Chronic red blood cell (RBC) transfusion therapy can prevent new and progressive stroke in Sickle Cell Disease (SCD), as shown by multiple clinical trials in SCD including the STOP 1&2, SIT, and SWITCH trials. Middle cerebral artery time-averaged mean velocity (MCA velocity), measured by Trans-cranial Doppler Ultrasound (TCD), is a clinical biomarker utilized to initiate chronic RBC transfusion to reduce stroke risk in children with SCD. However, MCA velocity measured by TCD lacks sensitivity and specificity in predicting stroke. Most SCD patients with high MCA velocity do not experience overt stroke, while 10-20% of SCD children will have new strokes on chronic RBC transfusion in spite of normalized MCA velocity.

Prior studies showed an association with high MCA velocity and MCA narrowing when TCD and angiography were performed after RBC transfusion to target a hematocrit (HCT) > 30%. HCT adjustments addressed effects of anemia and higher cardiac output (Adams et al, 1992; Stroke PMID 1636180). Nevertheless, HCT adjustments may not always be used for routine TCD. Thus, despite the substantial decrease in strokes with TCD screening, there may be significant lack of specificity that leads to unnecessary RBC transfusion resulting from TCD screening in non-optimal conditions. Due to these concerns, we undertook a detailed study assessing MCA velocity by TCD immediately pre- and post-transfusion in SCD patients.


The study was approved by the Children's Hospital Los Angeles Institutional Review Board. SCD and thalassemia patients who receive chronic transfusion had pre- and post-transfusion blood laboratory tests and MCA velocity measurements by TCD (GE Vivid I, FE Medical Systems Ultrasound Israel Ltd., Israel). The TCD ultrasound probe was placed on left and right trans-temporal window, landmarks were visualized, and MCA velocity data were collected in 1.7mm depth increments from 65mm to 35mm depth. Magnetic resonance angiography was not done at the time of these measurements. Data was recorded and analyzed in JMP Pro Version 13.0.


SCD patients (n=24) had a median and interquartile range of 15 years & 14-18 years, with 18 females & 6 males, received 90 transfusions, and 70% of the patients had 3 transfusion events with repeat studies. Seventeen patients received simple RBC transfusion and 7 received red cell exchange. The thalassemia control population received simple transfusion (n=11; 48 transfusion events).

Figure 1 shows the relation of MCA velocity compared to HCT in all subjects. There is a highly significant negative correlation showing lower HCT is associated with higher MCA velocity (Dots = TAMV measurements; lines = average TAMV measurements at different TCD depths; R2=0.12, F(1,4270)=556, p<.001). Figure 2 & 3 show that the MCA velocity measurement depends on depth of the TCD measurement. This accounts for a significant part of the variance evident in Figure 1. On average, a rise in HCT from a typical level in SCD of 21% to 31% would result in a 14% drop in TCD velocity. To illustrate the variable impact of acute PRBC transfusion among subjects, we show the MCA velocity response to RBC transfusion in two SCD patients (Figure 2&3: blue ♦=Pre-transfusion, red ♦=Post-transfusion). Patient 1 showed a 37% drop in MCA velocity following RBC transfusion on three occasions (∆HCT=7.5±1.7%; n=3; Figure 2) whereas Patient 2 showed only an 11% drop in MCA velocity following RBC transfusion (∆HCT=6.2±1.3%, n=5, Figure 3).


These data show a striking relationship between anemia and TCD velocity, and that RBC transfusion can alter MCA velocity. Even an 11% increase in hemoglobin such as that seen in the low-responder in Figure 3 could change a patient classification from high-risk to conditional TCD. These data also show the significant sensitivity of MCA velocity to the depth of measurement (Figure 2 & 3). The investigators in the STOP study went to extreme ends to standardize these measurements. Nonetheless, TCD measures may not be done with such rigor in clinical settings. It is clear that standard TCD screening has been of tremendous benefit to SCD patients, and many enjoy a good quality of life as a result. Our data suggest that the specificity of TCD screening could be improved if testing was done at an HCT> 30%. Quantifiable blood, patient, and measurement factors can impact TCD MCA velocity and clinical decisions that commit children to years of RBC transfusions.


Coates:ApoPharma: Consultancy, Honoraria; Celgene Corp.: Consultancy; Vifor Pharma: Consultancy; Sangamo: Consultancy, Honoraria.

Author notes


Asterisk with author names denotes non-ASH members.