{"paper_id":"22f607dd-839a-49d4-82cd-fb84394b97c4","body_text":"Cardiovascular disease (CVD) remains a leading cause of mortality in the United States, with significant disparities in outcomes between men and women [ 1 ]. Despite well-documented sex differences in cardiovascular pathophysiology, research efforts have historically prioritized male-focused investigations, resulting in critical knowledge gaps regarding female-specific risk factors [ 2 ]. Emerging evidence highlights unique mechanisms influencing CVD development in women, including hormonal fluctuations, pregnancy-related complications, and sex-specific responses to therapies [ 3 ]. These factors collectively underscore the urgent need to identify female-specific contributors to cardiovascular morbidity, particularly those associated with chronic inflammatory conditions.\nEndometriosis (EM), characterized by ectopic endometrial-like tissue implantation, affects approximately 5–10% of reproductive-aged women globally [ 4 ]. Beyond its classical manifestations—chronic pelvic pain, dysmenorrhea, and infertility—EM is increasingly recognized as a systemic disorder with multisystemic inflammatory and metabolic consequences [ 5 ]. Mechanistically, EM-induced chronic inflammation may disrupt vascular homeostasis and promote atherogenesis, positioning EM as a potential contributor to cardiovascular pathology.\nRecent meta-analyses have established EM as a potential risk factor for cardiovascular disease (CVD) but leave critical gaps unaddressed. Parsa et al. [ 6 ] identified an association between EM and hypertension-related CVD in a pooled analysis of six studies ( n  = 655,408), yet their work excluded emerging evidence from large-scale cohort studies published after April 2023 and did not explore risks for specific CVD subtypes such as ischemic heart disease (IHD) or cerebrovascular events. Similarly, Poeta Do Couto et al. [ 7 ] and Okoli et al. [ 8 ] reported elevated CVD risks in EM patients but faced limitations, including restricted data timelines (up to 2021–2023), incomplete subgroup stratification (e.g. arrhythmias, heart failure), and unresolved heterogeneity ( I 2  = 85–94%) due to methodological and demographic variability. Importantly, these analyses overlooked recent cohort studies published between 2023 and 2024 that reported novel insights into CVD subtype risks and geographic disparities. For instance, studies by Havers-Borgersen et al. [ 9 ] and Wang et al. [ 10 ] demonstrated elevated risks of atrial fibrillation and major adverse cardiovascular events in EM patients, while Blom et al. [ 11 ] highlighted regional differences in CVD outcomes across continents. These findings underscore the need to integrate updated evidence and refine analytical frameworks. To bridge these gaps, we conducted a comprehensive meta-analysis incorporating more recently cohort studies with data extending through December 2024. By expanding sample size, incorporating recent studies, and rigorously stratifying outcomes by CVD subtype, geographic region, and socioeconomic context, this analysis provides the most current and granular assessment of EM-associated CVD risks.\n\nWe adhered to standardized procedures for conducting and reporting meta-analyses, following the guidelines set out by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [ 12 ]. The protocol for this meta-analysis was registered with PROSPERO (CRD42024628888).\nStudies were considered eligible if they met the following criteria: (1) cohort studies or case-control studies (including nested case-control studies based on cohort data); (2) examined at least one association between EM and the risk of CVDs, including atrial fibrillation; and (3) reported risk ratios (or odds ratios, hazard ratios, or incidence rate ratios) with 95% confidence intervals (CIs).\nWe excluded conference abstracts, study protocols, duplicate publications, and studies that did not report outcomes of interest. In cases where multiple studies were derived from the same cohort, we selected the study with the largest sample size, longest follow-up period, or the most comprehensive data.\nWe systematically searched PubMed, Embase, and the Cochrane Library from their inception to December 6th, 2024, for potentially relevant studies, without language restrictions. Our search strategy combined medical subject headings (MeSH) and keywords. The search terms included: exposure (Endometriosis or Endometrioses or Endometrioma*) and outcome (Mortality or Death or Cardiovascular diseases or Cardiovascular or Cardiac or Coronary artery disease or Coronary heart disease or Ischemic heart disease or Myocardial infarction or Heart failure or Hypertension or High blood pressure or Atrial Fibrillations or Atrial Fibrillation* or Auricular Fibrillation*) and (Risk). A detailed description of the search strategy is provided in  Supplementary Tables S1 –S3. In addition, we reviewed previously published relevant meta-analyses to avoid missing potentially compatible literature [ 6–8 ].\nBasic characteristics of the included studies.\nNHIRD: National Health Insurance Research Database; ICES: Institute for Clinical Evaluative Sciences NACRS: National Ambulatory Care Reporting System database; SDS: Same Day Surgery Database; NHSII: Nurses’ Health Study II; THIN: The health improvement network.\nTwo independent reviewers (WLY and LYJ) screened the literature according to the predefined eligibility and exclusion criteria. Initially, titles and abstracts were reviewed to exclude studies that did not meet the inclusion criteria. Full texts of potentially eligible articles were then obtained and examined in detail to determine suitability for inclusion. Any disagreements were resolved through consultation with a third reviewer (LXL).\nData extraction was performed independently by the two reviewers (WLY and LYJ), following established guidelines for systematic reviews and meta-analyses. The extracted data included: first author, publication year, country, study setting, duration of follow-up, study design, data source, study type, diagnostic criteria, diagnostic methods, enrollment period, sample size, participant demographics and clinical characteristics, outcome variables, and confounders adjusted for in the analysis.\nThe quality of cohort studies was assessed using the Newcastle–Ottawa Scale (NOS) (Available from:  http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp ). The quality was rated on a scale of 0–9 points, with 4 stars for participant selection and exposure, 2 for comparability, and 3 for outcome assessment and follow-up. Studies were categorized as having low (0–3), moderate (4–6), or high (7–9) quality based on their star ratings.\nDue to the relatively low incidence of cardiovascular and cerebrovascular diseases in women with EM, odds ratios (OR) and relative risks (RR) were approximated as hazard ratios (HR) [ 13 ]. The adjusted HR and 95% CI from each study were used to assess the association between EM and the risk of all-cause CVDs, including atrial fibrillation.\nHeterogeneity across studies was assessed using the χ 2  test and  I 2  statistic. A fixed-effects model was applied when the  p  value for heterogeneity was greater than 0.1 and  I 2  was less than 50%. In cases of substantial heterogeneity ( I 2  > 50%) [ 14 , 15 ], a random-effects model was used. Sensitivity analyses were conducted by sequentially excluding one study at a time to examine the robustness of the overall results.\nPublication bias was visually assessed using funnel plots, and Egger’s regression test was performed to statistically evaluate the presence of bias. A subgroup analysis was conducted based on disease type, continent, and the development level of the country.\nAll statistical analyses were performed using Stata software version 14.0 (Stata Corp, College Station, Texas).\n\nA comprehensive search for cohort studies published prior to December 6, 2024, resulted in 1,671 records. After conducting a plagiarism check, 332 articles were excluded. Finally, 11 studies were identified as potentially relevant. A detailed flowchart of the selection process is shown in  Figure 1 .\nStudies screening process.\nThis meta-analysis included 11 cohort studies [ 9–11 , 14–21 ], with a total of 3,100,610 participants, published between 2016 and 2024. All studies were cohort studies, and the diseases were diagnosed using well-established criteria, preferably with diagnostic methods specified. The average age of participants across the studies ranged from 34.7 to 40.08 years. The enrollment periods varied, spanning from 1977 to 2021, with the earliest starting in 1977 and the latest ending in 2021. The duration of the studies ranged from a minimum of 13 years (2000–2013, excluding the start and end years) to a maximum of 44 years (1977–2021), with most studies having a follow-up period of approximately 20 years. The basic characteristics of the included studies are summarized in  Table 1 .\nThe average score was 7.4, with 6 studies receiving a score of 8, 3 studies scoring 7, and 2 study scoring 6. The detailed quality assessment for each study is provided in  Supplementary Table S4 .\nIn the 11 selected studies [ 9–11 , 14–21 ], we conducted a statistical analysis to examine the relationship between EM and the incidence of all-cause CVDs. We noted the different classifications of CVD and the diseases within its scope in these 11 articles, which we categorised with reference to the WHO ( https://www.who.int/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds) ) presentation. We therefore considered hypertension to be a comorbid disease of CVD, and we included hypertension as a search term for more comprehensive screening. For articles that did not give effect sizes directly, we combined the diseases subdivided under CVD and aggregated to give plausible effect sizes.\nWomen with a history of EM exhibited a significantly higher risk of developing overall CVDs compared to those without EM (HR = 1.22; 95% CI: 1.08–1.38;  I 2  = 94.6%,  p  < 0.001;  Figure 2 ). Sensitivity analysis confirmed the robustness of these findings, as no single study significantly altered the overall effect size, as shown in  Supplementary Figure S1 .\nMeta-analysis of the risk of all-cause CVDs.\nIn the subgroup analyses we performed a careful and rational grouping of the diseases within the CVD category, and the statistical analyses yielded the data in  Table 2 . The subgroup analysis revealed that patients with EM were at a slightly higher risk for angina compared to other CVDs. In terms of all-cause CVD, EM patients had a lower likelihood of developing heart failure (HF). Furthermore, compared with patients from North America and Asia, European patients with EM had a slightly lower risk of developing CVDs. Additionally, patients with EM from developed countries exhibited a lower risk of all-cause CVD compared to those from developing countries.\nSubgroup analysis for the risk of CVD in patients with EM.\nMI: Myocardial infarction; AF: Atrial fibrillation; HF: Heart failure; IHD: Ischemic Heart Disease; CAD: Coronary artery disease; MACCE: Major Adverse Cardiovascular and Cerebrovascular Events; PAD: Peripheral Artery Disease; CVA: Cerebrovascular Disease.\nVisual inspection of the funnel plot indicated no significant publication bias in the results regarding EM and the risk of all-cause CVDs ( Figure 3 ). Egger’s regression test ( p  = 0.525) further confirmed that publication bias was not present in the analysis.\nPublication bias of the risk of all-cause CVDs caused by EM.\n\nThis meta-analysis incorporated 11 cohort studies published between 2016 and 2024, involving 3,100,610 participants. Our findings demonstrate that women with EM exhibit a 22% increased risk of developing CVD compared to those without EM, reinforcing EM independently associating with an elevated risk of for CVD.\nOur results align with prior meta-analyses by Parsa et al. [ 6 ], Poeta Do Couto et al. [ 7 ], and Okoli et al. [ 8 ], all of which reported elevated CVD risks in EM patients. However, this study extends prior work by incorporating large-scale cohort data published after 2023, such as Havers-Borgersen et al. [ 9 ] and Wang et al. [ 10 ], which provided critical insights into understudied CVD subtypes (e.g. atrial fibrillation) and geographic disparities. Notably, Parsa et al. [ 6 ] identified hypertension-related CVD risks but lacked granular analyses of ischemic heart disease (IHD) or myocardial infarction (MI), while Poeta Do Couto et al. [ 7 ] and Okoli et al. [ 8 ] were constrained by limited data timelines (up to 2021–2023) and unresolved heterogeneity ( I 2  = 85–94%). Our inclusion of recent studies, such as Blom et al. [ 11 ] and Havers-Borgersen et al. [ 9 ], addresses these gaps, offering updated risk estimates and subgroup stratifications.\nA notable divergence arises in the analysis of Saavalainen et al. [ 15 ], whose data suggested protective effects for cerebrovascular disease (CVA) and IHD. While Peota et al. [ 15 ] excluded this study to enhance result robustness, we retained it to minimize selection bias, acknowledging its potential confounding by regional healthcare practices (e.g. higher EM treatment rates in Finland). Sensitivity analyses confirmed that excluding Saavalainen et al. [ 15 ] shifted pooled estimates closer to prior meta-analyses, underscoring the need for nuanced interpretation of geographic and methodological variability.\nThe retrograde menstruation hypothesis posits that ectopic endometrial tissue induces chronic inflammation, a plausible mechanism linking EM to CVD [ 22 ]. Elevated levels of reactive oxygen species and pro-inflammatory cytokines [ 23 , 24 ] may accelerate atherogenesis and destabilize coronary plaques, explaining the elevated risks of coronary artery disease (CAD) (HR = 1.47) and MI (HR = 1.29). Chronic inflammation may also disrupt cardiac electrophysiology, increasing susceptibility to arrhythmias [ 25 ], while inflammatory endothelial dysfunction could contribute to stroke risk [ 18 ]. However, EM patients typically experience an average diagnostic delay of 7 – 10 years, which may prolong the duration of chronic inflammatory exposure. Persistently elevated levels of pro-inflammatory factors (e.g. IL-6, TNF-α) may accelerate atherosclerosis progression [ 26 ].\nEM treatments further modulate CVD risk. Hysterectomy and bilateral salpingo-oophorectomy (BSO), common in EM management, are associated with elevated CAD risk [ 24 , 27 ] particularly in women under 50 [ 21 , 28–30 ]. Mu et al. [ 20 ] highlighted that BSO before age 50 exacerbates CVD morbidity, likely due to abrupt estrogen depletion, whereas hysterectomy without ovarian removal increases risk only in younger women [ 31 , 32 ]. These findings emphasize the need for age-stratified risk–benefit evaluations of surgical interventions.\nAlthough this study adjusted for certain medication variables, EM patients frequently receive long-term GnRH analog or progestin therapy, interventions that may indirectly exacerbate CVD risk through estrogen deprivation or pro-inflammatory effects. Research suggests that prolonged gonadotropin-releasing hormone (GnRH) analog therapy may accelerate atherosclerosis progression [ 33 ]. Research suggests that synthetic progestins may impair endothelial function through pro-inflammatory mechanisms [ 34–36 ]. Therefore, the mechanisms underlying EM-associated CVD risk may involve dual contributions from both the disease’s inherent chronic inflammation and treatment-related effects.\nPathological effects of EM on the ovaries may exacerbate cardiovascular risk through hormone-mediated pathways. Ovarian endometriosis (particularly deep infiltrating subtypes or cases with repeated surgical interventions) can lead to ovarian tissue destruction and diminished follicular reserve, resulting in decreased estrogen secretion or premature menopause (defined as menopause before age 45) [ 37 ]. Notably, premature menopause itself is an independent CVD risk factor HR = 1.50 (1.28–1.76) [ 38 ]. Importantly, several included studies (e.g. [ 11 , 18 ]) adjusted for menopausal status and hormone replacement therapy, suggesting that the EM-CVD association may persist independently of ovarian functional decline.\nAnimal experiments provide further evidence supporting the biological mechanisms by which EM directly drives cardiovascular injury. For instance, Mamillapalli et al. [ 39 ] demonstrated in a mouse EM model that ectopic endometrial tissue releases pro-inflammatory factors (e.g. IL-6, TNF-α), which can circulate systemically to promote aortic endothelial cell apoptosis and accumulation of oxidative stress markers (e.g. MDA, 8-OHdG), thereby accelerating atherosclerotic plaque formation. Furthermore, the study observed impaired vasodilation function in EM mice (confirmed by acetylcholine-induced vascular reactivity tests) and significantly reduced coronary microvascular density, suggesting that EM may indirectly increase myocardial ischemia risk through endothelial dysfunction.\nSubgroup analyses revealed significant geographic variation, with higher CVD risks in Asian (HR = 1.36) and North American cohorts (HR = 1.37) compared to European populations (HR = 0.93). These disparities may reflect differences in EM treatment accessibility and socioeconomic factors. For instance, over 50% of EM patients in the UK receive treatment [ 40 ], compared to 20% in Asian cohorts [ 14 ]. Higher healthcare costs in developed countries (e.g. $4,000 per patient in the U.S. [ 4 ]) may also limit treatment access in low-resource settings, exacerbating systemic inflammation and CVD progression.\nAlthough the overall CVD risk increase in EM patients is 22% (HR = 1.22), its public health impact should not be overlooked. Given EM’s high prevalence and elevated risks for specific subtypes (e.g. CAD: HR = 1.47), we recommend regular cardiovascular assessments for young EM patients (particularly those with chronic inflammation or geographically high-risk factors). This stratified management approach may optimize resource allocation while avoiding excessive interventions in low-risk populations.\nThis study underscores the imperative for cardiovascular monitoring in EM patients, particularly in high-risk subgroups (e.g. younger women, those undergoing BSO). However, limitations include geographic homogeneity (predominantly Asian and North American cohorts) and residual heterogeneity ( I 2  = 94.6%), potentially due to unmeasured confounders (e.g. socioeconomic status, lifestyle factors). Additionally, the effects of comorbidities like diabetes and hyperlipidemia were insufficiently detailed in included studies. More critically, factors such as diagnostic delay duration, surgical type (e.g. unilateral/bilateral oophorectomy), and lifestyle parameters (e.g. dietary patterns, physical activity levels) may influence the degree of systemic inflammation associated with endometriosis, thereby modulating cardiovascular risk. The included studies lacked recording of these data or failed to systematically adjust for them. Future studies should integrate dynamic monitoring of inflammatory biomarkers with lifestyle questionnaire data to delineate the independent contributions of these factors to CVD risk.\n\nThis meta-analysis demonstrates that endometriosis is independently associated with a 22% increased risk of cardiovascular disease, with elevated risks observed for myocardial infarcrisktion, coronary artery disease, and cerebrovascular events. Geographic and socioeconomic factors significantly modulate this risk, highlighting the need for tailored prevention strategies in high-burden regions. Clinicians should integrate EM into CVD risk assessments, particularly for women undergoing surgical interventions or residing in low-resource settings. Further research is warranted to elucidate the inflammatory pathways linking EM to CVD and optimize early intervention protocols.","source_license":"CC0","license_restricted":false}