Accuratezza di nuovi marker ecografici nella diagnosi di adenomiosi

3

5 Patients and Methods 7

10

13 Tables 17 Image 21

22 3

Objective : to evaluate the diagnostic accuracies of well-kno wn sonographic markers of adenomyosis and of two innovative ones, the question mark sign and the transvaginal ultrasound uterine tenderness .

78 patients scheduled for hysterectomy for uterin e benign diseases underwent preoperative transvaginal ultrasonography to evaluate the criteria of sonographic diagnosis of adenomyosis as reported by consensus statement MUSA. Adenomyosis was diagnosed in presence of two or more of the following parameters: asymmetry of the uterine wall s, hyperechoic striae, anechoic myometrial cysts, hyperechogenic islands, echogenic subendometrial lines and buds, interruption/irregularities of the junctional zone and translesional vascular flow. In addition the question mark sign and the transvaginal ultrasound uterine tenderness were evaluated, the first being the longitudinal section of the uterus with a morpholog y similar to a question mark and the other being the dynamic ultrasound evaluati on of uterine tenderness by the pressure of the transvaginal probe. Sonographic features were compared with histological examination.

the prevalence of adenomyosis in the sample is 33 .3%. Sensitivity, specificity, positive and negative predictive values and accuracy of transvaginal ultrasound in the diagnosis of adenomyosis are 83%, 96%, 91%, 89% and 92%. Asymmetry, hyperechoic striae and interruption of t he junctional zone were the 4 most accurate markers for the diagnosis of adenomyo sis. Myometrial heterogeneity was the most frequently encountered f eature (100%), but showed a low specificity (7%). The question mark sign and the transvaginal ultrasound uterine tenderness showed sensitivity, specificity, positive and nega tive predictive values and accuracy of 41%, 96%, 83%, 77 %, and 69% and 69%, 65%, 66%, 81% and 67% respectively.

the sonographic markers proposed by consensus sta tement MUSA were confirmed accurate in the diagnosis of adenomyosis in our sample. The question mark sign and the transvaginal ultrasound uterine tenderness showed good diagnostic capacities and may be a usef ul complement in the sonographic diagnosis of adenomyosis. 5

Adenomyosis is a benign condition of the uterus def ined by the presence of endometrial glands and stroma within the myometrium . Adenomyosis affects around 20% of women during their fertile age and ma y be associated to dysmenorrhea, menorrhagia and infertility 1. A diagnosis of certainty can be posed only by histo logical examination. Several studies showed that transvaginal sonography (TVS) can be considered the first-line imaging modality for studying adenom yosis, because it is as sensitive and as specific as magnetic resonance, 2–6 nevertheless univocal ultrasound parameters for the diagnosis of adenomyo sis are still lacking. 7 Recently the MUSA (Morphological Uterus Sonographic Assessment) consensus statement proposed terms, definitions and measurements that may be used to describe and report the sonographic feature s of the myometrium using gray-scale sonography, color/power Doppler and thre e-dimensional ultrasound imaging, with particular regard to two conditions: adenomyosis and fibroids. 8 Even if many ultrasound features are supposed to be associated with adenomyosis, the diagnostic weight of each one is n ot clear and some features may be more relevant than others in order to formul ate a diagnosis. 8 A particular shape of the uterine rime, called the question mark sign, has been recently described as a typical sign of adenomyosis associated with deep 6 infiltrating posterior endometriosis 9,10 and it was deliberately not included in the MUSA statement. 8 Transvaginal ultrasound is a dynamic examination 11 , permitting to the operator to evaluate the tenderne ss of an examined anatomical structures by a gentle pressure of the probe. An en larged and tender uterus, painful at mobilization may suggest adenomyosis 12 . The TVS uterine tenderness , that is the tenderness of the uterus during the g entle pressure with the transvaginal probe, could be useful to rule out the presence of adenomyosis, often associated to painful uterine mobilization. The aim of our study is to establish the diagnosti c accuracy of the ultrasound features associated to adenomyosis accor ding to the MUSA statement and of two new markers, the question mark sign , evaluated independently from the presence of endometriosis, a nd the TVS uterine tenderness . 7 Patients and methods This is a cross-sectional observational study enrol ling all consecutive premenopausal women with a diagnosis of a benign ut erine pathology, diagnosed by ultrasound or by hysteroscopy, and sch eduled for hysterectomy from November 2014 to June 2016 in the Department o f Gynaecology and Human Reproduction Pathophysiology, Sant’Orsola Hos pital, University of Bologna. Postmenopausal women and those with a pre- surgical diagnosis of a reproductive tract cancer were excluded. A data sheet with most relevant information on eac h patient’s medical history was filled in: age, BMI, last menstrual per iod, gravidity and parity, previous pelvic surgery, previous diagnosis of endo metriosis, presence of dysmenorrhea, menorrhagia and dyspareunia. Transvaginal ultrasound examination was carried out using a 4–9-MHz probe with a three-dimensional (3D) facility (Voluson E8, GE Medical Systems, Zipf, Austria). All transvaginal ultrasound scans were pe rformed in a standardized fashion by a single operator with more than 7 years of experience (L.Z.). Photos, clips and 3D scans were saved and stored fo r further examinations. The study of the uterine corpus was carried out as indi cated by the MUSA statement. 8 In addition the sonographer evaluated subjectively the globular- 8 shape of the uterus, the tenderness of the uterus a t the gentle pressure of the transvaginal probe and the presence of the question mark sign (Figure 1) .9, 10 A 3D volume of each uterus was stored following the m ethod of Exacoustos 13 in order to evaluate the junctional zone (JZ). 8 The diagnosis of adenomyosis was posed when at least two of the ultrasound features studied were present. Each patient underwent laparoscopic, laparotomic o r vaginal hysterectomy according to her clinical condition wi thin one month from the ultrasound examination. In each case the whole uter us was sent to histological examination, except one for which morcellation was needed. All histopathological examinations were performed by the same pathologist, skilled in gynaecologic pathology and blinded to the ultrasound findings. For each uterus a series of samples were taken, including all the wall from the serosa to the endometrium. Of these, at le ast three samples were taken both from the posterior and from the anterior wall. The diagnosis of adenomyosis was posed if endometrial stroma and gla nds were present into the myometrial layer. Adenomyosis was reported as diffu se or focal and evaluated by the grade of invasion: limited to the internal half of the myometrium (M1) or full-thickness (M2). In case of doubt, an immunoche mical test with CD10 antibodies was performed in order to highlight the ectopical endometrium. 14 For the purpose of this study, only the presence or abs ence of adenomyosis was considered, but not the depth of infiltration. 9 All the data were recorded in an electronic databa se. Patients were divided into two groups, according to the presence or absence of adenomyosis at the histological examination. Means and standard deviations were calculated for the continuous variables, using the Student's T -test. Relative frequencies were calculated for the categorical variables using the chi-squared test or Fisher’s exact test. P value less than .05 was cons idered statistically significant. In order to compare gravidity and parity of the two groups, each one was divided into three classes, considering the number of pregnancies and deliveries. Agreement between TVS and histological diagnosis was measured with Cohen’s Kappa coefficient. Sensitivity, specif icity, positive (PPV) and negative (NPV) predictive values, positive (LH+) and negative (LH-) likelihood ratios and accuracy (area under curve ROC) of each TVS variable were calculated. Analyses were performed by using the St atistical Package for the Social Sciences version 16 (SPSS Inc., Chicago, IL, USA). All the patients signed an informed consent and the study was approved by our local ethics committee (clinical trial ARC-ENDO n. 149/2014/O/Oss). 10

Seventynine patients were enrolled in this study an d one was excluded because the uterus was morcellated during laparoscopic hyst erectomy, due to its great size. The total number of patients considered in th e statistical analysis is 78. Forty/78 (51.2%) patients were operated on for leyomyomatosis, 24/78 (30.8%) for adenomyosis, 10/78 (12.8%) for uterine prolapse , 4/78 (0.5%) for fibroids with atypical ultrasound appearance. Hysterectomy w as performed through laparoscopy in 62/78 (79.5%) patients, laparotomy i n 6/78 (7.7%) and vaginal approach in 10/78 (12.8%) cases. Histology showed adenomyosis in 26/78 (33.3%) pati ents, among them 16/26 (61.5%) presented fibroids and 6/26 (23.1%) p resented adenomyomas. Among the 52/78 (66.6%) patients without adenomyosi s 41/78 (52.5%) had fibroids, 1/78 (1.2%) had a spindle-like cells neop lasia with myogenic differentiation and mitotic index <4 M/10 HPF, and 10/78 (12.8%) showed hysterocele not associated to myometrial pathology. Clinical features of patients are shown in Table 1 . There are no statistically significant differences between the t wo groups with and without adenomyosis for age, BMI, gravidity and parity. Stu dent's T-test and linear regression shows an inverse correlation between the uterine volume calculated by ultrasound and the diagnosis of adenomyosis. Nev ertheless, by dividing the 11 two populations into quartiles of volumes, no stati stically significant difference between the two groups can be found. Percentage frequencies of symptoms and association with previous pelvic surgeries or with endometriosis are summariz ed in Table 2. All the diagnosis of endometriosis nodules suspected by TVS were confirmed by histology. Menorrhagia was significantly more frequ ent in patients with adenomyosis. TVS diagnosed adenomyosis in 22/78 (28.2%) patient s: in 20/22 (90.9%) cases ultrasound diagnosis was confirmed by the pathologist, while 2/22 (9.1%) cases were false positives. Among the 5 6/78 (71.8%) patients without ultrasound features of adenomyosis, 4/56 (7 .1%) were false negatives, while 52/56 (92.9%) were true negatives. TVS diagnosed adenomyosis in 20/26 (76.9%) patients positives at histological examinat ion, with sensitivity, specificity, PPV, NPV 83%, 96%, 91% e 89% respectiv ely. Positive and negative likelihood ratio were 20 and 0.24. Global accuracy of TVS is 92.3%. Kappa analysis showed a good accordance between his tology and TVS (kappa=0.760). Table 3 shows the statistical significance of each ultrasound feature included in the study according to presence/absence of adenomyosis at histology. Hyperechoic islands and subendometrial l ines and buds were not present in the examined sample. Table 4 shows diagn ostic capacities of each ultrasound marker. Heterogeneous myometrium showed the highest sensitivity and PPV (both 100%). Most specific markers were JZm ax ≥8mm, fan-shaped 12 striations and question-mark sign (respectively 99%, 96% and 96%), with PPV respectively 100%, 88% and 83%. 13

This is the first study strictly applying the MUSA indications 8 to prospectively validate the importance of each of these ultrasound features in the diagnosis of adenomyosis in a sample of 78 hysterec tomies. In addition it showed two new ultrasound markers for the diagnosis of adenomyosis: the question mark sign and the TVS uterine tenderness , showing an accuracy of 69% and 67% respectively. In this study 2D, 3D and power Doppler ultrasound features were associated in order to diagnose adenomyosis, obtaining diagnos tic capacities superior than Kepkep et al. 15 and similar to those obtained in other studies, wh ich report sensitivity up to 89% and specificity up to 100%. 16-18 In accordance with Bazot et al. 18 TVS is very specific, but prone to produce false n egatives, which is the best condition for a test aiming to diagnose a beni gn pathology. In addition, our data show that TVS diagnostic capacity is reduced i n the presence of comorbidities, as it was already demonstrated 3,18 : all the diagnostic mistakes (4 false negatives, 2 false positives) were made in pa tients affected by fibroids. A recent meta-analysis of 14 trials and 1985 particip ants reported sensitivity and specificity of ultrasound-diagnosed adenomyosis to be as high as 82.5 and 84.6%, respectively 19 . Our data showed a similar sensitivity but a great er 14 specificity, maybe due to the presence of two new s onographic markers of endometriosis. The q uestion mark sign has been recently proposed by our group as a marker of adenomyosis strongly associated with deep infiltrating posterior endometriosis. 10 In the present study the question mark sign showed to be a marker of adenomyosis independent from the presence of endometriosis. Its strong association with adenomyosis is in contrast with MUSA consensus statement. 8 The question mark sign showed also great specifici ty (96%) and PPV (83%) with the best positive likelihood ratio among 2D ultrasound features. In this sample, only 4 out of the ten patients affe cted by deep infiltrating posterior endometriosis showed an associated question mark sign and Fisher’s exact test excluded a correlation (p=0.245). These results suggest that question mark sign might have a wider application in diagnosing adeno myosis than previously thought. As far as we are aware this is the first prosp ective study proposing TVS uterine tenderness as a marker of adenomyosis, showing a NPV of 81% an d an accuracy of 67.3%. Original descriptions of adenomy osis reported an association between the disease and “a great deal o f pain” 20 . Several later studies reported similar findings 21-23 , but others have not shown significant differences in the prevalence of adenomyosis in wom en with and without a history of pain 24-26 . One possible confounder in the interpretation of pain could be the coexisting presence of endometriosis, which is a common cause of pain in women of reproductive age. We believe that the u se of TVS, as a dynamic 15 examination 11 , permits to show if the pain is related to the gen tle pressure and mobilization of the uterus and permits as well to f ind the tenderness related to other location of endometriosis, if present. Among the ultrasound features of adenomyosi s the most specific were JZmax ≥8mm, fan-shaped striations and question-marked sign . Our results are comparable to previous studies for the high specifi city of fan-shaped striation 15- 18 and of myometrial cysts 27 , while Jzmax in our study showed better values than previously shown 27 . Heterogeneous myometrium is once again the most sensitive marker, but with very low specificity 18 . The main problem with the use of histology for the diagnosis of adenomyosis i s the heavy selection bias incurred 28 , indeed we had a very high percentage of leyomyomato sis, typically associated to heterogeneous myometrium at TVS. Prevalence of adenomyosis in the sample is 33.3 %, which is consistent with the Literature, where a mean value of 20-30% is reported in patients undergoing hysterectomy for various indications 29-31 . Differently from Literature 5,6,8 an association between increased uterine volume and adenomyosis was not found. Regarding thi s, it should be taken into account that patients without adenomyosis were ofte n affected by leyomyomatosis, which also increases uterine volume . Nevertheless, Exacoustos et al. demonstrated a significant correlation between dec reased uterine volume and adenomyosis, in comparison with uterus without fibroids 13 . Several strengths add power to this study: the use of histological confirmation of the diagnosis, the fact that all ul trasound scans were performed 16 using top-of-the-range equipment by a single operat or, thereby minimizing interobserver variability and the fact that all the demographic, ultrasound and anamnestic data were collected prospectively. In pa rticular it is remarkable that all the diagnosis of endometriotic nodules suspected by TVS were confirmed by histology. Moreover the choice of using wide inclus ion criteria reduced the selection bias and allowed to evaluate adenomyosis in presence of numerous potentially confounding variables, as fibroids, tha t often reduce diagnostic accuracy 3,18 . Wide inclusion criteria are also a potential weakne ss of this study, as confounding factors, such as fibroids and hormon al treatments prevented some features from being detectable in several pati ents. Another main

of this study is the only inclusion of p atients undergoing hysterectomy, creating a selection bias, as patient s who chose surgery are more symptomatic than those who do not. This study confirms TVS diagnostic accuracy in diagnosing adenomyosis, reinforcing TVS role as a first-line exam for its r eliability, safety and cheapness. Comparing the ultrasound features considered curren tly to be typical of adenomyosis in a sample full of confounding factors demonstrates their validity even in less selected patients, providing an update d and realistic idea of TVS diagnostic capacities that could be applied in ever yday clinical practice. The two new proposed features, the question mark sign and the TVS uterine tenderness, showed promising results and might prove to be use ful for the diagnosis of adenomyosis. Further prospective studi es are needed in order to prove their efficacy in wider samples. 17 Table 1. Population characteristic of 78 premenopausal patie nts according to presence/absence of adenomyosis at histology. Data are presented as mean ± standard deviation, or as % frequencies. Adenomyosis at histology Characteristic Yes ( n=26) No ( n=52) P Age (years) 48.2 ± 3.9 47.1 ± 3.9 n.s. BMI 24.6 ± 1.1 25.5 ± 1.0 n.s. Parity 0 38.5% 23.1% n.s. 1 23.1% 30.8% n.s. >1 38.4% 46.1% n.s. Gravidity 0 30.8% 19.2% n.s. 1 23.1% 26.9% n.s. >1 46.1% 53.9% n.s. Uterine volume 230 ± 189 295 ± 306 n.s. 18 Table 2. Clinical symptoms and association with previous pe lvic surgeries and endometriosis according to presence/absence of aden omyosis at histology. Data are presented as frequencies. Adenomyosis at histology Yes (n=26) No (n=52) P Dysmenorrhea 77% 50% n.s. Dyspareunia 46% 31% n.s. Menorrhagia 85% 50% 0.045 Previous pelvic surgery 46% 46% n.s. Presence of endometriosis 30% 15% n.s. 19 Table 3. Ultrasound features according to presence/absence of adenomyosis at histology. Data are presented as n(%). N.e. not evaluable. Adenomyosis at histology Features Yes (n=26) No (n=52) N.e. P Globular shape 10 (77%) 14 (54%) 0 n.s. Heterogeneus myometrium 13 (100%) 24 (92%) 0 n.s. Fan-shaped striations 7 (54%) 1 (4%) 0 0.001 Myometrial cysts 4 (31%) 2 (8%) 0 n.s. Ill-defined interface 11 (85%) 11 (42%) 1 0.037 Question mark sign 5 (38%) 1 (4%) 2 0.005 Walls asymmetry 8 (62%) 2 (8%) 20 0.001 TVS uterine tenderness 9 (69%) 9 (35%) 0 0.044 Doppler 4 (31%) 2 (8%) 2 0.011 JZ max ≥ 8 mm 4 (31%) 0 (0%) 13 0.015 ∆ JZ ≥ 4 mm 6 (46%) 2 (8%) 13 0.027 JZ interruption 7 (54%) 2 (8%) 13 0.008 20 Table 4. Diagnostic capacities of each ultrasound features associated to the diagnosis of adenomyosis. Feature Sensibility Specificity PPV NPV LR+ LR- Accuracy Globular shape 77% 46% 42% 80% 1.43 0.5 61.5% Asimmetry 80% 70% 72% 78% 2.67 0.29 78.9% Heterogeneous myometrium 100% 7% 35% 100% 1.08 0 53.9% Ill-defined interface 85% 56% 50% 88% 1.9 0.27 70.3% Fan-shaped striations 54% 96% 88% 81% 14 0.48 75.0% Myometrial cysts 30% 92% 67% 73% 4 0.75 61.5% Question-mark sign 41% 96% 83% 77% 10.42 0.61 68.8% TVS uterine tenderness 69% 65% 66% 81% 2 0.47 67.3% JZ max 40% 99% 100% 73% 4 0.60 70.0% JZ interruption 70% 88% 78% 82% 5.64 0.35 78.8% ∂JZ 60% 87% 75% 76% 4.5 0.46 73.8% Doppler 55% 88% 66% 82% 4.72 0.51 71.5% 21 Figure 1. Transvaginal sonography longitudinal section of a uterus showing the question mark sign which is described when the corpus uteri is flexed backwards, the fundus uteri is facing the posterior pelvic compartment and the cervix is directed frontally towards the urinary bl adder with the endometrial rhim resembling a question mark sign (a, b). Schema tic drawing of the question mark sign (c). a c b 22

1. Naftalin J, Hoo W, Pateman K, Mavrelos D, Hollan d T, Jurkovic D. How common is adenomyosis? A prospective study of p revalence using transvaginal ultrasound in a gynaecology clinic. Hu m Reprod. 2012; 27(12):3432-3439. 2. Dueholm M, Lundorf E, Hansen ES, Sorensen JS, Le dertoug S, Olesen F. Magnetic resonance imaging and transvaginal ultrasonography for the diagnosis of adenomyosis. Fertil Steril. 2001;76(3):588-594. 3. Bazot M, Cortez A, Darai E, et al. Ultrasonograp hy compared with magnetic resonance imaging for the diagnosis of ade nomyosis: correlation with histopathology. Hum Reprod. 2001;16(11):2427-2433. 4. Tahlan A, Nanda A, Mohan H. Uterine adenomyoma: a clinicopathologic review of 26 cases and a review o f the literature. Int J Gynecol Pathol. 2006;25(4):361-365. 5. Graziano A, Lo Monte G, Piva I, et al. Diagnosti c findings in adenomyosis: a pictorial review on the major concer ns. Eur Rev Med Pharmacol Sci. 2015;19(7):1146-1154. 6. Sakhel K, Abuhamad A. Sonography of adenomyosis. J Ultrasound Med. 2012;31(5):805-808. 23 7. Gordts S, Brosens JJ, Fusi L, Benagiano G, Brose ns I. Uterine adenomyosis: a need for uniform terminology and con sensus classification. Reprod Biomed Online. 2008;17(2):244-248. 8. Van Den Bosch T, Dueholm M, Leone FPG, et al. Te rms, definitions and measurements to describe sonographic features o f myometrium and uterine masses: A consensus opinion from the Morpho logical Uterus Sonographic Assessment (MUSA) group. Ultrasound Obs tet Gynecol. 2015;46(3):284-298. 9. Di Donato N, Bertoldo V, Montanari G, Zannoni L, Caprara G, Seracchioli R. Question mark form of uterus: a simp le sonographic sign associated with the presence of adenomyosis. Ultras ound Obstet Gynecol. 2015;46(1):126-127. 10. Di Donato N, Seracchioli R. How to evaluate ade nomyosis in patients affected by endometriosis? Minim Invasive Surg. 2014;2014:507230. 11. Testa AC, Van Holsbeke C, Mascilini F, Timmerma n D. Dynamic and interactive gynecological ultrasound examination. U ltrasound Obstet Gynecol. 2009;34(2):225-9. 12. Alabiso G, Alio L, Arena S, Barbasetti di Prun A, Bergamini V, Berlanda N, Busacca M, Candiani M, Centini G, Di Ce llo A, Exacoustos C, Fedele L, Fuggetta E, Gabbi L, Geraci E, Imperiale L, Lavarini E, Incandela D, Lazzeri L, Luisi S, Maiora na A, Maneschi F, Mannini L, Mattei A, Muzii L, Pagliardini L, Perand ini A, Perelli F, Pinzauti S, Porpora MG, Remorgida V, Leone Roberti Maggiore U, 24 Seracchioli R, Solima E, Somigliana E, Tosti C, Ven turella R, Vercellini P, Viganò P, Vignali M, Zannoni L, Zullo F, Zupi E; Endometriosis Treatment Italian Club. Adenomyosis: What the Patient Needs. J Minim Invasive Gynecol. 2016;23(4):476-88 13. Exacoustos C, Brienza L, Di Giovanni A, et al. Adenomyosis: Three- dimensional sonographic findings of the junctional zone and correlation with histology. Ultrasound Obstet Gynecol. 2011;37(4):471-479. 14. Dr ăghici IM, Dr ăghici L, Cojocaru M, Gorgan CL, Vrabie CD. The immunoprofile of interstitial Cajal cells within adenomyosis/endometriosis lesions. Rom J Morphol Em bryol. 2015;56(1):133-138. 15. Kepkep K, Tuncay YA, Göynümer G, Tutal E. Trans vaginal sonography in the diagnosis of adenomyosis: Which f indings are most accurate? Ultrasound Obstet Gynecol. 2007;30(3):341-345. 16. Reinhold, C.; McCarthy, S.; Bret, M.; Mehio, A. ; Atri, M.; Zakarian, R.; Glaude, Y.; Liang, L.; Seymour RJ. Diffuse adenomyo sis: comparison of endovaginal US and MR imaging with histopatholog ic correlation. Radiology. 1996;199(1):151-158. 17. Atzori E, Tronci C, Sionis L. Transvaginal Ultr asound in the Diagnosis of Diffuse Adenomyosis. Gynecol Obstet Invest. 1996;42(1):39-41. 18. Bazot M, Daraï E, Rouger J, Detchev R, Cortez A , Uzan S. Limitations of transvaginal sonography for the diagnosis of ade nomyosis, with 25 histopathological correlation. Ultrasound Obstet Gy necol. 2002;20(6):605-611. 19. Meredith SM, Sanchez-Ramos L, Kaunitz AM. Diagn ostic accuracy of transvaginal sonography for the diagnosis of adenom yosis: systematic review and metaanalysis. Am J Obstet Gynecol. 2009;201:107.e1–107 20. Cullen TS. Adenomyoma of the Uterus. W.B. Saund ers: Philadelphia & London, 1908. 21. Benson RC, Sneeden VD. Adenomyosis: a reapprais al of symptomatology. Am J Obstet Gynecol 1958; 76: 1044– 1057;

1057–1061. 22. Bird CC, McElin TW, Manalo-Estrella P. The elus ive adenomyosis of the uterus revisited. Am J Obstet Gynecol 1972; 112: 583–593. 23. Emge LA. The elusive adenomyosis of the uterus. Its historical past and its present state of recognition. Am J Obstet Gynec ol 1962; 83: 1541– 1563. 24. Parazzini F, Vercellini P, Panazza S, Chatenoud L, Oldani S, Crosignani PG. Risk factors for adenomyosis. Hum Reprod 1997; 12: 1275–1279. 25. Bergholt T, Eriksen L, Jacobsen M, Hertz JB. Pr evalence and risk factors of adenomyosis at hysterectomy. Hum Reprod 2001; 16: 2418– 2421. 26. Weiss G, Maseelall P, Schott LL, Brockwell SE, Schocken M, Johnston JM. Adenomyosis a variant, not a disease? Evidence from 26 hysterectomized menopausal women in the Study of Wo men’s Health across the Nation (SWAN). Fertil Steril 2009; 1: 201–206. 27. Exacoustos C, Brienza L, Di Giovanni A, et al. Adenomyosis: Three- dimensional sonographic findings of the junctional zone and correlation with histology. Ultrasound Obstet Gynecol. 2011;37(4):471-479. 28. Mehasseb MK, Habiba MA. Adenomyosis uteri: an u pdate. Obstetrician & Gynaecologist 2009; 11: 41–47. 29. Azziz R. Adenomyosis: current perspectives. Obs Gynecol. 1989;16(1):221-235. 30. Parazzini F, Vercellini P, Panazza S, Chatenoud L, Oldani S, Crosignani PG. Risk factors for adenomyosis. Hum Reprod. 1997;12(6):1275-1279. 31. Vercellini P, Parazzini F, Oldani S, Panazza S, Bramante T, Crosignani PG. Adenomyosis at hysterectomy: a study on frequen cy distribution and patient characteristics. Hum Reprod. 1995;10(5):1160-1162.