Objectives: Myocardial microvascular permeability-surface area product (MPSP) and fractional vascular volume (FVV), indices of endothelial function and microvascular perfusion, can be noninvasively evaluated by electron beam computed tomography (EBCT), but it remains unknown whether comparable assessments can be obtained with 64-slice multidetector CT (CT-64).

Methods: We studied 12 pigs with both EBCT and CT-64 in randomized order 1 week apart, before and during IV adenosine infusion. Myocardial attenuation changes in the cardiac wall were assessed after a central-venous injection of iopamidol. Time-attenuation curves were analyzed using both indicator-dilution and Patlak models to calculate MPSP and FVV.

Results: CT-64 and EBCT assessments of basal MPSP obtained by the Patlak method were similar (0.37 ± 0.03 vs. 0.37 ± 0.04 mL/min/g), as was its response to adenosine, and correlated significantly (r = 0.87). Patlak FVV was also similar between CT-64 and EBCT at baseline (0.08 ± 0.02 vs. 0.07 ± 0.02 mL blood/mL) and during adenosine, and correlated well (r = 0.93). MPSP and FVV estimated by the indicator-dilution method were not significantly correlated.

Conclusions: CT-64 assessments of myocardial MPSP and FVV may not be reliable when using indicator-dilution analysis, likely due to its sensitivity to scan duration. However, CT-64 assessments obtained using the Patlak model are feasible.