Inert gas measurements of myocardial perfusion in the presence of heterogeneous flow documented by microspheres.

Inert gas measurements of flow per unit weight (F/W) incorporating even saturation of heterogeneously perfused areas and two–decade resolution of coronary venous desaturation curves have been compared with radioactive microsphere measurements of F/W in closed–chest dogs with modest heterogeneity of F/W. Coefficients of variation for microsphere measurements in 96 left ventricular segments revealed global heterogeneity of F/W, of similar degree, in dogs with and without an abdominal aortocaval fistula (0.18 ±4- 0.07 vs 0.15 ± 0.04; p > 0.3). Endocardial–epicardial flow ratios were lower in the fistula dogs (0.77 ± 0.11 vs 1.05 ⊥ 0.08; p < 0.01), reflecting transmural as well as nontransmural heterogeneity of F/W. Inert gas measurements of average F/W, derived from the Kety–Schmidt equation using dissolved hydrogen (H2) as tracer, agreed within i 20% of average microsphere F/W in 18 of 20 comparisons in fistula and nonfistula dogs. Semilogarithmically plotted H2 desaturation data were curvilinear in both settings, but arbitrarily derived ‘slow–compartment’ H2 F/W agreed with average microsphere endocardial F/W only in the fistula dogs. We conclude that 1) methodologically adequate inert gas measurements give accurate values for average F/W in the presence of moderate heterogeneity of perfusion; and 2) although the presence of heterogeneous perfusion can be appreciated from the shape of inert gas desaturation curves, compartmental analyses of curves cannot ordinarily be interpreted in a specific transmural or other spatial sense.