Cross Sectional Study to Evaluate the Utility of Screening Tools to Identify Low Von Willebrand Factor Levels In Women with Menorrhagia (ZPMCB-VWD).

Abstract 3839

Menorrhagia is a common symptom of haemostatic disorders and often triggers an evaluation for a bleeding disorder. Clinical screening tools, to identify women with menorrhagia who have an underlying bleeding disorder, are very desirable. Scoring systems to quantify menstrual blood loss in women, the Pictorial Blood Loss Assessment Chart, (PBAC), and to assess the severity of bleeding symptoms in individuals diagnosed with von Willebrand disease (VWD), European Union Bleeding Score (EU-BS) (MCMDM-1VWD), were evaluated for their ability to predict the presence of low von Willebrand factor (VWF) level in a cohort of 150 women recruited from an outpatient gynecology clinic. One of us (SG or ML) completed a PBAC and an EU Bleeding questionnaire with the participant. Menorrhagia cases were defined by a PBAC ≥185 and controls by a PBAC<185 (n=75 in each group). Complete blood count (CBC), ABO blood group, VWF activity (VWF:RCo) and VWF antigen (VWF:Ag) were measured in all subjects. Due to the variability in the laboratory testing of VWF: RCo and to capture all women with low levels, low VWF level was defined as VWF:RCo or VWF:Ag’60 IU/dl in our study. PBAC score correlated with the presence of menorrhagia defined by the menorrhagia score of MCMDM-1VWD (Spearman's Rank correlation coefficient, r=0.51, p<0.0001). PBAC of ≥185 had a sensitivity and specificity of 86% and 64% for predicting a menorrhagia score≥1 (AUC=0.80, p<0.0001). When women were evaluated for bleeding symptoms using the EU questionnaire, the most common bleeding symptom in the entire cohort was bruising in 42% followed by menorrhagia in 35%. Compared to women without menorrhagia, women with menorrhagia had an increased frequency of other bleeding symptoms including bleeding with invasive procedures (bleeding with tooth extraction 11% vs. 1% and bleeding with surgery 12% vs. 4%), post partum hemorrhage 15% vs. 8% and epistaxis 11% vs. 4%. Of the 150 women, 20 had an overall EU score of ≥4, mean 7.5 (range 4–22) indicating significant bleeding symptoms including menorrhagia and bruising in 85%, bleeding after surgical procedures in 45% followed by post partum hemorrhage and epistaxis in 35 %. Comparing menorrhagia cases to controls, the statistically significant difference between the two groups was in the median overall EU-BS (1 vs. 0, p=0.007) and poor wound healing with frequent scarring (24% vs. 9.3%, p=0.027).We found that 21.3% women with menorrhagia had low VWF level (VWF:RCo or VWF:Ag ≤ 60 IU/dl) compared to 14.6% women without menorrhagia (p =0.28). Comparing the 27 subjects with low VWF level with the 123 subjects with normal VWF level, the median VWF:RCo was 47 vs. 101IU/dl (p<0.0001), median PBAC was 234 vs. 176 (p=0.34), median EU-BS was 0 in both groups. There was no predictive correlation between the PBAC score and VWF:RCo (Spearman's Rank correlation coefficient r=0.05, p=0.54) and between the EU-BS and VWF: RCo level (Spearman's Rank correlation coefficient, r=0.07, p=0.37). The only predictor of low VWF: RCo was blood group O (OR: 8.9, 95% CI 2.9–27.6). The majority (87%) of blood group O subjects with low VWF level did not have a bleeding phenotype (EU-BS ≤3).In our cohort, the predictor of post partum hemorrhage was menorrhagia (OR: 4.1, 95% CI-1.4-11.8, p=0.009) and the only predictor of surgical bleeding and bleeding with tooth extraction was bleeding from minor wounds (OR: 11,95% CI-2.5-47, p=0.001; OR: 34,95% CI-6.9-166, p<0.0001 respectively). We conclude that PBAC is an easy and quick method for assessing the severity of menorrhagia in an outpatient setting. A score of ≥185 can be reliably used to diagnose clinical menorrhagia and as a threshold score to screen for other bleeding symptoms using the EU questionnaire. Women with menorrhagia are more likely to have bleeding with other invasive procedures, post partum bleeding and epistaxis. Menorrhagia could be a surrogate marker for an underlying bleeding disorder or a connective tissue disorder as reflected by the higher EU-BS and symptoms of poor wound healing in women with increased menstrual blood loss. We could not determine a PBAC score or EU-BS that could be used to identify women with low VWF. Low VWF level correlated best with blood group O, however a majority of these women did not have a bleeding phenotype (EU-BS≤3). Studies in a large cohort of women to identify reliable predictors of clinical bleeding are warranted.