Variability in Bleeding Pattern of Severe Hemophilia.

Introduction

The aim of this study was to describe variability in bleeding pattern in a single center cohort of severe hemophilia patients treated with prophylaxis.

Methods

At the Van Creveldkliniek, all patients with severe hemophilia born between 1944 and 2002 were followed from 1972 onwards. Data on bleeding characteristics and treatment were collected yearly and Pettersson scores were performed with five-year intervals. Prophylactic dose was adjusted always according to bleeding pattern. Since the number of accepted bleeds has decreased over the years, each treatment dependent indicator was described stratified for age. Age specific quartiles for the different indicators were used as arbitrary cut-off values.

Results

Data on prophylaxis were available for 247 patients, and a total of 2760 follow up years on prophylaxis were collected. Variability in treatment characteristics is shown in Table 1.

Treatment characteristics and clinical manifestations per age group

The variation in annual clotting factor use was used as a marker of bleeding pattern. Age at first joint bleed was inversely related to annual clotting factor use in the youngest age group. Using cut off levels of the 25^th and 75^th percentiles (P25 and P75) of age at first joint bleed and annual clotting factor use, 10% of patients were identified as patients with a milder bleeding pattern (i.e. age at first joint bleed above P75 and annual clotting factor use below P25) and 9.3% as patients with a more severe bleeding pattern. Using annual clotting factor use and joint bleed frequency in the middle group, 8.0% of patients were identified as patients with a milder phenotype and 10.3% as patients with a more severe bleeding pattern. In the oldest group, Pettersson scores were positively associated with annual clotting factor use. Using these parameters, 14% of the patients were identified as patients with a milder bleeding pattern. Due to the ceiling effect of the Pettersson score, these parameters could not be used to identify patients with a more severe bleeding pattern.

Conclusion

There is considerable variation in bleeding pattern of patients with severe hemophilia. Combinations of bleeding and treatment characteristics could be used as indicators of phenotype identifying 8 to 14% patients with a milder bleeding pattern and 9 to 10% of patients with a more severe bleeding pattern.