Validation of a scoring system to triage women with heavy menstrual bleeding into hysterectomy or uterine sparing modalities: a retrospective study

Objective: To validate a scoring system aims to triage women with heavy menstrual bleeding (HMB) into hysterectomy or uterine sparing options and inform the development of a prospective clinical trial. Design: Retrospective study. Setting: UK single centre. Population: 327 women aged 25-54 referred with HMB between January 2021 and December 2021. Methods: Data was retrieved from patients‘ records and divided by outcome. The score was applied to the collected data, patients followed for two years. The score has 6 parameters, each scores 1, apart from uterine cavity length scores 2. Differences and associations between variables were examined. Main outcome measures: sensitivity and specificity of the scoring system in triaging women with HMB. Results: 38/327 (11. 6%) of patients had hysterectomy. There was a significant association between having each factor of the score and hysterectomy and a significant association of having a total score of >=3 and hysterectomy OR 6.23 [95%CI 4.80 – 7.67]. The score has a calculated sensitivity of 0.84 [0.68-0.93], specificity of 0.99 [0.97-0.99] and positive predictive value of 0.91 [0.75-0.98]. The strongest predictive factor for hysterectomy was large uterine cavity, followed by woman’s desire for hysterectomy, then adenomyosis (p<0.001). Other demographic and clinical variables did not predict hysterectomy as an outcome with the exception of dysmenorhea which when added to the score, it decreased positive predictive value affecting performance. Conclusions: This scoring system has high sensitivity and specificity in triaging women with HMB into hysterectomy and uterine sparing modalities. Further evaluation in a prospective clinical trial is now warranted.

Many women in their reproductive age suffer with heavy menstrual bleeding (HMB), described as excessive menstrual blood loss affecting the woman‘s physical, social, and emotional quality of life 1,2,3 . It affects around 25% of women in the UK and around a fifth of the referrals to gynaeoclogical clinics are due to HMB 4 . Different medical and surgical interventions are usually offered prior to hysterectomy. These interventions might temporality relieve some of the symptoms, however; several women continue to have HMB & other symptoms for years before definitive management is offered. The two most widely used uterine sparing medical and surgical interventions in modern gynecological practice for HMB are the Levonorgestrel IUS (Mirena IUS Bayer Healthcare pharmaceutical) and endometrial ablation (EA). They both proved to be very effective in managing the condition and controlling women‘s symptoms 5,6,7,8 . These procedures represent a lower cost with a shorter recovery time compared to hysterectomy. However, the ECLIPSE Trial highlighted on a longer term that 1:5 women needed hysterectomy after 5 years of having the Levonorgestrel IUS 5 . The same applies to EA, where follow up showed reduced treatment efficacy and women‘s satisfaction with a post ablation hysterectomy rate reaching 21% in some series 7,8,9 . Utilising these uterine sparing modalities prior to hysterectomy for women with HMB who also have additional gynaecological symptoms and pathologies, might result in failure of these interventions which can cause increased women‘s suffering by prolonging their treatment journey with additional procedure- related cost. Women who continue to suffer with these symptoms of heavy bleeding and pain; their quality of life and wellbeing can be further affected especially when suffering is prolonged. Certain conditions can lead to failure of EA leading to hysterectomy including fibroids, adenomyosis and endometriosis, in addition to large uterine cavity, women‘s desire for complete amenorrhoea and HMB causing severe anaemia needing parental therapy 9,10 . In this study, we have validated the use of a scoring tool which utilises 6 factors, in order to triage women with HMB into hysterectomy and uterine sparing modalities. This would aim to shorten the time to hysterectomy when other modalities would fail and avoid hysterectomy when other options would be effective. This would support the future counselling process and decision making.

Patient involvement: All gynaecological referrals to gynaecology clinics between January 2021 and December 2021 were identified which were 44225 patients. Following removing those who did not attend their appointments, oncology, fertility, colposcopy urogynaecology patients, and other irrelevant gynaecological referrals, a total of 361 patients were identified as being referred with HMB. Only 327 met the inclusion criteria of the study and were included. Those who had additional gynaecological indications for hysterectomy such as cervical intraepithelial neoplasia, malignant and premalignant uterine and ovarian conditions, uterine prolapse or women wanting to preserve fertility, were excluded from the study. Women with a very high BMI different uterine sparing modalities to avoid the complications of hysterectomy. Women with contraindications for hysterectomy such as previous multiple abdominal laparotomies or those with significant anaesthetic risk, were also excluded. Data including demographics, symptoms, investigations and treatment options was retrieved retrospectively from the patient‘s medical notes via the local online information system (Meditech). Collected data was documented on a spread excel sheet. The scoring system was applied to the collected data. The proposed scoring system uses 6 clinical parameters which are: presence of uterine fibroid of =>3cm, adenomyosis, uterine cavity length of >10.5 cm, the presence of chronic pelvic pain/ endometriosis/ dysmenorrhoea and dyspareunia, iron deficiency anaemia needing iron infusion/blood transfusion and woman‘s desire for hysterectomy. Each parameter scores 1 apart from the uterine cavity length which scores 2. The score was calculated following data collection on each patient. Patients were followed for two years. Core outcome sets: we calculated the number of patients needed hysterectomy. We also divided the patients into those who scored years. We measured the correlation between each factor in the scoring tool and the outcome of hysterectomy and measured the sensitivity and specificity of the scoring system in predicting outcome. Sample size estimation: This was calculated as suggested by Negida A et al 11 . Based on clinical experience that between 10 and 15% of women in this clinical group proceeded to hysterectomy an average prevalence of 12.5% was chosen and this yielded a total required sample size of 320 participants and 35 outcomes of hysterectomy to estimate sensitivity and specificity at 0.9. Statistical analysis: all analysis was performed using JASP 0.19.3 with an alpha level of <.05 throughout. Differences between scale variables were examined using the Welch corrected t-test (except for IMD (Index of multiple deprivation) where the ordinal data needed the use of Mann-Whitney test for non -parametric data and associations between categorical variables using the Chi-squared test. Associations with the binary outcome (hysterectomy or not) were assessed using a series of logistic regressions with hysterectomy as the outcome variable and the members of the score construct or other clinical and demographic variables as predictors.

38/327 of the patients who met the criteria (11.6%) had hysterectomy. Duration of HMB was converted into months. Where a range was given the midpoint was chosen. For one patient this was reported as since menarche and this was replaced as her age minus 13 expressed in months. 6 patients were reported as “years” or “many years” and these were replaced at approximately mean value + 3 SD or 10 years (120 months). Postcode was converted to IMD score (Index of multiple deprivation) to assess if there was any association between socio-economic status and having hysterectomy. There were no significant differences or associations between demographic variables and clinical variables outside the score construct and having a hysterectomy with the exception of those women experiencing dysmenorrhea alone which was associated with hysterectomy as an outcome (p=3 and hysterectomy, OR 6.23 [95%CI 4.80 – 7.67]. Each of the elements of the score system was independently associated with the outcome of hysterectomy (Table 1B). The scoring system has a calculated sensitivity of 0.84 [95%CI 0.68-0.93], specificity of 0.99 [95% CI 0.97-0.99] and positive predictive value of 0.91 [95%CI 0.75-0.98]. Women who ended up having hysterectomy had higher mean score than those who did not t(40.52) = 14.71, p<.001, d = 3.15 [95%CI 2.24– 3.90]. Number of women experiencing each outcome when classified by a score cutoff of <=3 are shown in table 1C. A series of logistic regression models was constructed to determine which factors predict hysterectomy using initially the six members of the score (and to determine their combined effect). A significant model is found with coefficients as in table 2A. χ 2 (317) = 180.45, p<.001, R 2 (McFadden) 0.77, R 2 (Nagelkerke) 0.83, R 2 (Tjur) 0.77, R 2 (Cox & Snell) 0.43. When considered as a group of 6, the strongest predictive factor is the presence of a large uterine cavity (p<.001) followed by the woman’s desire for hysterectomy (p<.001) and then the presence of adenomyosis. The remaining factors are then weaker but all significant (p<.03). (Table 2A). A second series of logistic regression models was constructed to determine if any of the other clinical or demographic factors predict hysterectomy using the other measured variables (and to determine their combined effect). A borderline significant model is found with coefficients as in table 2B. χ 2 (312) = 17.67, p=.05, R 2 (McFadden) 0.07, R 2 (Nagelkerke) 0.10, R 2 (Tjur) 0.06, R 2 (Cox & Snell) 0.05. None of the demographic or other factors were significantly related to the outcome of having a hysterectomy at two years with the exception of dysmenorrhea alone which is a significant predictor. (Table 2B). In a combined logistic regression model once the members of the score variable were entered, none of the other clinical or demographic factors were significant predictors of hysterectomy as an outcome. Distribution of scores calculated for each woman divided by outcome is shown in figure 1A. Taking estimated prevalence of hysterectomy (from data) calculated as 0.12 [95% CI 0.08 – 0.16] then the sensitivity and specificity of the cut off of 3 on this scoring system is as shown in table 3A with the ROC plot for this model is shown in figure 1B. Even though dysmenorrhea alone did not add significantly to the combined logistic model, its significance as an independent predictor of hysterectomy suggested it should be considered as a member of the score construct. Scoring was repeated including this factor. There was a significant association between scoring above / below the threshold with the outcome of having a hysterectomy χ 2 (1) = 243.07, p<.001 with sensitivity and specificity as shown in table 3B. Adding dysmenorrhoea as a member of the scoring system leads to an increase in sensitivity but a considerable reduction in positive predictive value suggesting it should not be retained as a predictive factor and the original members of the score construct should be retained unchanged. Table 1, Differences between those who did / did not have hysterectomy in A. Clinical and demographic variables, B. members of the score construct and C. number of patients experiencing each outcome depending on score status | N | 289 | 38 | | | Age | 42.89 (7.68) | 43.13 (6.91) | .31 | | BMI | 30.09 () | 30.41 (4.86) | .70 | | Previous CS | No 238 (82.35%) | No 30 (78.95%) | | | Number of CS | 1.67 (0.77) n = 51 | 1.88 (0.64) n = 8 | .37 | | Previous Tubal Ligation | No 263 (91.00%) | No 34 (89.47%) | .76 | | Duration HMB (months) | 24.67 (35.74) | 34.68 (36.55) | .12 | | Smoker | No 277 (95.85%) | No 36 (94.74%) | .75 | | IMD mean (SD), median | 3.45 (2.33), 3.0 | 3.32 (2.00), 3.0 | .97 (Mann-Whitney) | | Ethnicity | WB 97.65% | WB 100% | - | | PCOS | Yes 13 (4.51%) | Yes 3 (7.90%) | .37 | | Dysmenorrhea alone | Yes 44 (15.44%) | Yes 16 (42.11%) | <.001 | | B. SCORE | ||| | CPP | Yes 58 (20.21%) | Yes 22 (57/90%) | <.001 | | fibroid | Yes 29 (10.11%) | Yes 20 (52.63%) | <.001 | | adenomyosis | Yes 16 (5.58%) | Yes 18 (47.39%) | <.001 | | anaemia | Yes 12 (4.17%) | Yes 12 (31.58%) | <.001 | | Large uterine cavity | Yes 3 (1.05%) | Yes 18 (47.37%) | <.001 | | desire | Yes 22 (7.61%) | Yes 25 (65.79%) | <.001 | | Score mean (SD) median | 0.50 (0.71), 0.00 | 3.47 (1.20), 3.00 | =3 | 2 (0.69%) | 33 (86.84%) | <.001 | | C Outcomes | ||| | No hysterectomy | Hysterectomy | Totals | | | Score =3 | 2 | 33 | 35 | | Totals | 289 | 38 | 327 | Table 2. Coefficients for logistic regression model predicting outcome of hysterectomy (A) from the 6 members of the SCORE variable and (B) from the other demographic and clinical variables measured | Factor | B (SEB) | OR [95% CI] | p | | Intercept | -7.43 (1.24) | || | Large cavity | 5.90(1.27) | 360.87 [29.99 – 4342.51] | <.001 | | Desire | 4.36 (0.98) | 78.52 [11.53 – 534.59] | <.001 | | Adenomyosis | 3.67 (0.94) | 39.22 [6.24 – 243.49] | <.001 | | CPP | 1.85 (0.81) | 6.38 [1.32 – 30.89] | .021 | | Anaemia | 1.98 (0.99) | 7.22 [1.03 – 50.63] | .047 | | Fibroid | 2.02 (0.89) | 7.56 [1.31– 43.55] | .024 | | χ 2 (317) = 180.45, p<.001, R 2 (McFadden) 0.77, R 2 (Nagelkerke) 0.83, R 2 (Tjur) 0.77, R 2 (Cox & Snell) 0.43 | ||| | B. Other demographic and clinical factors | ||| | Factor | B (SEB) | OR [95% CI] | p | | Intercept | -3.21 (1.47) | || | Age | 0.03 (0.03) | 1.03 [0.98 – 1.08] | .29 | | Parity | 0.13 (0.18) | 0.87 [0.61 – 1.25] | .47 | | BMI | -0.006(0.03) | 0.99 [0.93 – 1.06] | .86 | | Number Prev CS | 0.17 (0.25) | 1.19 [0.73 – 1.92] | .49 | | Tubal ligation v not | -0.22 (0.67) | 0.80 [0.22– 3.01] | .75 | | Duration of HMB | 0.006 (0.004) | 1.01 [0.99 – 1.01] | .20 | | Smoker v not | 0.33 (0.83) | 1.40 [0.28– 7.08] | .69 | | IMD decile | -0.04 (0.08) | 0.96 [0.82 – 1.13] | .64 | | Dysmenorrhea | 1.36 (0.38) | 3.90 [1.85 – 8.19] | <.001 | | χ 2 (312) = 17.67, p=.05, R 2 (McFadden) 0.07, R 2 (Nagelkerke) 0.10, R 2 (Tjur) 0.06, R 2 (Cox & Snell) 0.05 | Figure 1. A. Distribution of the score variable between those who did / did not have hysterectomy. Heavy black line indicates cut off of 3 and B ROC plot showing performance of logistic model including the 6 factors of the SCORE Table 3. Performance of the score system with a cut off of >=3 for predicting hysterectomy (A) based on original members of the score system and (B) also including the predictor “dysmenorrhea alone” as a possible predictive factor | Measure | Value | 95% CI | Value | 95% CI | | Sensitivity | 0.87 | 0.71 – 0.95 | 0.92 | 0.78 – 0.98 | | Specificity | 0.99 | 0.97 – 1.00 | 0.98 | 0.95 – 0.99 | | For any particular positive test result the probability that it is a: | |||| | True positive (Positive predictive value) | 0.94 | 0.79 – 0.99 | 0.85 | 0.70 – 0.94 | | False positive | 0.06 | 0.01 – 0.20 | 0.15 | 0.06 – 0.30 | | For any particular negative test result the probability that it is a: | |||| | True negative (Negative predictive value) | 0.98 | 0.96 – 0.99 | 0.99 | 0.97– 0.99 | | False negative | 0.02 | 0.01 – 0.04 | 0.01 | 0.002– 0.03 | | Likelihood ratios Conventional: | |||| | Positive | 125.49 | 31.36 – 502.12 | 44.36 | 19.99 – 98.46 | | Negative | 0.13 | 0.06 – 0.30 | 0.08 | 0.02 – 0.24 | | Likelihood ratios Weighted by prevalence | |||| | Positive | 16.50 | 4.29 – 63.53 | 5.83 | 2.76 – 12.35 | | Negative | 0.02 | 0.007 – 0.04 | 0.01 | 0.003– 0.03 |

HMB, a common gynaecological condition affecting at least a quarter of women during their reproductive life and has several implications on women‘s quality of life by affecting their physical and mental health as well as their emotional and social wellbeing 8,9,10 . It can also be associated with severe pain symptoms, anaemia and fatigue 12.13,14 Different pathologies can be associated with HMB including uterine fibroids, adenomyosis and endometriosis which can predispose to pelvic pain, dysmenorrhoea, dyspareunia, anaemia, fatigue and pressure symptoms 9,10 . Medical options are usually offered to women with HMB and if failed or declined, EA would be offered if family is deemed complete. Hysterectomy is usually reserved as last option if all options fail 1 . Women who have tried different medical and surgical interventions and continue with HMB and its associated symptoms of pain, tiredness, anaemia and fatigue, their prolonged suffering results in reduced satisfaction till hysterectomy is offered which would also lead to additional cost 15 . Main findings: In this study, we have validated the use of a scoring system which utilises six factors identified from previous studies which are usually associated with HMB and when calculated together and the score is =>3, conservative measures might fail leading to hysterectomy. 4,6,9,14,15,16,17,18,19,20,21,22,23 This scoring system aims to shorten the time to hysterectomy when other modalities would fail and avoid hysterectomy when other medical and surgical interventions would usually work. Several researchers have identified the correlation between different factors and failed EA. Husu et al, reported a 7.1% hysterectomy rate following Novasure EA and suggested preoperative imaging to detect submucous myomas, large myomas, polyps and adenomyosis to optimise patient’s selection and reduce the need for subsequent hysterectomy 16 . Stevens et al in 2019 predicted the likelihood of failed EA for HMB. This was based on woman‘s age, parity, duration of HMB, associated dysmenorrhoea and previous caesarean section 4 EA was reported to be associated with post ablation pelvic pain in 20.8% in one series especially in women with pre ablation dysmenorroeha and endometriosis and recommended properly counselling women regarding this expected surgical outcome 1 7,18 . Late-Onset Endometrial Ablation Failure (LOEAF) was recently identified to be associated with hysterectomy in 25% of women regardless of the type of EA used and an unknown number of women in this series, have less than satisfactory results 18 . Understanding factors which lead to LOEAF with good patient selection would reduce LOEAF and improve patient‘s satisfaction 19 . The commonest causes of post ablation hysterectomy in different series were found to be recurrence of HMB due to inadequate destruction of the endometrium or its regrowth following ablation, chronic pelvic pain, leiomyomas and adenomyosis 17,18,19,20 . Adenomyosis is a common cause of chronic pelvic pain and HMB. Deep adenomyosis was identified to be present at hysterectomy specimens in a significant number of patients following failed Novasure endometrial ablation 21 . On the other hand, the Levonorgestrel IUS was found to be associated with failure in women with a uterine fibroid>=2.5 cm or a uterine size of>12 cm 22,23 . Large uterine cavity >10.5 cm was identified as a risk factor for failed EA and LOEAF and it can also be associated with a failed Levonorgestrel IUS 19,21,22,23,24 Large uterine cavity, chronic pelvic pain/dysmenorrhoea and dyspareunia or known endometriosis/adenomyosis were identified in this study when present together and the score is =>3, a hysterectomy could be the preferred option. One of the causes of Iron deficiency anaemia in women is HMB which can be life-threatening needing to be addressed proactively 25 . A consensus guidance covering screening and diagnosis of iron deficiency anaemia in women with HMB has been suggested to improve health outcomes in those women 26 . We considered HMB leading to iron deficiency anaemia needing parenteral therapy as a factor which can predict hysterectomy if other factors also existed 17 . Women‘s desire for definitive treatment was identified as a factor which would eventually lead to hysterectomy when present with other factors in a previous series 17 . This was especially identified in women aiming for complete amenorrhoea where other uterine sparing modalities such as the Levonorgestrel IUS and EA cannot guarantee. Hysterectomy is the most commonly performed major gynaecological surgical procedure 27 . Despite its invasive nature, it represents the most definitive treatment option for HMB in women when future fertility is not desired 27 . Due to its invasive nature with longer recovery time and complications, hysterectomy is thought to be only considered when other treatment options have failed or are contraindicated 1,27 . The HEALTH Trial; one of the largest randomised controlled trials to date in benign gynaecology; compared laparoscopic supracervical hysterectomy (LASH) with EA for HMB. It found that women allocated to LASH were more satisfied with their treatment with improved quality of life (QoL) compared to the EA group with no difference in serious adverse events between the two groups over 15 months follow up 3 . In this trial, patient‘s decision making preference was eliminated by being in the trial which could have led to regret and less satisfaction 28 . The trial excluded women with fibroids and large endometrial cavities and patients were only followed for 15 months which might not be sufficient to assess EA failure. A QoL questionnaire in women with HMB was suggested to identify women when HMB interferes with their QoL. This was found to be valid 29 . However, a recent systematic review of QoL measurement tools for HMB has highlighted that none of the available QoL tools for HMB is appropriate for use in practice. It also suggested the need to invest in developing and validating reliable tools that offer high qualitative and quantitative assessment 30 . There is no available scoring system to triage women with HMB into hysterectomy and uterine sparing modalities. This study has validated a scoring system and found a significant association between scoring => the threshold of 3 with the outcome of having a hysterectomy. Each member of the scoring tool was significantly associated with the outcome of hysterectomy. This scoring system as it is, was found to be highly sensitive and specific in triaging women with HMB into hysterectomy and uterine sparing options. The strongest predictive factor in the scoring system was found to be the presence of a large uterine cavity followed by the woman’s desire for hysterectomy then the presence of adenomyosis. No other clinical or demographic factors were found to be significant predictors. Dysmenorrhoea, which was found to be significantly associated with the outcome of hysterectomy, was not added to the final scoring tool as this would result in improving sensitivity on slight cost of specificity hence reducing the performance of the tool. Strengths of this study : This study has validated a scoring tool which utilised different evidence-based parameters and measured its sensitivity and specificity. It correctly excluded the factor which reduced the score performance. It also identified which factors in the tool predicted hysterectomy most. It reviewed and summarised recent evidence on failed conservative management for HMB. To our knowledge, this is the only available scoring system to triage women with HMB. This study would allow the development of a prospective clinical trial to test the tool for wider use in the future. Weaknesses and limitations : This study is a retrospective study over one year period with a small sample size in one UK centre. It only followed women for two years which could have missed some of the LOAF cases.

This scoring system as developed has high sensitivity and specificity in triaging women with HMB into hysterectomy and uterine sparing modalities with logistic regression confirming the validity of the scoring system in differentiating outcomes. Further evaluation of this tool in a prospective clinical trial is now warranted.

We would like to thank the coding department at South Tyneside & Sunderland NHS Foundation Trust (STSFT) for their help and support in identifying women referred to the gynaecology clinic and for their help in identifying the final list of patients. We would also like to thank the two research nurses Judith Ormond and Ashleigh Duffy for their help in triaging the first triage list. Disclosure: The word women is used as a collective term in this paper, but we acknowledge those that identify by other terms. Conflicts of Interest: The authors declare no conflicts of interest. Authors‘ contribution: Shamma Al-Inizi identified the concept of introducing a scoring system to triage women with HMB following publishing three articles on the subject. She planned this validation study with the support of the methodologist and statistician. She planned and obtained the final list of patients with the help of the coding department at the organisation. She finalised the data collection excel sheet. She reviewed 55% of the cases and collected their data. She wrote the first, second and final draft of the paper including its conclusions based on the statistical analysis. She also completed the Medline search. Dr Jon Rees, has advised on planning the validation study, the sample size and duration required to validate the scoring system. He has reviewed the data collection form and adjusted it and advised adding post codes to the demographics. He also advised converting each answer on the spreadsheet into number to facilitate the statistical analysis which led to the conclusions of the study. He has performed all the statistical analysis required for the study which led to its outcome. He has reviewed the first, second and final draft of the paper and has finalised it. Dr. Laura Hewitt, she helped in around 32% of the data collection and finalising the data collection spreadsheet. Dr. Humaira Ali, she helped in around 8% of the data collection and reviewing the paper. Safa Kindawi, she helped in 5% of data collection, converting results into numbers on the data collection spreadsheet and converting post codes into numbers. She also helped in the correction/writing of the paper including proof reading and Medline search. Ethical approval: the study did not require ethics approval from the regional or institutional ethics committees as it is a retrospective study which reviewed the online medical records of the patients.

Heavy menstrual bleeding: assessment and management. National institute for health and care excellence (NICE) guideline. NG88, 2018 updated 2021. https://www.nice.org.uk/guidance/ng88 Lui Z, Doan QV. Blumenthal P, Dubois RW. A systematic review evaluating health- related quality of life, work impairment, and health care costs and utilization in abnormal uterine bleeding. Value Health 2007; 10(3):183-194. https://doi:10.1111/.1524-433.200.0018.x. Copper K, Breeman S, Scott NW, Scotland G, Clark J, Hawe J et al. Laparoscopic supracervical hysterectomy versus endometrial ablation for women with heavy menstrual bleeding (HEALTH): a parallel-group, open-label, randomised controlled trial. 2019, Lancet, 394(10207):1425-1436. https://dx.doi.org/10.1016/S0140-673(19)31790-8 4. Stevens K.Y.R, Meulenbroeks D, Houterman S, Gijsen T, Weyers S, Schoot BC. Gynecol Surg. Weyers S. Prediction of unsuccessful endometrial ablation: a retrospective study 2019; 16 (7). https://doi.org/10.1186/s10397-019-1060-1. 5.Gupta J, Daniels J, Middleton L, Pattison HM, Prileszky G, Roberts TC et al. A randomised controlled trial of the clinical effectiveness and cost-effectiveness of the levonorgestrel-releasing intrauterine system in primary care against standard treatment for menorrhagia: the ECLIPSE trial. Health Technol Assess, 2015;19(88):i-xxv,1-118. https://DOI: 10.3310/hta19880 6.Kumar v, Chodankar R, Kumar Gupta J. Endometrial ablation for heavy menstrual bleeding. Women’s Health (Lond). 2016;12(1):45-52. https:// doi: 10.2217/whe.15.86 7.Matteson KA, Abed H, Wheeler TL, Sung VW, Rahn DD, Schaffer JI et al. A systematic review comparing hysterectomy with less-invasive treatments for abnormal uterine bleeding. J Minim Invasive Gynecol. 2012;19(1):13-28. https://doi;10.1016/j.jmig.2011.08.005 8. Kiran A, Geary RS, Gurol- Urganci I, Cromwell DA, Bansi-Matharu L, Shakespeare J, et al. Sociodemographic differences in symptom severity and duration among women referred to secondary care for menorrhagia in England and Wales: a cohort study from the National Heavy Menstrual Bleeding Audit. BMJ Open, 2018;8(2):e018444. https://doi:10.1136/bmjopen-2017-018444. 9. Al-Inizi S. Why do women with Heavy menstrual bleeding end in hysterectomy? 2023; Open Access, 2(1).108. https://doi:10.31021/jgrb.20232108. 10. Kirstin R, Davies M, Gerald J, Harkins GJ. Characteristics of patients undergoing hysterectomy for failed endometrial ablation. 2013, JSLS. 17(4):503-507. https://doi:10.4293/108680813x13693422520602. 11. Negida A, Fahim NK, Negida Y. Sample Size Calculation Guide - Part 4: How to Calculate the Sample Size for a Diagnostic Test Accuracy Study based on Sensitivity, Specificity, and the Area Under the ROC Curve. Adv J Emerg Med. 2019, 19;3(3):e33. https://doi: 10.22114/ajem.v0i0.158. 12. Royal College of Obstetricians and Gynaecologists, London School of Hygiene & Tropical Medicine (LSHTM), and Ipsos MORI, National Heavy Menstrual Bleeding Audit: Third Annual Report. London, UK: RCOG Press, 2013. https://www.hquip.org.uk 1. Wang W, Bourgeois T, Klima J, Berlan ED, Fischer AN, O’Brien SH. Iron deficiency, and fatigue in adolescent females with heavy menstrual bleeding. Haemophilia. 2013; 19(2):225-230. https://doi: 10.1111/hae.12046 2. Cohen BJ, Gibor Y. Anaemia and menstrual blood loss. Obstet Gynecol Surv 1980; 35:597-618. PMID: 6997784. 3. Al-Inizi S. How much cost saving can be achieved by introducing a scoring system to triage women with heavy menstrual bleeding? Int J Clin Stud Med Case Report.2023,25(4),2-6. https://doi:10.46998/IJCMCR.2023.25.000617. 4. Hsu A L, Bell C, Burkman R, Grow D. Hysterectomy after novasure endometrial ablation failure of a bridge to menopause. Fert Steril, 2007,88(1),S193. https://doi:1O,1O16/j.fertstert.2007.o7. 665. 5. Al-Inizi S, Kaur S, Amr M. Heavy menstrual bleeding: Can a scoring system predict the likelihood of hysterectomy? Med J Obs Gyn. 2023;11(1):1164. https://doi:10.13140/RG.2.2.27578.36800. 6. Thomassee M, Curlin H, Yunker A, Anderson TL. Predicting pelvic pain after endometrial ablation: which preoperative patient characteristics are associated? J Minim Invasive Gynecol. 2013;20(5):642-647. https:// doi: 10.1016/j.jmig.2013.04.006 7. Wortman M. Late-Onset endometrial ablation failure. Case Rep in Womens Health. 2017;12:15:11-28. https:// doi: 10.1016/j.crwh.2017.07.001 8. McCausland AM, McCausland VM. Depth of endometrial penetration in adenomyosis helps determine outcome of rollerball ablation. Am J Obstet Gynecol. 1996; 174(6):1786-1793;1793-1794. https:// doi: 10.1016/s0002-9378(96)70211-9 9. Seidman JD, Kjerulff KH. Pathologic findings from Maryland Women‘s Health Study: practice patterns in the diagnosis of adenomyosis. Int J Gynecol Pathol.1996;15(3):217-221. https:// doi: 10.1097/00004347-199607000-00005 10. Mengerink B, Van der Wurff, Haar JF, Van Rooij IA, Pinjnenborg J. Effect of undiagnosed deep adenomyosis after failed Novasure endometrial ablation. J Minim Invasive Gynecol. 2015,22(2):239-244. https:// doi: 10.1016/j. jmig.2014.10.006. 11. Singh K, Bharati G, Prasad D, Kumari S. Role of levonorgestrel releasing intrauterine device in management of heavy menstrual bleeding: a conservative approach. Int J Reprod Contracep Obstet Gynecol. 2017;6(2): 01-31. https:// doi.org/10.18203/2320-1770.ijrcog20170396. 12. Shazly SA, Famuyide AO, El-Nashar SA, Breitkopf DM, Hopkins MR, Laughlin-Tommaso S. Intraoperative predictors of long-term outcomes after radiofrequency endometrial ablation. J Minim Invasive Gynecol. 2016; 23(4):582-589. https:// doi: 10.1016/j.jmig.2016.02.002 13. Nelson A, Ritchie J. Severe anemia from heavy menstrual bleeding requires heightened attention. Am J Obstet Gynecol. 2015;213(1):97.e1-97.e6. https:// doi: 10.1016/j.ajog.2015.04.023 14. Mansour D, Hofmann A, Gemzell-Danielsson K. A review of clinical guidelines on the management of iron deficiency and iron-deficiency anaemia in women with heavy menstrual bleeding. Adv Ther. 2021;38(1):201-225. https://doi.org/10.1007/s12325-020-01564-y. 15. Van der Meij E, Emanuel M. Hysterectomy for heavy menstrual bleeding. Women‘s Health (Lond). 2016,1;12(1):63-69. https:// doi:10.2217/whe.15.87. 28.Singh S, Bougie O. HEALTH for heavy menstrual bleeding: real-world implications. Open Access.2019;394(10207):1390-1392. https://doi:10.1016/S0140-6736(19)32086-0.29.Calaf J, Cancelo MJ, Andeyro M, Jimenez JM, Perello J, Correa M, et al. Development and Psychometric Validation of a Screening Questionnaire to Detect Excessive Menstrual Blood Loss That Interferes in Quality of Life: The SAMANTA Questionnaire. J Womens Health. 2020;29(7):1021-1031. https://doi:10.108/ jwh.2018. 7446.30. Al Wattar B, Shofield S, Minns V, Khan K. Quality of life measurement tools for heavy menstrual bleeding: A systematic review and critical appraisal. Sex Reprod Healthc. 2025;44:101102. https://doi:10.1016/j.srhc.2025.101102 Information & Authors Information Version history Copyright This work is licensed under a Non Exclusive No Reuse License.

Authors Metrics & Citations Metrics Article Usage 197views 87downloads Citations Download citation Shamma Al-Inizi, Laura Hewitt, Humaira Ali, et al. Validation of a scoring system to triage women with heavy menstrual bleeding into hysterectomy or uterine sparing modalities: a retrospective study. Authorea. 25 November 2025. DOI: https://doi.org/10.22541/au.176409208.85780983/v1 DOI: https://doi.org/10.22541/au.176409208.85780983/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu.