Advances in non-invasive diagnosis of uterine adenomyosis: a narrative review

Adenomyosis is a benign gynecological condition characterized by the presence of endometrial tissue within the myometrium. Its estimated prevalence ranges from 5% to 70%, depending on the population studied and the diagnostic method used ( 1 , 2 ). It is more commonly diagnosed in women aged 40–50 years, but recent studies have shown an increased detection rate in younger women, mainly due to improvements in imaging techniques ( 3 ). Although recognized clinically, adenomyosis remains difficult to diagnose due to its variable presentation and similarity to conditions like endometriosis and fibroids ( 4 ). The pathogenesis of adenomyosis is still debated. The most accepted theories regarding the pathogenesis of adenomyosis are: (I) the invagination theory, where basal endometrial glands penetrate into the myometrium, triggering reactive hyperplasia of the latter; (II) the metaplasia theory; (III) the theory of myometrial invasion originating from deep endometriosis foci ( 2 , 5 ). The most common symptoms of adenomyosis include heavy menstrual bleeding (menorrhagia), severe menstrual cramps (dysmenorrhea), chronic pelvic pain, and, in some cases, infertility ( 6 ). Adenomyosis has also been associated with adverse obstetric outcomes, including an increased risk of miscarriage, preterm birth and pre-eclampsia ( 7 ). Studies suggest that the structural changes in the myometrium caused by adenomyosis may affect uterine contractility and placental function, contributing to these complications ( 4 - 8 ). However, many women may be asymptomatic, making the condition harder to diagnose based solely on clinical presentation. Historically, diagnosis required histology after (mainly) hysterectomy, an invasive approach typically reserved for cases where other diagnostic methods are inconclusive ( 9 ). The histological diagnosis of adenomyosis relies on identifying ectopic endometrial glands and stroma within the myometrium. It is challenging to define a precise cut-off due to the naturally irregular endometrial-myometrial interface and variable invasion depth. Operator-dependent criteria further complicate the assessment, often showing a poor correlation with clinical symptoms ( 10 ). Given the high prevalence of adenomyosis in reproductive-age women and associated symptoms like dysmenorrhoea and infertility, there is an urgent need for effective non-invasive diagnostic methods. Accurate and timely diagnosis, allowing fertility preservation and effective symptom management, is essential to improve patients’ quality of life and provide appropriate therapeutic options without resorting to invasive procedures ( 11 ). Recent advances in imaging, such as transvaginal ultrasound (TVUS) and magnetic resonance imaging (MRI), have enhanced non-invasive evaluation of adenomyosis. TVUS offers a cost-effective and accessible initial assessment, while MRI provides detailed and high-resolution images that can improve diagnostic accuracy ( 12 , 13 ). Advancements in imaging technology, like 3D ultrasound and MRI, are improving the early detection of adenomyosis, allowing for better patient management. 3D ultrasound enhances the visualization of the junctional zone (JZ), providing more detailed lesion information. MRI can detect early signs of adenomyosis, such as JZ thickening or small high-intensity foci in a low-intensity myometrial background, before significant uterine enlargement occurs. Despite challenges in diagnosis, such as variable imaging findings and a lack of standardized criteria, ongoing research and technological progress offer hope for more reliable early detection ( 14 ). New diagnostic techniques such as elastography and hysteroscopy are also gaining attention. Elastography measures tissue stiffness and may improve the ability to differentiate adenomyosis from other conditions. Hysteroscopy, traditionally used for direct visualisation of the uterine cavity, is now being investigated for its potential to visualise adenomyotic lesions and offer both diagnostic and therapeutic benefits ( 15 ). These techniques, their diagnostic accuracy and their clinical applications are discussed in detail in the following sections. Adenomyosis remains a complex and often under-diagnosed condition that has a significant impact on the reproductive health of women. Previous reviews have mainly focused on TVUS and MRI, with few systematic comparisons based on clinical trials using histological confirmation. Emerging techniques such as elastography and hysteroscopy with targeted biopsy have rarely been included. Efforts to correlate imaging findings with clinical symptoms are still in their early stages. A unified, standardised classification system is lacking. This review aims to address these gaps. This review analyzes clinical trial data on non-invasive diagnostic tools—TVUS, MRI, elastography, and hysteroscopy—compared to histology. It explores the benefits of combining imaging with minimally invasive biopsy and evaluates classification systems linking ultrasound features to symptom severity. The focus is on younger women prioritizing fertility preservation. The 2015 Morphological Uterus Sonographic Assessment (MUSA) consensus established sonographic criteria for adenomyosis and published a consensus on the terminology to be used when describing sonographic images of adenomyosis ( 16 ). The addition of color Doppler (CD) to TVUS improves diagnostic accuracy by providing insight into both myometrial structure and vascularization. This technique allows the detection of increased and irregular blood flow in the myometrium, which is a hallmark of adenomyosis, helping to differentiate it from other conditions such as uterine fibroids ( 17 ). An additional diagnostic contribution has come from the application of 3D TVUS, which has allowed better evaluation of the myometrial structure and the JZ, where irregular thickening and distortion are often indicative of adenomyosis. In addition, 3D imaging has proven beneficial in the identification of focal adenomyomas by providing precise visualisation of the lesions and their localisation within the myometrium, especially in the early stages of the disease or when the pathology is focal ( 18 ). A pilot study showed that inter-rater agreement using the MUSA criteria to describe ultrasound images of adenomyosis was poor for both highly experienced and moderately experienced graders, particularly for poorly defined lesions ( 19 ). In addition, inter-rater agreement for the diagnosis of adenomyosis between an expert and a non-expert rater showed good agreement for 2D TVUS images, but poor agreement for 3D TVUS images ( 20 ). In light of these findings, and in order to clarify the definitions of the various criteria, the MUSA criteria were revised in 2021 through a Delphi consensus ( 21 ). These criteria were divided into direct signs (indicative of endometrial gland invasion into the myometrium) and indirect signs (reflecting the reactive myometrial hypertrophy). ❖ Direct signs. Myometrial cysts. Hyperechoic islands within the myometrium. Linear striations in the myometrium. Interrupted JZ with distortion. ❖ Indirect signs. Asymmetrical myometrial thickening. Enlarged uterus. Heterogeneous myometrial texture. Increased vascularity in the affected area. Despite the MUSA consensus and the revised criteria standardizing the terminology and features to look for in the diagnosis of adenomyosis, a universally accepted classification system is still lacking. In 2018, Lazzeri et al. proposed a scoring system that demonstrated excellent interobserver agreement and also showed clinical relevance in follow-up assessments ( 22 ). In a double-blind study, the diagnostic agreement between two experienced sonographers was assessed by assigning a score from 1 to 4 for each type of disease considered. Finally, a total score was calculated to assess the extent of the disease: mild [1–3], moderate [4–6] and severe [>7]. There were no differences in agreement between the different levels of adenomyosis severity, except for adenomyoma score 4. This could be due to the small number of adenomas >40 mm in the patient group considered. In 2019, Van den Bosch et al. attempted to further classify adenomyosis based on its sonographic appearance and clinical presentation. This classification aims to provide a more structured approach to assessing the extent and type of adenomyosis, with the aim of guiding clinical management ( 23 ). The proposed classification distinguishes adenomyosis into different subtypes based on the distribution and severity of myometrial involvement: ❖ Diffuse adenomyosis: characterized by a widespread involvement of the myometrium with diffuse thickening and the presence of myometrial cysts throughout. ❖ Focal adenomyosis (adenomyomas): involves localized areas of adenomyosis, often forming nodular structures that can be mistaken for fibroids. ❖ Superficial adenomyosis: involves the outer myometrium or the subserosal region, often more challenging to detect via ultrasound but still significant in terms of symptomatology. ❖ Cystic adenomyosis: a rare form where large myometrial cysts are present, typically associated with severe pain and abnormal bleeding. This classification system not only assists in improving diagnostic clarity but also provides a foundation for individualized treatment plans based on the type and extent of adenomyosis. According to the most recent meta-analysis by Alcázar et al. , the pooled sensitivity and specificity of TVUS for detecting adenomyosis were 75% [95% confidence interval (CI): 63–84%] and 81% (95% CI: 60–92%), respectively. The positive likelihood ratio (LR+) was 3.9 (95% CI: 1.7–9), and the negative likelihood ratio (LR−) was 0.31 (95% CI: 0.21–0.47), confirming the utility of TVUS as a first-line diagnostic tool, although variability remains depending on operator expertise and patient characteristics ( 24 ). According to the classifications proposed by Kishi et al. ( 9 ) and Bazot et al. ( 25 ), adenomyosis features can be described based on the involvement of different myometrial layers, distinguishing between internal adenomyosis, external adenomyosis, and adenomyomas. These subtypes are not mutually exclusive and may coexist in the same patient, reflecting different manifestations of the disease. ❖ Internal adenomyosis: primarily affects the JZ and presents with direct MRI signs such as hyperintense myometrial foci and linear striations on T2-weighted images representing heterotopic endometrial glands. Indirect signs include localised or diffuse thickening of the JZ, with a threshold of 12 mm commonly used for diagnosis ( 22 , 23 , 26 ). ❖ Adenomyoma: as a mass-forming variant, adenomyomas appear as ill-defined T2 hypointense masses within the myometrium, often containing cystic or haemorrhagic foci. This subtype typically does not involve the JZ and may be intramural, submucosal or subserosal. ❖ External adenomyosis: involves the outer myometrium, sparing the JZ, and is often associated with endometriosis. It appears as a hypointense T2 mass with hyperintense or haemorrhagic foci, most commonly in the posterior myometrium. Several authors have proposed different MRI-based classifications of adenomyosis over the past two decades. However, a universally accepted standard has yet to be established. Gordts et al. ( 27 ) first identified three main patterns—JZ hyperplasia, adenomyosis, and adenomyoma. Kishi et al. ( 9 ) refined these into four subtypes, based on the involvement of different myometrial layers, while Gordts et al. ( 27 ) and Bazot et al. ( 25 ) further developed this by adding lesion characteristics such as affected area, lesion type (muscular or cystic), and volume. Kobayashi et al. ( 28 ) proposed a comprehensive MRI-based classification of adenomyosis, integrating key elements from previous classifications. This system includes five main parameters that help to stratify adenomyosis in terms of its extent, location, and associated conditions: ❖ Affected area: internal (affecting the JZ) or external (sparing the JZ, typically associated with endometriosis); ❖ Pattern: diffuse or focal; ❖ Size: the volume or extent of adenomyosis based on the proportion of the uterine wall involved (<1/3, 2/3 of uterine wall); ❖ Location: anterior, posterior, left-lateral, right-lateral, fundus; ❖ Concomitant pathologies: peritoneal endometriosis, ovarian endometrioma, deep infiltrating endometriosis (DIE), uterine fibroids, other uterine or pelvic pathologies. In the meta-analysis by Alcázar et al. ( 24 ), MRI demonstrated a sensitivity of 69% and a specificity of 80% for diagnosing adenomyosis, highlighting its diagnostic accuracy, particularly in differentiating adenomyosis from other uterine pathologies. Traditionally used for direct visualisation of the uterine cavity, hysteroscopy has become a tool in both the diagnosis and treatment of adenomyosis. Although hysteroscopy is not usually a first-line diagnostic technique, it offers unique advantages, particularly in specific cases of adenomyosis. Hysteroscopy allows direct visualisation of the endometrial surface and can detect subtle abnormalities that may indicate adenomyosis. According to Di Spiezio et al. ( 15 ) the following aspects are generally indicative of the pathological condition: ❖ Irregular endometrium with tiny openings seen on the endometrial surface; ❖ Pronounced hypervascularization; ❖ An endometrial “strawberry” pattern; ❖ Fibrous cystic appearance of intrauterine lesions (following 3–5 episodes of intramyometrial haemorrhage); ❖ Haemorrhagic cystic lesions assuming a dark blue or chocolate brown appearance. The ability to perform a targeted biopsy during hysteroscopy provides further diagnostic confirmation. To obtain an adequate biopsy using this approach, it is important to first take a sample that includes both the endometrium and the underlying myometrial layer. A second biopsy is then taken from the myometrial tissue only. When performing a resectoscopic biopsy, three key indicators are highly suggestive of adenomyosis: (I) irregular subendometrial myometrium, either spiral or fibrotic; (II) distortion of the normal myometrial architecture observed during resection; and (III) the presence of intramural endometriomas. In addition to its diagnostic capabilities, hysteroscopy also has a therapeutic role, particularly in the treatment of focal adenomyosis. Techniques such as: ❖ Adenomyomectomy: the surgical removal of localised adenomyotic nodules that can be visualised and removed during hysteroscopy. ❖ Endometrial ablation: in cases of diffuse adenomyosis, endometrial resection or ablation may help reduce symptoms such as abnormal uterine bleeding. Di Spiezio et al. ( 15 ) have highlighted the potential of hysteroscopy for both diagnosis and treatment, particularly for the management of focal adenomyosis through targeted intervention. They proposed the use of hysteroscopic enucleation for focal adenomyomas smaller than 1.5 cm and close to the endometrial cavity. This minimally invasive technique can be performed in an office setting using a mini-hysteroscope or resectoscope, although caution is required due to the lack of a clear cleavage plane for identification of healthy myometrial tissue. For adenomyotic nodules larger than 1.5 cm, resectoscopic treatment has been shown to be feasible, with resectoscopic excision also possible in an outpatient setting using a mini-resectoscope. Endometrial ablation is another option, mainly for women who do not wish to preserve their fertility. This involves removing the defective myometrial tissue under the endometrium. However, this is not suitable for deep diffuse adenomyosis as it may not relieve symptoms and may mask the development of deeper adenomyosis. Operative hysteroscopy is more appropriate for superficial adenomyosis or adenomyomas visible in the uterine cavity, where adeno-myomectomy can be performed similarly to myomectomy. Xia et al. presented data from 51 women who underwent hysteroscopic resection of adenomyosis, demonstrating the clinical feasibility and safety of the technique, with low recurrence rates of menorrhagia and dysmenorrhoea at 2 years ( 29 ). However, the procedure requires careful patient selection and expertise due to the limitations of the hysteroscopic equipment. If deeper cystic adenomyosis is identified preoperatively, the spirotome device has been shown to allow hysteroscopic access to cystic structures under ultrasound guidance, followed by further treatment with bipolar coagulation or resection ( 30 ). Although there is no universally accepted hysteroscopic treatment for adenomyosis, several methods have been proposed based on clinical experience: enucleation of focal adenomyomas smaller than 1.5 cm using a mini-hysteroscope ( 15 ), resectoscopic excision of larger nodules ( 29 ), endometrial ablation for diffuse superficial forms in non-fertility-seeking patients, and spirotome-assisted access to cystic adenomyosis under ultrasound guidance ( 30 ). However, the heterogeneity of approaches limits the ability to determine which method is superior. Elastography is an emerging technique that is gaining attention in the diagnosis of adenomyosis. It measures tissue stiffness and offers a complementary method to TVUS to differentiate between different uterine pathologies, such as adenomyomas and fibroids. As adenomyosis tends to make the myometrium stiffer than healthy tissue, elastography can identify areas of increased stiffness that may not be as clearly visible with conventional ultrasound. Preliminary studies, such as that by Stoelinga et al. ( 31 ) have shown that transvaginal elastography has the potential to enhance the characterisation of myometrial lesions and distinguish adenomyotic tissue from other conditions, such as fibroids, which have different stiffness characteristics. Shear wave elastography (SWE) is one of the most commonly used techniques to assess tissue stiffness. This technique provides quantitative measurements that help to identify areas affected by focal or diffuse adenomyosis. Pongpunprut et al. in their study of uterine elastography (UE) shear wave velocity (SWV), found that a cut-off point of 3.465 m/s was able to discriminate between uteri with and without adenomyosis with a sensitivity and specificity of 80%. However, no significant differences were observed between adenomyosis and fibroids ( 32 ). We present this article in accordance with the Narrative Review reporting checklist (available at https://gpm.amegroups.com/article/view/10.21037/gpm-24-52/rc ).

This work is presented as a narrative review, with the objective of summarizing and critically discussing the current evidence on non-invasive diagnostic techniques for adenomyosis, rather than conducting a systematic quantitative synthesis. Although some systematic elements have been adopted (i.e., defined inclusion criteria and a transparent search strategy), the nature of the studies and heterogeneity in the available data justified a narrative, rather than systematic, approach. The work complies with the narrative review checklist. A structured literature search was conducted in PubMed for studies published between January 2014 and September 2024, using the following term: ‘adenomyosis AND diagnosis’. Although the strategy used a basic keyword combination, we manually reviewed Medical Subject Headings (MeSH) terms during the screening phase to ensure the inclusion of relevant articles. The search was limited to English-language, full-text articles to ensure terminological consistency and readability. Table 1 will provide the search strategy summary. MRI, magnetic resonance imaging; TVUS, transvaginal ultrasound. Studies were selected based on the following Population/Patient/Problem, Intervention, Comparison, and Outcome (PICO) criteria: ❖ Population: women evaluated for suspected adenomyosis; ❖ Intervention: non-invasive diagnostic techniques (TVUS, MRI, hysteroscopy, elastography); ❖ Comparison: histopathological confirmation (where available); ❖ Outcome: diagnostic accuracy [sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV)]. Exclusion criteria included: reviews, meta-analyses, animal studies, editorials, guidelines, and studies without correlation to histopathology. Studies focusing only on fertility without diagnostic evaluation were also excluded. Two authors (R.G. and L.K.) independently screened titles, abstracts, and full texts. Disagreements were resolved through discussion or consultation with a third author. From the selected articles, the following data were extracted: authors, year, study design, location, patient characteristics [age, body mass index (BMI), parity], histological prevalence of adenomyosis, diagnostic technique used, and diagnostic accuracy metrics (sensitivity, specificity, NPV, PPV). Due to the heterogeneity of the study designs (prospective vs. retrospective, operator-dependent imaging, non-standardized criteria), a formal risk of bias assessment [e.g., Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2)] was not performed. This limitation, along with the variability in imaging interpretation and reference standards, further supports the choice of a narrative format for this review.

Figure 1 shows the literature search process. The selection of studies by title resulted in 156 studies. By evaluating the abstracts and full text, 12 studies were selected that met the inclusion and exclusion criteria. The process of the literature search (see Inclusion and exclusion criteria section and Table 1 ). The main characteristics of the included studies are presented in Table 1 . Of the 12 included studies, 8 were prospective and 4 were retrospective. No randomised controlled trials were identified. Four studies were conducted in Asia, six in Europe, one in the Americas and one in Africa. A total of 2,337 patients were included in the 12 selected trials. The mean age was 43.7 years, BMI 27 kg/m 2 , and parity 3.19. Seven trials used TVUS. One trial evaluated MRI, one evaluated elastography, and four trials evaluated both TVUS and MRI. All trials used histology as the reference standard. Where reported, outcome measures varied widely between the included trials. TVUS has a sensitivity ranging from 10.9% to 83.95% and a specificity ranging from 40% to 98.3%. CD has a sensitivity of 77.7% to 83.3% and a specificity of 72.7% to 78.5%. Only one study evaluated the use of 3D, which had a sensitivity of 69% and a specificity of 86%. MRI has a sensitivity of 29.7% to 91.6% and a specificity of 81% to 90%. The study that evaluated elastography showed a sensitivity of 89.7% and a specificity of 92.9%. Jain et al. ( 33 ) considered the presence of at least three ultrasound criteria and obtained a sensitivity of 77.78% and a specificity of 64.29%. Of the different criteria considered, heterogeneous myometrium had the highest sensitivity (80.56%) and myometrial cyst had the highest specificity. The addition of CD to 2D TVUS increased specificity from 64.2% to 78.5% and PPV from 84.8% to 90.3%. Zannoni et al. ( 34 ) showed that the most specific ultrasound criteria are junctional zone maximum thickness (JZmax) ≥8 mm, fan-shaped striations and “question mark” sign. Myometrial heterogeneity is the most sensitive marker (100%) but is associated with low specificity (7%). In the study by Rasmussen et al. ( 20 ), among the features assessed by 2D TVUS, uterine wall asymmetry and the presence of anechoic lacunae both had a sensitivity and specificity of over 60%. When two or more features of the 2D technique were combined with two or more features of the 3D technique, the specificity increased to 79%. Jain et al. ( 33 ) used a criterion of the presence of ≥3 features to diagnose adenomyosis, with a sensitivity of 91.7% and a specificity of 85.71%. The highest sensitivity of 97.22% was found for JZ thickness ≥12 mm and the highest specificity (100%) for hyperintense myometrial foci. The study by Tellum et al. ( 35 ) showed that the presence of an irregular or interrupted JZ was strongly correlated with the presence of adenomyosis (sensitivity 74% and specificity 83%), whereas a regular JZ was strongly correlated with the absence of adenomyosis (sensitivity 81% and specificity 75%). Details of the results of each study are shown in Tables 2-4 . BMI, body mass index; CS, cross sectional; E, endometriosis; HPE, histopathology examination; L, leiomyomas; MRI, magnetic resonance imaging; OB, observational study; PS, prospective study; RS, retrospective study; TA, transabdominal; TV, transvaginal; TVUS, transvaginal ultrasound. 3D, three-dimensional; CD, color Doppler; MRI, magnetic resonance imaging; NPV, negative predictive value; PPV, positive predictive value; TVUS, transvaginal ultrasound. † , subjectively evaluation; ‡ , 3D evaluation according to Exacoustos et al. 2011 ( 18 ). 2D, two-dimensional; 3D, three-dimensional; CD, color Doppler; JZ diff , represents irregularities of the JZ [is calculated as JZ max (all planes) − JZ min (all planes) ]; JZ, junctional zone; MRI, magnetic resonance imaging; N.A., not available; TVUS, transvaginal ultrasound; T2w, T2-weighted.

Our study confirmed the feasibility of non-invasive techniques for diagnosing adenomyosis, with an average sensitivity and specificity of 78.5% and 70.7% for TVUS and 64.8% and 87.5% for MRI, respectively ( 24 , 33 , 34 , 39 ). Among the specific ultrasound markers, a JZmax of 8 mm or greater, fan-shaped striations and the “question mark” sign were the most prominent, findings consistent with previous studies ( 12 , 34 ). Notably, heterogeneous myometrium remains the most sensitive marker, but with low specificity, highlighting the importance of combining multiple criteria for greater diagnostic accuracy ( 33 , 34 ). Moreover, TVUS allows dynamic, real-time evaluation of uterine tenderness; notably, pain elicited by gentle uterine mobilization can suggest adenomyosis ( 34 ). TVUS allows dynamic, real-time evaluation of uterine tenderness. The inclusion of histology as a gold standard in our study allowed a more accurate assessment of the sensitivity and specificity of these non-invasive techniques. In line with our findings, Shaikh et al. ( 43 ) demonstrated a sensitivity of 74.36% and specificity of 96.15% for TVUS using MRI as a reference, further supporting the diagnostic reliability of TVUS, especially in settings where MRI may not be available. Distinguishing adenomyosis from similar painful conditions, such as endometriosis, remains a significant diagnostic challenge. TVUS, as a dynamic examination, allows the examiner to detect pain on gentle pressure or mobilisation, providing additional information about possible coexisting endometriosis. Techniques such as TVUS and MRI provide an effective, fertility-sparing diagnostic approach with reliable accuracy. The role of hysteroscopy in the diagnosis of adenomyosis has been increasingly recognised, particularly in women with fertility problems who often undergo this procedure. Dakhly et al. evaluated the diagnostic role of two-dimensional TVUS and office hysteroscopy together with endomyometrial biopsy in identifying adenomyosis using hysterectomy specimens as the reference standard and found that combining endomyometrial biopsy with TVS significantly improved specificity to 89.23%, suggesting that targeted biopsies guided by imaging can improve diagnostic accuracy ( 40 ). Many patients undergo hysteroscopy primarily for infertility and recurrent pregnancy loss; however, assessing its diagnostic accuracy is challenging because these cases are rarely followed by hysterectomy. This finding is promising as it suggests a viable method of integrating histological confirmation—the traditional gold standard—with a more conservative, fertility-sparing approach. For younger patients for whom hysterectomy is not an option, this combination of TVUS and hysteroscopy with biopsy provides an accurate diagnosis and allows for tailored treatment strategies. In recent years, research on adenomyosis has increasingly focused on the relationship between clinical symptoms and ultrasound findings, laying the groundwork for a more integrated diagnostic approach. Martire et al. investigated this association in adolescents and found that dysmenorrhoea was associated with adenomyosis in the outer myometrium, whereas dyspareunia correlated with diffuse adenomyosis involving both inner and outer myometrial layers, mainly in the posterior uterine wall. Heavy menstrual bleeding was most often associated with presence of diffuse adenomyosis in the outer myometrium ( 44 ). Similarly, Naftalin et al. observed a positive correlation between the number of ultrasound features of adenomyosis and the severity of menstrual pain in premenopausal women. They proposed a grading system based on ultrasound features, with the aim of creating a standardised diagnostic score that correlates with clinical symptoms, thus supporting a more structured approach to assessing the severity of adenomyosis ( 41 ). Further studies by Alson et al. ( 45 ) and Puente et al. ( 26 ) highlight the clinical significance of adenomyosis in subfertile women undergoing assisted reproductive technology (ART). Alson et al. found that approximately one in 10 women scheduled for ART had direct ultrasound evidence of adenomyosis ( 45 ). Puente et al. found a higher prevalence of adenomyosis in cases of recurrent pregnancy loss and previous ART failure, emphasising the importance of non-invasive diagnostics in reproductive outcomes ( 26 ). Piccioni et al. highlighted age-related differences in the presentation of adenomyosis, finding that younger patients were more likely to present with severe dysmenorrhoea and focal, mild adenomyosis, whereas older women had a higher prevalence of menorrhagia, altered JZ and diffuse, severe adenomyosis ( 46 ). These findings suggest that a comprehensive understanding of adenomyosis must include associated gynaecological conditions such as uterine fibroids and/or endometriosis. The development of a standardised classification system that integrates ultrasound features with clinical and anamnestic variables would improve diagnostic accuracy and patient management. Moawad’s review ( 47 ) supports this need, highlighting the inconsistencies in current diagnostic criteria and advocating for a more unified approach to improve diagnostic confidence. Recent advances in 3D ultrasound imaging have facilitated the detection and characterisation of adenomyosis, particularly in differentiating it from related conditions. A key limitation of our review lies in the substantial heterogeneity among the included studies, which may have affected the diagnostic accuracy of imaging techniques. Clinical heterogeneity emerged from differences in patient age, parity, and comorbidities such as fibroids or endometriosis, all of which may influence the detectability of adenomyosis ( 26 , 46 ). Some studies also included asymptomatic women or those undergoing fertility workups, adding further variability. Methodological differences included retrospective versus prospective designs, inconsistent blinding procedures, and variability in the reference standard, with some studies relying on hysterectomy specimens while others used less definitive biopsy methods ( 33 , 35 , 36 , 40 ). Finally, the diagnostic criteria applied across studies were not uniform. While some followed MUSA guidelines ( 16 , 18 - 21 ), others used non-standardised combinations of sonographic or MRI features, with varying thresholds for JZ thickness ( 24 , 39 ). This heterogeneity highlights the need for standardised diagnostic criteria, uniform study designs, and greater methodological transparency in future research.

In addition to technical advancements, the current variability in diagnostic accuracy among studies reflects a broader need to address the heterogeneity in study design, patient selection, and diagnostic criteria. Future research should aim to adopt standardised classification systems ( 24 , 25 , 27 , 28 , 46 ), consistent methodological approaches (including uniform reference standards and blinding), and stratification based on disease stage and patient demographics ( 26 , 41 , 45 , 46 ). This will enable better comparison of results and facilitate the integration of imaging into personalised clinical management pathways. This review provides an up-to-date synthesis of non-invasive diagnostic tools for adenomyosis, including TVUS, MRI, elastography, and hysteroscopy with targeted biopsy ( 12 , 15 , 25 , 28 ). It adds value by highlighting the role of hysteroscopy in fertility-preserving strategies and discussing the correlation between imaging findings and clinical symptoms ( 41 , 46 ), as well as the need for a standardised classification system ( 19 , 20 , 24 ). However, heterogeneity in study design, diagnostic criteria, and patient populations limits comparability. The absence of meta-analysis and limited data on newer techniques like elastography reduce generalisability. Further prospective research is needed to validate emerging tools and support diagnostic standardisation.

The article’s supplementary files as