The hemodynamic effects of compliance, bulging, and curvature in a saphenous vein coronary artery bypass graft model.

The development of Intimal Hyperplasia (IH) in saphenous vein coronary artery bypass grafts (SV-CABG) is responsible for the short-term patency of these grafts. Previous studies of SV-CABG models were performed on rigid anastomotic vessels. However, the effects of compliance, bulging and curvature at the anastomosis on the general hemodynamic field, due to compliance and geometric mismatch between the vein and the artery have not been evaluated. We studied axial and transverse velocities by Laser Doppler Velocimetry on a compliant, in vitro, anatomical model of an end-to-side saphenous vein graft (SVG) to left anterior descending (LAD). The model incorporated a bulge at the sinus and curvature at the graft-host junction. Physiologic pressure and flow conditions pertaining to SV-CABG were applied. The presence of the bulge and curvature showed differences in the velocity profiles in comparison with previous rigid model studies. Dynamic separation zones were temporally augmented at the flow divider. The moving stagnation point at the floor of the host vessel was observed to move past the toe of the model during the accelerating portion of the cycle. These findings suggest that the presence of the bulge curvature and compliance may further favor conditions for the development of intimal hyperplasia (IH) at the floor of a CABG.

Extension of rapid phase-contrast magnetic resonance imaging using BRISK in multidirectional flow.

We developed BRISK-CON-VPS, a rapid phase-contrast cine approach that is a hybrid of the BRISK-VPS (Block Regional Interpolation Scheme for k-space) and conventional (CONV-VPS) scanning employing k-space views per segment (VPS). BRISK-CON-VPS allows data acquisition approximately four times faster than CONV-VPS imaging and has the advantage compared to BRISK-VPS that it can potentially be incorporated into real-time applications. In BRISK-CON-VPS contiguous regions of k-space are sampled using a views per segment factor that is varied as a function of distance from the k-space center. Computational fluid dynamics (CFD) data were used to simulate CONV-VPS, BRISK-VPS, and BRISK-CON-VPS. BRISK-CON-VPS was simulated by incrementing the VPS progressively with increasing distance from the k-space origin while BRISK-VPS was simulated using a uniform VPS applied to the sparse sampling scheme. Simulations showed that up to a base VPS of 5, both BRISK-CON-VPS and BRISK-VPS retained excellent axial-velocity accuracy. Secondary in-plane velocity flow fields were well represented with BRISK-CON-VPS and BRISK-VPS up to a base VPS of 3. CONV-VPS, BRISK-CON-VPS, and BRISK-VPS were applied in vivo and shown to provide comparable quantitative flow data. BRISK-CON-VPS accomplishes breath-hold acquisitions as efficiently as BRISK-VPS, but without requiring data interpolation or under-sampling k-space.