Hematological Evaluation and Management of Menorrhagia in Adolescents: Lessons from Adolescent Heavy Menstrual Bleeding Clinics.

We identified 77 adolescent females who underwent evaluation for HMB in our clinics. The median age of the cohort was 14 years (range: 10–19 years), and the median age at menarche was 12 years (range: 9–15, IQR 12–13). The median bleeding score was 4 (range: 1–7). Table 1 summarizes the clinical data of the study cohort. Characteristics of the entire study cohort and according to the presence of a bleeding disorder Patients without and with bleeding disorders were compared for continuous variables by using Mann-Whitney U test and for categorical variables by Fisher’s exact test. The resulting p values are presented in the table. IQR, interquartile range; BAT, bleeding assessment tool; PRBC, packed red blood cells; n , number of patients. Thirty-eight (49%) patients were hospitalized, of whom all but one presented with anemia. Thirty-three patients (43%) had symptomatic anemia including fatigue, paleness, dizziness, heart palpitations, pre-syncope, and syncope. A bleeding diathesis was diagnosed in 19 patients, including VWD ( n = 9), various factor deficiencies ( n = 3), and platelet disorders ( n = 7), delineated in Table 2 . Only 7/77 patients (9%) had a family history of a known bleeding disorder. A family history of menorrhagia was reported by 22/77 patients (28.5%) of which only 3 had an underlying bleeding disorder themselves. Characteristics of patients with bleeding disorders in our cohort VWD, von Willebrand disease; RiCof, ristocetin cofactor; rVWF, recombinant Von Willebrand factor; TXA, tranexamic acid; Epi, epinephrine; Col, collagen; Risto, ristocetin; HPS, Hermansky-Pudlak syndrome; factor VIII, FVIII. a The cutoff for a positive or abnormal bleeding score is ≥6 in adult females and ≥3 in children. Gynecologic and endocrine disorders detected in our cohort included PCOS (6 patients), anovulatory cycles (7 patients), and endometriosis (1 patient). Although thyroid function tests were performed, no cases of thyroid pathology were identified (data not shown). Hormonal therapy was administered at the discretion of the treating gynecologist. Notably, no patient required surgery. We did not find significant differences in the median age at presentation, median age at menarche, or median time from menarche to presentation between patients with or without a diagnosed bleeding disorder. Similarly, no differences were observed regarding family history of bleeding, hospitalization rate, hemoglobin level at presentation, or BAT score (see Table 1 ). No significant difference in the proportion of patients with symptomatic anemia at presentation was observed between the patients with and without a bleeding disorder ( p = 0.424). Notably, a higher BAT score correlated with lower hemoglobin levels ( r s = −0.56, p < 0.001). Hospitalized patients had a higher median BAT score compared to nonhospitalized patients (4 vs. 3, p < 0.001) ( Fig. 1 ); however, no significant difference in BAT score was observed between patients with and without an underlying bleeding disorder ( p = 0.194). Correlation between the bleeding score and hemoglobin level. Each data point represents an individual patient. The data were analyzed using the Spearman’s rank-order correlation test. In adolescents with HMB, higher bleeding scores correlated with lower hemoglobin levels. Among the total cohort of 77 patients, 65 received tranexamic acid. Of those who were hospitalized, 27/38 required packed red blood cell (PRBC) transfusion during hospitalization. Six out of the 27 patients who received PRBC transfusion were diagnosed with a bleeding disorder. The patient with Glanzmann thrombasthenia was administered platelet transfusion, while the patient with VWD type 3 received factor replacement therapy. At the time of referral to our clinic, 20/77 patients (26%) were already receiving oral iron supplementation, of whom only 4 had a bleeding disorder. Intravenous iron formulations, such as iron sucrose, ferric carboxymaltose, or ferric gluconate, were administered to 13 patients (17%). Overall, 48/77 patients (62%) initiated hormonal therapy, either with a combined estrogen-progestin (39/48, 81%) or progestin-only pills (9/48, 19%). Of the 19 patients with a bleeding disorder, 10 (52.6%) started hormonal treatment, with 9 receiving a combined estrogen-progestin pills and 1 receiving progestin-only pills. The majority of patients diagnosed with a bleeding disorder were treated with TXA during menses. This therapeutic management was effective in resolving symptoms of abnormal uterine bleeding.

This study presents data from collaborative adolescent hematology and gynecology clinics, evaluating the characteristics of adolescents with HMB with and without an underlying bleeding disorder. Overall, we could not differentiate between adolescents with HMB resulting from bleeding disorders to those related to other etiologies based on their clinical presentation. Previous reports have demonstrated the yield of BAT in identifying patients with VWD [ 15 ]. In our study, BAT score did not significantly differ between patients with and without an underlying bleeding disorder. This may stem from the fact that various bleeding disorders were identified within our cohort. In addition, it is possible that as all patients were pediatric – they did not experience suffice hemostatic challenges. Our findings align with the report by Fasulo et al. [ 16 ] in which BAT scores did not predict further bleeding episodes. Additional prospective studies are warranted to identify potential predictive markers for recognizing adolescents at risk of bleeding disorders among those presenting with HMB. In a multicenter prospective US cohort study of HMB in adolescents, Zia et al. [ 17 ] hypothesized that menstrual bleeding pattern would not predict the presence of a bleeding disorder. They found that the median time from the onset of first bleeding symptoms to diagnosis was 4 years for the entire cohort, with a shorter duration of 2 years for the anovulatory patient group and a longer duration of 6 years for the ovulatory group. A possible explanation for this finding is that adolescents with anovulatory HMB may experience more pronounced or challenging menstrual episodes, leading to earlier referral and diagnosis. In our study, no difference in age at referral was noted between girls with or without bleeding disorders. These results are in line with the findings of Lavin et al. [ 18 ] who reported no differences in age at diagnosis for women with HMB and VWD compared to those who did not seek medical counseling. Additionally, our data suggest that familial menorrhagia is more prevalent among adolescents without a diagnosed bleeding disorder. This observation could potentially indicate the presence of other hereditary or hormonal disorders that were not fully assessed in this study, such as PCOS. Although the BAT score has been shown to be an effective predictive tool for identifying patients with VWD [ 15 ], its utility appears to be limited in the setting of adolescents presenting with HMB. Further prospective studies are warranted to identify potential predictive markers for recognizing adolescents at risk of bleeding disorders among those presenting with HMB. The American College of Obstetricians and Gynecologists (ACOG) 2019 guidelines for the management of adolescents with HMB recommend a combination of hormonal therapy and antifibrinolytics as first-line treatment to control the acute presentation of HMB, irrespective of an underlying bleeding disorder [ 1 ]. However, ACOG suggests maintenance hormonal therapy or intrauterine device (IUD) insertion and does not address the role of prophylactic antifibrinolytics during menstruation. In our cohort, TXA was administered to most patients at the physician’s discretion. In contrast to ACOG, Zia et al. in their expert opinion panel, consider TXA as one of the first‐line agents for HMB control [ 19 ]. In the 2018 Cochrane systematic review by Bryant-Smith et al. [ 20 ], a notable decrease in blood loss was observed in patients with HMB treated with antifibrinolytics compared to placebo. When comparing TXA with progestogens, no clear evidence of difference in mean blood loss was found. However, TXA was associated with a higher likelihood of improvement. Conversely, the use of IUDs was associated with a significant reduction in blood loss compared to TXA [ 20 ]. We suggest a trial of perimenstrual antifibrinolytics as a single agent merit consideration before offering hormonal therapy or an IUD, as many young girls and their families might be reluctant to engage in hormonal therapy or undergo IUD insertion at a young age. A relatively high proportion of patients in our study received blood transfusions. This may partially stem from referral bias as the study was conducted in large tertiary centers. However, it should be highlighted that blood transfusion should only be administered as a last resort in most of patients with anemia secondary to HMB. In most cases, the use of antifibrinolytics and hormonal therapy as well as intravenous iron administration should suffice. Our study has several limitations. First, a population bias cannot be ruled out, as the cohort included only patients referred to our tertiary specialist clinics, potentially excluding individuals with milder HMB symptoms. Second, as most patients without a diagnosed bleeding disorder were not further followed at our clinic, the efficacy of treatment strategies cannot be assessed in this subgroup of patients. A major strength of this study lies in the large cohort of adolescents evaluated at specialized clinics, providing valuable real-world data on the diagnosis and management of HMB in this population. In conclusion, adolescents with HMB resulting from gynecologic etiologies had a similar presentation compared to those with bleeding disorders. Therefore, we advocate for conducting a comprehensive bleeding evaluation in all adolescents with HMB, even within their first year post-menarche. We encourage primary care physicians to refer these adolescents to specialized HMB clinics for a comprehensive multidisciplinary evaluation and treatment involving the young patient in a shared decision-making process.

Heavy menstrual bleeding (HMB), or menorrhagia, is defined as excessive menstrual blood loss that interferes with a woman’s physical, social, emotional, or material quality of life [ 1 ]. It is estimated to occur in approximately 37% of adolescent females [ 2 , 3 ]. Screening questionnaires for identifying women with menorrhagia include evaluation of menstrual characteristics, such as a duration of menses ≥7 days, the need to change sanitary products more frequently than every 2 h, flooding, a history of treatment of anemia, a family history of diagnosed bleeding disorders, or a personal history of excessive bleeding in other circumstances [ 1 ]. The main etiologies for HMB in adolescent girls are anovulation, mostly due to the immaturity of the hypothalamus-pituitary-gonadal axis, and inborn or acquired bleeding disorders, mainly von Willebrand disease (VWD) and platelet function disorders [ 4 ]. Additionally, endocrine disorders such as polycystic ovary syndrome (PCOS) and thyroid disorders, as well as infectious diseases, may contribute to HMB [ 5 ]. Bleeding disorders have been reported in 10–62% of females with HMB [ 4 ]. The most common inherited bleeding disorder is VWD, while 1% of the population has laboratory results compatible with VWD, only about 1/1,000 individuals have a bleeding phenotype [ 6 ], and HMB is the most common symptom in affected women [ 7 ]. About 3–36% of adolescent girls suffering from HMB have been diagnosed with VWD [ 8 ]. Notably, bleeding disorder of unknown cause may also contribute to HMB. However, data regarding the diagnosis and management of HMB with or without an underlying bleeding disorder in the adolescent population are scarce. This study aimed to evaluate the hemostatic workup, identified etiologies, and management of adolescents with HMB presenting at two tertiary centers. We sought to compare the differences between HMB patients who were found to have a bleeding disorder versus HMB patients without an identified bleeding disorder.

The authors have no conflict of interest to declare, except: Sarina Levy-Mendelovich receives grants from Pfizer, Novo Nordisk and honoraria from Pfizer. Gili Kenet receives grant and research support from BSF, Novo Nordisk, Pfizer, Roche, Tel Aviv University and honoraria for consultancy/lectures from Bayer, BioMarin, CSL, Pfizer, Sanofi Genzyme, Sobi, Spark, Takeda, Uniquore. Assaf Arie Barg receives honoraria for lectures from Roche.

This study was not supported by any sponsor or funder.

We conducted a cross-sectional study of adolescents who presented with HMB to two tertiary centers between 2014 and 2022. Data were collected by the authors (consulting pediatric hematologists with expertise in thrombosis and hemostasis). Data included demographic characteristics, age at menarche, age at presentation, personal and family history of bleeding, and presence of comorbidities. Information on hospitalization due to HMB or severe anemia and iron deficiency was documented. The laboratory evaluation data extracted included: complete blood count, prothrombin time (PT), international normalized ratio, activated partial thromboplastin time, fibrinogen levels, platelet light transmission aggregometry, von Willebrand factor (VWF) antigen (VWF:Ag) and activity (VWF:RCo), and coagulation factor VIII activity. For specific patients exhibiting prolonged PT or activated partial thromboplastin time, additional coagulation factor activity assays were performed, including factor VII or factors IX, XI, and XII, respectively. All treatment modalities administered were documented. In addition, a hormonal profile (including sex hormones and thyroid function) was tested. The study was approved by the Local Institutional Review Boards (IRB) prior to initiation in accordance with the Declaration of Helsinki. 1. HMB was defined by menstrual duration ≥7 days and/or a sensation of “flooding” or bleeding through a pad or tampon in less than 2 h [ 9 , 10 ]. 2. Anemia was defined as hemoglobin <11 g/dL and iron deficiency as serum ferritin levels ≤20 ng/mL [ 11 ]. Iron deficiency anemia was defined when both anemia and ferritin levels of ≤20 ng/mL were present [ 9 ]. 3. The bleeding score was determined using the ISTH-Bleeding Assessment Tool (ISTH-BAT), which includes data on the frequency and severity of hemorrhagic symptoms. It comprises 14 categories for assessing bleeding symptoms retrospectively. The normal range is 0–5 for adult females and 0–2 for children of both sexes [ 12 ]. 4. Laboratory confirmation of bleeding disorders: a clotting factor deficiency was defined as a particular clotting factor level falling below the age-adjusted reference range, and bleeding disorder were defined in concordance with the British Committee for Standards in Haematology [ 13 ]. VWD was classified in accordance with the ASH 2021 guidelines [ 14 ]. A platelet aggregation disorder was diagnosed when impaired aggregation was observed in repeated tests. HMB was defined by menstrual duration ≥7 days and/or a sensation of “flooding” or bleeding through a pad or tampon in less than 2 h [ 9 , 10 ]. Anemia was defined as hemoglobin <11 g/dL and iron deficiency as serum ferritin levels ≤20 ng/mL [ 11 ]. Iron deficiency anemia was defined when both anemia and ferritin levels of ≤20 ng/mL were present [ 9 ]. The bleeding score was determined using the ISTH-Bleeding Assessment Tool (ISTH-BAT), which includes data on the frequency and severity of hemorrhagic symptoms. It comprises 14 categories for assessing bleeding symptoms retrospectively. The normal range is 0–5 for adult females and 0–2 for children of both sexes [ 12 ]. Laboratory confirmation of bleeding disorders: a clotting factor deficiency was defined as a particular clotting factor level falling below the age-adjusted reference range, and bleeding disorder were defined in concordance with the British Committee for Standards in Haematology [ 13 ]. VWD was classified in accordance with the ASH 2021 guidelines [ 14 ]. A platelet aggregation disorder was diagnosed when impaired aggregation was observed in repeated tests. Statistical analysis was performed with IBM SPSS Statistics (version 23.0; IBM Corp.). Continuous variables were presented as median, range, and/or interquartile range (IQR); categorical variables were presented as counts and/or percentages. Patients without and with bleeding disorders were compared for continuous variables by using the Mann-Whitney U test and for categorical variables by the Fisher’s exact test. Correlation between continuous variables was assessed by using the Spearman’s rank-order correlation test. Two-tailed p values of less than 0.05 were considered statistically significant.

This study was performed in line with the principles of the Declaration of Helsinki. This study protocol was reviewed and approved by the Institutional Review Boards of Sheba Medical Center (Approval No. 7828-20-SMC) and Tel Aviv Sourasky Medical Center (Approval No. 0649-22-TLV). Written informed consent from the parent/legal guardian of participants was not required for this retrospective study in accordance with local/national guideline.

Marganit Benish and Adi Shitrit Yarhi: investigation, methodology, conceptualization, writing – original draft, and writing – review and editing. Ivan Budnik: formal analysis and writing – review and editing. Chagit Klieger, Noah Gruber, and Yael Harel: data curation and writing – review and editing. Gili Kenet: writing – review and editing. Assaf Arie Barg and Sarina Levy-Mendelovich: investigation, conceptualization, methodology, and supervision.