Effects of Chronic Transfusion Therapy on MRI and MRA in Children with Sickle Cell Anemia at Risk for Primary Stroke: Baseline Imaging from the Twitch Trial

Introduction :

Stroke is a major cause of morbidity and mortality in children with sickle cell anemia (SCA), and current efforts seek to prevent primary stroke in this high-risk population through transcranial Doppler (TCD) screening programs and prophylactic blood transfusions for children with abnormally high intracranial TCD velocities. During the SWITCH trial (ClinicalTrials.gov NCT00122980), a novel MRA vasculopathy grading (VOG) scale was developed, which documented baseline bilateral vessel stenosis and parenchymal injury in children receiving transfusions for secondary stroke prophylaxis, and provided a severity scale suitable for future clinical trials. We now describe the baseline brain magnetic resonance imaging/angiography (MRI/MRA) results and vasculopathy grading scores in children with SCA on primary stroke prophylaxis, and report correlations with intracranial TCD velocities and duration of transfusion.

Methods:

TCD With Transfusions Changing to Hydroxyurea (TWiTCH) is a Phase three randomized, controlled, multicenter clinical trial (ClinicalTrials.gov NCT01425307), comparing transfusions versus hydroxyurea for children with abnormal TCD velocities. Children were eligible for TWiTCH enrollment if they currently receive monthly transfusions for primary stroke prophylaxis, and had no known history of stroke, transient ischemic attack, or severe vasculopathy. All TWiTCH participants underwent baseline MRI/MRA examinations with central review after study enrollment. Baseline TCD examinations were also centrally reviewed and Time-Averaged Mean Velocities (TAMV) were recorded for the distal internal carotid artery (dICA), internal carotid bifurcation (BIF), first segment middle cerebral artery (MCA) = M1, mid MCA (MCA), and posterior cerebral artery (PCA). MRA vasculopathy grades were recorded bilaterally and compared to the TCD velocities and duration of transfusion.

Results:

Of 159 enrolled subjects, 143 had complete baseline MRI/MRA data. The mean age at study enrollment was 9.8 years, with a 2:3 male:female ratio, and mean age at the original abnormal (Index) TCD exam was 5.4 years. A total of 62 children had parenchymal abnormalities identified on baseline brain MRI, including 2 with infarcts, 54 with leukoencephalopathy, 5 with encephalomalacia, and 20 with lacunae lesions (not mutually exclusive). A total of 124 children had baseline MRA vasculopathy scores ≤3, while 19 had grades 4-6 (severe vasculopathy) that required removal from the study. There were significant positive associations between the baseline hemisphere vasculopathy grades and the corresponding TAMV velocities in the Left dICA (p=0.02), Left BIF (p˂.0001), Left MCA (p=.0002), Right BIF (p=.0003) and Right MCA (p=.0002). Likewise, when TCD velocities were categorized as low (<70 cm/s), normal (70-169 cm/s) or conditional/abnormal (≥170 cm/s), there were positive associations between left-sided vasculopathy and the left BIF (p=.04) and MCA (p=.003), along with right-sided vasculopathy with the right dICA (p=.04). The mean duration of transfusion for subjects with baseline MRA vasculopathy scores ≤3 was 1557 days, while those with severe vasculopathy (grades 4-6) was 1985 days; p= 0.053.

Conclusions:

Since the TWiTCH study excluded the enrollment of children on chronic transfusion therapy for primary stroke prevention with a known history of stroke, TIA, or severe vasculopathy, the prevalence of baseline brain MRI and MRA abnormalities was relatively higher than expected and 19 participants failed screening due to unrecognized severe vessel stenosis. The recent published SWiTCH vasculopathy grading scale was validated in the TWiTCH population, with more severe stenosis found to be associated with higher ipsilateral TCD velocities. Exit brain MRI/MRA exams will be compared to entry studies, to provide longitudinal data including comparisons between the effects of continued transfusions and hydroxyurea.