Rebuttal from Alicia D'Souza, Sanjay Sharma and Mark R. Boyett

Abstract

We read with great interest the arguments presented by Coote & White (2015) in favour of the traditional view that exercise training decreases resting heart rate by an increase in the tonic discharge of the vagal nerves. However, we are not persuaded. Coote & White (2015) quote Maciel et al. (1985); however, Maciel et al. (1985) showed that the intrinsic heart rate of trained athletes remains lower after full vagal nerve block. Although Ordway et al. (1982) reported that there was no training-induced bradycardia in six cardiac-denervated dogs, there is evidence to the contrary: Kavanagh et al. (1988) reported that in eight highly compliant cardiac transplant patients (with no evidence of cardiac reinnervation) there was. Coote & White (2015) state that the high frequency (HF) component of heart rate variability (HRV) is a robust surrogate for vagal tone and the increase in the HF component in athletes is evidence of increased vagal tone. However, we have argued that HRV cannot be used in any simple way to measure autonomic nerve activity (Monfredi et al. 2014). Our arguments are based on those of Zaza & Lombardi (2001), who showed that the HF component (like other measures of HRV) is highly sensitive to the heart rate. For example, on training Al-Ani et al. (1996) observed a 12 beats min–1 decrease in the resting heart rate (69 decreased to 57 beats min–1) and a 1.5-fold increase in the HF component, whereas Zaza & Lombardi (2001) calculated that a 10 beats min–1 decrease (70 to 60 beats min–1) will cause a 3-fold increase simply because of the decrease in heart rate alone. Coote & White (2015) mention the sensitivity of the baroreceptor–heart rate reflex; this is a commonly used index of autonomic balance, but this too relies on HRV and is therefore flawed (Zaza & Lombardi (2001). The data from Danson & Paterson (2003) quoted by Coote & White (2015) showing a training-induced upregulation of NOS-1 in intracardiac cholinergic ganglia are fascinating; the data indicate an improvement in vagal neurotransmission but this does not prove it is responsible for the training-induced bradycardia. For example, our own data demonstrate a training-induced reduction in the sinus node of mRNA for the M2 receptor and Kir3.1, Kir3.4, HCN4, Cav1.2 and Cav1.3 channels (responsible for the ACh-sensitive currents IK,ACh, If and ICa,L) (D'Souza et al. 2014); these changes potentially reduce the sensitivity of the sinus node to vagal tone. Overall, given that direct recording of cardiac vagal activity is currently unfeasible, the available evidence tips the balance in favour of electrophysiological remodelling of the sinus node to explain the resting bradycardia in athletes. A challenge for neuroscientists is to measure vagal activity to the sinus node directly and not rely on questionable surrogates.