Biomechanical Basis of Myocardium/Vessel Interaction: Implications for Pathophysiology and Therapy

Abstract

Ischemic heart disease is a major cause of morbidity and mortality worldwide. Interestingly, the onset of ischemia is transmurally heterogeneous, where the deeper (subendocardial) layers are more vulnerable to ischemia than the more superficial (subepicardial) ones (Hoffman 1987). This observation is especially puzzling in light of the opposite manifestation of coronary artery disease, which exclusively affects the epicardial coronary arteries, whereas intramural arteries are athero-protected (Geiringer 1951). Initiation of both atherosclerosis and ischemia depend highly on flow conditions; therefore, investigation of the hemodynamic determinants of both pathologies requires comprehension of the local coronary flow conditions, which are measured in the beating heart. Computer simulation is an attractive approach to study local coronary flow conditions. For hemodynamic simulation to be realistic, however, it must incorporate both a realistic description of the coronary network and the manner by which the contracting myocardium affects coronary flow—the myocardium/vessel interaction (MVI). Such an approach has several inherent challenges: First, the vast number of coronary blood vessels (Kaimovitz et al. 2005) is associated with an extensive computational cost to solve the network dynamic flow. To circumvent this difficulty, previous flow models (Bruinsma et al. 1988; Cornelissen et al. 2000; Flynn et al. 1992; Klocke et al. 1985; Manor et al. 1994) used lumped representations for the coronary vasculature. Although this approach is useful to reveal basic flow characteristics, it cannot address the physical relation between structure, vessel mechanics, and blood flow. Moreover, validation of a lumped model with experimental data is limited due to the inability of the model to describe flow conditions in specific vessels. Asecond challenge stems from paucity of experimental data required for both the flow model representation and for validation. Finally, the physical origins of the MVI, a key determinant in coronary flow analysis, are under a long-standing dispute and hitherto unknown. In fact, none of the mechanisms previously proposed (Downey and Kirk 1975; Krams et al. 1989a; Rabbany et al. 1989; Spaan et al. 1981; Zinemanas et al. 1994) to describe this mechanical interaction predict all of the characteristics of coronary flow (Westerhof et al. 2006), i.e., the blood flow velocities, pressures, and vascular diameters that correspond with the measured data.