(F-210) Computer prediction of coronary capillary transit time

Introduction:: Capillary transit time (CTT) is a crucial determinant of the exchange of oxygen, nutrients, and waste between blood and tissues in all organs. Despite advances in experimental techniques, it remains difficult to measure CTT in vivo under different physiological or pathological conditions. In the heart, blood vessels are embedded in the beating myocardium, which makes it even more challenging to measure CTT. Computational modeling can be used to estimate CTT and understand how this quantity is affected in different diseases and treatments. Here, we developed a novel computational framework that couples the coronary arterial-capillary-venous microcirculation with cardiac mechanics in a closed-loop system that allows us to isolate and investigate individual factors affecting CTT.


Materials and Methods:: The computational framework consists of the coronary arterial, venous and capillary networks (with cross-connections) and the Thebesian vessels (Figure 1). The entire network consists of 3,711 vessels and each vessel is represented by a nonlinear lumped model consisting of two vascular resistances and a compliance. Geometry (radius and length) and mechanical properties of the vessels are prescribed based on measurements. Resistance and capacitance for each vessel of order n are also scaled accordingly based on the branching ratios measured in pigs. Contraction of the left ventricle and left atrium in the systemic circulation are each described using a time-varying elastance model.

To estimate CTT in the computational framework, a tracer is “virtually” released at different time points in the cardiac cycle and at different post-arteriole capillary entrances. The time taken for each tracer to transit from an arteriole to a venule, which directly reflects the transit time of red blood cells (RBCs) through the capillary network, is recorded. The CTT is then computed by averaging the transit time of all the tracers.

The computational framework was first validated against experimental measurements and then applied to simulate different myocardial ischemia cases. Specifically, we simulate different degrees of occlusion in the order 10 artery (as indicated by the black cross symbol in Figure 1) by varying the artery’s resistance. Two degrees of occlusion indexed by a fractional flow reserve (FFR) of 0.6 (moderate occlusion) and 0.1 (severe occlusion) were considered.


Results, Conclusions, and Discussions:: Capillary transit time, which is the time taken for blood to transit the entire capillary network, plays a key role in tissue oxygenation. This quantity, however, is usually estimated simplistically by the ratio of total capillary blood volume to total flow rate. In the present study, we estimate the CTT by directly tracking “virtual tracers” as they traverse through the entire capillary and coronary network in a computational modeling framework that describe blood flow in the coronary arterial-capillary-venous network. The computational framework also overcomes limitations found in previous models, which do not consider the coronary venous and capillary networks and their anatomical structure (e.g., branching patterns and Thebesian vessels). The model predictions compare well with measurements of transit time as well as blood pressure and flow rate distribution in the coronary arterial network (Figure 2).



We show that, with increasing coronary artery stenosis (as quantified by fractional flow reserve, FFR), intravascular pressure and flow rate downstream are reduced but remain non-stationary even at 100% stenosis because some flow (~3%) is redistributed from the non-occluded to the occluded territories. Importantly, the model predicts that occlusion of a large artery results in higher CTT. For moderate stenosis (FFR > 0.6), the increase in CTT (from 1.21s without stenosis to 2.23s at FFR=0.6) is caused by a decrease in capillary flow rate. In severe stenosis (FFR = 0.1), the increase in CTT to 14.2s is due to both a decrease in flow rate and an increase in path length taken by “virtual tracers” in the capillary network (Figure 3).

This framework can be further extended to account for myocardial contractility and oxygen transport, and eventually to facilitate optimizations of coronary venous retroperfusion treatments.


Acknowledgements (Optional): : NIH R01 HL160997


References (Optional): : Chatzantonis, G., Bietenbeck, M., Florian, A. et al. “Myocardial transit-time” (MyoTT): a novel and easy-to-perform CMR parameter to assess microvascular disease. Clin Res Cardiol 109, 488–497 (2020).

Hudetz AG, Fehtr G, Knuese DE, Kampine JP. Erythrocyte flow heterogeneity in the cerebrocortical capillary network. Adu Exp Med Biol 345, 633–643 (1994).