Cerebral Vasculopathy Outcome in a Sickle Cell Anemia (SCA) Newborn Cohort Screened Early with Transcranial Doppler

Background: Stroke is the most severe complication in children with SCA and has been reported to concern 11% of patients by the age of 18 years. Transcranial Doppler detects patients at risk for stroke and transfusion program significantly reduces the risk of stroke in patients with abnormal TCD (Adams, NEJM 1998, 2005). However, the estimate of overt and silent strokes occurrence in a newborn SCA cohort explored early with TCD has not yet been reported.

Methods: In our Créteil pediatric center, TCD was systematically performed annually since 1992 as soon as 12–18 months of age and patients were assessed by MRI/MRA every 2 years after the age of 5 years or earlier in case of abnormal TCD. Time-averaged mean of maximum velocities higher than 200 cm/sec were considered as abnormal, resulting in initiation of a transfusion program (TP). A switch to hydroxyurea was proposed to patients with normalized velocities on TP (< 170 cm/sec) and normal MRA, but TP was reinitiated if abnormal velocities recurred. Patients with “conditional velocities” (170–199 cm/sec) were assessed with TCD every three months. Alpha genes and beta-globin haplotypes were determined. Baseline biological parameters (G6PD activity; WBC, PMN, Reticulocytes, Platelets counts; Hemoglobin, Hematocrit, HbF%; LDH levels; MCV; SpO₂) were obtained after 12 months of age, before the first TCD, a minimum of 3 months away from a transfusion, one month from a painful episode, and always before intensive therapy. One patient refused initiation of TP for abnormal TCD and was excluded from the analysis.

Results: This study included 217 SS/Sb0 children (114 M, 103 F), annually explored by TCD before age 5 (mean age at first TCD±SD: 2.2±1.1yr). They were followed for a total of 1627 patient-years. Eighty-two of 187 patients had alpha-Thal (44%); beta genotype, available in 157, was BaBa in 67 (43%), BeBe in 36 (23%), SeSe in 11 (7%) and “other” in 43 (27%); 21 of 182 (11.5%) had G6PD deficiency. TCD became “conditional” in 57 of 217 (26.3%) at the median age of 3.3 yr (range 1.2–8), and abnormal in 45 of 217 patients (20.7%) at the median age of 3.2 yr (range 1.3–8.3). Three patients had a stroke at the ages of 1.6, 4.2 and 4.4 yr: the first one had a first abnormal TCD at the age of 1.5 but had a stroke just before the TCD control at 1.6 yr; the second one had normal TCD on one side but unaivalable window on the other side and had a stroke before the first MRI/MRA; the third one had normal TCD just before the occurrence of an ACS and had a massive thrombotic stroke in intensive unit. The Kaplan-Meier (KM) estimate of stroke occurrence was 1.9% by the age of 18 yr. MRA, available in 126, was normal in 104 and showed stenoses in 22 (17.5%) at the median age of 4.8 yr (range 2.6–11.5). In stroke-free patients, MRI showed silent infarcts in 31 of 122 patients (25.4%) at the median age of 5.9 yr (range 3.5–15.3); 10 of the 31 had a history of abnormal TCD. KM estimate for abnormal TCD was 29.7% by the ages of 10 and 18 yr. KM was higher in case of G6PD deficiency: 52% vs. 26% (Log Rank, p=0.059), significantly lower in case of alpha-Thal: 16.3% vs. 36.9% (Log Rank p=0.005) and significantly dependent on the degree of hemolysis: LDH IU/l <750:11.6%; 750–1209:27%; >1209: 40.3% (Log-rank < 0.026). KM estimate of stenoses on MRA was 23.6% by the age of 18 yr, was highly significantly higher in case of G6PD deficiency: 53.3% vs 15.8% (Log Rank p<0.001), lower in case of alpha-Thal: 8.7% vs 29.9% (p=0.022), and significantly dependent on hemolysis: LDH IU/l <750: 0%; 750–1209:13.9%; >1209: 37.1% (Log Rank p=0.013). KM estimate of silent infarcts was significantly dependent on the presence of stenoses on MRA (Log Rank p=0.002), and on gender (male older than 8) (Log Rank p=0.038).

Conclusion:This monocenter prospective study reports a significant decrease in the risk of stroke at 18 yr (KM 1.9% vs. 11–11.5%, previously published by Ohene-Frempong [Blood, 1998] and Quinn [Blood, 2004]) demonstrating the strong efficacy of systematic and early screening by TCD in association with TP to prevent stroke. It confirms our recent study that G6PD deficiency, absence of alpha-Thal and hemolysis significantly and independently increase the risk for abnormal velocities (Blood 2008 in press), which are themselves highly significantly predictive of stenosis occurrence. Morever, this study shows that the risk of silent infarcts remains high, even in patients with normal TCD, and is higher in males than in girls after the age of 8 years.