Assessment of coronary stenoses with myocardial perfusion imaging during pharmacologic coronary vasodilatation. IV. Limits of detection of stenosis with idealized experimental cross-sectional myocardial imaging.

Because atherosclerosis may be reversible, a routine noninvasive screening test for the reliable diagnosis of mild coronary arterial lesions would allow potential prevention of coronary events in specific patients through intensive dietary management, drug therapy and physical training. To determine the minimal coronary stenosis detectable with myocardial perfusion imaging techniques, standardized stenoses ranging from 31.4 to 72.5 percent diameter narrowing were applied to the left circumflex coronary artery of 12 open chest dogs. Indium-113m-labeled human albumin microspheres were injected into the left atrium under control conditions and technetium-99m human albumin microspheres during maximal coronary vasodilatation induced with intravenous dipyridamole. Hearts were removed, sliced into 1 cm thick cross sections and imaged under a gamma camera. The results demonstrate that 40 percent diameter coronary stenoses can be identified by imaging relative subendocardial underperfusion during pharmacologic coronary vasodilatation. An imaging technique sensitive enough to identify mild coronary lesions for diagnostic screening purposes requires (1) a potent stimulus for coronary vasodilatation, such as intravenous dipyridamole; (2) an imaging agent taken up by the myocardium in proportion to coronary flow at flow rates up to four or more times resting coronary flow so that differences in regional maximal flows caused by mild stenoses can be identified; and (3) cross-sectional tomographic myocardial imaging to visualize relative endocardial-epicardial perfusion, the most sensitive indicator of the hemodynamic effects of coronary stenoses, and to exclude from the imaging field the interfering activity of lung, background and overlying heart structures.