Changes in Resting and Exercise Hemodynamics Early After Heart Transplantation: A Simulation Perspective

Abstract

Introduction: During heart transplantation (HTx), cardiac denervation is inevitable, thus typically resulting in chronic resting tachycardia and chronotropic incompetence with possible consequences in patient quality of life and clinical outcomes. To this date, knowledge of hemodynamic changes early after HTx is still incomplete. This study aims at providing a model-based description of the complex hemodynamic changes at rest and during exercise in HTx recipients (HTxRs). Materials and Methods: A numerical model of early HTxRs is developed that integrates intrinsic and autonomic heart rate (HR) control into a lumped-parameter cardiovascular system model. Intrinsic HR control is realized by a single-cell sinoatrial (SA) node model. Autonomic HR control is governed by aortic baroreflex and pulmonary stretch reflex and modulates SA node activity through neurotransmitter release. The model is tuned based on published clinical data of 15 studies. Simulations of rest and exercise are performed to study hemodynamic changes associated with HTxRs. Results: Simulations of HTxRs at rest predict a substantially increased HR [93.8 vs. 69.5 beats/min (bpm)] due to vagal denervation while maintaining normal cardiac output (CO) (5.2 vs. 5.6 L/min) through a reduction in stroke volume (SV) (55.4 vs. 82 mL). Simulations of exercise predict markedly reduced peak CO (13 vs. 19.8 L/min) primarily resulting from diminished peak HRs (133.9 vs. 169 bpm) and reduced ventricular contractility. Yet, the model results show that HTxRs can maintain normal CO for low- to medium-intensity exercise by increased SV augmentation through the Frank-Starling mechanism. Conclusion: Relevant hemodynamic changes occur after HTx. Simulations suggest that (1) increased resting HRs solely result from the absence of vagal tone; (2) chronotropic incompetence is the main limiting factor of exercise capacity whereby peripheral factors play a secondary role; and (3) despite the diminished exercise capacity, HTxRs can compensate chronotropic incompetence by a preload-mediated increase in SV augmentation and thus maintain normal CO in low- to medium-intensity exercise.

Keywords: cardiac denervation; computer simulation; exercise response; heart transplantation; hemodynamics; numerical model.

FIGURE 1

Schematic overview of the integrated numerical model, comprising (1) the cardiovascular system and its principal compartments; (2) the single-cell SA node model; and (3) the autonomic control governed by arterial baroreflex and pulmonary stretch reflex integrating receptor information on instantaneous aortic pressure (P_ao) and instantaneous lung volume (V_L). Based on autonomic activity, the SA node depolarization rate is modulated through changes in acetylcholine (ACh) and isoprenaline (Iso) concentrations. Autonomic control has also influence on the left and right ventricular contractility (E_lv, E_rv), systemic vascular resistance (R_as), and systemic venous unstressed volume (V_usv). Muscle mechanoreflex and metaboreflex modulate Ras to account for vasodilation in response to exercise. Exercise intensity determines the level of central command, respiration frequency, and muscle mechanoreflex and metaboreflex activity. Ultimately, the central command has a direct influence on autonomic control loops and the adrenal medulla releasing catecholamines (Iso_circ).

FIGURE 2

Summary of the screening process applied to identify eligible publications for data pooling. In the initial screening phase, publications were excluded if the title did reveal that the study includes hemodynamic variables neither for rest nor exercise and thus did not match the expected context. Finally, 15 publications (McLaughlin et al., 1978; Crisafulli et al., 1985; Kavanagh et al., 1988; Labovitz et al., 1989; Wilson et al., 1991; Marzo et al., 1992; Rudas et al., 1993; Kao et al., 1994; Doering et al., 1996; Geny et al., 1996; Notarius et al., 1998; Hayman et al., 2010; Peled et al., 2017; Nygaard et al., 2019; Nytrøen et al., 2019) could be identified as eligible for data pooling.

FIGURE 3

Comparison of selected hemodynamic variables derived from simulations to pooled published data (McLaughlin et al., 1978; Crisafulli et al., 1985; Kavanagh et al., 1988; Labovitz et al., 1989; Wilson et al., 1991; Marzo et al., 1992; Rudas et al., 1993; Kao et al., 1994; Doering et al., 1996; Geny et al., 1996; Notarius et al., 1998; Hayman et al., 2010; Peled et al., 2017; Nygaard et al., 2019; Nytrøen et al., 2019), for HTxRs and the healthy control group, both at rest and during exercise. HR, heart rate (A); CO, cardiac output (B); MAP, mean arterial pressure (C); SVR, systemic vascular resistance (D).

FIGURE 4

Comparison of exercise response between HTxRs and healthy controls, characterized by selected hemodynamic variables. The horizontal axis shows the relative exercise intensity with respect to the peak exercise intensity of the healthy control group (1,032 kpm/min). Note the substantial reduction of exercise capacity in HTxRs, reaching only about 50% (522 kpm/min) of the healthy control’s peak exercise intensity. We can see a marked impairment of chronotropic response and reduced peak HR in HTxRs (A). Furthermore, the simulation results show no significant differences in augmentation in CO but substantially decreased peak CO in HTxRs (B). Resting SV is notably reduced in HTxRs and undergoes a rapid increase in response to exercise (C). MAP response is similar in both groups; however, healthy controls reach higher peak values due to greater exercise intensity (D). Despite the reduced absolute values, HTxRs show a markedly stronger augmentation of LVEDVI in response to exercise (E). Simulations predict a similar resting LVEDP for both the HTxRs, however, the augmentation of LVEDP is markedly stronger than in age- and gender-matched healthy individuals (F). HR, heart rate; CO, cardiac output; SV, stroke volume; MAP, mean arterial pressure; LVEDVI, left ventricular end-diastolic volume index; LVEDP, left ventricular end-diastolic pressure.

FIGURE 5

Pressure–volume loops for a time period of 10 s obtained from simulations of healthy controls (A) and HTxRs (B), for rest (0%), medium (50%), and peak (100%) exercise intensity. Note that peak exercise in HTxRs corresponds to 50% of exercise intensity in healthy individuals.