An Initial Assessment of Clinical Quantitative Digital Subtraction Angiography (qDSA) Data
Abstract
Purpose
Quantitative Digital Subtraction Angiography (qDSA) is a method for determining blood velocity in which iodinated contrast agent is injected into pulsatile arterial blood flow and oscillations in contrast are tracked as they propagate downstream. qDSA aims to quantify embolization endpoints, complementing subjective visual flow assessments. This study reports an initial experience acquiring qDSA data in human subjects.
Methods
Three patients undergoing transarterial embolization (TAE) or transarterial chemoembolization (TACE) procedures were included in an IRB-approved study. Each procedure was split into three interventional imaging stages: pre-embolization, mid-embolization, and post-embolization. Mid-embolization was estimated by the interventional radiologist as the point of 50% flow reduction. At each stage, there were two x-ray acquisitions with contrast enhancement (iohexol 300 mgI/mL): standard 7.5 fps DSA, and 30 fps fluoroscopy for qDSA. Doppler ultrasound velocity measurements were performed at each stage. Time Attenuation Maps (TAMs) were generated from qDSA sequences with manually annotated vessel centerlines. A TAM displays qDSA image intensity sampled along the centerline (y-axis) over time (x-axis) and constitutes the primary dataset for velocity estimation.
Results
Contrast injection techniques for qDSA were adjusted downwards in rate and volume relative to typical DSA to maintain visible contrast oscillations, and adjusted by patient, arterial branch, and state of embolization (0.5-5 mL/s, 2-6 s). Small detector modes resulted in less noisy images which were more suitable for qDSA analysis. Care must be taken in C-arm positioning to avoid vessel overlap. Respiratory management allowed for more accurate centerline extraction. TAM artifacts included broad spatial and temporal changes in intensity due to contrast stagnation and localized contrast due to vessel overlap.
Conclusion
Proper C-arm positioning, small detector modes, respiratory management, and optimized contrast delivery allowed for data acquisition suitable to qDSA analysis. Future studies will analyze the accuracy of qDSA velocity estimations versus doppler ultrasound measurements in humans.