Cognitive, Brain and Intracranial Artery Involvement in Beta Thalassemia

Background: Brain involvement in beta thalassemia is scarcely known so far, and available data show different degrees of abnormalities in terms of neuroradiologic findings and cognitive impairment, mainly showing asymptomatic white matter lesions in transfusion dependent thalassemia (TDT; Karimi et al, Ann Hematol 2016) or in non transfusion dependent thalassemia (NTDT; Karimi et al, Ann Hematol 2012 ; Pazgal et al, Thrombosis Research 2016), arterial stenosis in NTDT (Musallam et al, EJH 2011) and a general impairment in cognitive function in TDT (Elalfy et al, Hematology 2017). To our knowledge there are no studies investigating neuroimaging and cognitive function in both groups of patients, comparing results to healthy subjects.

Methods: In our observational study thalassemic patients from 4 major Centers in the South of Italy, aged more than 16 years, both Transfusion Dependent (TDT) and Not Transfusion Dependent (NTDT), and healthy volunteers underwent 3T brain MRI including a FLAIR sequence and intracranial time-of-flight MR-angiography. Three neuroradiologists blinded of clinical data evaluated the MR-exams reporting the presence of artery stenosis and aneuryms and the presence of white matter signal abnormalities graded according to the Fazekas score; the number and the size of parenchymal lesions (<0.5cm, 0.5-1.5cm, >1.5cm) were recorded. Both patients and controls were administered the Wechsler Adult Intelligence Scale-Fourth Edition (WAIS-IV) to detect cognitive ability, and results were analyzed by a team of blinded psychologists and psychiatrists. Clinical and laboratory data were collected to correlate neurologic and psychometric findings with disease history and severity.

Results: Seventy-four thalassemic patients (46/74 females, mean-age 35.2±11.4 years, range 16-66; 51 TDT and 23 NTDT; mean hemoglobin 9.3±1.4 g/dl, range 7.5-12; 41 splenectomized; 34 on prophylactic aspirin) and 56 healthy controls (36/56 females, mean-age 34.2±11.7 years, range 17-66) were enrolled. No intracranial artery stenosis was detected. Patients and controls did not differ regarding the incidence rate of aneurysms (9.46% vs 9.67%). White matter lesions (WML) were found both in patients and controls, but the prevalence (35/74; 47.30% vs 28/56; 50.0%), the number (4.1 vs 4.6), the size and the Fazekas score did not differ between the groups. Compared to controls, patients presenting WML tend to be younger ( 35.32y compared to 39.89y, p= 0.033). No differences in WML were found between TDT and NTDT, nor between patients on aspirin and others, and splenectomy was associated with an increased lesion burden only in TDT ( mean 5.56 versus 1.53, p= 0.048); in splenectomized patients on aspirin the number and prevalence of WML was lower compared to those splenectomized not on prophylaxis, but the difference did not reach the statistical significance. Patients with WML did not differ from others in terms of comorbidities, hemoglobin, platelets, erythroblasts, ferritin, hemolysis markers and chelation therapies. The presence of WML did not correlate with cognitive impairment as measured by the Intelligence Quotient (IQ) derived from the WAIS-IV test, both in patients and in controls. IQ in patients was significantly lower than controls, and this finding remained significant after correction for education (IQ: 82.75±11.11 in patients, 93.96±SD13.22 in controls, p=0.032), regardless of the presence of WML, that was again comparable.

Conclusions: Beta-thalassemia and chronic blood transfusion treatment do not seem to imply significant brain and intracranial artery changes, although cognitive function seems to be affected. Further analysis are required to clarify if the general cognitive impairment derives from functional changes in cerebral connectivity, or if it is one of the effect of the chronic course of the disease.