Clinicopathological characteristics and imaging findings to identify adenomyosis‐related symptoms

Abstract Purpose The study aims to identify the clinicopathological risk factors and magnetic resonance (MR) imaging findings for adenomyosis‐related symptoms, including menorrhagia, dysmenorrhea, and infertility. Methods This was an observation‐based cross‐sectional study using data from the adenomyosis cohort study. The authors evaluated the clinicopathological variables and various MR imaging findings. Results Two hundred twenty patients with histologically confirmed adenomyosis were included in this study. Multivariate analysis showed that a middle/retroflexed uterus and adenomyosis lesions of 21 mm or more were significant independent predictors of dysmenorrhea. The history of dysmenorrhea and the maximum length from the cervix to the uterine fundus ≥103 mm were independent risk factors of menorrhagia. One of the key factors associated with non‐infertility included the absence of deep infiltrating endometriosis (DIE) and/or superficial peritoneal disease (SUP). Conclusions This study identified clinicopathological risk factors and imaging findings associated with adenomyosis‐related symptoms. The maximum length from the cervix to the uterine fundus and adenomyosis lesion thickness are independent predictors for the presence of menorrhagia and dysmenorrhea, respectively. Infertility may be associated with the coexistence of endometriosis rather than adenomyosis itself. This result is from an analysis of a small number of infertility patients and requires further study.

pain and infertility, [5][6][7] but some women are asymptomatic. 2,3 Many researchers are interested in why patients have different symptoms.
Menorrhagia and metrorrhagia are the main clinical manifestations. 8 In many cases, both symptoms occur simultaneously. 7 In clinical practice, physicians experience that diffuse adenomyosis causes more severe menstrual symptoms compared with focal localized adenomyosis. 9 In addition, the disease has a potential negative impact on female fertility. 8 Accumulating evidence has shown that adenomyosis is associated with the risk of pregnancy outcome and obstetric complications. 10,11 Many patients with adenomyosis also have endometriosis, so it is difficult to determine whether adenomyosis is the only cause of infertility. 12 At this time, no clinicopathological characteristics and imaging findings associated with infertility have been identified. Adenomyosis is a heterogeneous group of conditions that include a range of clinical presentations, and its biological behavior remains incompletely understood. 6 In particular, the prediction of the onset of infertility is challenging due to their heterogeneity and various confounding factors.
Therefore, the study aims to identify the clinicopathological risk factors and imaging findings for adenomyosis-related symptoms, including menorrhagia, dysmenorrhea, and infertility in patients who were histologically diagnosed with adenomyosis in a single university hospital.

| Patient selection and data collection
The study was approved by the medical ethics committee of the Nara Medical University (reference nos. 541, 951, and 2295). Written informed consent was obtained from each patient.
A single-center prospective cohort (DoG-NaMe) study was conducted by collecting data from patients admitted to the Department of Gynecology, Nara Medical University Hospital, Kashihara, Japan, from January 2008 to December 2020. The DoG-NaMe study consists of an endometriosis cohort, an adenomyosis cohort, and an ovarian cancer cohort. Women scheduled for surgery were primarily enrolled. We performed an observational cross-sectional study using data from the adenomyosis cohort study. Participants underwent surgery or active surveillance (including some hormone therapy) to manage adenomyosis. Indications for surgical treatment are progressive anemia (8. (7) incomplete data. Women with a history of hormone therapy were not excluded.
All participants were recommended to undergo MRI after routine transvaginal ultrasonography (TVS) for preoperative evaluation of adenomyosis. MRI scanning was performed except during the menstrual phase. TVS was performed by experienced operators with a special interest in gynecological diseases with a single ultrasound system (Voluson E8; GE Healthcare, Tokyo, Japan) using a transvaginal transducer (5-7.5 MHz). MRI was obtained on a 3T system using T1W and T2W sequences (Magnetom Verio, Siemens Healthcare, Erlangen, Germany). The protocol of our MRI examination was performed as described previously. 13 15,16 based on the affected area and the degree of myometrial infiltration. The intrinsic type is defined as adenomyosis that occurs in the uterine inner layer without affecting the outer structures of the myometrium. The extrinsic type is defined as adenomyosis that occurs in the uterine outer layer without affecting the inner structures. If either of the two gynecologists diagnosed the patient as neither intrinsic adenomyosis nor extrinsic adenomyosis, she was classified as "unclassifiable or unidentifiable."

| Definition of adenomyosis-related symptoms: dysmenorrhea, menorrhagia, and infertility
A detailed definition of dysmenorrhea and menorrhagia can also be found in Reference 14. All patients were divided into two groups in terms of the presence and severity of dysmenorrhea and menorrhagia. In this study, "moderate pain" and "severe pain" were classified as having dysmenorrhea using the Numeric Rating Scale (NRS-11). 17 Menorrhagia was categorized according to the Mayo Clinic definition. 18 There were 2 types of patients in the infertility group: Patients who failed to achieve a clinical pregnancy following ≥12 months of regular unprotected sexual intercourse 19 and those who have already been treated at fertility hospitals. When a woman became pregnant with common fertility treatments such as timed intercourse, she was classified as "not infertile". To better clarify the risk factors, the definition of infertility in this study was limited to patients who underwent in vitro fertilization. Male factor infertility was excluded.
Women who were unmarried, divorced, or did not wish to become pregnant were excluded from this analysis.

| Quantification of serum CA125
Blood samples were obtained from all participants to determine serum/plasma levels of hemoglobin and CA125 within 4 weeks prior to surgery. Blood sampling was performed except during the menstrual phase. Samples were centrifuged at 1500×g for 10 min at 4℃, separated into serum and plasma, and stored at −20℃. Serum CA125 concentrations were determined using an electrochemiluminescence Elecsys immunoassay (ECLIA) (Roche Diagnostics, Salzburg, Austria) at Nara Medical University Hospital. SPSS 25.0 (SPSS Inc., Chicago, IL) statistical software was used for the statistical analysis. The data are presented as mean ± SD or median (range) for the continuous and the categorical variables, respectively. The normality of the data was examined using the Shapiro-Wilk test, and based on the results, differences in parameters between the two groups were analyzed using a parametric or nonparametric test. Continuous variables were compared with Student's t test or Mann-Whitney U test if the variables were not normally distributed. Categorical variables were compared using chi-square test. Receiver operating characteristic (ROC) analysis and the area under curve (AUC) were used to identify the sensitivity and specificity of each parameter cutoff point. The Youden index was used to determine optimal cutoff values. A multivariable regression model was employed after testing for multicollinearity. The multicollinearity relationships between pairs of variables were examined using a correlation test. The existence of multicollinearity was determined by high values of correlation coefficient (>0.7). Univariate and multivariate logistic regression models were used to evaluate these relationships between the parameters and the risk factors of adenomyosis-related symptoms; then, odds ratio (ORs) and 95% confidence intervals (CIs) were calculated. Differences with p < 0.05 were considered statistically significant.  Figure 1 illustrates the detailed selection process for study design.

| RE SULTS
Of the 220 women, 68 and 152 were painless + mild (described as "Absence") and moderate + severe (described as "Presence") for the severity of dysmenorrhea, respectively. The prevalence of patients with dysmenorrhea was 69.1% (n = 152). To identify the potential risk factors of the presence of dysmenorrhea, Youden index was measured to determine the optimal cutoff threshold. ROC curves were created to identify the best cutoff value, sensitivity and specificity, area under the curve (AUC), and p-value of each variable that affects dysmenorrhea ( Table 1). The optimal cutoff values computed using the ROC were 43 years for age, 30 mm for the thickest myometrium [either], 57 mm for the thickest myometrium  surgery ≥43 years (p = 0.001), CA125 ≥30.5 U/ml (p = 0.005), and coexistence of OMA (p = 0.032) were significantly more common in women with dysmenorrhea than in those without ( Table 2). The median age of patients at surgery was 3 years younger in the dysmenorrhea group than in the painless group (42.9 ± 5.5 years vs. 45.3 ± 5.6 years; p = 0.003). We investigated the impact of the adenomyosis phenotype on dysmenorrhea, but there was no significant difference in the proportion of adenomyosis phenotype between the two groups (p = 0.371). Multivariate analysis revealed that only the midline/retroflexed uterus and the thickest adenomyosis lesion ≥21 mm were independent risk factors that affect dysmenorrhea with the ORs of 7.494 (p = 0.026) and 4.332 (p = 0.047), respectively (Table 2).
Next, of the 220 patients, 130 (59.1%) women had menorrhagia. Table 3 Table 4. There were no significant differences in terms of other variables between the two groups. Multivariate analysis indicated that the presence of dysmenorrhea and maximum length from cervix to uterine fundus ≥103 mm were independent potential risk factors for the presence of menorrhagia, with the respective ORs of 6.668 (p < 0.001) and 2.711 The optimal cutoff values were 41 years for age, 1.5 for gravidity, and 1.5 for parity ( Table 5). The absence of a history of infertility was significantly associated with older age (p < 0.001), less frequent OMA (p = 0.004) and DIE/SUP (p = 0.008), a higher gravidity (p = 0.015), and a higher parity (p = 0.021) by univariate analysis ( Table 6). The multivariate analysis revealed that 1) women with adenomyosis over 41 years were associated with the absence of infertility (p = 0.003), and 2) the absence of DIE and/or SUP was associated with not being infertile (p = 0.026) ( Table 6).

| DISCUSS ION
We attempted to identify the potential clinicopathological risk factors and imaging findings affecting the adenomyosis-related symptoms, including pain, bleeding, and infertility. The selection of the study population was based on strict inclusion and exclusion criteria.
In our cohort study, the mean age at surgery was 43.  35 The authors considered that in patients with intrinsic adenomyosis, microvessels arising from adenomyotic lesions continue to the eutopic endometrium, leading to heavy bleeding. 35 It is easy to imagine that the longer the uterus with intrinsic adenomyosis, the more severe the menorrhagia. 24,34 Our data could not conclude that patients with intrinsic adenomyosis had more bleeding than patients with extrinsic adenomyosis. In this study, the presence of dysmenorrhea had a profound effect on menorrhagia, with an odds ratio of 6.668 (Table 4), which may negate the effects of other predictors, including the adenomyosis phenotype.
Third, many researchers believed that patients with adenomyosis have significantly lower pregnancy and implantation rates, 36,37 and higher miscarriage rates, 38  there was no statistical difference in pregnancy rates between patients with localized adenomyosis and diffuse adenomyosis. 36,40 There is still limited evidence as to whether infertility depends on the phenotype of adenomyosis or is secondary to the associated endometriosis. Furthermore, even considering multicollinearity, the absence of endometriosis (DIE and/or SUP, or OMA) was found to be an independent factor predicting non-infertility. Infertility may be associated with the coexistence of endometriosis rather than adenomyosis itself.
Finally, the strength of our study is the analysis of data from a

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets generated during the current study are available from Hiroshi Kobayashi on reasonable request.

PATI E NT CO N S E NT FO R PU B LI C ATI O N
Written informed consent was obtained from each patient. Human rights statements and informed consent: All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1964 and its later amendments.