Timely Diagnosis and Management in Adult Patients With Ehlers‐Danlos Syndrome Experiencing Gynaecological Bleeding

Ehlers-Danlos syndrome (EDS) is a heterogeneous group of connective tissue disorders caused by genetic defects in collagen, characterised by joint hypermobility, skin abnormalities, poor wound healing and chronic pain [1]. Bleeding diathesis is highly prevalent in EDS, but can be overlooked due to lack of recognition in the hypermobile EDS diagnostic criteria and normalization within affected families [2]. Women are disproportionately affected and may experience delayed diagnosis, often suffering from heavy menstrual bleeding (HMB), postpartum haemorrhage (PPH) and urogenital prolapse [3-5]. Haemostatic therapies are not well-defined and often derived from treatment principles of other bleeding disorders. Little has been published on HMB management in EDS. Existing literature is focused on hormonal therapy, while tranexamic acid (TXA) and desmopressin have not been studied in adults with EDS [6, 7]. We aim to examine time to diagnosis in women with EDS compared with age-matched women with von Willebrand disease (VWD), rates of gynaecological bleeding, and medical and surgical management of HMB. We conducted a retrospective cohort study of all women aged ≥ 18 years diagnosed with EDS followed by the Northern Alberta Bleeding Disorders Program (2000–2023), matched by age (± 2 years) and sex to VWD patients at the time of study. EDS diagnosis was ascertained from consultation reports based on the 2017 International EDS Classification, and typically made by haematologists, medical geneticists, and/or internists [1]. A diagnosis of VWD was defined according to the ASH-ISTH-NHF-WFH 2021 guidelines [8]. Patients with other haemostasis defects were excluded. Institutional ethics approval was obtained. We collected the following data from electronic medical records: demographics, disease characteristics, gynaecological bleeding and iron deficiency anaemia (IDA). Time to diagnosis was calculated from the first documented bleeding event to the time of EDS diagnosis. A bleeding event was defined as any bleeding category from the ISTH-BAT, except for bruising, due to subjectivity and suboptimal documentation. HMB was defined as menses lasting > 7 days, changing protection more frequently than every 2 h, flooding, passage of large clots, or as labelled by the treating haematologist. Severe HMB was defined as HMB requiring Emergency Department (ED) visit, hospitalisation, blood transfusion, or gynaecologic procedures to control bleeding. IDA was defined as ferritin < 30 mg/L and haemoglobin < 120 g/L. Descriptive analyses were performed. Median and interquartile range (IQR) were used for continuous variables, and frequencies (percentages) for categorical variables. Differences were assessed using Mann–Whitney's U test for continuous variables, chi-squared test or Fisher's exact test, as appropriate, for categorical variables. A 2-sided p-value of < 0.05 was considered statistically significant. Analyses were performed using R statistical software. We included 42 women with EDS and 42 women with VWD. The median age was 38 (IQR 31–55) years. VWD subtypes included: type 1 VWD in 35 (83%), type 2 in 6 (14%), and unspecified in 1 (2%). The median baseline VWF antigen and VWF activity were 0.36 IU/mL (IQR 0.28–0.44) and 0.34 IU/mL (IQR 0.26–0.42), respectively. EDS subtypes included: hypermobile EDS in 29 (69%), unspecified in 8 (19%) and vascular EDS in 5 (12%).19 (45%) EDS patients had a genetics consultation, 17 (40%) had documented Beighton scores. Baseline characteristics were comparable between the two cohorts (Table 1), except for higher rates of ISTH-BAT scores > 5 (96% vs. 56%, p = 0.001) and lower proportion of O blood type in EDS (29% vs. 62%, p < 0.001). Psychiatric disorders were common (18; 43%) in the EDS group. Women with EDS were diagnosed at an older age than those with VWD (median 33 vs. 26 years, p = 0.04). Only 2 (5%) and 8 (19%) of the EDS and VWD groups were diagnosed under age 18, respectively. Median time from first bleeding symptom to diagnosis was 19.4 (IQR 9.9–37.0) years in EDS and 14.1 (IQR 5.5–30.7) years in VWD. Women in both groups experienced frequent bleeding events (median 4 vs. 3 in EDS vs. VWD groups, p = 0.13) prior to diagnosis. Delayed diagnosis occurred despite known family histories of bleeding disorders prior to haematology consultation (48% of EDS and 33% of VWD cohort). There were 71 ED visits and 4 hospitalisations for bleeding in the EDS cohort. All hospitalisations and transfusions, and 43/71 (61%) of ED visits occurred prior to EDS diagnosis. Women with EDS had 101 pregnancies and 61 (60%) live births. Gynaecological bleeding included any HMB (37; 88%), severe HMB (21; 50%), PPH (13; 31%), haemorrhagic ovarian cysts (6; 14%), and antepartum haemorrhage (4; 10%). Rate of documented gynaecology consultation was significantly lower in EDS versus VWD (31% vs. 67%, p = 0.001). There was no significant difference in the rates of gynaecological bleeding between EDS and VWD groups. 15 (36%) EDS patients received desmopressin responsiveness trials, using bleeding time as endpoints. HMB were medically managed with TXA (29; 78%), subcutaneous desmopressin (21; 57%), oral contraceptive pills (OCP, 18; 49%) and intrauterine system (IUS, 16; 43%). 7 (19%) patients did not receive any hormonal therapy. Surgical management included hysterectomy in 11 (30%), endometrial ablation in 4 (11%), dilation and curettage in 3 (8%). Rates of medical and surgical therapies were comparable among the 36 VWD patients who experienced HMB (Figure 1), except for a trend towards lower rate of IUS compared with EDS (7; 19%, p = 0.05). Hysterectomies and ablations were performed prior to a diagnosis of EDS and VWD in 78% and 43%, respectively (p < 0.01). Iron deficiency was prevalent in 21 (50%) patients with EDS prior to diagnosis and persisted in 14 (33%) at last clinic visit. The median ferritin was 26 mcg/L (IQR 10–71) and 38 mcg/L (IQR 25–80) prior to haematology consultation and at most recent visit, respectively. At last follow-up, those who received TXA (21% vs. 62%, p = 0.01) and desmopressin (19% vs. 48%, p = 0.10) were less likely to be iron-deficient, albeit non-significant with desmopressin. Our study demonstrates a high prevalence of diagnostic delays in women with EDS, with an older age at diagnosis compared to matched VWD patients. Despite a high burden of HMB in nearly 90% of our cohort, gynaecology consultation was underutilised in women with EDS. While the treatment strategies for HMB appear comparable between the groups, we found an alarmingly high rate of surgical procedures for HMB in EDS. The vast majority were performed prior to EDS diagnosis, thereby increasing risks of perioperative bleeding and wound dehiscence. Increased awareness amongst primary care providers (PCP) and early gynaecology involvement may improve timely diagnosis, prevent unnecessary hysterectomies, and monitor for other gynaecologic complications of EDS. While the burden of gynaecological bleeding has been well described, there is little data on management. Most reports were from paediatric/adolescent population, focused on hormonal therapies [6, 7]. There is scarce data on the utility of desmopressin in EDS, limited to case reports and a small retrospective study in children [9]. Our study demonstrates a high usage of TXA in nearly 80%, and desmopressin in 57%, higher than rates of OCP. In comparison, one survey from 386 women with hEDS showed higher rates of hormonal contraceptive (59%) but no mention of desmopressin [3]. Desmopressin raises von Willebrand factor and factor VIII, and promotes platelet-subendothelial collagen interaction, making it a good alternative for EDS patients unresponsive/intolerant to hormonal therapy. Our study also highlights suboptimal management of IDA in women with EDS. Persistent IDA was found in a third of our patients even after haematology consultation, with TXA (significant) and desmopressin use (non-significant trend) identified as predictors of persistent IDA. This may reflect confounding by indication. Alternatively, TXA/desmopressin use may be surrogates for clinic engagement and access to private insurance. Our study confirmed prior reports of diagnostic delays faced by women with EDS [5]. Only 5% of our cohort received EDS diagnosis in childhood, in contrast to 41% women and 72% men in a UK nationwide study [5]. Despite the multisystemic involvement, we found that women with EDS experienced lengthier diagnostic delays compared with VWD. Alarmingly, delayed diagnosis persisted even among those with a known family history of bleeding disorders. We hypothesise upstream systemic barriers, such as lack of PCP, insufficient physician knowledge in EDS diagnosis, delays in specialty referrals, dismissal of symptoms by clinicians and patients, and regional differences in the training, expertise and resources for EDS diagnosis. Comorbid physical and mental health issues such as fibromyalgia, chronic fatigue, and mood disorders may affect self-efficacy in seeking prompt care. Limitations include selection bias by sampling from specialised bleeding disorders clinic, misclassification bias in EDS diagnosis/subclassification as over half did not have medical genetics consult, lack of population controls to assess HMB management in those without bleeding disorders, missing data (e.g. race/ethnicity, socioeconomic status), and bias introduced by the imbalance of EDS and VWD groups in bleeding scores. Generalisability is limited due to our setting of a single Canadian province. Given our universal publicly funded health system, we postulate that diagnostic delays may be comparable or worse in resource-limited settings or jurisdictions lacking public healthcare. Our study highlights diagnostic challenges faced by women with EDS, who may not have the same recognition, referral and resources as women with VWD. Further studies are needed to better understand the patient journey. We call for further action in multidisciplinary collaboration between PCP, haematologists, patients/families and patient care organisations to improve education, awareness and advocacy. A.H. conceived the study idea, designed the study, performed data collection and interpretation, and wrote the first draft of the manuscript. H.S. conceived the study idea, designed the study, performed data analysis and interpretation, and critically revised the manuscript. All authors read and approved the final manuscript. The authors have nothing to report The authors have nothing to report. H.L.S. received honoraria from Pfizer, Roche, Sanofi/Sobi, and Shire/Takeda. The study was conducted in accordance with the principles of the Declaration of Helsinki, and approved by the University of Alberta Health Research Ethics Board. As this study involved analysis of anonymised data from patient records, the requirement for informed consent was waived by the Ethics Board. Data are available from the authors upon reasonable request.