Brain MRI Findings in Lebanese Sickle Cell Disease Patients.

Background: Sickle cell disease (SCD) is one of the most prevalent hemoglobinopathies in Lebanon. Acute stroke, defined as an acute neurological syndrome secondary to arterial occlusion or hemorrhage with resultant neurological symptoms and signs lasting for more than 24 hours, is the most catastrophic brain injury seen in this disorder and has a prevalence of around 10% by 50 years of age. Silent cerebral infarcts (SCI), defined as an abnormal magnetic resonance image (MRI) of the brain and no history of neurologic deficit, are seen in 20–44% of children with sickle cell anemia (SCA) and represent a major risk factor for overt stroke. Identifying asymptomatic patients at high risk of developing sickle-related brain injury helps physicians implement preventive therapies.

Aims:

1. Determine prevalence of brain imaging abnormalities in a group of Lebanese patients with SCD and

2. Identify predictive factors for these abnormalities.

Methods: A review of brain MRI for 53 SCD patients followed in 2 centers specialized in inherited hemoglobin disorders was undergone. Of those, 42 asymptomatic patients had surveillance brain MRI and 20 had imaging at least twice. T1- and T2- weighted MR imaging at 1.0 T plus FLAIR at 5-mm section thickness were used. All images were read independently by two radiologists with a 10 years experience in neuroradiology. Brain images considered abnormal included atrophy, lacunar infarction, leukoencephalopathy and encephalomalacia. Fisher’s Exact test (Wilcoxon two sample rank-sum test) was employed to study the association of binary data (continuous data) with the 2 groups. The SAS v8.2 (Cary, N.C.) was used to analyze the data. A p-value less than 0.05 was considered significant.

Results: Overall stroke prevalence in this group was 34% (18/53). 11 patients (21%), median age 14 years and 6 months, had overt strokes. Of those, 91% had SCA and 9% sickle beta thalassemia (ST). Of the remaining 42 patients, 6 (14%), median age 13 years and 5 months, all having SCA, had MRI findings compatible with SCI. 3 had brain atrophy and 3 encephalomalacia. None of the overt stroke patients had ever received hydroxyurea. As for SCI patients, 5 were on hydroxyurea at imaging time. No one of these 5 had undergone brain imaging prior to hydroxyurea treatment, and it is possible that these SCI are old ones. Bivariate analysis showed an association between the occurrence of overt stroke and premature death (p=0.006), regular blood transfusions (p=0.006), and osteomyelitis (p=0.009). Bivariate analysis showed an association between the occurrence of SCI and presentation of disease before 2 years of age (p=0.053, borderline) and acute chest syndrome (p=0.035). The small number of patients in this group did not allow for multivariate analysis.

Conclusion: Stroke prevalence rate in Lebanese SCD patients is lower than reports from others. This may be attributed to the small sample size and to the imaging technology utilized being less sensitive than newer MRI technology. A multicenter Lebanese SCD trial is warranted to assess the true stroke prevalence, the importance of modifying genes and gene-gene interaction in predicting stroke risk, and the efficacy of hydroxyurea for primary stroke prevention.