Characterization of patients that can continue conservative treatment for adenomyosis

21

22 Historically, hysterectomy has been the radical treatment for adenomyosis. 23 However, some patients do not wish to hysterectomy. Nevertheless, patients often 24 required hysterectomy during the course of conservative treatment, but the factors 25 involved remain unknown. The purpose of this study was to determine which patients 26 can continue conservative treatment for adenomyosis. 27

28 We selected women diagnosed with adenomyosis and provided with 29 conservative treatment at the Kindai University Hospital and Osaka Red Cross Hospital. 30 Age at diagnosis, parity, uterine size, subtype of adenomyosis, type of conservative 31 treatment, and timing of hysterectomy for cases with difficulty continuing conservative 32 treatment were examined retrospectively. 33

34 A total of 885 patients were diagnosed with adenomyosis, and 124 started 35 conservative treatment. Conservative treatment was continued in 96 patients (77.4%) 36 and hysterectomy was required in 28 patients (22.6%). The cumulative hysterectomy 37 rate was 32.4%, and all women had hysterectomy within 63 months. In the decision tree 38 analysis, 82% (23/28) of women aged 46 years or younger were able to continue 39 conservative treatment when parity was zero or one. If parity was two and over, 95% 40 (20/21) of those aged 39 years and older had hysterectomy. 41

42 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 3 Patients that continue conservative treatment for approximately 5 years are 43 more likely to have successful preservation of the uterus. Multipara and higher age of 44 diagnosis are factors that contribute to hysterectomy after conservative treatment. Parity 45 and age at diagnosis may be stratifying factor in future clinical trials on hormone 46 therapy. 47 48 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 4 1. Introduction 49 Adenomyosis is a benign disorder in which the endometrium and endometrial 50 stromal cells proliferate in the muscle layer of the uterus.1 2 Associated symptoms of 51 anemia, abdominal pain, and chronic pelvic pain due to excessive menstruation and 52 dysmenorrhea are common in women of reproductive age and significantly impair 53 quality of life.3 Traditionally, adenomyosis was often first diagnosed by pathological 54 examination after hysterectomy and was considered a disorder that affected the peri-55 menopausal period.4 5 However, with the widespread use of ultrasonography and 56 magnetic resonance imaging (MRI) in recent years, it has become possible to accurately 57 diagnose adenomyosis by imaging, and it is now diagnosed in relatively young 58 women .6-9 59 Historically, the radical treatment for adenomyosis has been hysterectomy.7 60 However, conservative treatments of adenomyosis, such as hormone therapy and 61 adenomyomectomy, are preferred by patients who are young and wish to preserve 62 fertility, or do not want hysterectomy or may be at high risk for perioperative 63 complications.10 For women who do not wish to become pregnant immediately, 64 conservative treatment mainly involves hormone therapy, which is continued until 65 menopause.11 However, even with hormone therapy for adenomyosis, patients often 66 experience persistent symptoms, including pain and drug side effects, such as irregular 67 bleeding or osteoporosis, that result in the need for hysterectomy.10 12-14 To date, it has 68 been unclear which women can continue conservative treatment for adenomyosis. The 69 identification of factors related to the success or failure of conservative treatment would 70 greatly contribute to the choice of treatment strategy, and significantly benefit the 71 quality of life of women and the health care economy. 72 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 5 The purpose of this study was to evaluate the treatment course of patients with 73 adenomyosis who have requested conservative treatment, and to determine which 74 women can continue conservative treatment. 75 76 2. Materials and Methods 77 2.1. Cases 78 From January 2008 to December 2017, patients diagnosed with adenomyosis 79 and started conservative treatment at Kindai University Hospital and Osaka Red Cross 80 Hospital were selected and studied retrospectively. Exclusion criteria was the absence 81 of symptoms due to adenomyosis (e.g., if the patient is being monitored for 82 endometriosis or other comorbidities) or pre-treatment imaging, request for 83 hysterectomy at the first visit, presence of submucosal myoma or 3 cm/three or more 84 intramuscular myomas that may cause hypermenorrhea. 85 86 2.2. Diagnosis 87 The diagnosis of adenomyosis was made using patients’ symptoms, such as 88 dysmenorrhea and hypermenorrhea, and imaging techniques, such as MRI or 89 transvaginal ultrasound. The criteria for diagnosis by MRI were the presence of an 90 enlarged myometrium with an indistinct limbus and a heterogeneous internal signal on 91 T2-weighted images or thickening of the junctional zone (>12 mm).15 16 The diagnostic 92 criteria for transvaginal ultrasonography were asymmetrical enlargement of the 93 myometrium and an asymmetrical decrease in echogenicity of the lesion.1 10 Most of the 94 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 6 cases were diagnosed by MRI, but only two cases were diagnosed by transvaginal 95 ultrasonography without pre-treatment MRI. Age was defined as the time when 96 adenomyosis was diagnosed on imaging. 97 98 2.3. Size Measurement 99 Measurements of the size of the uterus and the myometrium were performed 100 using MRI (figure1). In sagittal sections of MRI T2-weighted images, the length from 101 the cervix to the bottom of the uterus was defined as the long axis diameter of the uterus 102 (a), the maximum diameter perpendicular to long axis diameter was defined as the short 103 axis diameter of the uterus (b), and the thickness of the uterine muscle layer within the 104 short axis diameter of the uterus was defined as the muscle layer thickness (c). The 105 maximum transverse diameter of the uterus in the axial section of MRI T2-weighted 106 images was defined as the transverse diameter of the uterus (d). In the two cases 107 measured by transvaginal ultrasonography, (a), (b), and (c) were measured at the 108 position of maximum sagittal section. 109 110 2.4. Type of Adenomyosis 111 Adenomyosis was classified into four subtypes based on MRI imaging 112 features.17 Subtype I adenomyosis involved adenomyotic lesions that extended from the 113 endometrium and did not extend to the entire myometrium. Subtype II adenomyosis 114 was defined as adenomyotic lesions that extended from the perimetrium and did not 115 extend into the junctional zone. Subtype III adenomyosis was an isolated adenomyotic 116 lesion in the myometrium that did not extend into the junctional zone and the 117 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 7 perimetrium. Subtype IV adenomyosis was defined as a lesion that could not be 118 classified as types I–III, where the lesion involved the entire muscle layer. Two cases 119 diagnosed by transvaginal ultrasonography were not evaluated. 120 121 2.5. Type of conservative treatment 122 Hormone therapy (gonadotropin releasing hormone agonist (GnRHa), progestins, 123 levonorgestrel-releasing intrauterine system (LNG-IUS, Mirena intrauterine delivery 124 system®, Bayer Yakuhin, Ltd), oral contraceptives (OCs), and danazol (BONZOL 125 tablets®, Mitsubishi Tanabe Pharma Corporation) and adenomyomectomy were 126 provided as conservative treatment for adenomyosis. Hysterectomy was performed after 127 consultation with the patient when the symptoms worsened, or it became difficult to 128 continue hormone therapy. Treatment was started on the date of the first visit, and the 129 end of treatment was set at the date of the hysterectomy surgery or at the end of the 130 observation period. 131 132 2.6. Statistical analysis 133 Statistical analysis was performed using Graphpad Prism ver. 8.2.0 (GraphPad 134 Software, San Diego, CA, USA). The cumulative hysterectomy rate was determined by 135 the Log-rank test, and comparison between the two groups used the Mann-Whitney U 136 test and χ2 test, with P<0.05 as a significant difference. Decision tree analysis was 137 performed using weka (https://doi.org/10.1016/j.knosys.2019.04.013). 138 139 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 8 2.7. Ethics statement 140 This study was conducted with the approval of the ethics committees of Kindai 141 University Hospital and Osaka Red Cross Hospital (The approval numbers are R02-090 142 for Kindai University Hospital and J-0156 for Osaka Red Cross Hospital). 143 144 2.8. Patients and public involvement 145 Patients and the public were not involved in this study, including data 146 collection, analysis and interpretation. 147 148 3. Results 149 A total of 885 patients were diagnosed with adenomyosis and started on 150 treatment; 694 with no symptoms or no pre-treatment imaging, 51 who requested a 151 hysterectomy at the time of first visit, and 16 with submucosal myoma or 3 cm/three or 152 more intramuscular myomas that may cause hypermenorrhea were excluded, and 124 153 patients were started on conservative treatment (figure 2). Baseline characteristics of the 154 124 patients are presented in table 1. The median treatment period was 28 months (1–155 132 months), median age was 41 years (24–53 years), median parity was 1 (0–3), 156 median long axis diameter of the uterus was 9.7 cm (6.3–17.7 cm), median short axis 157 diameter was 6.7 cm (3.5–12.9 cm), median transverse diameter was 6.8 cm (2.8–14.2 158 cm) and the median muscle layer thickness was 3.9 cm (1.3–8.8 cm). Adenomyosis 159 subtypes I, II, III and IV were identified in 33 (26.6%), 28 (22.6%), 3 (2.4%) and 60 160 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 9 (48.4%) of these patients, respectively. Conservative treatment with hormone therapy 161 alone was provided for 117 patients (94.4%), adenomyomectomy alone was performed 162 for three patients (2.4%), and a mixture of these two procedures were provided for four 163 patients (3.2%). The breakdown of hormone therapy is presented in figure 3. 164 Ninety-six women (77.4%) were able to continue conservative treatment 165 throughout the treatment period, and 28 patients (22.6%) required hysterectomy during 166 the course of conservative treatment (figure 4). The cumulative total hysterectomy rate, 167 determined from the log-rank test of 124 patients who started conservative treatment, 168 was 32.4% and the 28 that required hysterectomy (Group A) all had hysterectomy 169 within 63 months (figure 4). Of the 96 patients who were able to continue conservative 170 treatment, 26 were able to continue conservative treatment for adenomyosis beyond 63 171 months (Group B), and all of them ultimately did not require hysterectomy (figure 4). 172 The characteristics of Group A and Group B are presented in Table 2. Group A 173 had a significantly higher age (Group A: 43 years, Group B: 37 years, p<0.001), higher 174 gravidity (Group A: 2, Group B: 0, p<0.001) and parity (Group A: 2, Group B: 0, 175 p<0.001), and a significantly higher proportion of multipara (Group A: 82.1%, Group 176 B: 42.3%, p<0.001) compared with Group B. The long axis diameter (Group A: 11.1 177 cm, Group B: 9.0 cm, p<0.001), short axis diameter (Group A: 9.0 cm, Group B: 7.7 178 cm, p=0.002), transverse diameter Group A: 8.0 cm, Group B: 6.6 cm, p=0.012), and 179 muscle layer thickness (Group A: 4.6 cm, Group B: 3.6 cm, p=0.018) were significantly 180 larger in Group A than those in Group B. The proportion of subtype IV adenomyosis 181 and other complications of endometriosis were not significantly different between the 182 two groups. 183 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 10 To determine the critical factors involved in whether conservative treatment for 184 symptomatic adenomyosis can be continued or not, we performed a decision tree 185 analysis of Groups A (group of discontinued conservative treatment) and B (group of 186 continued conservative treatment) using all the factors presented in Table 2, as shown in 187 figure 5. Interestingly, only parity and age, rather than factors related to adenomyotic 188 lesions, such as uterine size or adenomyosis subtype classification, had a decisive 189 impact on the success rate of conservation treatment. The first and most important 190 factor was parity, with 74% (23/31) of women with a parity of zero or one continuing 191 conservative treatment, compared to only 13% (3/23) with a parity of two or more 192 continuing conservative treatment. A total of 80% of patients were divided into two 193 groups based on whether or not they could continue treatment with parity alone. For 194 example, three cases of hysterectomy occurred in patients aged 47 years and older who 195 had a parity of zero or one. When parity was two or more, only two patients younger 196 than 38 years continued conservative treatment. 197 198 4. Discussion 199 We retrospectively examined the course of attempted uterine preservation in 200 patients with symptomatic adenomyosis to determine in which patient conservative 201 treatment could be continued and in which patient hysterectomy was necessary. This 202 study was unique in that (i) we extracted continued and discontinued conservative 203 treatment cases from the curves of the cumulative hysterectomy rate increase in women 204 who attempted uterine preservation, and (ii) we identified factors that distinguish 205 between uterine preservation and non-preservation were clarified by decision tree 206 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 11 analysis. For the first time, this study showed that parity and age may be important 207 factors for the consideration of conservative treatment for adenomyosis. Women in this 208 study were relatively young, with a median age of 41 years, making them younger than 209 those in reports from the early 2000s, but more consistent with recent reports.18-21 The 210 median parity was also low (at one), which may reflect the recent increase in aging of 211 primipara and the trend of low fertility (United nations: World Population Prospects 212 2019). In previous reports examining the benefit of hormone therapy in adenomyosis, 213 the mean pre-treatment uterine volume was 86 cm3 22, 96.5 cm3 3, 113.8 cm3 23, 158.9 214 ml 24, 278 cm3 25, and 311 cm3 26. The median uterine volume calculated from the long, 215 short, and transverse uterine diameters in the present study was 217 (71–1400) cm3, so 216 the size of the uterus was consistent with those previously reported. Adenomyosis was 217 classified as subtype IV in half of the cases, which tended to be more severe than 218 previously reported.17 This may be due to the fact that the two centers participating in 219 the study were core hospitals in the region, and therefore accepting patients with 220 advanced or difficult diagnosis. In conservative treatment for adenomyosis, 221 adenomyomectomy is indicated when hormone therapy is difficult to continue or when 222 the patient is undergoing infertility treatment. Because the uterine myometrium must be 223 repaired after removal of the adenomyotic lesion, it is limited to lesions that are 224 localized and capable of preserving the normal muscle layer 27. In the present study, 225 adenomyomectomy was chosen for a very small number of cases during infertility 226 treatment or when there was a desire for surgery. Multiple methods of hormone therapy 227 were used in most cases, including GnRHa, OCs, progestins, LNG-IUS and danazol.228 Multiple reports have shown that the smaller the size of the uterus at the start of 229 hormone therapy, the more successful hormone therapy has been in treating 230 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 12 adenomyosis 3 22-25 28 29. In this study, the size of the uterus at the start of treatment was 231 also significantly smaller in Group B, which was able to continue with conservative 232 treatment of adenomyosis (table 1). However, previous reports have had mixed follow-233 up periods and may have included women who ultimately needed hysterectomy. In the 234 present study, of the 124 patients who started conservative treatment, the failure to 235 continue treatment and necessary hysterectomy were most frequent within the first year. 236 This frequency then decreased, and treatment continued in all patients without much 237 change until the fifth year. Women who were able to continue conservative treatment 238 beyond 63 months did not require a hysterectomy. This novel analysis and the above 239

may provide guidance for planning the treatment of adenomyosis. Furthermore, 240 our study exploring factors involved in the acceptability of conservative treatment found 241 that patients undergoing conservative treatment for at least 5 years should be compared 242 with those who have had hysterectomy. 243 Decision tree analysis was able to extract the fewest factors needed to most 244 clearly separate the two patient groups (Group A and B) in terms of sensitivity and 245 specificity. Surprisingly, our current study revealed that uterine size and adenomyosis 246 subtype classification17 were not among the factors. The most important factor was 247 parity, and most patients with a parity of two or more were found to eventually require 248 hysterectomy. This may reflect the psychological factor of patients with two or more 249 children wanting to prioritize parenthood, rather than continuing conservative treatment, 250 which is also associated with symptoms such as irregular bleeding and pain. It has been 251 reported that patients who had undergone hysterectomy for any condition, not just 252 adenomyosis, were significantly more likely to have had a parity of two or more.4 In 253 addition, parity was reported to correlate with the incidence of adenomyosis11 30, which 254 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 13 may have influenced this result. The second most important factor in the decision tree 255 analysis was age. This may also be related to the intensity of the desire to preserve the 256 uterus and the frequency of adenomyosis. Parity and age of diagnosis may be stratifying 257 factor in future clinical trials on hormone therapy. 258 One limitation of this study was the small number of cases. We screened 885 259 cases of adenomyosis, but only 124 patients matched the criteria for inclusion in the 260 analysis. The number of cases was further reduced to 26, because we found that only 261 patients successfully treated for more than 63 months could be considered to have 262 successful uterine preservation. Therefore, it is expected that about 5000 patients with 263 adenomyosis would be needed to perform a similar analysis with more than 100 cases 264 per group. Furthermore, this study was a retrospective study of routine practice over a 265 10-year period, and the diversity in treatments available over this period is also a 266 limitation. In Japan, progestins and the levonorgestrel-releasing intrauterine system 267 were approved within the last 5 years for the treatment of adenomyosis, the increased 268 frequency of their use may have influenced the results. Additionally, criteria for the 269 diagnosis of adenomyosis are not yet clear, so it is possible that the patient may not be 270 diagnosed even if she has symptoms. It is hoped that further research will establish clear 271 criteria for the diagnosis and treatment of adenomyosis. 272 5. Conclusions 273 Uterine preservation in patients with adenomyosis is more likely to be 274 successful if they can continue conservative treatment for approximately 5 years. In 275 addition, multipara and higher age at diagnosis are factors for hysterectomy during 276 conservative treatment of adenomyosis. The results of this study may be useful in 277 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 14 decision-making and for informed consent when treating patients with adenomyosis. 278 Parity and age at diagnosis may be stratifying factor in future clinical trials on hormone 279 therapy. 280 281 282 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 15

283 1. Levy G, Dehaene A, Laurent N, et al. An update on adenomyosis. Diagn Interv 284 Imaging. 2013;94(1):3-25. 285 2. Vannuccini S, Luisi S, Tosti C, Sorbi F, Petraglia F. Role of medical therapy in 286 the management of uterine adenomyosis. Fertil Steril. 2018;109(3):398-405. 287 3. Osuga Y, Fujimoto-Okabe H, Hagino A. Evaluation of the efficacy and safety of 288 dienogest in the treatment of painful symptoms in patients with adenomyosis: a 289 randomized, double-blind, multicenter, placebo-controlled study. Fertil Steril. 290 2017;108(4):673-678. 291 4. Vercellini P, Parazzini F, Oldani S, Panazza S, Bramante T, Crosignani PG. 292 Adenomyosis at hysterectomy: a study on frequency distribution and patient 293 characteristics. Hum Reprod. 1995;10(5):1160-1162. 294 5. Parazzini F, Mais V, Cipriani S, Busacca M, Venturini P, On behalf of GISE. 295 Determinants of adenomyosis in women who underwent hysterectomy for 296 benign gynecological conditions: results from a prospective multicentric study 297 in Italy. Eur J Obstet Gynecol Reprod Biol. 2009;143(2):103-106. 298 6. Ryan GL, Stolpen A, Van Voorhis BJ. An unusual cause of adolescent 299 dysmenorrhea. Obstet Gynecol. 2006;108(4):1017-1022. 300 7. Farquhar C, Brosens I. Medical and surgical management of adenomyosis. Best 301 Pract Res Clin Obstet Gynaecol. 2006;20(4):603-616. 302 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 16 8. Dueholm M, Lundorf E. Transvaginal ultrasound or MRI for diagnosis of 303 adenomyosis. Curr Opin Obstet Gynecol. 2007;19(6):505-512. 304 9. Pinzauti S, Lazzeri L, Tosti C, et al. Transvaginal sonographic features of 305 diffuse adenomyosis in 18-30-year-old nulligravid women without 306 endometriosis: association with symptoms. Ultrasound Obstet Gynecol. 307 2015;46(6):730-736. 308 10. Garcia L, Isaacson K. Adenomyosis: review of the literature. J Minim Invasive 309 Gynecol. 2011;18(4):428-437. 310 11. Vannuccini S, Petraglia F. Recent advances in understanding and managing 311 adenomyosis. F1000Res. 2019;8. 312 12. Struble J, Reid S, Bedaiwy MA. Adenomyosis: A Clinical Review of a 313 Challenging Gynecologic Condition. J Minim Invasive Gynecol. 314 2016;23(2):164-185. 315 13. Benetti-Pinto CL, Mira TAA, Yela DA, Teatin-Juliato CR, Brito LGO. 316 Pharmacological Treatment for Symptomatic Adenomyosis: A Systematic 317 Review. Rev Bras Ginecol Obstet. 2019;41(9):564-574. 318 14. Younes G, Tulandi T. Conservative Surgery for Adenomyosis and Results: A 319 Systematic Review. J Minim Invasive Gynecol. 2018;25(2):265-276. 320 15. Dueholm M, Lundorf E, Hansen ES, Sørensen JS, Ledertoug S, Olesen F. 321 Magnetic resonance imaging and transvaginal ultrasonography for the diagnosis 322 of adenomyosis. Fertil Steril. 2001;76(3):588-594. 323 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 17 16. Novellas S, Chassang M, Delotte J, et al. MRI characteristics of the uterine 324 junctional zone: from normal to the diagnosis of adenomyosis. AJR Am J 325 Roentgenol. 2011;196(5):1206-1213. 326 17. Kishi Y, Suginami H, Kuramori R, Yabuta M, Suginami R, Taniguchi F. Four 327 subtypes of adenomyosis assessed by magnetic resonance imaging and their 328 specification. Am J Obstet Gynecol. 2012;207(2):114.e111-117. 329 18. Levgur M, Abadi MA, Tucker A. Adenomyosis: symptoms, histology, and 330 pregnancy terminations. Obstet Gynecol. 2000;95(5):688-691. 331 19. Templeman C, Marshall SF, Ursin G, et al. Adenomyosis and endometriosis in 332 the California Teachers Study. Fertil Steril. 2008;90(2):415-424. 333 20. Taran FA, Weaver AL, Coddington CC, Stewart EA. Understanding 334 adenomyosis: a case control study. Fertil Steril. 2010;94(4):1223-1228. 335 21. Pontis A, D'Alterio MN, Pirarba S, de Angelis C, Tinelli R, Angioni S. 336 Adenomyosis: a systematic review of medical treatment. Gynecol Endocrinol. 337 2016;32(9):696-700. 338 22. Li L, Leng J, Jia S, Lang J. Treatment of symptomatic adenomyosis with the 339 levonorgestrel-releasing intrauterine system. Int J Gynaecol Obstet. 340 2019;146(3):357-363. 341 23. Sheng J, Zhang WY, Zhang JP, Lu D. The LNG-IUS study on adenomyosis: a 342 3-year follow-up study on the efficacy and side effects of the use of 343 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 18 levonorgestrel intrauterine system for the treatment of dysmenorrhea associated 344 with adenomyosis. Contraception. 2009;79(3):189-193. 345 24. Lee KH, Kim JK, Lee MA, et al. Relationship between uterine volume and 346 discontinuation of treatment with levonorgestrel-releasing intrauterine devices in 347 patients with adenomyosis. Arch Gynecol Obstet. 2016;294(3):561-566. 348 25. Fawzy M, Mesbah Y. Comparison of dienogest versus triptorelin acetate in 349 premenopausal women with adenomyosis: a prospective clinical trial. Arch 350 Gynecol Obstet. 2015;292(6):1267-1271. 351 26. Zhang P, Song K, Li L, Yukuwa K, Kong B. Efficacy of combined 352 levonorgestrel-releasing intrauterine system with gonadotropin-releasing 353 hormone analog for the treatment of adenomyosis. Med Princ Pract. 354 2013;22(5):480-483. 355 27. Abbas S, Raybould JE, Sastry S, de la Cruz O. Respiratory viruses in transplant 356 recipients: more than just a cold. Clinical syndromes and infection prevention 357 principles. Int J Infect Dis. 2017;62:86-93. 358 28. Shaaban OM, Ali MK, Sabra AM, Abd El Aal DE. Levonorgestrel-releasing 359 intrauterine system versus a low-dose combined oral contraceptive for treatment 360 of adenomyotic uteri: a randomized clinical trial. Contraception. 361 2015;92(4):301-307. 362 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 19 29. Neriishi K, Hirata T, Fukuda S, et al. Long-term dienogest administration in 363 patients with symptomatic adenomyosis. J Obstet Gynaecol Res. 364 2018;44(8):1439-1444. 365 30. Vercellini P, Viganò P, Somigliana E, Daguati R, Abbiati A, Fedele L. 366 Adenomyosis: epidemiological factors. Best Pract Res Clin Obstet Gynaecol. 367 2006;20(4):465-477. 368 369 370 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint a b c A d B Figure.1 Measurement of uterine size. T2-Weighted Image (T2WI) of MRI. A: We used the sagittal T2WI of the uterus to measure (a); the uterine long axis diameter, (b); the uterine short axis diameter and (c); the muscle layer thickness. B: We used the axial T2WI of the uterus to measure (d); the uterine transverse diameter. . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 761 cases were excluded; • 694 cases had no symptom without treatment or no uterine size before treatment • 51 cases wanted to receive hysterectomy at first visit • 16 cases had myoma (subendometrial or over 3cm or three pieces) 124 cases were started conservative treatment 885 cases were diagnosed adenomyosis Figure.2 Cases flow chart. Of the 885 patients diagnosed with uterine adenomyosis, conservative treatment was initiated in 124 patients. . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 53 75 18 23 1 Number of cases GnRHaprogestinsLNG-IUS OCs DNZ 0 20 40 60 80 Figure.3 Number of cases treated with hormone therapy. GnRHa: gonadotropin releasing hormone agonist, LNG-IUS: levonorgestrel- releasing intrauterine systems, OCs: Oral contraceptives, DNZ: danazol. Y-axis shows the number of cases. . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint 0 50 100 150 0 10 20 30 40 50 Time (month) Cumulative Hysterectomy Rate (%) Cumulative Hysterectomy Rate Treatment period Cumulative Hysterectomy Rate 0 10 20 30 50 40 0 50 63 100 150 (months) (%) 32.4 Treatment period (month) 12 36 63 84 108 132 Continuing Conservative treatment number 87 54 26 15 11 2 Total number of Hysterectomy 13 23 28 0 0 0 Figure.4 Cumulative hysterectomy rate. Kaplan-Meier analysis of the treatment period. The X-axis is the duration of treatment and Y-axis is the cumulative hysterectomy rate. The cumulative hysterectomy rate was 32.4% and reached a plateau after 63 months. The median treatment period was 28 months (1–132 months). . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint Age Group A : 5 Group B : 23 47 or older46 or younger Parity 38 or younger 0 or 1 Age 2 or more Group A : 28 cases Group B : 26 cases Group A : 3 Group B : 0 Group A : 8 Group B : 23 Group A : 20 Group B : 3 Group A : 0 Group B : 2 Group A : 20 Group B : 1 39 or older Figure.5 Decision tree analysis. Group A; cases that required hysterectomy, Group B; cases that continued conservative treatment of adenomyosis. Accuracy: 77.8% . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint Table 1 Clinical characteristics of the 124 patients undergoing conservative treatment for adenomyosis n=124 Age 41 (24-53) Parity 1 (0-3) Gravida 1 (0-6) Size of Uterus Long axis diameter (cm) 9.65 (63-177) Short axis diameter (cm) 6.65 (35-129) Transvers diameter (cm) 6.75 (28-142) Muscle layer thickness (cm) 3.9 (13-88) Type of Adenomyosis Type Ⅰ 33 (26.6%) Type Ⅱ 28 (22.6%) Type Ⅲ 3 (2.4%) Type Ⅳ 60 (48.4%) Type of Treatment Hormonal Therapy 117 (94.4%) Adenomyomectomy 3 (2.4%) Hormonal & Adenomyomectomy 4 (3.2%) . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint Table 2 Comparison of clinical characteristic of at baseline between cases failed conservative therapy (Group A) and continued uterine conservative therapy (Group B). *the Mann-Whitney U test, †χ2 test Group A (n=28) Group B (n=26) P value Age† 43 (33-53) 37 (27-46) <0.001 Gravida† 2 (0-6) 0 (0-3) <0.001 Parity† 2 (0-3) 0 (0-2) <0.001 Multipara† 23 (82.1%) 11 (42.3%) <0.001 Long axis diameter (cm)* 11.1 (76-177) 9.0 (64-130) <0.001 Short axis diameter (cm)* 7.7 (47-129) 6.0 (35-99) 0.002 Transvers diameter (cm)* 8.0 (41-142) 6.6 (37-92) 0.012 Muscle layer thickness (cm)* 4.6 (25-73) 3.5 (13-64) 0.018 Type Ⅳ adenomyosis* 18 (64.3%) 10 (38.5%) 0.059 Another of endometriosis* 11 (39.3%) 13 (50.0%) 0.766 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprintthis version posted March 6, 2021. ; https://doi.org/10.1101/2021.03.03.21252870doi: medRxiv preprint