Background Kaplan-Meier (KM) estimated cumulative incidence of silent cerebral infarcts (SCI) in a SCA-newborn cohort with sickle cell anemia (SCA) early screened by transcranial-color-Doppler-ultrasound (TCD-US) was shown to be high, despite initiation of chronic transfusion (CT), in children detected at risk by TCD, i.e., 28.2% by age 8, 37.4% by age 14, and an incidence of 3.4/100 patient-years (Bernaudin et al Blood 2011). We hypothesized that adding early extracranial Internal Carotid Artery (eICA) assessment would contribute to reducing SCI incidence.

Patients and Methods TCD-US was systematically performed since May 1992 in our pediatric sickle cell anemia (SCA) cohort. Cerebral MRI/MRA was assessed every two years after age 5, or earlier in children on CT for abnormal time-averaged mean velocity (TAMV). eICA-color-Doppler-ultrasound and neck-MRA were added since June 2011. The highest TAMV in middle, anterior, posterior, basilar, intra and extracranial internal carotid cerebral arteries were recorded. Abnormal intracranial and eICA TAMV were defined as ≥ 200 cm/s and ≥ 160 cm/s, respectively. Stenosis was defined as a ≥ 25% decrease in the lumen of arteries. SCI was defined as an hyperintensity focus of at least 3 mm diameter, visible in two planes on FLAIR MRI. Genetic markers (alpha and beta genes, G6PD activity) and baseline biological parameters were recorded during the 2nd year of life before intensive therapy, and away from crisis and transfusion. Hospitalizations, annual check-up data, Doppler, MRI/MRA assessments and events (abnormal-TAMV, intra or extracranial stenosis and SCI) were prospectively recorded.

CT was systematically applied in children with TAMV ≥ 200 cm/s for any artery and children with normalized TAMV and no stenosis were switched to hydroxyurea (HU) with CT overlap until reaching the maximal tolerated dose. For children with eICA TAMV 160-199 cm/s, neck-MRA was performed, and CT applied in those with eICA-stenosis while the others were given HU. Moreover, HU was initiated in children with frequent crises or baseline hemoglobin < 7g/dL. Stem cell transplantation was performed in children with cerebral vasculopathy and in those with frequent crises despite HU treatment.

Cerebral arteriopathy was defined by history of abnormal intracranial or eICA TAMV or of intra or extracranial stenosis.

Results We analyzed data updated in 7/2019 from stroke-free SCA children, born between 11/1992 and 8/2015, all assessed with cerebral MRI/MRA, including neck-MRA (n=338). Median (range) age at first and last cerebral MRI was 5.3 (1.8-16.6 yr) and 11.1 (2.8-19.9 yr), respectively, providing 3866 patient-years of MRI follow-up. SCI occurred in 65/388 SCA-children at median age of 6.4 (1.8-17.9 yr). Incidence of SCI was 65/3866 or 1.7/100 patient-years. The KM-estimated cumulative incidence of SCI was 15.1% (95%CI: 10.9-19.3%) by age 8 and 25.7% (95%CI: 19.7-31.7%) by age 14.

Intracranial and isolated eICA-arteriopathies were present in 110/338 and 54/338 patients, respectively. SCI were significantly associated with the presence of intra or extracranial arteriopathy (Fisher test, p=0.013) as 41/65 children with SCI had intra or extracranial arteriopathy (20 isolated intra, 15 isolated extra and 6 intra and extracranial), while in the other 24 patients, 5 had history of conditional TAMV, 11 of acute anemia and 8 of recent acute chest syndrome.

At first cerebral MRI, the multivariate Cox regression analysis retained as significant and independent predictive risk factors for SCI, baseline high reticulocyte count [HR=1.005 (95%CI: 1.002-1.008); p=0.004] and LDH [HR=1.001 (95%CI :1.001-1.002); p=0.001].

Discussion This long-term longitudinal study shows a reduction of the risk of SCI with time. The addition of eICA-assessment allowed the detection of isolated extracranial arteriopathy in 54 patients, with 15 who had SCI. The earlier HU initiation reducing the risk of ACS, acute anemia and hemolytic rate, along with detection of isolated eICA assessment and the use of transplantation in children with cerebral arteriopathy have most likely contributed to this overall improvement. Nevertheless, we suggest that initiating CT in patients with eICA-TAMV 160-199 cm/s, even in the absence of eICA stenosis, as done for intra- and extracranial-TAMV ≥ 200 cm/s, could further reduce SCI risk.

Bernaudin:blueBirdBio: Consultancy; AddMedica: Honoraria; GBT: Consultancy, Membership on an entity's Board of Directors or advisory committees; vertex CRISPR: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution