroscopic endometrial ablation. Clinical Trial Registration: It was registered in https://www.clinicaltrials.gov database with Identifier: NCT05483348.

endometrial; ablation; hysteroscopic; rollerball; uterine; adenomyosis 1. Introduction Endometrial ablation (EA) is a safe and efficacious endoscopic surgical procedure that has become a well- established alternative option in selected women with heavy and abnormal uterine bleeding [1]. In the United Kingdom, hysterectomy rates were significantly reduced over the past 20 years due to both improved medical treatment and in- creased use of EA techniques [ 2]. EA improves treatment access for those women who have abnormal uterine bleed- ing (AUB) and provides an alternative to radical procedures such as hysterectomy [ 3]. Hysterectomy was found to be associated with a higher risk than endometrial ablation of surgery for pelvic floor repair and stress urinary inconti- nence [3]. Despite the high success rate of endometrial ablation for uterine bleeding management, some patients experi- ence persistent symptoms after the procedure, necessitat- ing a hysterectomy. The characteristics of patients whose endometrial ablation fails had been studied by Kristin A Ri- ley in a cohort study of 51 patients who had multiple vari- ants assessed, including age, body mass index (BMI), parity and tubal ligation. The only statistically significant vari- ant was age; it also revealed that endometriosis was the most common finding among patients who presented for hysterectomy after endometrial ablation; other findings in- cluded adenomyosis and leiomyomata [ 4]. Another retro- spective cohort study of 968 women who had endometrial ablation between January 2007 and July 2009 compared the risk of treatment failure between women who had regular and irregular heavy uterine bleeding, concluding that pre- operative bleeding pattern has no effect on failure rates or the need for gynecological procedures after endometrial ab- lation [5]. A study that included 128 women who had en- dometrial ablation or resection was conducted to determine the perioperative factors that influenced the success of this hysteroscopic intervention It was found that the presence of submucosal myoma was found to increase the risk of sub- sequent surgery [6]. The first-generation hysteroscopic techniques namely, roller ball ablation, bipolar endometrial resection and laser ablation, all required visualization of the cavity for the en- tirety of the procedure [ 7]. Second generation techniques have been developed which do not require hysteroscopy and can be easily performed without requiring high lev- els of skill include bipolar radiofrequency, microwave and thermal balloon ablation, endometrial cryotherapy, and hy- drothermal ablation [8]. The success of endometrial ablation was seen in 80.6% of cases [ 9]. On the other hand, failure rates of en- dometrial ablation, which were manifested as persistence of pain or bleeding, have reached 10–20% that required pa- tients to undergo further interventions to handle their mal- ady [10]. The objective of our study was to determine additional patient characteristics and the clinical predictive factors of failure of endometrial ablation and the need of subsequent gynecological procedure. The primary outcomes were to determine the pre- operative clinical predictive factors of failure of endome- trial ablation in the management of uterine bleeding. Sec- ondary outcomes were to search for any intra-operative pre- dictive factors of failure of the procedure. Better under- standing of the predictive factors of success or failure will help in proper counselling of patients and in decreasing the incidence of endometrial ablation failure and avoid further re-ablation or hysterectomy. 2. Methods 2.1 Study Design A retrospective cohort study was conducted in a single tertiary center, Jordan University Hospital (JUH). Such a design enabled us to compare two populations of similar baseline characteristics for the exception of the one studied. 2.2 Population Patients with heavy menstrual bleeding for 6 months who had trans-cervical hysteroscopic endometrial ablation in the period September 2016 to September 2021 and fol- lowed up for at least one year. Heavy bleeding was sub- jectively indicated by the patient as having prolonged cy- cles, using excessive pads, staining her underwear, affect- ing her usual activities, taking supplements to correct her low hemoglobin and the presence of clots. This bleeding pattern should be experienced by the patient for at least 3 months. Patients with hemoglobin of less than 10.5 g/dL were considered anemic. 2.3 Data Collection Clinical data included age, BMI, parity, tubal liga- tion, uterine size/length (in weeks as determined by biman- ual examination and uterine sound), endometrial thickness (in millimeters as measured by transvaginal ultrasound scan (U/S)), procedure duration, fluid deficits, presence of uter- ine abnormality as endometriosis, adenomyosis, leiomy- omata and polyps or history of dysmenorrhea, menorrha- gia, number of cesarean sections, history of uterine surgery; myomectomy, dilatation and curettage (D&C). Patients who had resection or subsequently had a diag- nosis of endometrial cancer or hyperplasia (both simple and those with atypia), those with intramural fibroids, patients with submucosal fibroids larger than 5 cm, patients with previous ablation, patients who had ablation using bipolar diathermy and those with incomplete or missing records or follow up were excluded from the study’s target population. 2.4 Ethical Considerations Patients’ data was treated with the utmost of confi- dentiality. Moreover, all records were accessed through JUH’s computers and under JUH’s medical staff supervi- sion ensuring that no data was copied or transferred to pri- vate storage sources. Any and all data that contributes to the recognition of patients’ identities, such as names, phone numbers or addresses were either turned into codes or not be used at all. The study obtained approval of the insti- tutional review board (IRB) committee at JUH (decision number 58/2022, dated 17/02/2022). It was also registered in https://www.clinicaltrials.gov database with Identifier: NCT05483348. All women had general and gynecologic examination preoperatively and either had diagnostic hysteroscopy or trans-vaginal ultrasound (U/S) scan done by consultant gy- necologists to accurately measure the uterine size, endome- trial thickness and the size of the fibroids or polyps, or both before the hysteroscopic ablation. A written informed consent was obtained from all pa- tients included in this study. The consent included full explanation of the procedure, benefits, potential complica- tions and risks. 2.5 Surgical Technique Hysteroscopic ablation was performed in the post- menstrual period when the endometrium was very thin. In our practice, we not use pre-operative endometrial thin- ning agents. The operative hysteroscopy in our study was done using rigid, with 30 degrees angulated continuous flow hysteroscopy utilizing a monopolar rollerball. This necessitated the use of glycine 1.5% as a uterine distending medium in 3.0 Liter-bags. The monopolar resectoscope was sterilizable and reusable while the bipolar one was a single- use instrument, thus minimizing financial costs. The flow rate used was at 250 mL/minute. Myomectomy cases were performed using angled wire cutting loop electrodes. All patients had standardized 80 millimeter mercury (mmHg) of intra-uterine pressure during surgery with the use of the same automated glycine infusion pump. Patients had cervi- cal dilatation done to 10 mm using Hegar dilators unless the cervix was already found to be dilated. Patients with uterine polyps or fibroids (submucosal (SM) type 0 and 1 only) [11] 2 Table 1. Comparison of failed and successful endometrial ablations in relation to different factors. V ariables Failure Success p value Effect size Mean Standard deviation Mean Standard deviation Age 45.39 8.45 46.36 6.68 0.431 - BMI 30.75 3.47 29.8 6.71 0.413 - Parity 4.1 2.49 4.43 2.27 0.283 - Miscarriage 0.81 0.87 1.23 1.6 0.51 - N of cesarean sections 2 1.66 1.31 1.55 0.135 - Uterine length (weeks) 9.57 2.48 8.59 1.91 0.057 - Endometrial thickness (mm) 10.77 5.21 11.02 7.04 0.935 - Procedure duration (min) 36.3 14.94 35.42 13.63 0.893 - Fluid deficit (cc) 575 298.33 335.3 203.99 0.002 0.34 Blood loss (cc) 65.22 49.35 49.93 46.09 0.047 0.21 Notes: p values for the Parity and uterine length were calculated by independent sample t-test. All other items were subjected to Mann-Whitney Utest or Mann–Whitney–Wilcoxon (MWW/MWU) test. BMI, body mass index; mm, millimeter; min, minute; cc, cubic centimeter; N, number. had their lesions first resected using a resectoscope loop fol- lowed by rollerball endometrial ablation (REA). Before the REA, all patients had curettage of the whole endometrial cavity to obtain tissue for histopathological examination and to thin the endometrial to enhance the thermal effect of the ablation. There were selected cases that had bipo- lar ablation and they were excluded from the study to have a homogenous study population. All procedures were per- formed by the same consultant gynecologist. Data was ex- tracted from patients’ clinical notes in the clinic, operative notes, admission and follow up visits’ notes. Data was extracted by using relevant International Classification of Diseases—Ninth revision (ICD-9) codes as well as Current Procedural Terminology (CPT) codes [12]. Patients included in this study experienced heavy uter- ine bleeding and underwent endometrial ablation. Success- ful cases were defined as patients’ satisfaction with the pro- cedure as indicated by amenorrhea or very light menstrual bleeding for 12 months at least. Failure was defined as patients’ un-satisfaction necessitating subsequent gyneco- logical procedures such as hysterectomy or repeat ablation within 12 months after the endometrial ablation. We did not assess the effect of the ablation on the post-operative hemoglobin level. We then determined those who were unsatisfied (recurrent heavy bleeding, dysmenorrhea) and those who ended up with repeated ablation, hormonal treat- ment and those who had hysterectomy after the ablation. 2.6 Statistical Analysis Data analysis was performed using the IBM SPSS, version 28.0 (IBM Corp., Armonk, NY , USA) developed by Norman H. Nie, Dale H. Bent and C. Hadlai Hull. Cate- gorical variables were summarized as frequencies and per- centages. Participants were categorized into two groups ac- cording to their failure or success of the primary endome- trial ablation. Pearson’s Chi-square ( χ2) test and Fischer’s exact test with crude odds ratios (COR) (95% confidence in- terval (CI)) was used to determine the association between categorical data. The phi coefficient ( φ) was used as a mea- sure of the strength of association. Normality was assessed using the Shapiro-Wilk test. Independent samples t-test and Mann-Whitney U tests were used to determine the association between continuous data. The effect size was calculated depending on the test use by effect size r for Mann-Whitney U. V ariables that showed potentially interesting associations from the univariate anal- ysis were included in the binary logistic regression model in order to control for possible confounding factors, which were summarized using adjusted odds ratios (AOR) (95% CI). Statistical significance was defined as a p-value of less than 0.05. 3. Results In total, there were 138 procedures were done in the study period. Ninety-five patients were included in this study; the age ranged from 29 to 52 years, with a mean age of 46.13 ± 7.11 years. Among the participants, the mean BMI was 29.97 ± 6.23, ranged from 18 to 56. Participants were categorized into two groups according to their fail- ure or success of the primary endometrial ablation. Overall, 24.2% (n = 23) had a failed endometrial ablation with a sub- sequent gynecological procedure and 75.8% (n = 72) had a successful endometrial ablation with no further procedures required. Tables 1,2 show a comparison between patients with successful and those with failed endometrial ablation groups. Our results showed a significant difference between patients with successful and failed endometrial ablations in regard to fluid deficit ( p = 0.002) and blood loss during the surgery ( p = 0.047); however, the age, BMI, parity, num- ber of miscarriages, number of cesarean sections, uterine length, endometrial thickness and procedure duration had no significant difference between the two groups (p > 0.05). 3 Table 2. A comparison between the clinical characteristics of failed and successful endometrial ablation. V ariables Total number (%) Endometrial ablation OR (95% CI) p value Failure Success Smoker (yes) 3/78 (3.8) 2/18 (11.1) 1/60 (1.7) 7.38 (0.63–86.60) 0.131 History of coagulation disorders (yes) 1/92 (1.1) 0/22 (0) 1/70 (1.4) 0.99 (0.96–1.01) 1 Use of anticoagulants (yes) 1/92 (1.1) 0/22 (0) 1/70 (1.4) 0.99 (0.96–1.01) 1 Previous tubal ligation (yes) 9/95 (9.5) 2/23 (8.7) 7/72 (9.7) 0.88 (0.17–4.59) 1 Uterine abnormalities 1.55 (0.26–9.16) 0.637 Polyps 43/84 (51.2) 7/21 (33.3) 36/63 (57.1) Fibroids 42/84 (50) 14/21 (66.7) 28/63 (44.4) Adenomyosis 6/84 (7.1) 5/21 (23.8) 1/63 (1.6) No abnormalities 6/84 (7.1) 2/21 (9.5) 4/63 (6.3) Endometriosis 3/84 (3.6) 0/21 (0) 3/63 (4.8) Thickened endometrium 3/84 (3.6) 0/21 (0) 3/63 (4.8) Leiomyomata 1/84 (1.2) 1/21 (4.8) 0/63 (0) Menstrual abnormalities 0.38 (0.08–1.83) 0.349 Menorrhagia 79/89 (88.8) 18/21 (85.7) 61/68 (89.7) Dysmenorrhea 7/89 (7.9) 2/21 (9.5) 5/68 (7.4) Intermenstrual spotting 1/89 (1.1) 0/21 (0) 1/68 (1.5) History of uterine surgery 0.66 (0.23–1.90) 0.440 Dilation and curettage 51/91 (56) 15/22 (68.2) 36/69 (52.2) Evacuations 7/91 (7.7) 0/22 (0) 7/69 (10.1) Myomectomy 4/91 (4.4) 0/22 (0) 4/69 (5.8) Other 7/91 (7.7) 0/22 (0) 7/69 (10.1) No 31/91 (34.1) 6/22 (27.3) 25/69 (36.2) Indication for primary endometrial ablation Menorrhagia 36/95 (37.9) 12/23 (52.2) 24/72 (33.3) 2.18 (0.84–5.66) 0.139 Heavy menstrual bleeding 31/95 (32.6) 4/23 (17.4) 27/72 (37.5) 0.35 (0.11–1.14) 0.081 Fibroid 17/95 (17.9) 8/23 (34.8) 9/72 (12.5) 3.73 (1.24–11.29) 0.026 Irregular vaginal bleeding 14/95 (14.7) 2/23 (8.7) 12/72 (16.7) 0.48 (0.10–2.30) 0.506 Polyps 10/95 (10.5) 0/23 (0) 10/72 (13.9) – 0.112 Postmenopausal bleeding 9/95 (9.5) 2/23 (8.7) 7/72 (9.7) 0.88 (0.17–4.59) 1 Infertility 3/95 (3.2) 1/23 (4.3) 2/72 (2.8) 1.59 (0.14–18.40) 0.569 There might be more than one indication. OR, odds ratio; CI, confidence interval. Table 1 shows the details of the successful and failed en- dometrial ablations. There was no significant relationship between the uterine abnormalities in general and the outcome of the en- dometrial ablation ( p = 0.637); however, there was a sta- tistically significant, moderate association between adeno- myosis and the outcome of the endometrial ablation ( p = 0.003, φ = 0.37) and were more likely to have a failed endometrial ablation COR = 19.38, 95% CI, 2.11–177.73) (Table 2). In addition, patients with combined adenomyosis and dysmenorrhea had a statistically significant association with the outcome of the endometrial ablation ( p = 0.016, φ = 0.28) and were more likely to have a failed endometrial ablation (COR = 4.67, 95% CI, 1.35–16.09). Menorrhagia and heavy irregular menstrual bleeding were among the most common indications for endometrial ablation in 37.9% and 32.6% of the patients, respectively. Moreover, the patients who underwent endometrial ablation due to fibroids (17.9%) had a significant weak relationship with the outcome of the endometrial ablation ( p = 0.26, φ = 0.25), they were more likely to have a failed endometrial ablation (COR = 3.73, 95% CI, 1.24–11.29). Other indi- cations included irregular vaginal bleeding, polyps, post- menopausal bleeding, and infertility and they no longer sought pregnancy and would only treat their heavy periods. We also found that patients who had both adenomyosis and dysmenorrhea were at higher statistically significant risk to have ablation failure than those with adenomyosis alone, p value = 0.016, 95% CI, 1.51–50.69 and p value= 0.003, φ = 0.37, respectively. Overall, 24.2% had a failed endometrial ablation and had to undergo an additional procedure to alleviate the symptoms. The most common were repeated hysteroscopic ablation and hysterectomy, in 15 and 11 patients, respec- tively (Table 3). In order to account for any confounding effects, the variables having significant associations at the p < 0.1 level from the univariate analysis were included in the logistic 4 Table 3. Subsequent gynecological procedures after failed ablation (N = 23). Surgeries N (%) Hysterectomy 11 (47.8) Hysteroscopy and repeated ablation 15 (65.2) Dilation and curettage (endometrial biopsy) 9 (39.1) Other 5 (21.7) Some patients underwent more than one procedure. N, number. regression model. These were the uterine length in weeks, fluid deficit, blood loss, diagnosis of adenomyosis, heavy menstrual bleeding and fibroid(s). The significant pre- dictors for endometrial ablation failure were adenomyosis (AOR = 50.83, 95% CI, 3.64–706.75, p = 0.003) and fluid deficit (AOR = 1.003, 95% CI, 1.000–1.006, p = 0.044). The uterine length in weeks, blood loss, indication for pri- mary endometrial ablation being heavy menstrual bleed- ing, or presence of fibroid were not significant predictors in the regression model, which had a Pseudo Nagelkerke R Squared of 0.49. Table 4 demonstrates the regression model

for an endometrial ablation failure prediction. Table 4. The results of regression model for endometrial ablation failure. V ariables AOR 95% CI p value Uterine length (weeks) 1.069 0.704–1.622 0.755 Fluid deficit (mL/cc) 1.003 1.000–1.006 0.044 Blood loss (mL/cc) 1.004 0.991–1.018 0.505 Adenomyosis 50.825 3.655–706.745 0.003 Heavy menstrual bleeding 0.295 0.043–2.015 0.213 Fibroid 3.224 0.585–17.754 0.179 AOR, adjusted odds ratios; CI, confidence interval; mL, milliliter; cc, cubic centimeter. 4. Discussion The overall success rate of our endometrial ablation procedures using the firstly generation rollerball ablative technique, REA, was very much good, acceptable and very helpful in a third world country with limited resources. When compared with hysterectomy, the treatment of AUB with endometrial ablation has been associated with reduced operative time, decreased morbidity, and cost effectiveness [13,14]. Over half of patients treated with microwave endome- trial ablation (MEA) achieved amenorrhea, and the proce- dure was suitable for women with myomas and irregular uterine cavities and the success rate of MEA at 12 months (87.0%; 95% CI, 81.7%–91.2%) did not differ significantly (p = 0.40) from that of REA (83.2%; 95% CI, 74.7%– 89.7%) [15]. Our study showed an overall failure rate of 24.2%. All patients were operated upon using either general or spinal anesthesia with no reported serious complications like per- foration, bleeding or significant fluid absorption. When second-generation versus first-generation techniques were compared, there was no evidence of differences in amen- orrhea at 1 year and 2 to 5 years’ follow-up and there was subjective improvement at 1 year follow-up based on a Pic- torial Blood Assessment Chart (PBAC) [ 16]. Moreover, patient satisfaction was not different between second- and first-generation techniques at 1 year, 2 years and 5 years’ follow-up. Second-generation ablation techniques were as- sociated with shorter operating times and more often were performed under local instead of general anaesthesia [ 16]. There was uncertainty whether perforation rates differed between second- and first-generation techniques [ 16]. Tri- als reported little or no difference between second- and first- generation techniques in requirement for additional surgery (ablation or hysterectomy) at 1 and 5 years’ follow-up [ 16]. These findings support our practice in using first genera- tion ablation as newer generations (second and third) are expensive and not readily available in our country. Simi- larly, levonorgestrel intra-uterine system (LNG-IUS) is still expensive in Jordan and not covered by many medical in- surance companies. In a meta-analysis in 2020, endome- trial ablation and LNG-IUS were found to be two excellent treatment options for heavy menstrual bleeding, although women treated with the LNG-IUS were at higher risk of experiencing side effects compared to women treated with endometrial ablation/resection [17]. This is very encourag- ing to us to further implement endometrial ablation in our practice taking into consideration the global and, particu- larly, the local economic restrictions. Similarly, in a re- view, when compared to endometrial ablation, it was not clear whether the LNG IUS offered any benefits with re- gard to reduced heavy bleeding while satisfaction rates and quality of life measures were found to be similar. In the same review, some minor adverse effects were more com- mon with the LNG IUS but it appeared to be more cost effective than endometrial ablation techniques [ 18]. Hys- terectomy was found to cause serious complications for a minority of women and most women might be well advised to try a less radical treatment as first-line therapy and it was found that both LNG-IUS and conservative surgery (abla- tion/resection) appeared to be safe, acceptable and effective [19]. Unfortunately, there were no studies in Jordan that compared these different modalities regarding particularly cost-effectiveness and patient satisfaction. There is an on- going a multicenter randomized controlled trial to evaluate if the combination of endometrial ablation and an LNG-IUS is superior to endometrial ablation alone in terms of reduc- ing subsequent rates of hysterectomy at two years following the initial ablative procedure [ 20]. Our study found a significant association between greater operative blood loss and fluid absorption (fluid deficit), as independent factors, with endometrial ablation 5 failure. The presence of adenomyosis and uterine fibroids as indication of endometrial ablation was also found to be significantly associated with the outcome of the en- dometrial ablations. However, the patient’s age, BMI, par- ity, miscarriage, previous cesarean sections, uterine length (in weeks), endometrial thickness (in millimeters mm), dysmenorrhea and procedure duration (minutes) were not found to influence the effectiveness of the procedure. Some but not all of these findings were consistent with previous studies. Previous Cesarean delivery was not associated with an increased risk of failure of endometrial ablation, but dys- menorrhea, a submucous myoma and longer uterine depth are [21]. Their definition of failure was those patients who had ablation and hysterectomy. In our study, we did not specify hysterectomy as the only destination for failure def- inition. Moreover, dysmenorrhea cases in our study were only seven in both failed and successful cases. This low number might have influenced the statistical result. We had a total of 6 cases of adenomyosis who had ab- lation; one was successful while the remaining 5 were a fail- ure. There was no statistically significant difference in the outcome in our study concerning the success or failure of ablation. This might be due to the low number of cases or to the degree of junctional zone changes. In fact, the effect of transcervical endometrial resection might depend upon the degree of junctional zone changes, and patients with intrinsic adenomyosis were more likely to undergo re- intervention surgery than patients with either linear or ser- rated junctional zone [ 22]. However, hysteroscopic roller- ball endometrial ablation as a surgical management in pa- tients with adenomyosis associated with menorrhagia was found to be an effective and safe procedure which could reduce the need for the unnecessary major surgery of hys- terectomy [23]. In a systemic review and meta-analysis [ 10], younger age, prior tubal ligation and preexisting dysmenorrhea were found to be associated with failure of endometrial ablation. Obesity and large submucous fibroids showed conflicting

and there was a need for further research to estimate the influence of these factors. This meta-analysis involved studies that were different from ours. The studies involved in this meta-analysis were using second generation endome- trial ablation [10]. Our study did not show a statistically significant dif- ference in ablation failure or success regarding preoperative menstrual irregularities and intermenstrual spotting. These findings were similar to a retrospective cohort study of a relatively large number of patients (968) [ 5]. As the first- and second- generation ablation tech- niques are effective and safe, we can expand our hystero- scopic endometrial ablation service to decrease the need for hysterectomy and particularly useful for patients with contraindications for medical therapy [ 24]. Both seem to be equally effective in reducing heavy bleeding and there was no evidence that rates of satisfaction differed signif- icantly [ 25]. Despite second-generation techniques were often found to be easier to perform with shorter operative times with the ability to use local rather than general anes- thesia and complications appeared to be less after second- generation techniques, the easiness of use can be a pitfall [25]. In fact, our experience with the first-generation tech- nique showed that in expert hands and with strict adherence to energy settings and fluid management, significant com- plications could almost be eliminated. Endometrial resection and ablation were effective al- ternative to hysterectomy for heavy menstrual bleeding and both were shown to have high satisfaction rates. Hysterec- tomy was associated with longer operating time, longer re- covery period and higher rates of postoperative complica- tions [26]. These features further support the incorporation of endometrial ablation service particularly in developing and under-resourced countries with limited financial, tech- nical and medical capabilities to deal with complications. In our practice, for endometrial ablation, we use the monopolar rollerball rather than the resectoscope. For bipo- lar resectoscope, in a randomized clinical trial, no sig- nificant difference was found in the postoperative PBAC score, amenorrhea rates, patient satisfaction, and need for re-intervention between ball endometrial ablation and trans- cervical resection of the endometrium using resectoscope [27]. This might have further explained the success and low complication rates in our study. We did not assess post-ablation pain in our study as we followed patients for one year. It was reported in 20.8% of cases and the median number of days for the development of pain after ablation was 301 days [ 28]. The strengths of our study lied in the consistency of the study population and of the surgical technique. More- over, the low cost of the procedure was a feature as no pre- operative thinning agents were used and, instead, whole en- dometrial cavity curettage was utilized to obtain tissue for assessment and to mechanically and instantly thin the en- dometrium to increase the effectiveness of the thermal dam- age on the endometrium. Study Limitations The study was limited by its retrospective nature and being a single-center study. The post-operative follow up was another limiting feature. 5. Conclusions Among pre-operative factors, adenomyosis, fluid deficit and combined adenomyosis and dysmenorrhea were found to predict failure of hysteroscopic endometrial abla- tion. Abbreviations BMI, body mass index; EA, endometrial ablation; AUB, abnormal uterine bleeding; JUH, Jordan university hospital; g, gram; mL, milliliter; dL, deciliter; D&C, dilata- 6 tion and curettage; IRB, institutional review board; U/S, ul- trasound; mmHg, millimeter mercury; REA, rollerball en- dometrial ablation; ICD, international classification of dis- eases; CPT, current procedural terminology; COR, crude odds ratio; AOR, adjusted odds ratio; CI, confidence in- terval; mm, millimeter; min, minute; cc, cubic centime- ter; MEA, microwave endometrial ablation; PBAC, picto- rial blood assessment chart; LNG-IUS, levonorgestrel intra- uterine system; SM, submucosal. Availability of Data and Materials All data are available from the corresponding author on reasonable request. Author Contributions NA—design, concept and writing the manuscript; AA—statistical analysis and extraction of data; RN— statistical analysis and extraction of data; LA—statistical analysis and extraction of data; MA—statistical analysis and extraction of data; HA—drafting and revision. All au- thors read and approved the final manuscript. Ethics Approval and Consent to Participate The study was conducted in accordance with the Dec- laration of Helsinki, and the protocol was approved by the Ethics Committee of Jordan University hospital decision number 58/2022, dated 17/02/2022. Acknowledgment Not applicable. Funding This research received no external funding. Conflict of Interest The authors declare no conflict of interest.

[1] Laberge P , Leyland N, Murji A, Fortin C, Martyn P , Vilos G, et al. Endometrial Ablation in the Management of Abnormal Uter- ine Bleeding. Journal of Obstetrics and Gynaecology Canada. 2015; 37: 362–376. [2] Reid PC. Endometrial ablation in England–coming of age? An examination of hospital episode statistics 1989/1990 to 2004/2005. European Journal of Obstetrics and Gynecology and Reproductive Biology. 2007; 135: 191–194. [3] Cooper K, Lee A, Chien P , Raja E, Timmaraju V , Bhattacharya S. Outcomes following hysterectomy or endometrial ablation for heavy menstrual bleeding: retrospective analysis of hospital episode statistics in Scotland. BJOG: An International Journal of Obstetrics & Gynaecology. 2011; 118: 1171–1179. [4] Riley KA, Davies MF, Harkins GJ. Characteristics of Patients Undergoing Hysterectomy for Failed Endometrial Ablation. Journal of the Society of Laparoendoscopic Surgeons. 2013; 17: 503–507. [5] Smithling KR, Savella G, Raker CA, Matteson KA. Preoper- ative uterine bleeding pattern and risk of endometrial ablation failure. American Journal of Obstetrics and Gynecology. 2014; 211: 556.e1–556.e6. [6] Gemer O, Kruchkovich J, Huerta M, Kapustian V , Kroll D, Anteby E. Perioperative predictors of successful hysteroscopic endometrial ablation. Gynecologic and Obstetric Investigation. 2007; 63:205–208. [7] Papadopoulos NP , Magos A. First-generation endometrial abla- tion: roller-ball vs loop vs laser. Best Practice & Research Clin- ical Obstetrics & Gynaecology. 2007; 21: 915–929. [8] Madhu CK, Nattey J, Naeem T. Second generation endometrial ablation techniques: an audit of clinical practice. Archives of Gynecology and Obstetrics. 2009; 280: 599–602. [9] Takahashi WH, Lopes RG, Depes Dde B, Martins e Castello Branco HK. Results of hysteroscopic endometrial ablation after five-year follow-up. Revista Brasileira de Ginecologia e Obste- tricia. 2012; 34: 80–85. (In Portuguese) [10] Beelen P , Reinders IMA, Scheepers WFW, Herman MC, Ge- omini PMAJ, van Kuijk SMJ, et al . Prognostic Factors for the Failure of Endometrial Ablation. Obstetrics & Gynecology. 2019; 134: 1269–1281. [11] Munro MG, Critchley HOD, Broder MS, Fraser IS. FIGO classi- fication system (PALM-COEIN) for causes of abnormal uterine bleeding in nongravid women of reproductive age. International Journal of Gynecology and Obstetrics. 2011; 113: 3–13. [12] National Center for Health Statistics. International Classifica- tion of Diseases, Ninth Revision, Clinical Modification (ICD- 9-CM). 2021. Available at: https://www.cdc.gov/nchs/icd/icd9c m.htm (Accessed: 20 October 2022). [13] Bonafede MM, Miller JD, Lukes A, Meyer NM, Lenhart GM. Comparison of direct and indirect costs of abnormal uter- ine bleeding treatment with global endometrial ablation and hysterectomy. Journal of Comparative Effectiveness Research. 2015; 4: 115–122. [14] Dwyer N, Hutton J, Stirrat GM. Randomised controlled trial comparing endometrial resection with abdominal hysterectomy for the surgical treatment of menorrhagia. BJOG: An Interna- tional Journal of Obstetrics and Gynaecology. 1993; 100: 237– 243. [15] Cooper JM, Anderson TL, Fortin CA, Jack SA, Plentl MB. Mi- crowave Endometrial Ablation vs. Rollerball Electroablation for Menorrhagia: a Multicenter Randomized Trial. The Journal of the American Association of Gynecologic Laparoscopists. 2004; 11: 394–403. [16] Bofill Rodriguez M, Lethaby A, Grigore M, Brown J, Hickey M, Farquhar C. Endometrial resection and ablation techniques for heavy menstrual bleeding. The Cochrane Database of Sys- tematic Reviews. 2019; 1: CD001501. [17] Bergeron C, Laberge PY , Boutin A, Thériault M, V alcourt F, Lemyre M, et al. Endometrial ablation or resection versus lev- onorgestrel intra-uterine system for the treatment of women with heavy menstrual bleeding and a normal uterine cavity: a system- atic review with meta-analysis. Human Reproduction Update. 2020; 26: 302–311. [18] Lethaby A, Hussain M, Rishworth JR, Rees MC. Progesterone or progestogen-releasing intrauterine systems for heavy menstrual bleeding. The Cochrane Database of Systematic Reviews. 2015; 4: CD002126. [19] Marjoribanks J, Lethaby A, Farquhar C. Surgery versus medical therapy for heavy menstrual bleeding. The Cochrane Database of Systematic Reviews. 2003; CD003855. [20] Oderkerk TJ, Beelen P , Geomini PMAJ, Herman MC, Lee- mans JC, Duijnhoven RG, et al. Endometrial ablation plus lev- onorgestrel releasing intrauterine system versus endometrial ab- lation alone in women with heavy menstrual bleeding: study protocol of a multicentre randomised controlled trial; MIRA2 trial. BMC Women’s Health. 2022; 22: 257. 7 [21] Peeters JAH, Penninx JPM, Mol BW, Bongers MY . Prognostic factors for the success of endometrial ablation in the treatment of menorrhagia with special reference to previous cesarean section. European Journal of Obstetrics & Gynecology and Reproductive Biology. 2013; 167: 100–103. [22] Rasmussen CK, Hansen ES, Al-Mashadi Dahl S, Ernst E, Due- holm M. The effect of transcervical endometrial resection on clinical symptoms related to intrinsic adenomyosis and junc- tional zone changes. European Journal of Obstetrics & Gyne- cology and Reproductive Biology: X. 2019; 3: 100029. [23] Preutthipan S, Herabutya Y . Hysteroscopic rollerball endome- trial ablation as an alternative treatment for adenomyosis with menorrhagia and/or dysmenorrhea. The Journal of Obstetrics and Gynaecology Research. 2010; 36: 1031–1036. [24] Hokenstad AN, El-Nashar SA, Khan Z, Hopkins MR, Famuyide AO. Endometrial Ablation in Women with Abnormal Uterine Bleeding Related to Ovulatory Dysfunction: a Cohort Study. Journal of Minimally Invasive Gynecology. 2015; 22: 1225– 1230. [25] Bongers MY . Hysteroscopy and heavy menstrual bleeding (to cover TCRE and second-generation endometrial ablation). Best Practice & Research Clinical Obstetrics & Gynaecology. 2015; 29: 930–939. [26] Fergusson RJ, Lethaby A, Shepperd S, Farquhar C. Endometrial resection and ablation versus hysterectomy for heavy menstrual bleeding. The Cochrane Database of Systematic Reviews. 2013; 29: CD000329. [27] Subbaiah M, Selvest N, Maurya D. Comparison of bipolar ball endometrial ablation and transcervical resection of the en- dometrium in the treatment of heavy menstrual bleeding: a randomized clinical trial. Gynecology and Minimally Invasive Therapy. 2021; 10: 143. [28] Thomassee MS, Curlin H, Y unker A, Anderson TL. Predicting pelvic pain after endometrial ablation: which preoperative pa- tient characteristics are associated? Journal of Minimally Inva- sive Gynecology. 2013; 20: 642–647. 8