Intracranial blood flow velocities in patients with sickle cell disease and β‐thalassemia intermedia

To the Editor: Thalassemia intermedia (TI) and sickle cell disease (SCD) represent two common forms of hereditary anemias that share many clinical manifestations and pathophysiologic mechanisms. Decreased nitric oxide production, endothelial dysfunction, and formation of reactive oxygen species play a role in increasing the thrombosis risk in both of these diseases [1,2]. Transcranial Doppler ultrasonography (TCD) detects patients with SCD who are at an increased risk of developing stroke. Despite the increased risk of cerebral thrombosis in TI, the role of TCD in identifying high-risk patients has not been properly evaluated. We aimed at comparing the mean maximal blood flow velocities (MBFV) in two populations of patients with SCD and TI, and we looked at the role of MBFV in predicting magnetic resonance imaging (MRI) abnormalities. We also evaluated the role of age and anemia in altering the intracranial blood flow. Between January 2011 and October 2011, a total of 60 patients with SCD and 20 patients with TI were enrolled in this study. After obtaining informed consent, the right and left, middle and anterior cerebral arteries (RMCA, LMCA, RACA, and LACA) were insonated using a temporal window approach. The MBFV of each artery was recorded separately. Demographic and recent (less than 3 months) laboratory information were collected for each patient. Sixteen out of the 20 patients with TI underwent brain MRI for another study. The MRI results were used as they were reported. All statistical analysis was done using SPSS version 19. Table I shows the baseline characteristics of the two groups of patients along with their respective MBFVs. There is no statistically significant difference between the distribution of sex and hemoglobin levels among the two groups; however, a statistically significant difference was detected regarding the age distribution. Fifty-four out of the 60 sickle cell patients have SS (90%) while 6/60 (10%) have Sb. Maximal blood flow velocity was significantly different among the two groups in the RACA (P 5 0.006), borderline significant in the LMCA (P 5 0.056), and not significant in the RMCA and LACA. Supporting Information Figs. 1 and 2 represent the linear trend lines correlating MBFV in the four arteries with age and hemoglobin in the TI and SCD group, respectively. Children with sickle cell have been found to have higher MBFV than the normal population in several studies [2]. Several mechanisms have been implicated including anemia, deficiency of nitric oxide synthesis, and endothelial dysfunction. The pathophysiology of SCD and TI includes disruption of the aforementioned mechanisms, thus one would expect the MBFV to be similar in these two diseases. In this study, the mean MBFV in 60 patients with SCD were higher than those of 20 patients with TI, despite similar hemoglobin levels. The difference was only statistically significant for the RACA and borderline significant for the LMCA. The fact that the results did not reach statistical significance could be due to many reasons. First, our TI sample was relatively small. TI is a rare disease and the results presented herein are for proof of concept. Secondly, patients with SCD were all on hydroxyurea therapy which has been shown to lower MBFV [3]. We also aimed at selecting the older patients with SCD to try to annul the confounding effect of age on MBFV since our TI population was significantly older than the SCD population. The average age of our population was close to 14 years with a wider distribution as compared to an average age of 10 years in the STOP study. Older patients with SCD have been reported to have lower MBFV than the pediatric patients in several studies. An ROC analysis for a cut-off MBFV that correlates with a 100% sensitivity and 73% specificity for the diagnosis of an internal carotid artery or middle cerebral artery stenosis on MR imaging was found to be 123.5 cm/sec which is lower than the value of 200 cm/ sec used for children [4]. MBFVs have also been shown to decrease with age in normal controls [5]. It is safe to assume that age has the same effect on our TI population, and in fact, Figs. 1 and 2 show that age, in addition to hemoglobin level, has a similar effect in both disease, despite the fact that the linear regression model did not reach statistical significance. The mean MBFV values for TI were also higher than those reported for the normal population [6]. Out of the 20 patients with TI, 16 had previous brain MRI with 9 patients having asymptomatic abnormal white matter lesions (silent infarcts). No statistically significant correlation was found between MBFV and the presence or absence of silent infarcts (P 5 0.21). However, we noticed a cluster of abnormal MRIs in the lower range of the MBFV (highlighted in Table 2, Supporting Information). Whether this was an incidental finding or not merits consideration in larger studies. Patients with TI have been shown to be at an increased risk of silent brain infarcts [1]. TCD has been used in SCD to predict the risk of stroke in children and abnormal stenosis in brain vasculature in adults. However, the role of TCD in TI has not been properly elucidated. This pilot study shows that patients with TI have higher MBFV than reported controls, but lower than those in SCD. Larger scale studies that are longitudinal in nature are required to elucidate the role of TCD in predicting the risk of silent infarcts in patients with TI.