BACKGROUND:

Stroke, occurring in about 10 percent of pediatric patients with sickle cell disease, is one of the most devastating complications, with a high recurrence rate after the first episode. Several non-randomized studies have shown reduction in stroke recurrence when periodic blood transfusions are administered to maintain hemoglobin S under 30 percent. Periodic blood transfusions are associated with significant risks of iron overload and other complications and must be accompanied by parenteral iron chelation therapy. However, this has become a standard of care for prevention of recurrent stroke in SS children. Thus, a randomized trial of blood transfusion for secondary prevention would not be feasible because it would be considered unethical. Based on various studies, the recurrence rate is reduced from 46 to 67 percent to approximately 7 percent on transfusion therapy. Because most stroke patients are left with some neurological deficit, and face a lifetime of disability, primary prevention would have a significant impact on the management of patients. However, because of complications of blood transfusions, the hypothesis should be proven by a randomized clinical trial.

A primary prevention trial had not been possible because an acceptable means of detecting those children at risk of stroke was not available. The advent of TCD to identify arterial abnormalities for the prediction of stroke has provided a means of detection. TCD abnormalities have a high specificity (100 percent) and high sensitivity (90 percent) for detecting angiographically proven narrowing of arterial diameter. Thus, TCD examination of the basal cerebral arteries is predictive of who will develop a stroke.

DESIGN NARRATIVE:

Randomized, Phase III, multicenter. Approximately 3,000 children from 12 clinics were screened with transcranial Doppler (TCD). A total of 130 were randomized to receive either standard supportive care or periodic blood transfusions if they were found to be at high risk of stroke on the basis of elevated cerebral blood flow as measured by TCD screening tests. Primary endpoints included clinically evident symptoms of cerebral infarction with consistent findings on magnetic resonance imaging (MRI), and/or symptomatic intracranial hemorrhage. Secondary endpoints included asymptomatic brain lesions detected by MRI in brain areas not involved in primary endpoints. Hematologic characteristics of the high risk group were analyzed and serum and DNA samples frozen for future analysis. Recruitment ended in October 1997 with the accrual of 130 subjects. The clinical phase ended in 1999.