Nitrogen-13-ammonia and oxygen-15-water estimates of absolute myocardial perfusion in left ventricular ischemic dysfunction

Abstract

Measurements of resting myocardial blood flow (MBF) in patients with chronic left ventricular ischemic dysfunction by 15O-water with 13N-ammonia and PET have yielded conflicting results. The aim of this study was to perform a head-to-head comparison of both tracers in the same patient population and to answer the question of whether distinctive tracer properties account for differences in estimates of MBF in chronically dysfunctional myocardium by both tracers.

Methods: A total of 30 patients with chronic dysfunction of the anterior myocardial wall due to significant left anterior descending coronary artery disease underwent PET measurements of absolute MBF in the anterior wall by use of 15O-water and 13N-ammonia before coronary revascularization by either coronary artery bypass graft (n = 24) or percutaneous transluminal coronary angioplasty (n = 6). Improvement of regional contractile function was assessed by two-dimensional echocardiography at a mean of 7.5 +/- 2.1 mo after revascularization. As judged from the changes in anterior myocardial wall motion after revascularization, patients were considered to have either reversibly (n = 16) or persistently (n = 14) dysfunctional myocardium. Estimates of MBF by 15O-water and 13N-ammonia, obtained in every patient before revascularization, were compared among the two patient groups by use of previously validated methods.

Results: With 13N-ammonia, resting regional MBF was significantly higher in reversibly as opposed to persistently dysfunctional segments [84 +/- 8 versus 48 +/- 6 ml (min x 100 g)(-1), mean +/- s.e.m., p < 0.01]. By contrast, no such difference was found when using 15O-water to measure MBF [74 +/- 6 versus 86 +/- 9 ml (min x 100 g)(-1), p = ns]. This was mainly due to the fact that the perfusable tissue fraction (PTF), a fitted parameter of the 15O-water model, was significantly higher in reversibly as opposed to persistently dysfunctional segments (0.63 +/- 0.03 versus 0.50 +/- 0.03, p < 0.05). As a consequence, the 15O-water perfusable tissue index (PTI), which is the ratio of the PTF to the anatomical tissue fraction, was greater in reversibly dysfunctional as opposed to persistently dysfunctional segments (1.07 +/- 0.07 versus 0.79 +/- 0.05, p < 0.01).

Conclusion: This study demonstrates significant differences in MBF estimates between 15O-water and 13N-ammonia in chronically dysfunctional ischemic myocardium. Our results indicate that the 15O-water method yields higher absolute MBF values than the 13N-ammonia approach. Our results also support the use of PTI as a marker of myocardial tissue viability.