RESUMO
The present investigation was intended to evaluate myocardial inert gas desaturation curves for manifestations of heterogeneous coronary perfusion. The test gas was hydrogen (H(2)) and blood H(2) analyses were performed with a gas chromatograph capable of detecting small but prolonged venous-arterial H(2) differences produced by areas of reduced flow. Curves were initially obtained after 4-min left ventricular infusions of H(2)-saturated saline in six patients with arteriographically proven coronary artery disease, three patients with normal coronary arteries, and nine closed-chest dogs. The dogs were studied before and after embolic occlusion of a portion of the left coronary artery. Although the slopes of their semilogarithmically plotted venous desaturation curves varied with time before embolization, they showed more distinct deviations from single exponentials after embolization (after H(2) concentrations had fallen below 15% of their initial values). The human curves divided similarly, those from coronary artery patients deviating appreciably from single exponentials. A similar separation was also evident in studies of coronary venous-arterial H(2) differences after 20 min of breathing 2% H(2): data were obtained in four dogs before and after coronary embolization, and in three normal patients, and five patients with coronary artery disease. Additional data indicated that the findings were not the result of right atrial admixture in sampled coronary venous blood, although admixture occurred frequently when blood was sampled in the first 2 cm of the coronary sinus (as seen in the frontal projection). Finally, average coronary flows calculated from a given set of data varied significantly with different methods of calculation. Areas of below-average flow seemed likely to be overlooked when single rate constants of desaturation, relatively insensitive analytical techniques, or relatively short periods of saturation and (or) desaturation are employed.
