Brain MRI/MRA Findings after Hydroxyurea Treatment in Children with Sickle Cell Anemia

Background: Up to 40% of children with sickle cell anemia (SCA) will have abnormalities on brain imaging due to their hematologic disorder, much of which is subclinical. Common abnormalities on brain magnetic resonance imaging/magnetic resonance angiography (MRI/MRA) include leukoencephalopathy from microvascular ischemic insult and vascular stenosis from endothelial damage. Silent cerebral ischemic insults are progressive; further neurologic abnormalities, including overt stroke, are more common among children with ischemic findings on MRI compared to children without them. Furthermore, poor performance on neuropsychological testing, lower IQ, and higher rates of grade retention are common in children with SCA and cerebral ischemic disease. The effect of hydroxyurea treatment on the development and progression of vascular stenosis and leukoencephalopathy is unclear. This study aimed to longitudinally evaluate the development of intracerebral abnormalities through serial MRI/MRA in children with SCA (HbSS and HbSβ⁰-thalassemia) who receive long-term therapy with hydroxyurea.

Methods: Children with SCA and no prior history of overt stroke enrolled in the Hydroxyurea Study of Long-Term Effects (HUSTLE NCT00305175) underwent brain MRI/MRA, transcranial doppler (TCD) examinations, and laboratory evaluations immediately before hydroxyurea initiation and after 3 and 6 years of treatment to maximum tolerated dose. MRI/MRAs were reviewed for the presence or absence of vascular stenosis and leukoencephalopathy. Leukoencephalopathy was defined as white matter T2 hyperintensity. Proportions of abnormal MRI/MRA findings were compared between baseline and 3 years and baseline and 6 years using McNemar’s test and the Wilcoxon-Mann-Whitney exact test was used to explore associations with laboratory parameters.

Results: Forty-six children with SCA, mean age 9.4 years (range 1-17.3), had an MRI/MRA at baseline and 3 years post-initiation of hydroxyurea. Ten children had an additional MRI/MRA after 6-years of hydroxyurea therapy. Frequencies of leukoencephalopathy and vascular stenosis are shown in the table. Prevalence of leukoencephalopathy before hydroxyurea therapy was higher than that reported in the literature for untreated children with SCA of similar age. There were no significant differences between baseline imaging findings and those at 3 and 6 years.

Children with leukoencephalopathy at baseline and 3 years were older than those without (mean 10.7 vs. 7.7 years, p=0.01; 10.5 vs. 7.5 years, p=0.02 respectively). Lower HbF at baseline was associated with the presence of leukoencephalopathy at year 3 (median HbF 4.6% vs. 12.4%, p=0.008), but there was no association between HbF at 3-years and the presence of leukoencephalopathy at 3-years (median HbF 16.4% vs. 13.2%, p=0.55). When stratified by age, these findings were similar. TCD velocities and other hematologic parameters were not associated with MRI/MRA abnormalities. The small number of vascular stenosis cases precluded further analyses of this outcome.

Conclusions: In this longitudinal study, children treated with hydroxyurea for 3-6 years did not demonstrate an increased frequency of vascular stenosis or leukoencephalopathy on brain MRI/MRA during treatment. Older age at hydroxyurea initiation and lower pre-hydroxyurea HbF percentage were associated with the presence of leukoencephalopathy at baseline and 3-years. These findings suggest that hydroxyurea may mitigate the expected progression of vascular stenosis and leukoencephalopathy in children with SCA, and that this therapy should be initiated early on, before the development of cerebrovascular disease, particularly among those with a low HbF percentage.