Low-dose mifepristone for the management of refractory heavy menstrual bleeding in women with bleeding disorders: a retrospective study.

Between January 2010 and October 2024, 1,987 patients with HMB received treatment at our clinic. Among these, 217 (10.9%) were diagnosed with BD. Seventy-five patients received mifepristone for refractory HMB after failing or being contraindicated for COC. After excluding five lost to follow-up, five declining participation, and two with poor adherence, 63 patients were included in the final analysis (Fig.  1 ). Fig. 1 Study flow for participants in a study evaluating mifepristone for refractory heavy menstrual bleeding (HMB) associated with bleeding disorders (BD) treatment. HMB, heavy menstrual bleeding; BD, bleeding disorders; LNG-IUS, levonorgestrel intrauterine system; HSCT, hematopoietic stem cell transplantation Study flow for participants in a study evaluating mifepristone for refractory heavy menstrual bleeding (HMB) associated with bleeding disorders (BD) treatment. HMB, heavy menstrual bleeding; BD, bleeding disorders; LNG-IUS, levonorgestrel intrauterine system; HSCT, hematopoietic stem cell transplantation The BD cohort comprised two major groups. The first included 48 patients with thrombocytopenia: 19 with aplastic anemia (AA), 16 with acute myeloid leukemia (AML), 7 with myelodysplastic syndrome (MDS), and 6 with immune thrombocytopenia(ITP). The second group consisted of 13 patients receiving antithrombotic therapy, including 1 patient with AA who developed post-transfusion thrombosis, 1 renal transplant recipient taking aspirin for stent-related anastomotic stenosis, 1 case of systemic lupus erythematosus–associated antiphospholipid syndrome, 1 patient with end-stage renal failure requiring dialysis and antiplatelet therapy to prolong clotting time, and 9 patients with idiopathic thrombotic events. In addition, the cohort included 1 patient with allergic purpura and 1 with von Willebrand disease (VWD) (Table  1 ).A small number of patients had concomitant gynecologic conditions, including uterine fibroids ( n  = 4) and adenomyosis ( n  = 2); however, these were not considered the primary cause of bleeding, and HMB was judged to be predominantly related to BD. Patients were managed according to their bleeding status at presentation (Fig.  2 ). Those presenting with active acute bleeding first received short-term hemostatic interventions to achieve stabilization, followed by transition to maintenance mifepristone therapy. Patients without acute bleeding initiated mifepristone directly as monotherapy. Table 1 Basic information of participants Characteristic Age(years), median (IQR) 31.0(23.0–39.0) Age at Menarche(years), median (IQR) 14.0(13.0–14.0) BMI, median (IQR) 22.7(20.2–25.0) Parity, n (%)  Nulligravidas 36(57.1)  Parous women 27(42.9) Menstrual cycle, n (%)  Regular 47(74.6)  Irregular 16(25.4) BD, n (%)  Aplastic Anemia 19(30.2)  Thrombocytopenia 6(9.5)  Myelodysplastic Syndrome 7(11.1)  Acute Myeloid Leukemia 16(25.4)  Antithrombotic therapy 13(20.6)  Allergic Purpura 1(1.6)  Von Willebrand Disease 1(1.6) Gynecologic Disease, n (%)  Uterine Leiomyoma 4(6.3)  Adenomyosis 2(3.2)  Scarred Uterus 3(4.8)  Polycystic Ovary Syndrome 2(3.2)  Endometrial Polyps 1(1.6)  No comorbidities 51(81.0) Platelet count(/L), median (IQR) 16.0(7.0-41.5)×10 9 Anemia, n (%)  No 3(4.8)  Mild 7(11.1)  Moderate 26(41.3)  Severe 25(39.7) Time of Administration(months), n (%)   6 16(25.4) Emergency hemostasis, n (%)  Hormonal Agents 20(31.7)  Foley Balloon Uterine Tamponade 12(19.0)  Combined 3(4.8)  None 28(44.4) Basic information of participants Fig. 2 Clinical management algorithm. COC, combined oral contraceptives Clinical management algorithm. COC, combined oral contraceptives There were thirty-four patients presented with acute bleeding, including 27 with thrombocytopenia, 6 undergoing antithrombotic therapy, and 1 with VWD. Among the 27 thrombocytopenic patients, 21 were treated with COC or norethisterone, and bleeding was partially controlled in 19 patients. However, breakthrough bleeding occurred repeatedly during the dose reduction phase. There were 15 patients with acute severe bleeding received intrauterine Foley balloon tamponade for immediate hemostasis, including 8 thrombocytopenic patients, 2 of whom were refractory to estrogen-progestin therapy, 6 were receiving antithrombotic therapy with contraindications to oral estrogen-progestin, and 1 patient with VWD. Bleeding was successfully controlled in all cases. The balloons (mean volume 15 ± 3 mL) were retained for 12 to 24 h and gradually deflated under hemodynamic monitoring. After stabilization, the patients transitioned to mifepristone 25 mg once daily alone, and reduced to 12.5 mg once daily after 1 month. Among the 29 patients who initiated mifepristone outside the menstrual phase (25 mg once daily), 27 achieved amenorrhea, while the remaining 2 reported markedly reduced menstrual flow. After one month, the dose was reduced to 12.5 mg once daily for maintenance. Median treatment duration was 5 months (IQR 3.0–6.0). Mifepristone treatment was associated with marked reductions in menstrual blood loss, with PBAC scores decreasing from 385.0 (IQR 230.0–455.0) at baseline to 0 (IQR 0–0) at 1 month (Bonferroni-adjusted p  < 0.01), and this effect was sustained through 6 months (Table  2 ). Amenorrhea, defined as complete absence of menstrual bleeding, occurred in 92.1% (58/63) of patients at 1 month and in 100% of evaluable patients at 3 ( n  = 37) and 6 months ( n  = 16). Minimal spotting not requiring sanitary products may still result in a PBAC score of 0, which explains the small discrepancy between PBAC values and the amenorrhea rate. Hemoglobin levels increased in parallel (Table  2 ). Quality-of-life scores (MMAS) also improved, rising from 290.0 (IQR 185.0–390.0) to 600.0 (IQR 600.0–600.0) (Bonferroni-adjusted p  < 0.01). Table 2 Comparison before and after the use of mifepristone for BD patients Variable n(%) baseline 1 month 3 month 6 month P 1 cycle PBAC, median (IQR) 63(100) 385.0(230.0 ~ 455.0) 0(0 ~ 0) < 0.01 Hb(g/L), median (IQR) 63(100) 51.0(65.0 ~ 84.0) 84.0(73.0 ~ 108.0) < 0.01 3 cycles PBAC, median (IQR) 37(58.7) 425.0(258.8 ~ 470.0) 0(0 ~ 0) 0(0 ~ 0) < 0.01 Hb(g/L), median (IQR) 37(58.7) 60.0(49.8 ~ 82.5) 80.5(73.8 ~ 109.0) 84.0(73.0 ~ 108.0) < 0.01 6 cycles PBAC, median (IQR) 16(25.4) 385.0(246.3 ~ 450.0) 0(0 ~ 0) 0(0 ~ 0) 0(0 ~ 0) < 0.01 Hb(g/L), median (IQR) 16(25.4) 63.0(46.8 ~ 83.5) 82.0(77.3 ~ 107.8) 105.5(86.0 ~ 121.5) 99.5(85.5 ~ 133.5) < 0.01 PBAC Pictorial blood assessment chart, Hb Hemoglobin Comparison before and after the use of mifepristone for BD patients PBAC Pictorial blood assessment chart, Hb Hemoglobin There were no meaningful changes in coagulation parameters ( n  = 32) included prothrombin time(PT), activated partial thromboplastin time(APTT), or fibrinogen levels before and after treatment (Table  3 ; Fig.  3 ). No thromboembolic events were observed during therapy.Three patients received 12 months without evidence of endometrial hyperplasia or malignancy by endometrial biopsy. All patients underwent liver function testing every one to three months after starting treatment. Telephone follow-up was conducted for 27 patients, and liver function tests at local hospitals were found to be normal. The remaining 39 patients were monitored at our hospital. Before treatment, their median alanine aminotransferase(ALT) was 16 U/L (IQR 11–27) and median aspartate aminotransferase (AST) was 18 U/L (IQR 15–23). After treatment, median ALT was 17 U/L (IQR 13–25) and median AST was 20 U/L (IQR 16–26). These changes were not statistically significant (Table  3 ; Fig.  3 ).Seven patients had mildly elevated liver enzymes before starting treatment. Six were undergoing myeloablative chemotherapy before hematopoietic stem cell transplantation(HSCT), and one was in the acute phase of hepatitis E. All received mifepristone together with hepatoprotective therapy. After treatment, liver enzyme levels normalized in six patients; the remaining patient showed improvement, with ALT declining from 53 U/L to 47 U/L.Three additional patients developed mild liver enzyme elevations after initiating mifepristone, at treatment durations of 1, 2, and 9 months. The two with shorter exposure were also receiving myeloablative chemotherapy. All three were treated with hepatoprotective agents, and their liver enzyme levels normalized within two months. Table 3 Comparison in coagulation and liver function parameter valuables n Pre-treatment Post-treatment P PT(s), median (IQR) 32 11.5(11.1–12.1) 11.3(10.7–11.9) 0.41 APTT(s), median (IQR) 32 30.9(27.1–33.7) 30.8(27.6–33.2) 0.87 FIB(mg/dL), median (IQR) 32 331.0(279.0-371.0) 326.0(287.0-373.0) 0.81 ALT(U/L), median (IQR) 39 16(11–27) 17(13–25) 0.82 AST(U/L), median (IQR) 39 18(15–23) 20(16–26) 0.55 PT Prothrombin time, APTT Activated partial thromboplastin time, FIB Fibrinogen, ALT Alanine aminotransferase, AST Aspartate aminotransferase Comparison in coagulation and liver function parameter PT Prothrombin time, APTT Activated partial thromboplastin time, FIB Fibrinogen, ALT Alanine aminotransferase, AST Aspartate aminotransferase Fig. 3 Comparison in coagulation and liver function parameters. A  PT, prothrombin time; B  APTT, activated partial thromboplastin time; C  FIB, fibrinogen; D  ALT, alanine aminotransferase; E  AST, aspartate aminotransferase Comparison in coagulation and liver function parameters. A  PT, prothrombin time; B  APTT, activated partial thromboplastin time; C  FIB, fibrinogen; D  ALT, alanine aminotransferase; E  AST, aspartate aminotransferase Long-term outcomes were defined as clinical status assessed at ≥ 3 months after treatment initiation, including hemoglobin normalization, maintenance of amenorrhea, or transition to definitive therapies (HSCT, hysterectomy).Among the 63 patients, 33 (52.4%) achieved hemoglobin normalization and clinical stability, enabling further treatment. Eight patients with hematologic malignancies underwent HSCT while maintaining mifepristone-induced amenorrhea during conditioning; therapy was discontinued after engraftment with platelet count > 50 × 10 9 /L. Most patients subsequently experienced ovarian failure. Eight patients opted for levonorgestrel intrauterine system (LNG-IUS) insertion, with biopsies confirming no endometrial atypia. LNG-IUS retention was 100%, with amenorrhea in six and minimal spotting in two. Two patients with uterine fibroids and thrombotic disease underwent hysterectomy after 6 months of treatment; pathology showed proliferative or secretory endometrium without atypia. Two patients received etonogestrel implants(one VWD patient and one ITP patient combined with brain fog), one patient endometrial ablation, three patient discontinued mifepristone for thromboembolism resolution, recovery from allergic purpura and menopause. The remaining 29 patients (46.0%) continued mifepristone with planned escalation (Fig.  4 ). Fig. 4 Long-term outcomes. HSCT, hematopoietic stem cell transplantation; LNG-IUS, levonorgestrel intrauterine system Long-term outcomes. HSCT, hematopoietic stem cell transplantation; LNG-IUS, levonorgestrel intrauterine system

This retrospective cohort investigation was approved by the Institutional Review Board of Peking University People’s Hospital (2021PHB093-001) in full compliance with the Declaration of Helsinki and Good Clinical Practice guidelines. We reviewed electronic medical records at department of gynecology, Peking University People’s Hospital from 2010 to 2024. Eligible participants were females aged 10–55 years with BD especially included those with significant thrombocytopenia or receiving antithrombotic therapy and refractory HMB treated with mifepristone. Refractory HMB was validated independently by both a hematologist and a gynecologist.The primary objective was to evaluate bleeding control, assessed by change in pictorial blood assessment chart (PBAC) score. Secondary objectives included hemoglobin recovery, amenorrhea rate, quality-of-life improvement, and safety outcomes. All patients with HMB underwent an initial hemostatic system screening. Forty-eight patients with thrombocytopenia, laboratory tests included blood smear, bone marrow aspiration, platelet function tests, coagulation profile, liver function tests, and routine biochemical parameters. Although thrombocytopenia is conventionally defined as a platelet count < 100 × 10⁹/L, a lower threshold of < 50 × 10⁹/L was used to identify a clinically high-risk subgroup, as patients below this level generally exhibit greater bleeding severity and suboptimal response to standard hemostatic therapies. This cutoff was therefore applied as the inclusion criterion for the thrombocytopenic subgroup. Thirteen individuals requiring anticoagulation therapy following thrombotic events, renal dialysis, or cardiac stent placement. Thrombotic or structural diagnoses were confirmed by appropriate imaging modalities (e.g., Doppler ultrasound, computed tomography, or angiography). Additionally, these patients underwent coagulation analysis and coagulation factor testing. Exclusion criteria were defined as follows: (1) patients whose HMB was primarily attributable to structural or gynecologic causes unrelated to BD (e.g., fibroids or adenomyosis); (2) patients with BD but without HMB; (3) patients who had previously undergone endometrial ablation or uterine artery embolization; (4) patients who had endometrial pathology biopsy indicating simple hyperplasia, complex hyperplasia, or atypical hyperplasia; (5) patients with a history of gynecological malignancies or precancerous lesions involving the uterus, cervix, endometrium, ovaries, or breasts; (6) patients with a positive pregnancy test during the screening period or lactation; (7) patients with incomplete clinical data; and (8) patients who lost to follow-up. Baseline demographic and clinical data were extracted from medical records, including hemoglobin levels, coagulation parameters, liver function and PBAC scores, which quantified blood loss based on the proportion of sanitary pad staining and clot size. A cumulative monthly PBAC score > 100 corresponded to menstrual blood loss > 80 mL, consistent with HMB [ 18 , 19 ]. We also collected Menorrhagia Multi-Attribute Scale (MMAS) scores, which assessed six domains: practical difficulties, social life, psychological health, physical health, work and daily routine, and family relationships. Scores ranged from 0 (severe impact) to 100 (no impact). The MMAS was validated with strong reliability and internal consistency [ 20 ]. These measures were recorded at baseline and after treatment. When necessary, telephone interviews were conducted to verify incomplete records. Mifepristone therapy was initiated during any menstrual phase in patients without acute hemorrhage. The standard regimen was 25 mg once daily for 1 month, with tapering to 12.5 mg once daily upon achieving sustained amenorrhea. For acute HMB, additional interventions were used. Fifteen patients required intrauterine Foley balloon placement (12–24 h) alongside mifepristone, achieving bleeding control within 7 days. Nineteen patients received one COC tablet or 5 mg norethisterone every 8 h. After bleeding was controlled, patients began to receive 25 mg mifepristone once daily for 1 to 2 weeks. Afterward, COC or norethisterone was discontinued. All patients eventually transitioned to mifepristone monotherapy for maintenance. Follow-up assessments were conducted at 1, 3, and 6 months through clinic visits or follow-up by telephone. Blood tests, PBAC diaries and MMAS scores were collected. Data were analyzed using SPSS version 27.0 (IBM Corp., Armonk, NY, USA). Continuous variables are presented as medians (interquartile range, IQR). Paired comparisons between baseline and follow-up were analyzed using the Wilcoxon signed-rank test. Repeated measurements across multiple time points were assessed using the Friedman test for paired samples, followed by Bonferroni-adjusted post hoc pairwise comparisons to control the family-wise error rate. All reported P values for repeated measures were Bonferroni-adjusted unless otherwise specified. Because this study was a retrospective single-arm cohort including all eligible patients treated during the study period, an a priori sample size or power calculation was not feasible. Therefore, the analyses should be considered exploratory and hypothesis-generating.

In this retrospective cohort, low-dose mifepristone was associated with marked reductions in menstrual blood loss, increased hemoglobin levels, and acceptable tolerability, without clinically significant changes in coagulation or liver function parameters. These findings suggest that mifepristone may serve as a potential bridge therapy for women with BD and refractory HMB, particularly those with severe thrombocytopenia or receiving antithrombotic treatment, and may facilitate subsequent long-term menstrual management. Prospective controlled studies are warranted to confirm these observations.

To date, no studies have published about the use of mifepristone for HMB associated with BD. Many high-quality clinical trials on HMB therapies have excluded patients with BD, leaving a significant evidence gap for this group, who often face acute and severe bleeding. Mifepristone has shown effectiveness in inducing amenorrhea in perimenopausal AUB [ 12 ] and uterine fibroids [ 13 ] when given at low doses (5–25 mg daily).In this context, our study explores whether low-dose mifepristone is a potential bridge therapy for refractory HMB in women with BD. Current pharmacologic options for HMB in women with BD include COC, antifibrinolytics, and desmopressin. However, their effectiveness may be limited in patients with severe thrombocytopenia or coagulation defects, and breakthrough or withdrawal bleeding may still occur. For women receiving antithrombotic therapy, the use of COC or tranexamic acid may additionally raise concerns regarding thrombotic risk, creating a therapeutic dilemma between bleeding control and thrombosis prevention. The LNG-IUS is widely recommended as an effective long-term strategy for menstrual regulation in women with BD [ 21 , 22 ]. Nevertheless, its role in acute or severe bleeding situations may be less optimal, as immediate hemostatic control is often required. In our cohort, many patients presented with active heavy bleeding, profound thrombocytopenia, or were scheduled for HSCT, circumstances in which device insertion could be technically challenging and potentially associated with early expulsion or infection. Moreover, higher expulsion rates of the LNG-IUS have been reported in women with BD [ 23 ], likely related to ongoing heavy bleeding and impaired hemostasis. Although some studies suggest similar expulsion rates compared with contraceptive users [ 24 ], the acute hemorrhage and severe thrombocytopenia in our cohort led to deferral of LNG-IUS as first-line therapy. Therefore, in these high-risk clinical settings, short-term pharmacologic suppression was considered a practical bridging approach before transition to definitive long-term options such as LNG-IUS or surgery. While GnRH agonists can also induce amenorrhea, their initial flare-up effect may transiently exacerbate bleeding, limiting their suitability in unstable patients. In addition, newly available oral GnRH antagonists offer rapid ovarian suppression without a flare phenomenon and may represent another promising alternative; however, evidence in women with BD remains limited, and their availability and accessibility in routine clinical practice are still restricted in many settings. Collectively, these considerations highlight the need for additional safe and rapidly acting medical options in this vulnerable population. Mifepristone exerts its therapeutic effect through both central and peripheral mechanisms. Centrally, it suppresses the hypothalamic–pituitary–ovarian axis [ 20 , 25 ], while peripherally it alters endometrial vascularity and inhibits endometrial proliferation [ 26 – 28 ]. Together, these actions contribute to sustained amenorrhea and effective bleeding control. An added benefit is its lack of adverse effects on coagulation. In our study, coagulation parameters remained stable throughout treatment, in line with findings from multicenter pharmacological studies in abortion cohorts, which also report no significant coagulation-related effects [ 16 , 17 ]. Our study achieved a 92.1% amenorrhea rate within one month of mifepristone initiation, suggesting considerable therapeutic potential.These results align with findings from non-BD populations, such as a study by Shi et al. (93.8% amenorrhea after two months) [ 12 ] in perimenopausal AUB patients, as well as trials in fibroid and endometriosis populations, which also demonstrated significant effects. For example, a placebo-controlled trial of 50 mg/day mifepristone for fibroid patients reported an 86% amenorrhea rate by month 2 and 100% by month 3, compared to no effect in the placebo group [ 13 ].Although our study used a lower dose, it achieved a similar three-month amenorrhea rate and a slightly faster response (91.2% vs. 86%) within the first month. This suggests that lower doses of mifepristone might offer therapeutic benefits without unnecessary exposure. Evidence from endometriosis trials supports a dose-response relationship [ 29 ]. In a large study by Carbonell et al., daily doses of 2.5, 5, and 10 mg produced progressively higher six-month amenorrhea rates (78.7%, 97.8%, and 98.9%), compared with 1.1% in the placebo arm. In our cohort, all participants achieved amenorrhea, which may be attributed to the relatively higher doses administered. Despite its efficacy, our study observed that mifepristone takes approximately 7 days to control bleeding. For acute and severe bleeding, adjunctive therapies such as intrauterine tamponade remain necessary for immediate control, after which mifepristone sustains long-term suppression. In this study, it functions as a bridge therapy, stabilizing patients during high-risk bleeding phases, particularly for hematologic patients requiring HSCT, who need to remain amenorrheic during chemotherapy and transplant period, while also facilitating transition to interventions such as the LNG-IUS. Our study demonstrated that mifepristone pretreatment may also reduce LNG-IUS expulsion risk, supporting a staged management strategy. Safety remains a central consideration. A systematic meta-analysis confirms that mifepristone is effective in inducing amenorrhea, with a proven safety profile when used for up to six months [ 15 ]. While long-term use may cause benign and reversible endometrial changes——progesterone receptor modulator–associated endometrial changes (PAECs) [ 25 , 30 ], our study found no pathological endometrial changes, even among patients treated for up to 12 months. Mifepristone is well-tolerated in terms of liver safety; while some studies report mild-to-moderate transaminase elevations, these typically resolve after discontinuation of the drug [ 25 , 31 ]. In our cohort, liver function remained stable, and enzyme fluctuations were likely due to myeloablative chemotherapy. Limitation of this study is that it is a small, retrospective, and exploratory analysis. Future prospective, large-scale randomized controlled trials are needed to further validate these findings.

Bleeding disorders (BD), whether inherited or acquired, disrupt normal hemostasis [ 1 ] and often manifest as abnormal uterine bleeding (AUB). These conditions encompass abnormalities in platelet count, as well as iatrogenic causes such as the use of antithrombotic medications.The National Institute for Health and Care Excellence in the United Kingdom defines heavy menstrual bleeding (HMB) as the excessive loss of menstrual blood, which substantially impacts a woman’s physical, social, emotional, and material well-being [ 2 ]. HMB affects approximately 30% of women globally, posing a significant challenge as it negatively impacts physical activity, work efficiency, and sexual health, often necessitating gynecologist consultation [ 3 , 4 ]. Among women with BD, the occurrence of HMB is significantly higher [ 5 , 6 ]. Various research findings indicate a rise in BD among adolescents with HMB, with a prevalence range of 10% to 48% [ 7 – 11 ].The consequences extend beyond quality of life, contributing to anemia, fatigue, and even hemodynamic instability, often requiring urgent medical intervention. Conventional therapies such as combined oral contraceptives(COC) and tranexamic acid provide limited benefit for women with BD because of their underlying hemostatic defects. Breakthrough bleeding during COC use, as well as withdrawal bleeding after dose reduction or discontinuation, may trigger recurrent episodes of acute heavy bleeding in patients with platelet disorders. For women who experience HMB during antithrombotic therapy, COC and tranexamic acid carry an added risk of precipitating thrombotic events, creating a therapeutic dilemma between bleeding control and thrombosis prevention. As a result, patients with BD urgently require a treatment that can reliably and rapidly reduce menstrual blood loss without compromising safety. Mifepristone, a selective progesterone receptor modulator, has shown effectiveness in inducing amenorrhea and improving hemoglobin levels in perimenopausal AUB [ 12 ] and uterine fibroids [ 13 ] when given at low doses (5–25 mg daily). Recent evidence further supports its role in reducing fibroid-related bleeding. A 2024 meta-analysis demonstrated that mifepristone across different dosing regimens significantly reduced menstrual blood loss and improved anemia with an acceptable safety profile [ 14 ].Previous studies have demonstrated that mifepristone is effective with minimal adverse effects [ 15 ], and it has no measurable impact on coagulation function [ 16 , 17 ]. These attributes make mifepristone an appealing option for women with BD who require rapid and dependable menstrual suppression. Given these properties, we conducted the first exploratory study evaluating low-dose mifepristone as a bridge therapy specifically for women with BD experiencing refractory HMB—with deliberate inclusion of those with significant thrombocytopenia and those receiving antithrombotic therapy, two high-risk groups routinely excluded from clinical trials. The study investigated whether mifepristone could effectively control bleeding, increase hemoglobin levels, and had no significant impact on coagulation function.