Evolution of Extracranial Internal Carotid Artery Disease in Children With Sickle Cell Anemia

Cerebral arteriopathy in patients with sickle cell anemia mainly affects the intracranial anterior circulation. However, the extracranial internal carotid artery (eICA) can also be stenosed and responsible for ischemic lesions. In children with sickle cell anemia, we perform routine annual Doppler ultrasound assessment of the eICA and magnetic resonance imaging with 3-dimensional time-of-flight magnetic resonance angiography of the Willis circle and neck arteries in those with abnormal velocity. Our aim was to report the evolution of eICA stenoses from 2011 to the present as a function of therapy in a retrospective case-series study. We hypothesized that chronic transfusion (CTT) would be more effective than hydroxyurea and simple observation on the evolution of eICA stenosis.

Eligibility criteria were a history of eICA velocity ≥160 cm/s with a minimum Doppler and magnetic resonance imaging follow-up of 1 year. eICAs were graded for stenosis according to NASCET (The North American Symptomatic Carotid Endarterectomy Trial). Magnetic resonance imaging was investigated for ischemic lesions. Treatment with hydroxyurea and CTT were obtained from the chart review.

Fifty-four patients were included. Eight patients had a stroke history. The median (range) follow-up was 4.7 years (1.1-9.2 years). On the first neck magnetic resonance angiography, stenosis was present in 48/54 (89%) patients. Kinking was found in 39/54 (72%) patients. On the last neck magnetic resonance angiography, the proportion of patients with eICA stenosis decreased to 39/54 (72%). ICA occlusion occurred in 5 patients despite CTT. Three patients had carotid webs without intracranial stenosis. The proportion of patients with improvement in stenosis score was 8% with no treatment intensification, 20% with hydroxyurea, and 48% with CTT (P=0.016). The mean (SD) change per year in stenosis score was 0.40 (0.60) without intensification, 0.20 (0.53) with hydroxyurea, and -0.18 (0.55) with CTT (P=0.006). Ischemic lesions were present initially in 46% of patients, and the incidence of progressive ischemic lesions was 2.5 events/100 patient-years. Cox regression analysis showed that the initial score for eICA stenosis was a significant predictive factor for the risk of new silent cerebral infarct events.

Our study reinforces the need to assess cervical arteries for better prevention of cerebral ischemia and encourage initiation of CTT in sickle cell anemia children with eICA stenosis.