Point:Counterpoint: Cardiac denervation does/does not play a major role in exercise limitation after heart transplantation POINT: CARDIAC DENERVATION DOES PLAY A MAJOR ROLE IN EXERCISE LIMITATION AFTER HEART TRANSPLANTATION

With more than 100,000 procedures performed worldwide, more than 50% of patients undergoing orthotopic heart transplantation (HTx) survive longer than 10 years and enjoy significant and lasting improvements in their quality of life. However, exercise capacity by objective exercise testing has been shown to be markedly reduced. Despite normal ventricular ejection fraction at rest and a peak oxygen consumption that is higher than compared with preoperative values, exercise performance is typically only 40–60% of age-, sex-, and weight-matched controls. This phenomenon is best explained by the consequences of cardiac denervation at the time of explantation of a donor heart with subsequent HTx. In nontransplanted subjects, the capacity for performing aerobic exercise depends on the ability of the heart to augment its output to exercising muscles and the ability of these muscles to use oxygen from the delivered blood. The increase in cardiac output (typically 4to 6-fold) is caused by tachycardia operating in concert with increased myocardial contractility (20– 50% augmentation of stroke volume) brought about by the greater sympathetic activity, together with the Frank-Starling mechanism (1). The autonomic nervous system plays a key role in the regulation of heart rate. The initial rise in heart rate is due to withdrawal of vagal tone, resulting in an increase of 30–50 beats/min. Thereafter, increments are attributed to an increase in activity to cardiac sympathetic nerves (20). HTx recipients modulate their cardiac output in response to exercise in a different manner than in the normal heart, with no direct parasympathetic or sympathetic innervation. Due to lack of efferent vagal control, resting heart rate is elevated and with limited ability to increase heart rate early in exercise. Thus initial increases in cardiac output are completely dependent on augmented stroke volume (18). Increases in heart rate during later phases of exercise rely on an increase in circulating catecholamines because of sympathetic denervation (18). Importantly, the expected maximal inotropic left ventricular stimulation may also suffer from denervation due to at least two mechanisms. First, intact sympathetic nerve terminals with local norepinephrine release are required for maximum ventricular inotropic stimulation (20). Second, according to the frequency-force relationship, the time interval between beats is a determinant of the force of myocardial contraction. While the background for this last phenomenon is uncertain, it has been linked with calcium availability to contractile elements (4). Several reports have demonstrated an association between subnormal exercise capacity and chronotropic incompetence in the form of reductions in the rate of rise in heart rate, increase in heart rate, and peak exercise heart rate (2, 5, 8–10, 14, 15, 19, 21, 22, 26). The largest studies have also shown that heart rate is a powerful and independent predictor of exercise capacity (8, 9, 14). Assessing maximal symptom-limited graded upright bicycle testing with simultaneous invasive hemodynamic monitoring, Kao et al. (10) described the consequences of a 30% lower heart rate response and 79% lower heart rate reserve in 30 HTx recipients compared with healthy controls. The lower stroke and cardiac index throughout exercise, coupled with inability to compensate with the Starling mechanism, resulted in 43% lower peak weight-adjusted oxygen consumption than in controls. The authors conclude that the chronotropic incompetence “is no doubt a result of the effects of denervation” and a major limiting factor to exercise capacity. The lack of direct innervation of the sinoatrial node as the cause for the attenuated response to peak heart rate to exercise in HTx is strongly supported by the observations that the rise in circulating catecholamines is normal or increased at peak exercise, and the responsiveness of the sinoatrial node to beta-adrenergic stimulation is also normal or increased (7, 23). Serial assessment of exercise capacity in HTx recipients demonstrates some improvement in heart rate and peak V̇O2 during the first postoperative year (9). Nevertheless, they remain less fit at 1 year also compared with patients undergoing coronary artery bypass graft surgery (23). Interestingly, when extending comparisons to renal transplant recipients without denervated hearts, a near-normal exercise capacity is reached within a few months after surgery (6). The latter observation shows that contrary to speculations, the use of standard immunosuppressive medication does not necessarily relate to below-normal exercise capacity. Furthermore, while regular exercise and rehabilitation can improve peak V̇O2 and workload in various patient groups, improvements in heart rate reserve and peak heart rate in HTx recipients are either modest or negligible. In the mentioned comparison with bypass-treated patients, heart rate reserve was unchanged among HTx recipients while it increased significantly in the former group (16). In the only controlled trial of postoperative rehabilitation after HTx, Kobashigawa et al. (13) demonstrated a 2.5 ml kg 1 min 1 higher increase in peak V̇O2 after 6 mo in the exercise group than in the nonexercising group without any difference in improvement of peak heart rate at the end of the study. Thus, whereas exercise training within the first year after HTx modestly increases the capacity for physical work compared with a more sedentary daily life, this must either be due to a physiological training effect such as improved muscle-skeletal weakness or merely an ability to exercise with greater effort. Although conflicting results have been obtained, some studies describe the abnormal chronotropic response to exercise initially after HTx to return toward normal after 1–2 yr, suggesting sympathetic reinnervation (9, 21). Conclusively observed in animals (12), reinnervation in humans has also been inferred from invasive measurements of transcardiac epinephrine spillover (26), noninvasive imaging with radiolabeled catecholamine analogs (24), and by heart rate variability analyses (11), supported by observations such as typical anginal pain in those with graft vasculopathy (25) and regrowth of nerves across the aortic anastomosis determined by microscopy (17). Taken together, experimental and clinical evidence suggests that some degree of sympathetic reinnervation takes place over time at the sinus and the ventricular level, but not in all patients. J Appl Physiol 104: 559–564, 2008; doi:10.1152/japplphysiol.00694.2007.