Abstract
Heavy menstrual bleeding (HMB) is often the initial manifestation of a bleeding disorder in adolescent females. Von Willebrand disease (VWD) is the most common bleeding disorder, yet diagnosing VWD in HMB is challenging given that menstruation, acute illness, and physiologic stress all elevate von Willebrand factor (VWF) laboratory values. Thus, the interpretation of initial VWF results may be challenging, and the need for repeat VWF testing in this population may be unclear.
The aims of this study were to understand VWD diagnostic testing patterns in adolescent females evaluated within the Texas Children's Hospital (TCH) network and to develop a prediction model to determine whether repeat testing was indicated based on initial VWF testing results.
Female patients aged 9 to 21 years who had any VWF laboratory test over a 5-year period (January 1, 2017 through December 31, 2021) at a TCH facility (greater Houston metroplex) were identified. Data collection included patient demographics, VWF-related testing (antigen and activity [ristocetin cofactor or glycoprotein Ib]), factor VIII activity, hemoglobin concentration and serum ferritin. Both initial and subsequent laboratory evaluations were collected. The site of clinical testing was also identified, including both acute care settings (emergency department, inpatient, and critical care), and outpatient settings (adolescent medicine, gynecology, hematology, otolaryngology, other outpatient TCH primary care clinics). To ensure that subsequent laboratory testing was not mistaken as the initial testing within the cohort, patients who underwent VWF laboratory testing from 2010 to 2016 were excluded. VWF levels obtained after the administration of VWF replacement products were also removed from the data set.
Based on the ASH ISTH NHF WFH 2021 guidelines on the diagnosis of VWD, and with the presumption that adolescents with HMB met the criteria for abnormal bleeding, we analyzed the need to repeat VWF testing to identify any patient with a low VWF antigen level consistent with Type 1 VWD, defined as <50%. A Bayesian linear regression model was developed to predict follow-up VWF antigen levels using the initial levels. Prediction intervals were analyzed to identify thresholds for patients who may not require retesting. This study was approved by the Baylor College of Medicine Institutional Review Board.
A total of 6,125 adolescent females underwent VWD testing at a TCH facility during the study period; 1,204 (19.7%) had repeat VWF antigen values performed and were included in the analysis (TABLE). Of those, 1,037 subjects (86.1%) had normal VWF values (≥50%) on both initial and repeat testing. Seventy-five subjects (6.2%) had low VWF antigen values on both initial and repeat testing. Ninety-two subjects had discrepant initial and repeat testing. Forty-seven (3.9%) had low initial VWF antigen values that normalized on subsequent testing. Forty-five (3.7%) had normal initial VWF values that subsequently fell below 50%.
Based on our prediction model, in subjects with initial VWF antigen values of ≥80% there was ≥97% probability of having normal repeat VWF values. Values of ≥100% carry a 99.4% probability of having normal repeat VWF testing (FIGURE). No association was identified between age and VWF antigen level. Subjects assessed in outpatient adolescent medicine or gynecology clinics were more likely to have low VWF values (on any testing) whereas subjects with normal values were more frequently assessed in the acute care settings (P<0.001). The median serum ferritin and hemoglobin values in the cohort were 13 ng/ml and 12.8 g/dl, respectively, without a significant difference between those with normal versus low VWF antigen values.
Repeat testing in adolescent females whose initial VWF antigen testing values are 80% and higher is unlikely to identify more patients with VWF values below 50%. In this cohort, subjects undergoing hemostatic evaluation, presumably for HMB, by adolescent medicine or gynecology providers were more likely to identify abnormal VWD testing compared to the acute care setting. Iron deficiency screening should be considered for all adolescent females with HMB and not just those girls with a confirmed bleeding disorder. Additional studies are needed to determine the effects of iron deficiency and anemia on VWD laboratory testing in adolescent females with HMB and to validate the accuracy of the prediction model.
Disclosures
Powers:Pharmacosmos LLC: Consultancy; Women's Health Research Collaborative (WHrC)/ The Kinetix Group: Membership on an entity's Board of Directors or advisory committees.
