Original Study Hemostatic Abnormalities in Young Females with Heavy Menstrual Bleeding Rosa Dıaz MD 1,*, Jennifer E. Dietrich MD, MSc 1,2, Donald Mahoney Jr. MD 1, Donald L. Yee MD 1, Lakshmi V. Srivaths MD 1 1 2
Texas Children’s Hospital, Houston, TX Baylor College of Medicine, Houston, TX
a b s t r a c t Objective: To study the prevalence of hemostatic abnormalities, including bleeding disorders and risk factors, in young females referred to a multidisciplinary clinic for evaluation of heavy menstrual bleeding (HMB). Methods: Retrospective chart review was undertaken for 131 post-menarchal girls with HMB, 7 to 17 years of age, enrolled in the institutional ‘Menorrhagia Data Registry’ protocol. The diagnostic approach included: (1) complete blood count, prothrombin time, partial thromboplastin time, fibrinogen, von Willebrand panel (2) platelet aggregometry, specific clotting factor assay, fibrinolytic pathway analysis, and factor XIII level as needed. The prevalence of hemostatic abnormalities and the prognostic significance of clinical variables associated with hemostatic abnormalities in young girls with HMB were evaluated. Results: A hemostatic abnormality was identified in 69 (53%) young girls with HMB. Of these, 27 (21%) had an underlying bleeding disorder and 42 (32%) had a risk factor for bleeding, namely low von Willebrand factor activity. A larger number of girls with underlying bleeding disorder had personal history of other bleeding symptoms (48% vs 31%) and bleeding after surgical or dental procedure (25% vs 8%) when compared to females without hemostatic abnormality. Furthermore, girls with risk factor for bleeding (low vWF activity) were more likely to have bleeding after surgical or dental procedure (15% vs 8%) and family history of bleeding (79% vs 60%) than patients without hemostatic abnormality. Conclusions: There is high prevalence of hemostatic abnormalities, including bleeding disorders and risk factors, in young girls with HMB. These findings support comprehensive and systematic hemostatic evaluation in this group of patients. Key Words: Bleeding disorders, Young females, Heavy menstrual bleeding
Introduction
Heavy menstrual bleeding (HMB) is a common gynecologic complaint among young females.1 In many, anovulation and immaturity of the hypothalamic-pituitaryovarian axis are thought to be the underlying cause for heavy periods, which can often be managed with hormonal therapy. However, several studies have shown that 10%-62% of young girls with HMB have an underlying bleeding disorder that may benefit from the addition of non-hormonal therapies or may be relevant to future surgeries or pregnancies.2e7 Bleeding disorders associated with HMB include disorders of primary and secondary hemostasis such as clotting factors deficiencies, von Willebrand disease (vWD), platelet disorders, and rare fibrinolytic pathway defects. It is important to identify female adolescents with bleeding abnormalities because the obstetric and gynecologic morbidity related to bleeding disorders go beyond troublesome heavy periods. Iron deficiency anemia, fatigue, less
Dr. Dietrich is a consultant for CSL Behring and Bayer and conducts research for Duramed. Dr. Diaz, Dr. Mahoney, Dr. Yee, and Dr. Srivaths indicate no conflicts of interest. * Address correspondence to: Rosa Dıaz, MD, Department of Pediatrics, Section of Hematology/Oncology, Texas Children’s Hospital/Baylor College of Medicine, 1102 Bates Ave, Suite 1025, Houston, TX 77030; Phone: (832) 824-4836; fax: (832) 8251453 E-mail address: [email protected] (R. Dıaz).
time spent on desired activities, and difficulty performing schoolwork are common complaints among these young women.8 In addition, 8-18% of women with bleeding disorders have surgery for HMB and other complications including surgery for hemorrhagic ovarian cyst and hysterectomy which may be associated with significant bleeding.9 Despite the relatively high frequencies of bleeding disorders and risk factors in adolescents with HMB, such underlying hemostatic abnormalities are often not considered and a systematic evaluation not performed.10e12 This is in part due to the lack of consensus regarding which patients need to be tested, difficulty in discerning ‘normal’ from pathologic bleeding and the lack of an optimal diagnostic approach. The aim of the present study is to evaluate the prevalence of hemostatic abnormalities in young females with HMB by utilizing a step-wise, systematic diagnostic approach. We hypothesized that hemostatic abnormalities, including bleeding disorders and risk factors for bleeding, are prevalent in a substantial number of girls with HMB and that step-wise comprehensive laboratory testing is warranted in these patients. Patients and Methods Data Source and Study Population
Data collected from post-menarchal young females enrolled in the institutional Menorrhagia Data Registry
1083-3188/$ - see front matter Ó 2014 North American Society for Pediatric and Adolescent Gynecology. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jpag.2013.12.011
R. Dıaz et al. / J Pediatr Adolesc Gynecol 27 (2014) 324e329
approved by the institutional review board, referred to the multi-specialty Young Women’s Bleeding Disorder Clinic served by hematology and gynecology faculty and Hematology and Gynecology Clinics at Texas Children’s Hospital (TCH) for evaluation of HMB from 2009 to 2011, was analyzed retrospectively for this study. HMB is defined as more than 7 days of bleeding, greater than 80 ml of blood loss per cycle, pictorial blood assessment chart (PBAC) score of more than 100, or changing pads or tampons every 1-2 hours. All patients enrolled in this registry were referred by primary care physicians and/or gynecologists. Eligibility criteria included: (1) female gender, (2) post-menarche, (3) referred to TCH for HMB, and (4) non-pregnant. Data collected included demographics, clinical history and physical exam findings, laboratory study and imaging results, therapy details, and patient outcomes. Patients underwent concurrent evaluation by gynecology for pregnancy, anatomic pelvic abnormalities, pelvic infections, and endocrine abnormalities as possible causes for HMB, and only included in our study if these tests were normal or not thought to account for heavy menses. Definitions and Diagnostic Studies
All patients underwent complete blood count (CBC) to detect thrombocytopenia, coagulation studies for prothrombin time (PT), partial thromboplastin time (PTT), and fibrinogen to detect coagulopathy, and von Willebrand panel (vWP). Specific coagulation factor assays were done if PT, PTT, and/or fibrinogen were abnormal, to delineate specific clotting factor deficiency. Whole blood platelet aggregometry with secretion analysis was done in patients if the first level of testing was negative. In addition, fibrinolytic pathway testing, and factor XIII levels were performed to identify less common disorders for persistent, severe HMB, when deemed necessary by the treating physician. Studies for platelet aggregation and vWP were performed by the TCH Pathology Services. The vWP performed included individual tests for factor VIII (clot based assay), von Willebrand antigen (vWF: Ag), vW ristocetin co-factor activity (vWF: RCo) and multimers (MM). vWD was diagnosed based on the criteria described in the National Heart, Lung, and Blood Institute’s guideline “The diagnosis, evaluation, and management of von Willebrand disease 2008.13” Briefly, vWD is classified into 3 major categories: partial quantitative deficiency (type 1), qualitative deficiency (type 2) and total deficiency (type 3). Patients were diagnosed with type 1 vWD if they had vWF: Ag and/or vWF: RCo !30 IU/dL and normal MM. Those with vWF: RCo and/or vWF: Ag 30-50 IU/dL and normal MM were categorized as having low vWF activity, considered a risk factor for bleeding. Whole blood platelet aggregometry was performed using the Chronolog Whole-Blood Lumi-Aggregometer by measuring change in electrical impedance in whole-blood. The stimulus was brought about by using plateletaggregating agonists, including adenosine diphosphate (ADP), arachidonic acid, collagen, ristocetin, and thrombin. Adenosine triphosphate (ATP) release was measured by luminescence in whole-blood. The ATP released by the platelet dense granules binds with CHRONO-LUMEÔ
325
(luciferin-luciferase, Chrono-Log Corp, Havertown, PA) and generates a light (luminescence) that is measured by a stable, high-gain photomultiplier tube. The diagnosis of platelet function defect (PFD) required detection of at least 2 abnormalities in platelet aggregation and/or secretion, as described by Hayward et al.14 Testing for factor XIII deficiency was done by estimation of factor XIII activity using a chromogenic assay (normal range 57%-192%; Quest Diagnostics Nichols Institute, San Juan Capistrano, CA). Fibrinolytic pathway testing was done to evaluate for congenital deficiency of plasminogen activator inhibitor type 1 (PAI-1) and alpha-2 antiplasmin. Fibrinolysis panel including testing for PAI-1 activity (1.7-15 U/mL) was performed by the University of Washington Medical Center, Seattle, WA. Alpha-2 antiplasmin testing was performed by the Blood Center of SE Wisconsin using a chromogenic activity assay (normal range 75%-126%). Bleeding History
Data including demographic information, family history of bleeding, clinical symptoms and signs of bleeding, and laboratory test results on all patients were collected. Characteristics of menses including duration of menstrual period, number of pads and/or tampons used per day, presence and size of clots, flooding, and the PBAC score were collected. Information about other bleeding symptoms including epistaxis, gum bleeding, easy bruising, and excessive bleeding with dental procedures or other surgeries, hospitalization for management of severe menstrual bleeding and anemia and the need for red blood cell transfusions were gathered. Study Outcomes
The primary outcome for the study was prevalence of hemostatic abnormalities, including bleeding disorders and risk factors for bleeding, in young females with HMB. The secondary outcome included the prognostic significance of clinical variables associated with HMB, namely characteristics of menstrual bleeding, other bleeding symptoms, bleeding after surgical or dental procedure, family history of bleeding or bleeding disorder, personal history of anemia, red cell transfusion or hospitalization for HMB, which may help identify girls with HMB and hemostatic abnormalities. Statistical Analysis
The statistical analysis was performed using GraphPad Prism version 6.00 for Windows, GraphPad Software, San Diego, California, graphpad.com. Odds ratio and P values were calculated using the Fisher exact test from a contingency table. A P-value !.05 was considered statistically significant. Results Patient Characteristics
A total of 131 young females met eligibility criteria. The patients had a median age of 12 years (IQR 11-13) at the
326
R. Dıaz et al. / J Pediatr Adolesc Gynecol 27 (2014) 324e329
time of diagnosis with HMB. Thirty-seven percent (n 5 48) were white, 35% (n 5 46) were Hispanic, 17% (n 5 22) were Black, and 2% (n 5 3) were Asian. Two girls had other ethnicity including American Indian and Alaskan native and Hawaiian and other Pacific islander. Ethnicity was not specified for 10 patients. Primary Outcome: Prevalence of Hemostatic Abnormalities in Girls with HMB
We identified a hemostatic abnormality in 69 (53%) of the girls referred to our clinic due to HMB. Of these, 27 (21%) had an underlying bleeding disorder and 42 (32%) had a risk factor for bleeding, namely low vWF activity. PFD was the most commonly detected bleeding disorder in the patients (n 5 14, 11%). This was followed in frequency by vWD, clotting factor deficiency, thrombocytopenia, and PAI-1 deficiency, as shown in Table 1. Prognostic Significance of Clinical Variables: Bleeding Disorder vs No Hemostatic Abnormality
The median age of menarche and onset of HMB were similar for girls with bleeding disorders (11 years, IQR 10-12 and 12 years, IQR 11-14, respectively) and those without any hemostatic abnormality (12 years, IQR 11-13 and 12 years, IQR 11-13, respectively). In addition, menstrual bleeding characteristics such as duration of periods, number of pads/ tampons per day, passing clots during menses, flooding, and spotting were similar for both groups. We interrogated the prognostic utility of non-menstrual bleeding history for distinguishing girls with bleeding disorders from those without hemostatic abnormalities. Girls with underlying bleeding disorder were more likely to have personal history of other bleeding symptoms (48% vs 31%), and bleeding after surgical or dental procedure (25% vs 8%) when compared to those without hemostatic abnormality, as shown in Table 2. However, these results were not statistically significant. Within the entire group of patients, over half of the patients (n 5 66, 52%) were diagnosed with anemia, 18% (n 5 22) required red blood cell transfusion and 24% (n 5 31) needed hospitalization for management of bleeding symptoms and/or anemia. Prevalence of anemia, number of hospitalizations or transfusion history did not Table 1 Prevalence of Hemostatic Abnormalities among Young Girls with HMB Hemostatic Abnormalities Bleeding disorder PFD VWD Clotting factor deficiency Thrombocytopenia PAI-1 deficiency Risk factor for bleeding Low VWF activity No hemostatic abnormality
Table 2 Characteristics of Menses and Other Clinical Variables in Adolescents with Bleeding Disorder and without Hemostatic Abnormality Bleeding Disorder, N5 27 n (%) Clots Flooding Spotting Easy bruising Other bleeding symptoms Bleeding after procedure Family history Anemia PRBC transfusion Hospitalization
19 12 5 8 13
(70) (44) (19) (30) (48)
4/16 (25) 19 15 3 7
(70) (56) (11) (27)
No Hemostatic Abnormality, N5 62 n (%) 38 32 10 17 19
(61) (52) (16) (27) (31)
2/25 (8) 37 35 12 14
(60) (56) (19) (23)
OR (95% CI)
1.5 0.7 1.2 1.1 2.1
(0.5-3.9) (0.3-1.8) (0.3-3.8) (0.4-3.0) (0.8-5.3)
3.8 (0.6-24.0) 1.6 0.9 0.5 1.2
(0.6-4.2) (0.3-2.4) (0.1-2.0) (0.4-3.4)
P Value .477 .646 .766 1.000 .150 .186 .474 1.000 .538 .788
CI, Confidence interval; n, Number of girls with HMB; OR, Odds ratio; PRBC, Pure red blood cell
differ between girls with bleeding disorders and those without hemostatic abnormalities. Low vWF Activity: A Risk Factor for Bleeding?
We identified low vWF activity in 42 (32%) of the girls in the study (vWF: Ag and/or vWF: RCo median 37 IU/dL, range 31 to 49 IU/dL). First, we compared the personal and family history of bleeding of girls with bleeding disorder to patients with low vWF activity. Young females with bleeding disorder were more likely to pass clots (70% vs 52%) during their menses and have spotting (19% vs 5%) when compared to those with low vWF activity (Table 3). In addition, girls with bleeding disorders were more likely to have anemia (56% vs 38%) and bleeding after surgical or dental procedure (25% vs 15%). However, these results were not statistically significant. Easy bruising, other bleeding symptoms, family history of bleeding, number of hospitalizations and blood transfusion requirements were similar for both groups. Next, we compared girls with low vWF activity to those without hemostatic abnormalities. The patients with risk factor for bleeding were more likely to have bleeding after surgery or dental procedure (15% vs 8%), and more frequently had family history of bleeding (79% vs 60%, Table 4). These results were not statistically significant. All other variables such as passing clots, flooding, spotting, easy bruising, personal history of other bleeding symptoms, anemia, hospitalizations, and need for pure red blood cell (PRBC) transfusion, were similar for both groups, as summarized in Table 4.
Number of Girls (%); N 5 131 14 9 2 1 1
(11)* (7) (1)y (1) (1)z
42 (32) 62 (47)
N, Total number of girls with HMB * Four patients had combined bleeding disorders: PFD plus thrombocytopenia, low VWF activity or clotting factor deficiency. y One patient had combined hypofibrinogenemia and low VWF activity. z One patient had PAI-1 deficiency combined with low VWF activity.
Discussion
Our retrospective review showed that the frequency of hemostatic abnormalities is substantial among young girls referred for evaluation of HMB. An underlying bleeding disorder or risk factor was detected in 21% and 32% of girls evaluated, respectively, with PFD being the most commonly diagnosed bleeding disorder. The overall prevalence in our group of patients is consistent with previous reports. Mikhail et al3 by retrospective review identified a bleeding disorder in 41% of the 61 patients studied which included
R. Dıaz et al. / J Pediatr Adolesc Gynecol 27 (2014) 324e329 Table 3 Characteristics of Menses and Other Clinical Variables in Adolescents with Bleeding Disorder and Risk Factor for Bleeding Bleeding Disorder N 5 27 n (%) Clots Flooding Spotting Easy bruising Other bleeding symptoms Bleeding after procedure Family history Anemia PRBC transfusion Hospitalization
19 12 5 8 13
(70) (44) (19) (30) (48)
4/16 (25) 19 15 3 7
Risk Factor N 5 42 n (%) 22 20 2 15 16
(52) (48) (5) (36) (38)
3/20 (15)
(70) (56) (11) (27)
33 16 7 10
(79) (38) (17) (24)
OR (95% CI) 2.2 0.9 4.5 0.8 1.5
P Value
(0.7-6.0) (0.3-2.3) (0.8-25.4) (0.3-2.1) (0.6-4.0)
.209 .810 .108 .794 .460
1.9 (0.3-10.0)
.675
0.6 2.0 0.6 1.1
(0.2-1.9) (0.7-5.4) (0.1-2.6) (0.3-3.4)
.568 .215 .729 1.00
CI, Confidence interval; N, Total number of girls with HMB; OR, Odds ratio; PRBC, Pure red blood cell
vWD (36%) and PFD (7%). They defined vWD as vWF: RCo ! 40% and/or vWF: Ag !50%, and PFD as decreased aggregation to 1 or more agonists. A recent study by Vo et al6 of 105 patients reported a prevalence of 62% for bleeding disorders (PFD in 44% and vWD in 9% of the girls) in adolescents with HMB, wherein vWD was defined as vWF: RCo and/or vWF: Ag !40% and PFD was detected primarily by platelet electron microscopy with only some patients undergoing platelet aggregometry. A study by Philipp et al5 of 25 adolescents reported a hemostatic abnormality in 48% of the patients (PFD in 44% and vWD in 7% of the females) but did not clearly provide the definitions for bleeding disorders. Lastly, a study of 160 girls evaluated for menorrhagia reported only 16% with bleeding disorders, but did not utilize a uniform hemostatic evaluation and did not clearly define the diagnostic criteria.7 In comparison with these studies, we employed more stringent criteria to diagnose vWD and PFD based on recent publications,13,14 which likely accounts for the large number of patients identified with low vWF activity (32%), and lower prevalence of PFD (11%) and vWD (7%) in our study when compared to previous reports. Utilizing clinical variables such as characteristics of menses, personal history of other bleeding symptoms and family history of bleeding, to differentiate between ‘normal’ Table 4 Characteristics of Menses and other Clinical Variables in Adolescents with Risk Factor for Bleeding and Non-Hemostatic Abnormality Risk Factor N 5 42 n (%) Clots Flooding Spotting Easy bruising Other bleeding symptoms Bleeding after procedure Family history Anemia PRBC transfusion Hospitalization
22 20 2 15 16
(52) (48) (5) (36) (38)
3/20 (15) 33 16 7 10
(79) (38) (17) (24)
No Hemostatic Abnormality N 5 62 n (%) 38 32 10 17 19
(61) (52) (16) (27) (31)
2/25 (8) 37 35 12 14
(60) (56) (19) (23)
Odds Ratio (95%CI) 0.7 0.8 0.3 1.5 1.4
(0.3-1.5) (0.4-1.8) (0.1-1.2) (0.6-3.4) (0.6-3.2)
2.0 (0.3-13.5) 2.5 0.5 0.8 1.1
(1.0-6.1) (0.2-1.1) (0.3-2.3) (0.4-2.7)
P Value .421 .841 .116 .393 .526 .642 .055 .075 .800 1.000
CI, Confidence interval; n, Number of girls with HMB; OR, Odds ratio; PRBC, Pure red blood cell
327
and pathologic bleeding in young girls with chief complaint of heavy menses may provide an important tool to delineate the population that needs hemostatic evaluation. The secondary outcome measured by our study included a comparison of the bleeding characteristics, family history of bleeding and history of anemia, transfusion, and hospitalization between girls with and without underlying hemostatic abnormality. We did not find differences in menstrual characteristics, including length of period, passing clots, flooding, and spotting, in girls with and without bleeding disorder or hemostatic abnormality. This result is in contrast to a study by Vo et al6 in which the authors found statistically significant differences in reported regularity of patient’s periods, description of period flow, and number of days of each period that the bleeding was described as ‘heavy’ when they compared the bleeding profiles of females with and without bleeding disorders. Vo et al6 used a modified Ruta Menorrhagia Severity Scale to assess the menstrual bleeding profiles. The scale assessed the adolescent’s perception of whether periods are regular or irregular, heavy or light, but did not precisely measure length of period or severity. The difference in data collected by Vo et al and by our study may explain the discrepancy in results. We also collected data about personal history of other bleeding symptoms in addition to HMB and bleeding after surgery or procedures and found that girls with underlying bleeding disorders reported these more frequently than those without hemostatic abnormality. However, the difference between the 2 groups was not statistically significant. Our findings differ from those reported by Friberg et al15 in which girls with bleeding disorders had statistically significant higher frequencies of almost all other bleeding symptoms, including epistaxis, hematuria, and bleeding after surgery, when compared with those without bleeding disorders. However, that study compared 8 girls with known diagnosis of bleeding disorder with 1016 girls without bleeding disorder, and the authors did not pursue testing to confirm or refute the presence or absence of a bleeding disorder. At the present time, we cannot conclude that there is an association between personal history of other bleeding symptoms and/or bleeding after surgery or procedure and ultimate diagnosis of a bleeding disorder or hemostatic abnormality. However, our findings and the clear discrepancy among published reports deserve further study because there is great clinical value in detecting variables in a patient’s history that may help the physician identify the subset of patients who need to be tested for hemostatic abnormalities. An interesting finding of our study was a large number of girls with risk factor for bleeding, namely, low vWF activity (42 patients; 32%). Diagnosing patients with vWF: Rco and/or vWF: Ag !30 IU/dL with vWD is appropriate as some studies have shown that these patients are likely to have vWF gene mutations, significant bleeding, and strongly positive family history.17,18 However, vWF: Rco and/or vWF: Ag in the range of 30 to 50 IU/dL pose a significant diagnostic challenge as (1) mild bleeding symptoms can be common among healthy people and findings of moderately low vWF may be coincidental, (2) moderately low vWF activity are usually not co-inherited within
328
R. Dıaz et al. / J Pediatr Adolesc Gynecol 27 (2014) 324e329
Fig. 1. Step-wise algorithm for bleeding disorders and risk factors in young girls with HMB.Ă
families and not strongly associated with intragenic vWF gene mutation, and (3) ABO blood group strongly influences vWF activity.16e22 However, recently published guidelines by the European group on vWD classify patients with vWF: Rco and/or vWF: Ag more than 30 IU/dL but less than 40 IU/dL as having type 1 vWD, although acknowledging that these patients usually experience milder bleeding and management in general includes avoidance of anti-platelet drugs and antifibrinolytics as needed.23 At our center, we used more stringent criteria (vWF: Rco and/ or vWF: Ag !30 IU/dL and normal MM) to diagnose patients with vWD, and categorized low vWF activity as a bleeding risk factor and not a bleeding disorder. We found that patients with low vWF activity were more likely to bleed after surgical or dental procedures (OR 2.0, P 5 .642), which supports the observation by others and provides additional evidence for categorizing low vWF activity as a risk factor for bleeding and providing hemostatic management for such patients with bleeding symptoms. Our study limitations include the following: patient population was derived from a specialty clinic, which might have resulted in selection bias. We acknowledge that the true prevalence of bleeding disorders in girls with HMB would be ideally obtained in a primary care setting and our prevalence rate may be an over-estimate. Furthermore, the hemostatic evaluation that the patients underwent was not uniform among providers given the retrospective design of this study. Nevertheless, at a minimum, all patients underwent evaluation with CBC, PT/PTT, fibrinogen and vWP. A total of 78 (60%) of patients had platelet aggregometry testing with secretion analysis. We acknowledge that using a tiered approach may have potentially led to under diagnosing and/or missed patients with combined bleeding disorders. In summary, we conclude that the prevalence of hemostatic abnormalities, including bleeding disorders and risk factors, is substantial among young females with HMB. Based on these findings we suggest systematic and stepwise hemostatic evaluation in young girls presenting with HMB as detailed in Fig. 1. This approach is reasonable because
abnormal levels of vWF: Rco and/or vWF: Ag was the most commonly detected abnormality in our group of patients. Using platelet aggregometry with secretion analysis as a second tier test makes sense in light of recent reports of low sensitivity of this test in patients with inherited platelet disorders, thus making it a less than ideal screening test.14,24 In addition, platelet aggregometry is a labor intensive and costly test that should be preferably performed in specialized centers. Tests for rare bleeding disorders such as fibrinolytic pathway defects and factor XIII deficiencies may be done at the third tier level in patients with persistent and severe HMB after ruling out common bleeding disorders. Although our study showed that certain clinical factors such as personal history of other bleeding symptoms and bleeding after surgical or dental procedure were more frequently reported by girls with HMB and hemostatic defects, we cannot conclude that using these variables will help the physician identify which girls may benefit from hemostatic testing. A formalized bleeding tool is needed to help narrow the subset of young girls with HMB who need hemostatic evaluation. Until such a tool is developed, hemostatic evaluation is required for all young girls with HMB to avoid missing an abnormality because this may hold important implications for management of HMB, surgical prophylactic needs, prevention of complications such as anemia, and other obstetric and gynecologic complications related to excessive menstrual bleeding and the overall quality of life.
References 1. Boswell HB: The adolescent with menorrhagia: why, who and how to evaluate for a bleeding disorder. J Pediatr Adolesc Gynecol 2011; 24:228 2. Dilley A, Drews C, Miller C, et al: von Willebrand disease and other inherited bleeding disorders in women with diagnosed menorrhagia. Obstet Gynecol 2001; 97:630 3. Mikhail S, Varadarajan R, Kouides P: The prevalence of disorders of haemostasis in adolescents with menorrhagia referred to a haemophilia treatment centre. Haemophilia 2007; 13:627 4. Jayasinghe Y, Moore P, Donath S, et al: Bleeding disorders in teenagers presenting with menorrhagia. Aust N Z J Obstet Gynaecol 2005; 45:439 5. Philipp CS, Faiz A, Dowling N, et al: Age and the prevalence of bleeding disorders in women with menorrhagia. Obstet Gynecol 2005; 105:61
R. Dıaz et al. / J Pediatr Adolesc Gynecol 27 (2014) 324e329 6. Vo KT, Grooms L, Klima J, et al: Menstrual bleeding patterns and prevalence of bleeding disorders in a multidisciplinary adolescent haematology clinic. Haemophilia 2013; 19:71 7. Rodriguez V, Alme C, Killian JM, et al: Bleeding disorders in adolescents with menorrhagia: an institutional experience. Haemophilia 2013; 19:e101 8. Wang W, Bourgeois T, Klima J, et al: Iron deficiency and fatigue in adolescent females with heavy menstrual bleeding. Haemophilia 2012; 19:225 9. James AH: More than menorrhagia: a review of the obstetric and gynaecological manifestations of bleeding disorders. Haemophilia 2005; 11:295 10. Chi C, Shiltagh N, Kingman CE, et al: Identification and management of women with inherited bleeding disorders: a survey of obstetricians and gynaecologists in the United Kingdom. Haemophilia 2006; 12:405 11. Kulp JL, Mwangi CN, Loveless M: Screening for coagulation disorders in adolescents with abnormal uterine bleeding. J Pediatr Adolesc Gynecol 2008; 21:27 12. Quiroga T, Mezzano D: Is my patient a bleeder? A diagnostic framework for mild bleeding disorders. Hematology Am Soc Hematol Educ Program 2012; 2012:466 13. U.S. Department of Health and Human Services, National Institutes of Health, National Heart, Lund, and Blood Institute: The Diagnosis, Evaluation, and Management of von Willebrand Disease (NIH publication No. 08e5832, December 2007). Available: http://www.nhlbi.nih.gov/guidelines/vwd/vwd.pdf. Accessed October 15, 2013 14. Hayward CP, Moffat KA, Raby A, et al: Development of North American consensus guidelines for medical laboratories that perform and interpret platelet function testing using light transmission aggregometry. Am J Clin Pathol 2010; 134:955
329
€ AK, Lindgren A, et al: Bleeding disorders among young women: 15. Friberg B, Orno a population-based prevalence study. Acta Obstet Gynecol Scand 2006; 85:200 16. Sadler JE: Von Willebrand disease type 1: a diagnosis in search of a disease. Blood 2003; 101:2089 17. James PD, Notley C, Hegadorn C, et al: The mutational spectrum of type 1 von Willebrand disease: Results from a Canadian cohort study. Blood 2007; 109:145 18. James PD, Paterson AD, Notley C, et al: Association of Hemophilia Clinic Directors of Canada: Genetic linkage and association analysis in type 1 von Willebrand disease: results from the Canadian type 1 VWD study. J Thromb Haemost 2006; 4:783 19. Mannucci PM, Lombardi R, Bader R, et al: Heterogeneity of type I von Willebrand disease: evidence for a subgroup with an abnormal von Willebrand factor. Blood 1985; 66:796 20. Gill JC, Endres-Brooks J, Bauer PJ, et al: The effect of ABO blood group on the diagnosis of von Willebrand disease. Blood 1987; 69:1691 ~ a P, Martınez F, Haya S, et al: Significant linkage and non-linkage of type 1 21. Casan von Willebrand disease to the von Willebrand factor gene. Br J Haematol 2001; 115:692 22. Castaman G, Eikenboom JC, Bertina RM, et al: Inconsistency of association between type 1 von Willebrand disease phenotype and genotype in families identified in an epidemiological investigation. Thromb Haemost 1999; 82:1065 23. Castaman G, Goodeve A, Eikenboom J: European Group on von Willebrand Disease: Principles of care for the diagnosis and treatment of von Willebrand disease. Haematologica 2013; 98:667 24. Sokkary NA, Venkateswaran L, Dietrich JE, et al: Platelet function disorders and menorrhagia in adolescents: a review of laboratory diagnosis. J Pediatr Adolesc Gynecol 2012; 25:233
Texas Children’s Hospital, Houston, TX Baylor College of Medicine, Houston, TX
a b s t r a c t Objective: To study the prevalence of hemostatic abnormalities, including bleeding disorders and risk factors, in young females referred to a multidisciplinary clinic for evaluation of heavy menstrual bleeding (HMB). Methods: Retrospective chart review was undertaken for 131 post-menarchal girls with HMB, 7 to 17 years of age, enrolled in the institutional ‘Menorrhagia Data Registry’ protocol. The diagnostic approach included: (1) complete blood count, prothrombin time, partial thromboplastin time, fibrinogen, von Willebrand panel (2) platelet aggregometry, specific clotting factor assay, fibrinolytic pathway analysis, and factor XIII level as needed. The prevalence of hemostatic abnormalities and the prognostic significance of clinical variables associated with hemostatic abnormalities in young girls with HMB were evaluated. Results: A hemostatic abnormality was identified in 69 (53%) young girls with HMB. Of these, 27 (21%) had an underlying bleeding disorder and 42 (32%) had a risk factor for bleeding, namely low von Willebrand factor activity. A larger number of girls with underlying bleeding disorder had personal history of other bleeding symptoms (48% vs 31%) and bleeding after surgical or dental procedure (25% vs 8%) when compared to females without hemostatic abnormality. Furthermore, girls with risk factor for bleeding (low vWF activity) were more likely to have bleeding after surgical or dental procedure (15% vs 8%) and family history of bleeding (79% vs 60%) than patients without hemostatic abnormality. Conclusions: There is high prevalence of hemostatic abnormalities, including bleeding disorders and risk factors, in young girls with HMB. These findings support comprehensive and systematic hemostatic evaluation in this group of patients. Key Words: Bleeding disorders, Young females, Heavy menstrual bleeding
Introduction
Heavy menstrual bleeding (HMB) is a common gynecologic complaint among young females.1 In many, anovulation and immaturity of the hypothalamic-pituitaryovarian axis are thought to be the underlying cause for heavy periods, which can often be managed with hormonal therapy. However, several studies have shown that 10%-62% of young girls with HMB have an underlying bleeding disorder that may benefit from the addition of non-hormonal therapies or may be relevant to future surgeries or pregnancies.2e7 Bleeding disorders associated with HMB include disorders of primary and secondary hemostasis such as clotting factors deficiencies, von Willebrand disease (vWD), platelet disorders, and rare fibrinolytic pathway defects. It is important to identify female adolescents with bleeding abnormalities because the obstetric and gynecologic morbidity related to bleeding disorders go beyond troublesome heavy periods. Iron deficiency anemia, fatigue, less
Dr. Dietrich is a consultant for CSL Behring and Bayer and conducts research for Duramed. Dr. Diaz, Dr. Mahoney, Dr. Yee, and Dr. Srivaths indicate no conflicts of interest. * Address correspondence to: Rosa Dıaz, MD, Department of Pediatrics, Section of Hematology/Oncology, Texas Children’s Hospital/Baylor College of Medicine, 1102 Bates Ave, Suite 1025, Houston, TX 77030; Phone: (832) 824-4836; fax: (832) 8251453 E-mail address: [email protected] (R. Dıaz).
time spent on desired activities, and difficulty performing schoolwork are common complaints among these young women.8 In addition, 8-18% of women with bleeding disorders have surgery for HMB and other complications including surgery for hemorrhagic ovarian cyst and hysterectomy which may be associated with significant bleeding.9 Despite the relatively high frequencies of bleeding disorders and risk factors in adolescents with HMB, such underlying hemostatic abnormalities are often not considered and a systematic evaluation not performed.10e12 This is in part due to the lack of consensus regarding which patients need to be tested, difficulty in discerning ‘normal’ from pathologic bleeding and the lack of an optimal diagnostic approach. The aim of the present study is to evaluate the prevalence of hemostatic abnormalities in young females with HMB by utilizing a step-wise, systematic diagnostic approach. We hypothesized that hemostatic abnormalities, including bleeding disorders and risk factors for bleeding, are prevalent in a substantial number of girls with HMB and that step-wise comprehensive laboratory testing is warranted in these patients. Patients and Methods Data Source and Study Population
Data collected from post-menarchal young females enrolled in the institutional Menorrhagia Data Registry
1083-3188/$ - see front matter Ó 2014 North American Society for Pediatric and Adolescent Gynecology. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jpag.2013.12.011
R. Dıaz et al. / J Pediatr Adolesc Gynecol 27 (2014) 324e329
approved by the institutional review board, referred to the multi-specialty Young Women’s Bleeding Disorder Clinic served by hematology and gynecology faculty and Hematology and Gynecology Clinics at Texas Children’s Hospital (TCH) for evaluation of HMB from 2009 to 2011, was analyzed retrospectively for this study. HMB is defined as more than 7 days of bleeding, greater than 80 ml of blood loss per cycle, pictorial blood assessment chart (PBAC) score of more than 100, or changing pads or tampons every 1-2 hours. All patients enrolled in this registry were referred by primary care physicians and/or gynecologists. Eligibility criteria included: (1) female gender, (2) post-menarche, (3) referred to TCH for HMB, and (4) non-pregnant. Data collected included demographics, clinical history and physical exam findings, laboratory study and imaging results, therapy details, and patient outcomes. Patients underwent concurrent evaluation by gynecology for pregnancy, anatomic pelvic abnormalities, pelvic infections, and endocrine abnormalities as possible causes for HMB, and only included in our study if these tests were normal or not thought to account for heavy menses. Definitions and Diagnostic Studies
All patients underwent complete blood count (CBC) to detect thrombocytopenia, coagulation studies for prothrombin time (PT), partial thromboplastin time (PTT), and fibrinogen to detect coagulopathy, and von Willebrand panel (vWP). Specific coagulation factor assays were done if PT, PTT, and/or fibrinogen were abnormal, to delineate specific clotting factor deficiency. Whole blood platelet aggregometry with secretion analysis was done in patients if the first level of testing was negative. In addition, fibrinolytic pathway testing, and factor XIII levels were performed to identify less common disorders for persistent, severe HMB, when deemed necessary by the treating physician. Studies for platelet aggregation and vWP were performed by the TCH Pathology Services. The vWP performed included individual tests for factor VIII (clot based assay), von Willebrand antigen (vWF: Ag), vW ristocetin co-factor activity (vWF: RCo) and multimers (MM). vWD was diagnosed based on the criteria described in the National Heart, Lung, and Blood Institute’s guideline “The diagnosis, evaluation, and management of von Willebrand disease 2008.13” Briefly, vWD is classified into 3 major categories: partial quantitative deficiency (type 1), qualitative deficiency (type 2) and total deficiency (type 3). Patients were diagnosed with type 1 vWD if they had vWF: Ag and/or vWF: RCo !30 IU/dL and normal MM. Those with vWF: RCo and/or vWF: Ag 30-50 IU/dL and normal MM were categorized as having low vWF activity, considered a risk factor for bleeding. Whole blood platelet aggregometry was performed using the Chronolog Whole-Blood Lumi-Aggregometer by measuring change in electrical impedance in whole-blood. The stimulus was brought about by using plateletaggregating agonists, including adenosine diphosphate (ADP), arachidonic acid, collagen, ristocetin, and thrombin. Adenosine triphosphate (ATP) release was measured by luminescence in whole-blood. The ATP released by the platelet dense granules binds with CHRONO-LUMEÔ
325
(luciferin-luciferase, Chrono-Log Corp, Havertown, PA) and generates a light (luminescence) that is measured by a stable, high-gain photomultiplier tube. The diagnosis of platelet function defect (PFD) required detection of at least 2 abnormalities in platelet aggregation and/or secretion, as described by Hayward et al.14 Testing for factor XIII deficiency was done by estimation of factor XIII activity using a chromogenic assay (normal range 57%-192%; Quest Diagnostics Nichols Institute, San Juan Capistrano, CA). Fibrinolytic pathway testing was done to evaluate for congenital deficiency of plasminogen activator inhibitor type 1 (PAI-1) and alpha-2 antiplasmin. Fibrinolysis panel including testing for PAI-1 activity (1.7-15 U/mL) was performed by the University of Washington Medical Center, Seattle, WA. Alpha-2 antiplasmin testing was performed by the Blood Center of SE Wisconsin using a chromogenic activity assay (normal range 75%-126%). Bleeding History
Data including demographic information, family history of bleeding, clinical symptoms and signs of bleeding, and laboratory test results on all patients were collected. Characteristics of menses including duration of menstrual period, number of pads and/or tampons used per day, presence and size of clots, flooding, and the PBAC score were collected. Information about other bleeding symptoms including epistaxis, gum bleeding, easy bruising, and excessive bleeding with dental procedures or other surgeries, hospitalization for management of severe menstrual bleeding and anemia and the need for red blood cell transfusions were gathered. Study Outcomes
The primary outcome for the study was prevalence of hemostatic abnormalities, including bleeding disorders and risk factors for bleeding, in young females with HMB. The secondary outcome included the prognostic significance of clinical variables associated with HMB, namely characteristics of menstrual bleeding, other bleeding symptoms, bleeding after surgical or dental procedure, family history of bleeding or bleeding disorder, personal history of anemia, red cell transfusion or hospitalization for HMB, which may help identify girls with HMB and hemostatic abnormalities. Statistical Analysis
The statistical analysis was performed using GraphPad Prism version 6.00 for Windows, GraphPad Software, San Diego, California, graphpad.com. Odds ratio and P values were calculated using the Fisher exact test from a contingency table. A P-value !.05 was considered statistically significant. Results Patient Characteristics
A total of 131 young females met eligibility criteria. The patients had a median age of 12 years (IQR 11-13) at the
326
R. Dıaz et al. / J Pediatr Adolesc Gynecol 27 (2014) 324e329
time of diagnosis with HMB. Thirty-seven percent (n 5 48) were white, 35% (n 5 46) were Hispanic, 17% (n 5 22) were Black, and 2% (n 5 3) were Asian. Two girls had other ethnicity including American Indian and Alaskan native and Hawaiian and other Pacific islander. Ethnicity was not specified for 10 patients. Primary Outcome: Prevalence of Hemostatic Abnormalities in Girls with HMB
We identified a hemostatic abnormality in 69 (53%) of the girls referred to our clinic due to HMB. Of these, 27 (21%) had an underlying bleeding disorder and 42 (32%) had a risk factor for bleeding, namely low vWF activity. PFD was the most commonly detected bleeding disorder in the patients (n 5 14, 11%). This was followed in frequency by vWD, clotting factor deficiency, thrombocytopenia, and PAI-1 deficiency, as shown in Table 1. Prognostic Significance of Clinical Variables: Bleeding Disorder vs No Hemostatic Abnormality
The median age of menarche and onset of HMB were similar for girls with bleeding disorders (11 years, IQR 10-12 and 12 years, IQR 11-14, respectively) and those without any hemostatic abnormality (12 years, IQR 11-13 and 12 years, IQR 11-13, respectively). In addition, menstrual bleeding characteristics such as duration of periods, number of pads/ tampons per day, passing clots during menses, flooding, and spotting were similar for both groups. We interrogated the prognostic utility of non-menstrual bleeding history for distinguishing girls with bleeding disorders from those without hemostatic abnormalities. Girls with underlying bleeding disorder were more likely to have personal history of other bleeding symptoms (48% vs 31%), and bleeding after surgical or dental procedure (25% vs 8%) when compared to those without hemostatic abnormality, as shown in Table 2. However, these results were not statistically significant. Within the entire group of patients, over half of the patients (n 5 66, 52%) were diagnosed with anemia, 18% (n 5 22) required red blood cell transfusion and 24% (n 5 31) needed hospitalization for management of bleeding symptoms and/or anemia. Prevalence of anemia, number of hospitalizations or transfusion history did not Table 1 Prevalence of Hemostatic Abnormalities among Young Girls with HMB Hemostatic Abnormalities Bleeding disorder PFD VWD Clotting factor deficiency Thrombocytopenia PAI-1 deficiency Risk factor for bleeding Low VWF activity No hemostatic abnormality
Table 2 Characteristics of Menses and Other Clinical Variables in Adolescents with Bleeding Disorder and without Hemostatic Abnormality Bleeding Disorder, N5 27 n (%) Clots Flooding Spotting Easy bruising Other bleeding symptoms Bleeding after procedure Family history Anemia PRBC transfusion Hospitalization
19 12 5 8 13
(70) (44) (19) (30) (48)
4/16 (25) 19 15 3 7
(70) (56) (11) (27)
No Hemostatic Abnormality, N5 62 n (%) 38 32 10 17 19
(61) (52) (16) (27) (31)
2/25 (8) 37 35 12 14
(60) (56) (19) (23)
OR (95% CI)
1.5 0.7 1.2 1.1 2.1
(0.5-3.9) (0.3-1.8) (0.3-3.8) (0.4-3.0) (0.8-5.3)
3.8 (0.6-24.0) 1.6 0.9 0.5 1.2
(0.6-4.2) (0.3-2.4) (0.1-2.0) (0.4-3.4)
P Value .477 .646 .766 1.000 .150 .186 .474 1.000 .538 .788
CI, Confidence interval; n, Number of girls with HMB; OR, Odds ratio; PRBC, Pure red blood cell
differ between girls with bleeding disorders and those without hemostatic abnormalities. Low vWF Activity: A Risk Factor for Bleeding?
We identified low vWF activity in 42 (32%) of the girls in the study (vWF: Ag and/or vWF: RCo median 37 IU/dL, range 31 to 49 IU/dL). First, we compared the personal and family history of bleeding of girls with bleeding disorder to patients with low vWF activity. Young females with bleeding disorder were more likely to pass clots (70% vs 52%) during their menses and have spotting (19% vs 5%) when compared to those with low vWF activity (Table 3). In addition, girls with bleeding disorders were more likely to have anemia (56% vs 38%) and bleeding after surgical or dental procedure (25% vs 15%). However, these results were not statistically significant. Easy bruising, other bleeding symptoms, family history of bleeding, number of hospitalizations and blood transfusion requirements were similar for both groups. Next, we compared girls with low vWF activity to those without hemostatic abnormalities. The patients with risk factor for bleeding were more likely to have bleeding after surgery or dental procedure (15% vs 8%), and more frequently had family history of bleeding (79% vs 60%, Table 4). These results were not statistically significant. All other variables such as passing clots, flooding, spotting, easy bruising, personal history of other bleeding symptoms, anemia, hospitalizations, and need for pure red blood cell (PRBC) transfusion, were similar for both groups, as summarized in Table 4.
Number of Girls (%); N 5 131 14 9 2 1 1
(11)* (7) (1)y (1) (1)z
42 (32) 62 (47)
N, Total number of girls with HMB * Four patients had combined bleeding disorders: PFD plus thrombocytopenia, low VWF activity or clotting factor deficiency. y One patient had combined hypofibrinogenemia and low VWF activity. z One patient had PAI-1 deficiency combined with low VWF activity.
Discussion
Our retrospective review showed that the frequency of hemostatic abnormalities is substantial among young girls referred for evaluation of HMB. An underlying bleeding disorder or risk factor was detected in 21% and 32% of girls evaluated, respectively, with PFD being the most commonly diagnosed bleeding disorder. The overall prevalence in our group of patients is consistent with previous reports. Mikhail et al3 by retrospective review identified a bleeding disorder in 41% of the 61 patients studied which included
R. Dıaz et al. / J Pediatr Adolesc Gynecol 27 (2014) 324e329 Table 3 Characteristics of Menses and Other Clinical Variables in Adolescents with Bleeding Disorder and Risk Factor for Bleeding Bleeding Disorder N 5 27 n (%) Clots Flooding Spotting Easy bruising Other bleeding symptoms Bleeding after procedure Family history Anemia PRBC transfusion Hospitalization
19 12 5 8 13
(70) (44) (19) (30) (48)
4/16 (25) 19 15 3 7
Risk Factor N 5 42 n (%) 22 20 2 15 16
(52) (48) (5) (36) (38)
3/20 (15)
(70) (56) (11) (27)
33 16 7 10
(79) (38) (17) (24)
OR (95% CI) 2.2 0.9 4.5 0.8 1.5
P Value
(0.7-6.0) (0.3-2.3) (0.8-25.4) (0.3-2.1) (0.6-4.0)
.209 .810 .108 .794 .460
1.9 (0.3-10.0)
.675
0.6 2.0 0.6 1.1
(0.2-1.9) (0.7-5.4) (0.1-2.6) (0.3-3.4)
.568 .215 .729 1.00
CI, Confidence interval; N, Total number of girls with HMB; OR, Odds ratio; PRBC, Pure red blood cell
vWD (36%) and PFD (7%). They defined vWD as vWF: RCo ! 40% and/or vWF: Ag !50%, and PFD as decreased aggregation to 1 or more agonists. A recent study by Vo et al6 of 105 patients reported a prevalence of 62% for bleeding disorders (PFD in 44% and vWD in 9% of the girls) in adolescents with HMB, wherein vWD was defined as vWF: RCo and/or vWF: Ag !40% and PFD was detected primarily by platelet electron microscopy with only some patients undergoing platelet aggregometry. A study by Philipp et al5 of 25 adolescents reported a hemostatic abnormality in 48% of the patients (PFD in 44% and vWD in 7% of the females) but did not clearly provide the definitions for bleeding disorders. Lastly, a study of 160 girls evaluated for menorrhagia reported only 16% with bleeding disorders, but did not utilize a uniform hemostatic evaluation and did not clearly define the diagnostic criteria.7 In comparison with these studies, we employed more stringent criteria to diagnose vWD and PFD based on recent publications,13,14 which likely accounts for the large number of patients identified with low vWF activity (32%), and lower prevalence of PFD (11%) and vWD (7%) in our study when compared to previous reports. Utilizing clinical variables such as characteristics of menses, personal history of other bleeding symptoms and family history of bleeding, to differentiate between ‘normal’ Table 4 Characteristics of Menses and other Clinical Variables in Adolescents with Risk Factor for Bleeding and Non-Hemostatic Abnormality Risk Factor N 5 42 n (%) Clots Flooding Spotting Easy bruising Other bleeding symptoms Bleeding after procedure Family history Anemia PRBC transfusion Hospitalization
22 20 2 15 16
(52) (48) (5) (36) (38)
3/20 (15) 33 16 7 10
(79) (38) (17) (24)
No Hemostatic Abnormality N 5 62 n (%) 38 32 10 17 19
(61) (52) (16) (27) (31)
2/25 (8) 37 35 12 14
(60) (56) (19) (23)
Odds Ratio (95%CI) 0.7 0.8 0.3 1.5 1.4
(0.3-1.5) (0.4-1.8) (0.1-1.2) (0.6-3.4) (0.6-3.2)
2.0 (0.3-13.5) 2.5 0.5 0.8 1.1
(1.0-6.1) (0.2-1.1) (0.3-2.3) (0.4-2.7)
P Value .421 .841 .116 .393 .526 .642 .055 .075 .800 1.000
CI, Confidence interval; n, Number of girls with HMB; OR, Odds ratio; PRBC, Pure red blood cell
327
and pathologic bleeding in young girls with chief complaint of heavy menses may provide an important tool to delineate the population that needs hemostatic evaluation. The secondary outcome measured by our study included a comparison of the bleeding characteristics, family history of bleeding and history of anemia, transfusion, and hospitalization between girls with and without underlying hemostatic abnormality. We did not find differences in menstrual characteristics, including length of period, passing clots, flooding, and spotting, in girls with and without bleeding disorder or hemostatic abnormality. This result is in contrast to a study by Vo et al6 in which the authors found statistically significant differences in reported regularity of patient’s periods, description of period flow, and number of days of each period that the bleeding was described as ‘heavy’ when they compared the bleeding profiles of females with and without bleeding disorders. Vo et al6 used a modified Ruta Menorrhagia Severity Scale to assess the menstrual bleeding profiles. The scale assessed the adolescent’s perception of whether periods are regular or irregular, heavy or light, but did not precisely measure length of period or severity. The difference in data collected by Vo et al and by our study may explain the discrepancy in results. We also collected data about personal history of other bleeding symptoms in addition to HMB and bleeding after surgery or procedures and found that girls with underlying bleeding disorders reported these more frequently than those without hemostatic abnormality. However, the difference between the 2 groups was not statistically significant. Our findings differ from those reported by Friberg et al15 in which girls with bleeding disorders had statistically significant higher frequencies of almost all other bleeding symptoms, including epistaxis, hematuria, and bleeding after surgery, when compared with those without bleeding disorders. However, that study compared 8 girls with known diagnosis of bleeding disorder with 1016 girls without bleeding disorder, and the authors did not pursue testing to confirm or refute the presence or absence of a bleeding disorder. At the present time, we cannot conclude that there is an association between personal history of other bleeding symptoms and/or bleeding after surgery or procedure and ultimate diagnosis of a bleeding disorder or hemostatic abnormality. However, our findings and the clear discrepancy among published reports deserve further study because there is great clinical value in detecting variables in a patient’s history that may help the physician identify the subset of patients who need to be tested for hemostatic abnormalities. An interesting finding of our study was a large number of girls with risk factor for bleeding, namely, low vWF activity (42 patients; 32%). Diagnosing patients with vWF: Rco and/or vWF: Ag !30 IU/dL with vWD is appropriate as some studies have shown that these patients are likely to have vWF gene mutations, significant bleeding, and strongly positive family history.17,18 However, vWF: Rco and/or vWF: Ag in the range of 30 to 50 IU/dL pose a significant diagnostic challenge as (1) mild bleeding symptoms can be common among healthy people and findings of moderately low vWF may be coincidental, (2) moderately low vWF activity are usually not co-inherited within
328
R. Dıaz et al. / J Pediatr Adolesc Gynecol 27 (2014) 324e329
Fig. 1. Step-wise algorithm for bleeding disorders and risk factors in young girls with HMB.Ă
families and not strongly associated with intragenic vWF gene mutation, and (3) ABO blood group strongly influences vWF activity.16e22 However, recently published guidelines by the European group on vWD classify patients with vWF: Rco and/or vWF: Ag more than 30 IU/dL but less than 40 IU/dL as having type 1 vWD, although acknowledging that these patients usually experience milder bleeding and management in general includes avoidance of anti-platelet drugs and antifibrinolytics as needed.23 At our center, we used more stringent criteria (vWF: Rco and/ or vWF: Ag !30 IU/dL and normal MM) to diagnose patients with vWD, and categorized low vWF activity as a bleeding risk factor and not a bleeding disorder. We found that patients with low vWF activity were more likely to bleed after surgical or dental procedures (OR 2.0, P 5 .642), which supports the observation by others and provides additional evidence for categorizing low vWF activity as a risk factor for bleeding and providing hemostatic management for such patients with bleeding symptoms. Our study limitations include the following: patient population was derived from a specialty clinic, which might have resulted in selection bias. We acknowledge that the true prevalence of bleeding disorders in girls with HMB would be ideally obtained in a primary care setting and our prevalence rate may be an over-estimate. Furthermore, the hemostatic evaluation that the patients underwent was not uniform among providers given the retrospective design of this study. Nevertheless, at a minimum, all patients underwent evaluation with CBC, PT/PTT, fibrinogen and vWP. A total of 78 (60%) of patients had platelet aggregometry testing with secretion analysis. We acknowledge that using a tiered approach may have potentially led to under diagnosing and/or missed patients with combined bleeding disorders. In summary, we conclude that the prevalence of hemostatic abnormalities, including bleeding disorders and risk factors, is substantial among young females with HMB. Based on these findings we suggest systematic and stepwise hemostatic evaluation in young girls presenting with HMB as detailed in Fig. 1. This approach is reasonable because
abnormal levels of vWF: Rco and/or vWF: Ag was the most commonly detected abnormality in our group of patients. Using platelet aggregometry with secretion analysis as a second tier test makes sense in light of recent reports of low sensitivity of this test in patients with inherited platelet disorders, thus making it a less than ideal screening test.14,24 In addition, platelet aggregometry is a labor intensive and costly test that should be preferably performed in specialized centers. Tests for rare bleeding disorders such as fibrinolytic pathway defects and factor XIII deficiencies may be done at the third tier level in patients with persistent and severe HMB after ruling out common bleeding disorders. Although our study showed that certain clinical factors such as personal history of other bleeding symptoms and bleeding after surgical or dental procedure were more frequently reported by girls with HMB and hemostatic defects, we cannot conclude that using these variables will help the physician identify which girls may benefit from hemostatic testing. A formalized bleeding tool is needed to help narrow the subset of young girls with HMB who need hemostatic evaluation. Until such a tool is developed, hemostatic evaluation is required for all young girls with HMB to avoid missing an abnormality because this may hold important implications for management of HMB, surgical prophylactic needs, prevention of complications such as anemia, and other obstetric and gynecologic complications related to excessive menstrual bleeding and the overall quality of life.
References 1. Boswell HB: The adolescent with menorrhagia: why, who and how to evaluate for a bleeding disorder. J Pediatr Adolesc Gynecol 2011; 24:228 2. Dilley A, Drews C, Miller C, et al: von Willebrand disease and other inherited bleeding disorders in women with diagnosed menorrhagia. Obstet Gynecol 2001; 97:630 3. Mikhail S, Varadarajan R, Kouides P: The prevalence of disorders of haemostasis in adolescents with menorrhagia referred to a haemophilia treatment centre. Haemophilia 2007; 13:627 4. Jayasinghe Y, Moore P, Donath S, et al: Bleeding disorders in teenagers presenting with menorrhagia. Aust N Z J Obstet Gynaecol 2005; 45:439 5. Philipp CS, Faiz A, Dowling N, et al: Age and the prevalence of bleeding disorders in women with menorrhagia. Obstet Gynecol 2005; 105:61
R. Dıaz et al. / J Pediatr Adolesc Gynecol 27 (2014) 324e329 6. Vo KT, Grooms L, Klima J, et al: Menstrual bleeding patterns and prevalence of bleeding disorders in a multidisciplinary adolescent haematology clinic. Haemophilia 2013; 19:71 7. Rodriguez V, Alme C, Killian JM, et al: Bleeding disorders in adolescents with menorrhagia: an institutional experience. Haemophilia 2013; 19:e101 8. Wang W, Bourgeois T, Klima J, et al: Iron deficiency and fatigue in adolescent females with heavy menstrual bleeding. Haemophilia 2012; 19:225 9. James AH: More than menorrhagia: a review of the obstetric and gynaecological manifestations of bleeding disorders. Haemophilia 2005; 11:295 10. Chi C, Shiltagh N, Kingman CE, et al: Identification and management of women with inherited bleeding disorders: a survey of obstetricians and gynaecologists in the United Kingdom. Haemophilia 2006; 12:405 11. Kulp JL, Mwangi CN, Loveless M: Screening for coagulation disorders in adolescents with abnormal uterine bleeding. J Pediatr Adolesc Gynecol 2008; 21:27 12. Quiroga T, Mezzano D: Is my patient a bleeder? A diagnostic framework for mild bleeding disorders. Hematology Am Soc Hematol Educ Program 2012; 2012:466 13. U.S. Department of Health and Human Services, National Institutes of Health, National Heart, Lund, and Blood Institute: The Diagnosis, Evaluation, and Management of von Willebrand Disease (NIH publication No. 08e5832, December 2007). Available: http://www.nhlbi.nih.gov/guidelines/vwd/vwd.pdf. Accessed October 15, 2013 14. Hayward CP, Moffat KA, Raby A, et al: Development of North American consensus guidelines for medical laboratories that perform and interpret platelet function testing using light transmission aggregometry. Am J Clin Pathol 2010; 134:955
329
€ AK, Lindgren A, et al: Bleeding disorders among young women: 15. Friberg B, Orno a population-based prevalence study. Acta Obstet Gynecol Scand 2006; 85:200 16. Sadler JE: Von Willebrand disease type 1: a diagnosis in search of a disease. Blood 2003; 101:2089 17. James PD, Notley C, Hegadorn C, et al: The mutational spectrum of type 1 von Willebrand disease: Results from a Canadian cohort study. Blood 2007; 109:145 18. James PD, Paterson AD, Notley C, et al: Association of Hemophilia Clinic Directors of Canada: Genetic linkage and association analysis in type 1 von Willebrand disease: results from the Canadian type 1 VWD study. J Thromb Haemost 2006; 4:783 19. Mannucci PM, Lombardi R, Bader R, et al: Heterogeneity of type I von Willebrand disease: evidence for a subgroup with an abnormal von Willebrand factor. Blood 1985; 66:796 20. Gill JC, Endres-Brooks J, Bauer PJ, et al: The effect of ABO blood group on the diagnosis of von Willebrand disease. Blood 1987; 69:1691 ~ a P, Martınez F, Haya S, et al: Significant linkage and non-linkage of type 1 21. Casan von Willebrand disease to the von Willebrand factor gene. Br J Haematol 2001; 115:692 22. Castaman G, Eikenboom JC, Bertina RM, et al: Inconsistency of association between type 1 von Willebrand disease phenotype and genotype in families identified in an epidemiological investigation. Thromb Haemost 1999; 82:1065 23. Castaman G, Goodeve A, Eikenboom J: European Group on von Willebrand Disease: Principles of care for the diagnosis and treatment of von Willebrand disease. Haematologica 2013; 98:667 24. Sokkary NA, Venkateswaran L, Dietrich JE, et al: Platelet function disorders and menorrhagia in adolescents: a review of laboratory diagnosis. J Pediatr Adolesc Gynecol 2012; 25:233
