Adenomyosis and Endometrial Polyps: A Systematic Review of Coexistence Patterns and Diagnostic Contexts

Adenomyosis (AM) and endometrial polyps (EPs) are common benign uterine disorders encountered in overlapping clinical contexts, including abnormal uterine bleeding, infertility evaluation, pelvic pain, and peri- or postmenopausal assessment. 1–5 AM is characterized by endometrial glands and stroma within the myometrium and is commonly associated with dysmenorrhoea, abnormal uterine bleeding, chronic pelvic pain, and reproductive difficulties. 6–8 EPs are focal intracavitary lesions that may cause abnormal uterine bleeding, infertility, or postmenopausal bleeding or may remain asymptomatic. 8 , 9 Despite arising in different anatomical compartments and typically being identified through different diagnostic pathways, AM and EPs may occur in the same patient and contribute to overlapping clinical presentations. 10 , 11 The clinical challenge is not simply whether these lesions coexist, but whether identification of one lesion adequately explains the patient’s presentation. In women with established AM, persistent or disproportionate bleeding, infertility, or a suspected intracavitary abnormality may indicate an additional EP. Conversely, in women with EPs, substantial dysmenorrhoea, uterine enlargement, persistent symptoms after treatment, or myometrial imaging abnormalities may suggest coexisting AM. As transvaginal ultrasonography permits assessment of both the endometrial cavity and the myometrium during the same examination, recognition of AM–EP coexistence may not require additional disease-specific diagnostic steps in many patients. More importantly, identifying concomitant disease may influence symptom attribution and subsequent clinical management. Clinically useful evidence must therefore clarify the settings in which coexistence is repeatedly detected and the circumstances in which consideration of the alternative lesion may be warranted. Published evidence on AM–EP coexistence is fragmented across hysterectomy specimens, hysteroscopic or polypectomy cohorts, resectoscopy populations, fertility-preserving surgery, and studies using imaging, intraoperative assessment, or mixed clinical records. These settings differ in age distribution, symptoms, procedural indication, pre-test probability, and the opportunity to identify both lesions. Two complementary but distinct questions must also be preserved: how often EPs are identified among women with AM, and how often AM is identified among women with EPs. Although the same coexisting cases may contribute to both calculations within a study, the denominators and ascertainment opportunities differ. Direction-specific estimates therefore answer different clinical questions and should not be treated as interchangeable expressions of a single prevalence. This systematic review synthesized bidirectional evidence on AM–EP coexistence while preserving direction-specific denominators. We aimed to determine whether coexistence was repeatedly observed across different gynaecological pathways, characterize direction-specific patterns and dispersion, and examine how source population, clinical entry pathway, and ascertainment method shaped the reported estimates. By linking each estimate to its original clinical context, we sought to provide a framework for targeted, symptom- and pathway-driven evaluation while avoiding both premature diagnostic closure and indiscriminate dual-pathology screening.

The database searches identified 2839 records. After duplicate removal, 2510 records underwent title and abstract screening, and 74 reports were assessed in full text. Sixteen observational studies involving 14,917 women were included. 18–33 Detailed reasons for full-text exclusion are provided in Supplementary file 3 . Twelve studies contributed data on the prevalence of endometrial polyps among women with adenomyosis (Panel A), 11 contributed data on the prevalence of adenomyosis among women with endometrial polyps (Panel B), and seven informed both analytical directions. The included studies were conducted across multiple countries and clinical settings. Most recruited surgically or pathologically selected populations, including hysterectomy specimens, hysteroscopic polypectomy cohorts, resectoscopy populations, fertility-preserving surgery, and other gynaecological procedural pathways. Methods used to ascertain AM and EPs ranged from dual histopathological confirmation to imaging, hysteroscopy, intraoperative findings, physical examination, and mixed clinical records. No study sampled an unselected general population. Detailed study characteristics are presented in Table 1 . Table 1 Characteristics of Included Studies Study ID (Author, Year) Country/Region Study Period Study Design Source Population Sample Size (N) Age (Years) Diagnostic Method (AM/EP) Goksever Celik et al, 2025 20 Turkey 2011–2018 Multicenter retrospective cohort Hysterectomy cohort with histologically confirmed AM 1455 53.4±9.7 Path/Path Matalliotakis et al, 2026 24 Greece 2000–2022* Multicenter retrospective study Abdominal hysterectomy with salpingo-oophorectomy cohort 1692 50.16±12.74 Path/Path Ranaei et al, 2019 28 Iran 2012–2017 Cross-sectional Hysterectomy cohort 1900 48.6 Path/Path Matalliotakis et al, 2022 25 Greece 1985–2020* Multicenter retrospective study Surgical cohort with histologically confirmed AM 647 54.1±10.4 Path/Path Pervez et al, 2013 26 Pakistan 2008–2012 Retrospective study Hysterectomy specimens 861 44.5±24.74 Path/Path Monica Preethi et al, 2022 27 India 2020–2022 Retrospective study Hysterectomy specimens 247 41–50 (61.8%) Path/Path Sordia-Hernandez et al, 2012 31 Mexico 2007–2009 Retrospective case-control study Hysterectomy specimens 794 AM 43.0; non-AM 44.1 Path/Path Sidera et al, 2025 30 Greece 2000–2020 Retrospective study Gynecologic surgical intervention cohort 3490 CPs 50.90±10.17; EPs 59.08±10.78; both 53.49±10.23 Path/Path Weigel et al, 2025 33 USA 2009–2018 Cross-sectional Asymptomatic postmenopausal women undergoing hysterectomy for uterovaginal prolapse 317 With polyps 66.1±7.9; without 65.2±7.8 Path/Path Topcu et al, 2014 32 Turkey 2009–2014 Retrospective study Hysterectomy cohort with preoperative EP on endometrial sampling 203 Hyperplasia and polyp 60; polyp only 49 Path/Path Shi et al, 2023 29 China 2010–2017 Retrospective cohort Women with adenomyosis undergoing fertility-sparing laparoscopy 158 Anterior 34.28 ± 4.2; posterior 34.61 ± 4.0; both 33.86 ± 5.3 PE + TVS ± Histology/MRI/Intraop Indraccolo et al, 2011 22 Italy 2007–2008 Retrospective observational study Women undergoing resectoscopy 959 50 (19–88) Path/Path Firtina Tuncer et al, 2025 19 Turkey 2010–2024 Retrospective cohort Women undergoing hysteroscopic polypectomy 1021 42 (IQR 36–46) TVS ± MRI/Path Erkal et al, 2026 18 Turkey 2019–2022 Retrospective study Women who were diagnosed with AUB and treated with an LNG-IUS 233 Hysterectomy 45.35 (44–48); without 41.76 (38–46) Path/Path La Torre et al, 2026 23 Italy 2021–2024 Retrospective cohort Women who underwent total or subtotal hysterectomy 734 Not reported Path/Path Huang et al, 2025 21 China 2020–2023 Retrospective study Women undergoing hysteroscopic polypectomy with background endometrial biopsy 206 NPE 45; PHE 45; CE 44; NEH 50 Unclear/Path Notes : Studies contributing to both analytical panels are listed once in Table 1 and appear in the relevant panel (s) in Tables 3 and 4 . * Study period was harmonized for multicenter studies with different hospital-specific enrollment windows and is shown as the overall covered range. Age is presented as reported in the original studies. Abbreviations : AM, adenomyosis; AUB, abnormal uterine bleeding; EP, endometrial polyp; Intraop, intraoperative assessment; IQR, interquartile range; JBI, Joanna Briggs Institute; LNG-IUS, levonorgestrel-releasing intrauterine system; MRI, magnetic resonance imaging; Path, histopathology; PE, physical examination; TVS, transvaginal ultrasonography. Characteristics of Included Studies Notes : Studies contributing to both analytical panels are listed once in Table 1 and appear in the relevant panel (s) in Tables 3 and 4 . * Study period was harmonized for multicenter studies with different hospital-specific enrollment windows and is shown as the overall covered range. Age is presented as reported in the original studies. Abbreviations : AM, adenomyosis; AUB, abnormal uterine bleeding; EP, endometrial polyp; Intraop, intraoperative assessment; IQR, interquartile range; JBI, Joanna Briggs Institute; LNG-IUS, levonorgestrel-releasing intrauterine system; MRI, magnetic resonance imaging; Path, histopathology; PE, physical examination; TVS, transvaginal ultrasonography. Domain-level appraisal showed that analytical denominators and the data required to calculate direction-specific prevalence estimates were generally clearly reported. The principal concerns were selective sampling frames, limited representativeness, and incomplete or inconsistent reporting of disease ascertainment. Detailed domain-level assessments are presented in Table 2 , and study-specific concerns are summarized in Supplementary file 4 . Table 2 Methodological Quality Assessment of Included Studies Study ID (Author, Year) Panel Contribution Analytical Population Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Goksever Celik et al, 2025 20 Panel A only Adenomyosis population N Y U Y Y Y Y Y Matalliotakis et al, 2022 25 Panel A only Adenomyosis population N Y U Y Y Y Y Y Pervez et al, 2013 26 Panel A only Adenomyosis population N Y U Y Y Y Y Y Ranaei et al, 2019 28 Panel A only Adenomyosis population N Y U Y Y Y Y Y Shi et al, 2023 29 Panel A only Adenomyosis population N Y U Y Y Y Y Y Firtina Tuncer et al, 2025 19 Panel B only Endometrial polyp population N Y U Y Y Y Y Y Huang et al, 2025 21 Panel B only Endometrial polyp population N Y U Y Y U N Y Sidera et al, 2025 30 Panel B only Endometrial polyp population N Y U Y N Y Y Y Topcu et al, 2014 32 Panel B only Endometrial polyp population N Y U Y Y Y Y Y Indraccolo et al, 2011 22 Both panels Mixed population N Y U Y Y Y Y Y Matalliotakis et al, 2026 24 Both panels Mixed population N Y U Y Y Y Y Y Monica Preethi et al, 2022 27 Both panels Mixed population N Y U Y Y Y Y Y Sordia-Hernandez et al, 2012 31 Both panels Mixed population N N U Y N Y Y Y Erkal et al, 2026 18 Both panels Mixed population N Y U Y Y Y Y Y La Torre et al, 2026 23 Both panels Mixed population N Y U Y Y Y Y Y Weigel et al, 2025 33 Both panels Mixed population N Y U Y Y Y Y Y Notes : Q1–Q8 represent the applicable domains of the JBI prevalence checklist. The response-rate item was considered not applicable; therefore, scores are presented out of eight. Abbreviations : Y, yes; N, no; U, unclear. Methodological Quality Assessment of Included Studies Notes : Q1–Q8 represent the applicable domains of the JBI prevalence checklist. The response-rate item was considered not applicable; therefore, scores are presented out of eight. Abbreviations : Y, yes; N, no; U, unclear. Across 12 studies, the prevalence of EPs among women with AM ranged from 4.3% to 34.6% ( Table 3 ; Figure 2 ). Most pathology-based hysterectomy or surgical estimates fell between 4.3% and 22.9%. The highest estimate, 34.6%, was observed among asymptomatic postmenopausal women undergoing hysterectomy for uterovaginal prolapse. Estimates from other procedural pathways were 20.3% among women undergoing fertility-preserving laparoscopy and 24.9% among women undergoing resectoscopy. Figure 2 Prevalence of endometrial polyps among women with adenomyosis. Black squares represent study-specific prevalence estimates, and horizontal lines represent Wilson score 95% confidence intervals. No pooled estimate was calculated because the included studies differed structurally in source population, analytical denominator, clinical entry pathway, and disease ascertainment. A forest plot of endometrial polyp prevalence across studies, spanning about 4.3 percent to 34.6 percent. A forest plot showing study specific prevalence estimates with confidence intervals. The left side lists studies: Goksever Celik et al, 2025; Ranaei et al, 2019; Matalliotakis et al, 2022; Pervez et al, 2013; Matalliotakis et al, 2026; Sordia Hernandez et al, 2012; Monica Preethi et al, 2022; Weigel et al, 2025; Shi et al, 2023; Erkal et al, 2026; La Torre et al, 2026; Indraccolo et al, 2011. The x axis label is, Prevalence percent, ranging from 0 to 100 with tick marks every 10. Each study is shown as a square with a horizontal line for the 95 percent confidence interval. Right side values are: 12.0 with 95 percent confidence interval 10.5 to 13.8; 11.9 with 9.1 to 15.3; 11.9 with 9.6 to 14.6; 5.4 with 3.4 to 8.6; 22.9 with 20.6 to 25.4; 4.3 with 2.0 to 9.0; 9.2 with 5.3 to 15.3; 34.6 with 26.2 to 44.0; 20.3 with 14.7 to 27.2; 13.0 with 4.5 to 32.1; 15.3 with 11.4 to 20.2; 24.9 with 18.9 to 31.9. Table 3 Prevalence of Endometrial Polyps Among Women with Adenomyosis Study ID (Author, Year) Source population Denominator (N) Co-occurrence (n) Prevalence (%) Goksever Celik et al, 2025 20 Hysterectomy cohort with histologically confirmed AM 1455 175 12.0% Ranaei et al, 2019 28 Hysterectomy cohort with pathologically confirmed AM 413 49 11.9% Matalliotakis et al, 2022 25 Surgical cohort with histologically confirmed AM 647 77 11.9% Pervez et al, 2013 26 Hysterectomy specimens 296 16 5.4% Matalliotakis et al, 2026 24 Abdominal hysterectomy with salpingo-oophorectomy cohort 1197 274 22.9% Sordia-Hernandez et al, 2012 31 Hysterectomy specimens 140 6 4.3% Monica Preethi et al, 2022 27 Hysterectomy specimens 131 12 9.2% Weigel et al, 2025 33 Asymptomatic postmenopausal women undergoing hysterectomy for uterovaginal prolapse 107 37 34.6% Shi et al, 2023 29 Women with adenomyosis undergoing fertility-sparing laparoscopy 158 32 20.3% Erkal et al, 2026 18 Women who were diagnosed with AUB and treated with an LNG-IUS 23 3 13.0% La Torre et al, 2026 23 Women who underwent total or subtotal hysterectomy 255 39 15.3% Indraccolo et al, 2011 22 Women undergoing resectoscopy 169 42 24.9% Notes : Prevalence was calculated as coexisting AM and EP cases divided by the number of women with AM. Abbreviations : AM, adenomyosis; AUB, abnormal uterine bleeding; EP, endometrial polyp; LNG-IUS, levonorgestrel-releasing intrauterine system. Prevalence of endometrial polyps among women with adenomyosis. Black squares represent study-specific prevalence estimates, and horizontal lines represent Wilson score 95% confidence intervals. No pooled estimate was calculated because the included studies differed structurally in source population, analytical denominator, clinical entry pathway, and disease ascertainment. Prevalence of Endometrial Polyps Among Women with Adenomyosis Notes : Prevalence was calculated as coexisting AM and EP cases divided by the number of women with AM. Abbreviations : AM, adenomyosis; AUB, abnormal uterine bleeding; EP, endometrial polyp; LNG-IUS, levonorgestrel-releasing intrauterine system. Across 11 studies, the prevalence of AM among women with EPs ranged from 3.9% to 91.9% ( Table 4 ; Figure 3 ). Estimates in hysteroscopic polypectomy cohorts were 3.9% and 14.5%, and the resectoscopy cohort yielded an estimate of 13.0%. Hysterectomy-based estimates were highly variable, ranging from 9.7% to 91.9%. The two highest study-specific estimates, 55.9% and 91.9%, arose from selected gynaecological surgical or pathology-enriched populations. Figure 3 Prevalence of adenomyosis among women with endometrial polyps. Black squares represent study-specific prevalence estimates, and horizontal lines represent Wilson score 95% confidence intervals. No pooled estimate was calculated because the included studies differed structurally in source population, analytical denominator, clinical entry pathway, and disease ascertainment. A forest plot of adenomyosis prevalence across studies, spanning very low to very high estimates. A forest plot showing study specific prevalence estimates with confidence intervals. Left column header reads, Study. Right column header reads, Prevalence (95 percent confidence interval), percent. The x-axis label is, Prevalence (percent), ranging from 0 to 100 with ticks every 10. The y-axis lists 11 studies. Point estimate and confidence interval values shown at right are: Sidera et al, 2025, 55.9 (51.5 to 60.3). Matalliotakis et al, 2026, 91.9 (88.3 to 94.5). Sordia-Hernandez et al, 2012, 9.7 (4.5 to 19.5). Monica Preethi et al, 2022, 52.2 (33.0 to 70.8). Topcu et al, 2014, 21.7 (16.6 to 27.8). Weigel et al, 2025, 34.9 (26.5 to 44.4). Huang et al, 2025, 3.9 (2.0 to 7.5). Firtina Tuncer et al, 2025, 14.5 (12.5 to 16.8). Erkal et al, 2026, 25.0 (8.9 to 53.2). La Torre et al, 2026, 42.9 (33.2 to 53.1). Indraccolo et al, 2011, 13.0 (9.7 to 17.1). Table 4 Prevalence of Adenomyosis Among Women with Endometrial Polyps Study ID (Author, Year) Source Population Denominator (N) Co-occurrence (n) Prevalence (%) Sidera et al, 2025 30 Gynecologic surgical intervention cohort 488 273 55.9% Matalliotakis et al, 2026 24 Abdominal hysterectomy with salpingo-oophorectomy cohort 298 274 91.9% Sordia-Hernandez et al, 2012 31 Hysterectomy specimens 62 6 9.7% Monica Preethi et al, 2022 27 Hysterectomy specimens 23 12 52.2% Topcu et al, 2014 32 Hysterectomy cohort with preoperative EP on endometrial sampling 203 44 21.7% Weigel et al, 2025 33 Asymptomatic postmenopausal women undergoing hysterectomy for uterovaginal prolapse 106 37 34.9% Huang et al, 2025 21 Women undergoing hysteroscopic polypectomy with background endometrial biopsy 206 8 3.9% Firtina Tuncer et al, 2025 19 Women undergoing hysteroscopic polypectomy 1021 148 14.5% Erkal et al, 2026 18 Women who were diagnosed with AUB and treated with an LNG-IUS 12 3 25.0% La Torre et al, 2026 23 Women who underwent total or subtotal hysterectomy 91 39 42.9% Indraccolo et al, 2011 22 Women undergoing resectoscopy 324 42 13.0% Notes : Prevalence was calculated as coexisting AM and EP cases divided by the number of women with EPs. Abbreviations : AM, adenomyosis; AUB, abnormal uterine bleeding; EP, endometrial polyp; LNG-IUS, levonorgestrel-releasing intrauterine system. Prevalence of adenomyosis among women with endometrial polyps. Black squares represent study-specific prevalence estimates, and horizontal lines represent Wilson score 95% confidence intervals. No pooled estimate was calculated because the included studies differed structurally in source population, analytical denominator, clinical entry pathway, and disease ascertainment. Prevalence of Adenomyosis Among Women with Endometrial Polyps Notes : Prevalence was calculated as coexisting AM and EP cases divided by the number of women with EPs. Abbreviations : AM, adenomyosis; AUB, abnormal uterine bleeding; EP, endometrial polyp; LNG-IUS, levonorgestrel-releasing intrauterine system. Among the seven studies contributing data to both analytical directions, six reported a higher prevalence of AM among women with EPs than the corresponding prevalence of EPs among women with AM. Panel A estimates were comparatively more concentrated, whereas Panel B showed substantially greater dispersion across clinical pathways. This difference was characterized by relatively low Panel B estimates in hysteroscopic polypectomy cohorts and the concentration of its highest estimates in selected surgical or pathology-based populations.

This systematic review was prospectively registered in the International Prospective Register of Systematic Reviews (PROSPERO; CRD420261304541) and was conducted and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 (PRISMA 2020) statement. 12 The protocol specified a GRADE assessment; however, this was not undertaken because the review did not generate a common summary estimate to which certainty could be assigned. The completed PRISMA 2020 checklist is provided as Supplementary file 1 . Original human studies were eligible if they enrolled women with adenomyosis and/or endometrial polyps and reported sufficient data to estimate at least one of the following outcomes: the prevalence of EPs among women with AM or the prevalence of AM among women with EPs. Studies were eligible when the relevant numerator and analytical denominator were reported directly or could be derived from the full text. Diagnosis of AM or EPs could be based on imaging, hysteroscopy, intraoperative assessment, histopathology, clinical records, or a combination of these methods, provided that direction-specific coexistence data were available. We excluded: (1) non-original publications, including reviews, commentaries, editorials, case reports, and conference abstracts; (2) in vitro or animal studies; (3) studies that did not report the outcome of interest and from which a direction-specific prevalence estimate could not be calculated; (4) duplicate or overlapping reports, in which case the most complete report was retained; and (5) reports for which the full text could not be obtained. PubMed, Embase, the Cochrane Library, and Scopus were searched from the earliest date available in each database to February 10, 2026, with an updated search conducted on July 14, 2026. The search was restricted to human studies, with no restrictions on language, publication year, or geographical region. Controlled vocabulary and free-text terms relating to adenomyosis, endometrial polyps, and their synonyms were combined. The complete database-specific search strategies and numbers of records retrieved are provided in Supplementary file 2 . Two reviewers independently screened all retrieved records for eligibility. Duplicates were removed using EndNote, after which titles and abstracts were screened and potentially eligible reports underwent full-text assessment. Disagreements were resolved through discussion. No automated screening tools were used. The selection process is presented in the PRISMA 2020 flow diagram ( Figure 1 ), 13 and reasons for full-text exclusion are listed in Supplementary file 3 . Figure 1 PRISMA 2020 flow diagram of the study selection process. Records identified through the database searches were screened after duplicate removal, followed by full-text eligibility assessment and final study inclusion. A flowchart of study selection via databases, showing identification, screening and inclusion steps. The flowchart illustrates the study selection process via databases and registers. It begins with the identification of records from databases, totaling 2,839, including PubMed with 381, Embase with 1,445, Cochrane Library with 43 and Scopus with 970. Duplicate records removed amount to 329. Screening follows with 2,510 records screened, excluding 2,435 based on title and abstract. Reports sought for retrieval are 75, with 1 not retrieved. Reports assessed for eligibility are 74, with 58 excluded at full text. Finally, 16 studies are included in the review. PRISMA 2020 flow diagram of the study selection process. Records identified through the database searches were screened after duplicate removal, followed by full-text eligibility assessment and final study inclusion. Two reviewers independently extracted data using a standardized form and cross-checked all entries. Disagreements were resolved through discussion. Extracted information included first author, publication year, country or region, study period, study design, source population, clinical entry pathway, sample size, age, methods used to ascertain AM and EPs, and the numerators and denominators required to calculate direction-specific prevalence estimates. 14–16 When potentially eligible reports contained insufficient information to determine eligibility or calculate an estimate, the corresponding authors were contacted for clarification or supplementary data . The primary outcome was the prevalence of coexisting AM and EPs, evaluated in two directions. Panel A represented the prevalence of EPs among women with AM, whereas Panel B represented the prevalence of AM among women with EPs. Studies reporting both analytical denominators contributed separately to both panels. For each study, prevalence was calculated as the number of women with coexisting AM and EPs divided by the number of women in the corresponding direction-specific analytical population and multiplied by 100. Study-specific Wilson score 95% confidence intervals were calculated from the extracted numerators and denominators. The two panels were interpreted separately because they used different analytical denominators and answered different clinical questions. Two reviewers independently appraised the included studies using the Joanna Briggs Institute Critical Appraisal Checklist for Prevalence Studies. 17 Each applicable domain was rated as “Yes”, “No”, “Unclear”, or “Not applicable”, and disagreements were resolved through discussion. The appraisal examined the appropriateness of the sampling frame and recruitment process, adequacy of sample size, description of participants and setting, validity and consistency of disease ascertainment, completeness of coverage, and appropriateness of statistical analysis. Study-specific concerns relating to selective source populations, incomplete reporting of case ascertainment, and limited representativeness are summarized in Supplementary file 4 . Appraisal findings were not used to exclude studies or statistically weight their estimates. Before synthesis, the feasibility of quantitative pooling was assessed by comparing source population, clinical entry pathway, analytical denominator, procedural indication, and the methods and intensity of AM and EP ascertainment. Study-specific estimates arose from clinically distinct settings, including hysterectomy and other pathology-confirmed surgical cohorts, hysteroscopic polypectomy or resectoscopy populations, fertility-preserving surgery, imaging-based evaluation, intraoperative assessment, and mixed clinical records. These settings differed not only in participant characteristics but also in the opportunity to identify intracavitary and myometrial disease. The resulting estimates therefore represented clinically non-equivalent prevalence constructs rather than exchangeable estimates of a single common target parameter. A pooled prevalence was consequently not calculated; the study-level basis for this decision is documented in Supplementary file 5 . A structured narrative synthesis was performed separately for Panels A and B. Study-specific numerators, denominators, prevalence estimates, and Wilson score 95% confidence intervals were organized according to source population and diagnostic pathway, with particular attention to hysteroscopic, hysterectomy-based, fertility-preserving, and other surgically or pathologically enriched settings. Among studies contributing to both panels, an exploratory descriptive comparison was undertaken to identify recurrent direction-specific patterns. No formal statistical comparison between panels was performed because the two estimates used different denominators and the sets of contributing studies were not identical. Because no common summary effect was estimated, formal tests for publication bias, small-study effects, and leave-one-out sensitivity analyses were not performed.

Adenomyosis and endometrial polyps are repeatedly identified together in selected gynaecological populations, but the observed coexistence pattern is directionally asymmetric and strongly shaped by clinical entry pathway and disease ascertainment. The evidence does not support a universal coexistence rate or routine dual-pathology screening; instead, it supports targeted consideration of coexisting AM in women with EPs, or coexisting EPs in women with AM, when bleeding, pain, imaging findings, infertility, or persistent symptoms after treatment remain incompletely explained. However, interpretation is limited by the predominance of surgically selected populations, heterogeneity in disease ascertainment, and the lack of prospective studies in unselected clinical populations.

This systematic review identified three principal findings. First, AM and EPs were repeatedly co-detected across diverse gynaecological pathways, indicating that coexistence is not confined to a single clinical scenario. Second, the two analytical directions showed an asymmetric pattern: AM prevalence among women with EPs was generally higher and substantially more dispersed than EP prevalence among women with AM, with six of seven bidirectional studies showing this pattern. Third, the observed magnitude of coexistence was closely related to clinical entry pathway and ascertainment intensity, particularly in Panel B, where estimates ranged from relatively low values in hysteroscopic polypectomy cohorts to markedly higher values in selected surgical or pathology-enriched populations. The principal contribution of this review is therefore not a universal coexistence rate, but a direction- and pathway-specific interpretation of when the second lesion is more likely to be detected. The recurrently higher prevalence of AM among women with EPs suggests that clinicians should remain attentive to features of coexisting AM when evaluating women with EPs, particularly symptoms or myometrial findings suggestive of adenomyosis on routine ultrasonography. This should not be interpreted as a recommendation for additional systematic screening, because the observed estimates were strongly influenced by clinical selection and ascertainment intensity. The directional asymmetry should first be understood in relation to denominator structure. Within studies reporting both directions, the number of women with coexisting AM and EPs was the same, but the denominators were the total numbers of women with AM and women with EPs, respectively. A higher Panel B estimate therefore does not indicate that EPs cause AM or that the biological association is stronger in that direction; it indicates that the coexistence cases constituted a larger proportion of the EP denominator within that particular clinical population. Ascertainment opportunities also differed between the two lesions. EPs are focal intracavitary abnormalities that may be identified by hysteroscopy, imaging, or histopathology, whereas AM detection depends more strongly on the intensity of myometrial assessment. Complete histopathological examination after hysterectomy may identify AM that was previously unrecognized, while routine hysteroscopic or polypectomy pathways may not systematically evaluate the myometrium. This difference likely contributed to the greater dispersion observed in Panel B. Clinical selection further shaped the estimates. Hysterectomy cohorts included women selected for major surgery because of symptoms, prolapse, treatment failure, or coexisting uterine disease, whereas hysteroscopic cohorts generally represented women entering care through an intracavitary diagnostic or therapeutic pathway. These populations differed in age, menopausal status, symptom burden, disease severity, and pre-test probability of AM. Consequently, apparently similar denominators such as “women with EPs” represented clinically distinct patient groups. These structural differences also meant that a single pooled prevalence would not have had a consistent clinical interpretation. Broader adenomyosis literature has similarly shown that AM is frequently identified alongside other uterine or pelvic disorders, particularly leiomyomas and endometriosis, and in some hysterectomy series with endometrial hyperplasia or malignant disease. 34 , 35 Much of this evidence, however, has been derived from hysterectomy-based histopathological cohorts, in which comprehensive examination of the uterus increases the likelihood of detecting both clinically suspected and previously unrecognized lesions. These studies support viewing AM as part of a multi-pathology uterine phenotype in selected women, while also illustrating why coexistence observed in surgical specimens cannot be directly extrapolated to unselected outpatient populations. The literature on EPs shows a parallel dependence on clinical setting. EPs are commonly investigated in women with abnormal uterine bleeding, during infertility work-up, and in peri- or postmenopausal or surgical populations. 36 In abnormal-bleeding cohorts, hysteroscopy preferentially identifies intracavitary causes of bleeding; in infertility settings, uterine-cavity assessment may detect small or asymptomatic polyps; and in hysterectomy or postmenopausal cohorts, pathology provides more complete lesion ascertainment and a different spectrum of coexisting disease. Therefore, the clinical meaning of the denominator “women with EPs” differs according to why and how women entered evaluation. Previous research has more frequently examined EPs in relation to endometriosis than to AM. A recent bidirectional systematic review and meta-analysis reported increased relative risks of EPs among women with endometriosis and of endometriosis among women with EPs, with particularly high pooled proportions in infertile and surgically confirmed populations. 37 Although endometriosis and AM share certain hormone-responsive, inflammatory, and reproductive features, the comparative risk estimates reported for endometriosis cannot be transferred directly to AM. At a biological level, AM and EPs may arise within overlapping hormone-responsive, inflammatory, and tissue-remodelling environments. 38–42 AM has been associated with altered oestrogen signalling, inflammation, neuroangiogenesis, disruption of the endometrial–myometrial interface, and hormone-dependent myometrial remodelling, 43 , 44 whereas EP formation has been linked to focal endometrial overgrowth, altered receptor expression, local inflammation, and dysregulated proliferation. 36 , 45 These overlapping processes provide biological plausibility for coexistence but do not establish that either lesion causes the other. Biological coexistence is also difficult to separate from shared clinical detection. Abnormal bleeding, infertility, pelvic pain, and peri- or postmenopausal symptoms may lead to imaging, hysteroscopy, or surgery, creating repeated opportunities to identify both lesions. The observed coexistence is therefore likely to reflect a combination of shared biological susceptibility and shared routes to investigation. The findings are best translated into a directional clinical question: after one lesion has been identified, when should the other still be considered? After identification of an EP, coexisting AM may warrant consideration when dysmenorrhoea or chronic pelvic pain is prominent, the uterus is enlarged or myometrial abnormalities are present on imaging, heavy bleeding persists after polypectomy, or infertility remains insufficiently explained by the intracavitary lesion. In these situations, targeted myometrial assessment using standardised transvaginal ultrasonography, with magnetic resonance imaging when clinically indicated, may be appropriate. Conversely, after AM has been established, an EP should remain in the differential when intermenstrual or postmenopausal bleeding is present, imaging suggests a focal intracavitary abnormality, bleeding is disproportionate to the apparent myometrial disease or persists despite AM-directed treatment, or infertility remains multifactorial. Targeted uterine-cavity evaluation may then be considered according to age, symptoms, prior imaging, and reproductive goals. This framework should not be interpreted as a recommendation for universal bidirectional screening. The included studies did not test whether routine investigation of the second lesion improves clinical or reproductive outcomes. Rather, the findings support avoiding premature diagnostic closure when the established diagnosis does not adequately explain bleeding, pain, uterine findings, infertility, or treatment response. Estimates from hysterectomy and pathology-enriched cohorts should not be transferred directly to routine outpatient populations. This review has several strengths. It evaluated AM–EP coexistence in both clinical directions, preserved the appropriate direction-specific denominators, and reported study-level estimates with 95% confidence intervals. Estimates were linked to their source populations, clinical entry pathways, and ascertainment methods, allowing differences in clinical meaning to be distinguished rather than obscured. The review also identified a previously underemphasized directional pattern across studies contributing to both panels and assessed study-specific methodological concerns using the JBI prevalence checklist. This review also has limitations. Most included studies were derived from surgically or pathologically selected populations, particularly hysterectomy- or hysteroscopy-based cohorts, which limits transportability to broader outpatient populations. Important clinical descriptors, including menopausal status, fertility setting, symptom profile and coexisting uterine pathology, were incompletely reported across studies. Methods used to ascertain adenomyosis were not uniform and were sometimes insufficiently described. In addition, grey literature was not systematically included; one report could not be retrieved despite attempts to obtain the full text and a formal statistical assessment of reporting bias was not undertaken. These limitations restrict generalization to unselected outpatient populations and prevent determination of whether detecting both lesions improves symptom control, fertility outcomes, treatment selection, or recurrence risk. Future studies should recruit consecutive women within clearly defined clinical scenarios, such as abnormal uterine bleeding, infertility evaluation, or peri- and postmenopausal assessment. Investigators should report explicit denominators, indications for evaluation, menopausal and fertility status, symptom profiles, coexisting uterine disease, and standardised criteria for diagnosing both AM and EPs. Prospective studies using reproducible imaging definitions with pathological verification, where appropriate, would improve comparability and help determine when identification of coexistence changes management or outcomes. Future studies should move beyond prevalence alone and determine whether AM–EP coexistence defines a clinically distinct phenotype. Comparative prospective cohorts should evaluate symptom burden, infertility, treatment response, recurrence, and the incremental value of investigating the second lesion after one diagnosis has been established.