Change over Time of Transcranial Doppler Cerebral Blood Flow Velocity and the Relationship with Turbulence on Magnetic Resonance Angiography in Patients with Sickle Cell Anemia.

Cerebrovascular disease (CVD) involving the circle of Willis and documented using time-averaged maximum velocity (TAMM) on transcranial Doppler (TCD) and on magnetic resonance angiography (MRA), is common in sickle cell anemia (SCA). Few studies have examined the natural history of CVD. The relationship between TAMM and MRA turbulence is also unclear. We aimed to examine time trends in TAMM and any relationship with clinical stroke and silent infarction (SI) on MRI. We also wanted to determine (a) the degree of MRA turbulence associated with TCD V>200 cm/sec (b) the TCD signature of MRA occlusion and (c) the proportion of patients with no TCD signal who did not have MRA occlusion. TCD was performed in a cohort of 169 patients with highest TAMM measured in either internal carotid/middle cerebral artery recorded (time point 1). Studies were categorised as follows: 0=normal TAMM, 1=170–200cm/sec (conditional TAMM), 2=>200cm/sec, 3=<50cm/sec. The median age was 8 (range 1–25) years. Six (4%) patients had an abnormal scan with TAMM>200cm/sec; 3 had not been screened for SCD neonatally. 11 (7%) patients had abnormally low TAMM <50cm/sec. In the neonatally screened cohort, the median for the highest TAMM ever recorded was 124 (range 33–258 cm/s) at a median age of 9 (1–25 years). 91 patients had repeat TCD at time point 2 (median 5 years later). 63 patients remained normal throughout and 16 with initially normal TAMM changed to TAMM<50cm/s at time point 2. 3/5 patients with TAMM>200cm/s at time point 1 had TAMM<50cm/s at time point 2 and 3/5 patients with TAMM<50cm/s at time point 1 remained low at time point 2. Patients with abnormally high or low TAMM at time point 1 were more likely to have abnormally low TAMM at time point 2 (x², p= 0.001). Although clinical stroke (x², p=0.001)and SI were commoner in those with abnormal TCD, the latter relationship was not significant (x², p=0.94). 92 patients had both TCD and MRA within 1 month. The results were analysed for symptomatic (n=63) and asymptomatic (n=29) groups. There was no difference between groups for age (median 13; range 0–27 years). MRAs were examined for evidence of MCA turbulence graded as none, mild, moderate, severe or occlusion. There was a significant association between TAMM and MRA turbulence for the whole (ANOVA, right p=0.0001, left p=0.001), and symptomatic groups (ANOVA, right p=0.01, left p=0.025). All patients with TAMM>200 cm/sec had severe MCA turbulence but 2/3 with TAMM>170<200 cm/sec had normal MRA. The highest TAMM seen in a patient with occlusion was 80 cm/sec. Eighteen patients had no TCD signal, of whom 11 had abnormal MRA (3 occlusion). There is progressive CVD in children with SCD over time as evidenced by an increase in the number of children with abnormally low TCD. Patients with low TAMM or no signal may require MRA to document the severity of their CVD. Patients with TAMM>80 cm/sec are unlikely to have occlusion. As silent infarction is not predicted by TCD, MRA has a role as an additional tool to identify high risk patients, particularly in those with low TAMM. TAMM>200cm/sec appears to be a screen for severe turbulence, although this condition may represent specific pathophysiology rather than part of a spectrum.