ISTH 2013: Report from the International Society of Thrombosis and Haemostasis Congress

At the 24th Congress of the International Society of Thrombosis and Haemostasis (ISTH 2013), held June 29 through July 4 in Amsterdam, specialists discussed new findings in hematology. Dr. David Green was in attendance to report the following highlights:

Predicting Bleeding in Patients with Atrial Fibrillation Taking Warfarin

A number of scoring systems are used to assess thrombosis and bleeding risks in patients with atrial fibrillation. The most popular is the CHADS₂ score, which was developed to predict thrombosis risk but has also been used to estimate propensity to bleed in patients taking warfarin. All of the scores — CHADS₂, HAS-BLED, ATRIA, and HEMORR₂HAGES — consider older age, hypertension, and previous stroke. However, all but CHADS₂ include abnormal renal or hepatic function, factors known to increase the risk for bleeding.

To determine whether the CHADS₂ score is as predictive of bleeding as the other scores, Barnes and colleagues calculated bleeding rates for each risk score in 2577 patients with atrial fibrillation taking warfarin for ≥1 year. These patients experienced a total of 116 major bleeds (defined as a decrease of 2 g/dL of hemoglobin or transfusion of 2 units of red cells). The C-statistic, a measure that quantifies the predictive validity of the scores, was poorer for CHADS₂ versus the other scores (0.53 vs. 0.66—0.75; P<.0001), indicating that CHADS₂ is an inferior measure of bleeding risk. Therefore, the CHADS₂ score should not be used to predict the risk for hemorrhage in patients with atrial fibrillation taking warfarin. Although the other scoring systems improve bleeding prediction, they are not perfect, and they do not replace meticulous evaluation of each patient for bleeding risk.

Assessing Bleeding Risk in Patients with Severe Factor XI Deficiency

The fact that bleeding intensity is so variable in patients with factor XI deficiency is a clinical conundrum. Factor XI is activated by factor XII and thrombin, and it, in turn, activates factor IX in the coagulation cascade. However, factor IX can also be activated by the tissue factor–factor VII complex, so one might predict that low levels of factor XI would not be associated with severe bleeding.

To assess hemostasis in patients with factor XI deficiency, Zucker and colleagues evaluated 16 patients with average factor XI levels ranging from 2% to 6% of normal. The patients were divided evenly into two groups, based on whether or not they had a history of bleeding following tooth extractions. Compared with nonbleeders, bleeders had less-stable clots and reduced fibrin network density; most also had retarded clot formation in the presence of thrombomodulin, which binds and inhibits thrombin. The results of the study suggest that some patients with factor XI deficiency also have a defect in clot formation and that studies of clot stability and dissolution might assist in predicting bleeding risk.

Prophylaxis of Severe Hemophilia with a Long-Acting Recombinant Factor VIII

Because native factor VIII has a half-life of only 12 hours, prevention of bleeding in patients with severe hemophilia requires intravenous infusions of factor VIII at least 3 or 4 times per week. Now, Mahlangu and colleagues conducted a phase III study of long-lasting recombinant factor VIII that was reformulated by linking one molecule of factor VIII to the Fc domain of immunoglobulin G1 (rFVIIIFc). A total of 118 patients with severe hemophilia A were randomized to receive rFVIIIFc as individualized prophylaxis (26–65 IU/kg every 3–5 days), weekly prophylaxis (65 IU/kg), or episodic treatment (10–50 IU/kg).

Results were as follows:

• The geometric mean half-life of rFVIIIFc was 19.0 hours; 30% of patients achieved effective blood levels at a dosing interval of 5 days.

• The annualized bleeding rate (bleeds per year) was lower in 24 patients who received weekly prophylaxis than in 23 who received episodic treatment (3.6 vs. 33.6).

• More patients who received prophylaxis versus episodic treatment were free of bleeding episodes (45% vs. 17%).

• Bleeding was controlled with one infusion in 87% of bleeds and with 2 infusions in 98% of bleeds.

• rFVIIIFc was effective in preventing excessive bleeding in all 9 patients who underwent surgery.

• No patient experienced episodes of thrombosis, anaphylaxis, inhibitor development, or serious adverse events.

Long-acting rFVIIIFc appears to safely decrease the need for frequent intravenous infusions in patients with severe factor VIII–deficiency hemophilia.