Hypercoagulability in Pediatric Patients with Sickle Cell Disease and Correlation to Transcranial Doppler and Disease Severity

Abstract 3221

Children with sickle cell disease (SCD) are at high risk for thrombotic stroke. Transcranial doppler ultrasound (TCD) is utilized to predict children at highest risk. Anemia and and vessel occlusion are assocated with elevated TCD velocities and risk of stroke. There is increasing evidence that a prothrombotic state contributes to the complications of SCD including stroke. We hypothesized that increased thrombin generation and platelet activation are associated with increased TCD velocities. We conducted a cross-sectional cohort study of children with SCD at Duke University Medical Center. Children had been clinically well for at least 2 weeks, had not been transfused for at least 3 months, and had TCD performed at study enrollment (n= 28) or within the prior 6 months (n=7). TCD time-averaged mean velocities (TAMV) were calculated for each patient and used for analysis. Blood was collected at the time of study entry per study protocol. Thrombin generation was measured utilizing: D-dimer; thrombin antithrombin (TAT); and calibrated automated thrombography (CAT) which evaluates phases of thrombin production including lag phase, time to peak thrombin, endogenous thrombin potential (ETP), and peak thrombin. Platelet activation was assessed by measurement of soluble glycoprotein V (sGPV). Clinical data were abstracted from the medical record including confirmation of SS genotype, use of hydroxyurea, prior sickle cell-related clinical events, and hematological data including hemoglobin, platelets and white blood cell count. SCD severity scores were calculated to classify patients as either high risk or low/moderate risk (van den Tweel et al, 2010). Linear regression analyses were conducted to assess correlation. Children with high risk severity scores were compared to children with low/moderate risk severity scores. Statistical analysis was performed using Graphpad 5.0; p<0.05 was considered significant. We evaluated 35 patients (median age 10 years, range 3 – 17). Linear regression revealed a correlation between the TCD velocities for TAT (r²=0.147, p=0.03), ETP (r²=0.13, p=0.049), and peak thrombin (r²=0.12, p=0.05). There was no significant correlation between TCD velocity and D-dimer, lag phase or time to peak thrombin. There also was no correlation between TCD velocities and hemoglobin, platelets or sGPV. Patients taking hydroxyurea were found to have lower thrombin generation (D-dimer: 882 ± 93 ng/ml vs. 1679 ± 428, p<0.0001) and TAT (6.8 ± 0.67 ng/ml vs. 10.59 ± 2.22, p<0.0001). There were 8 (23%) patients classified as high risk and 27 (77%) classified as low/moderate risk. Comparison of comparison of D-dimer, TAT, and each phase of thrombin generation between patients with low/moderate severity scores and those classified as high severity revealed no significant differences. In summary, increased TAT, ETP, and peak thrombin levels correlated with increased TCD velocities. Markers of thrombin generation were lower in children taking hydroxyurea. The lag phase and time to peak thrombin which both reflect the ‘thrombin burst’, were not predictive of TCD velocities. In addition, hematological values and platelet activation did not correlate with TCD velocities. Further studies are warranted to understand the contribution of thrombin generation to stroke risk, to determine if markers of thrombin generation are independent markers of stroke risk, and determine if hydroxyurea can ameliorate this risk.