Introduction: The majority of carriers of hemophilia A have historically been considered to have normal hemostasis (FVIII:C >50%) and thus not have an increased bleeding tendency. However, our research group has demonstrated that some hemophilia A carriers can experience increased bleeding compared to normal women despite this normal FVIII activity. In our recently published study in adult hemophilia A carriers there was no correlation between FVIII:C, as measured by a one-stage coagulation assay, and bleeding phenotype, as measured by the MCMDM-1 VWD bleeding score. In this follow up study we sought to determine which of these hemophilia A carriers with normal FVIII:C are at risk for bleeding. The goal of this study is to determine the relationship between FVIII assays (one-stage and chromogenic) and thrombin generation with both menstrual bleeding and an overall bleeding tendency.

Methods: We recruited mothers of children with hemophilia A in the Emory University pediatric bleeding disorders clinic. We included only adult (>18 year of age) obligate hemophilia A carriers that did not have a concomitant bleeding disorder or chronic disease that would increase their bleeding tendency. We gathered basic demographic information and evaluated the overall bleeding tendency using the ISTH Bleeding Assessment Tool (ISTH BAT), a semi-quantitative assessment of bleeding scored from 0 to 56. We considered a score of 6 or higher consistent with pathologic bleeding. In addition, we inventoried menstrual bleeding with the Pictorial Bleeding Assessment Chart (PBAC), a visual representation of menstrual blood loss. We considered a PBAC greater than 100 consistent with heavy menstrual bleeding. Plasma samples were collected from each subject at the time of the survey in sodium citrate tubes without corn trypsin inhibitor. Samples were double spun at 2,500 rpm for 15 minutes and stored at -80°C.Laboratory evaluation for each subject included FVIII:C, measured by one-stage coagulation assay (aPTT reagent with micronized silica) and chromogenic assay (COATEST SP4 FVIII kit, Chromogenix), as well as thrombin generation (Calibrated Automated Thrombogram, PPP-Reagent Low, FluCa Kit) to evaluate endogenous thrombin potential (ETP, nM*min), peak thrombin (nM) and lag time (min). We performed linear regression using GraphPad Prism; p<0.05 was considered significant.

Results: Over a three-month period, we approached 32 adult obligate hemophilia A carriers; 23 agreed to participate in the survey, 16 consented to blood draw. One carrier was excluded in extreme outlier analysis and due to an autoimmune disorder, leaving 15 evaluable subjects. Our cohort is relatively young with a median age of 41 years (range 23-51), predominantly Caucasian (53%, 8/15) and the majority carry a severe mutation (10/15 severe, 2/15 moderate, 3/15 mild).

Reproductive bleeding (post-partum hemorrhage and/or menstrual bleeding) was commonly reported (93%, 14/15). The median ISTH BAT bleeding score for subjects was 2 (range 1-8). The median PBAC score was 148 (range 10-608). The carriers had a relatively normal FVIII:C by one-stage assay with a median of 0.78 U/mL (range 0.30 - 1.06) and by chromogenic assay with a median of 1.15 U/mL (range 0.66 - 1.9).

FVIII:C by one-stage assay was inversely correlated to PBAC (r2 =0.4; p=0.01, figure 1A). Conversely, FVIII:C by chromogenic assay did not correlate with PBAC, however trended toward significance (r2 = 0.19; p=0.10, figure 1A). Additionally, there was no correlation found between FVIII:C (one-stage or chromogenic assay) and ISTH BAT bleeding score (figure 1B). There was also no correlation found between bleeding score or PBAC and the parameters of the thrombin generation assay or between severity of the closest male relative's hemophilia and any of the assays performed.

Conclusions: In our cohort of adult obligate hemophilia A carriers, as the FVIII:C (one-stage assay) decreased, the menstrual bleeding tendency increased as measured by the PBAC. We were unable to find a correlation between other measures of hemostasis, including the chromogenic assay and thrombin generation, and commonly used bleeding assessment tools. Further larger studies are warranted to help determine which hemophilia A carriers would be at risk for bleeding.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.