In this study, we discovered that heart rate fluctuations possess robust scale-invariant patterns from minutes up to 24 h, and that lesioning the SCN leads to a breakdown of the scale-invariant patterns over a specific range of time scales (from ~3.6 to 24 h), with very different effects at <~3.6 h. These findings provide clear evidence that the SCN is one of the key intrinsic factors contributing to scale-invariant patterns in heart rate fluctuations. Moreover, because the SCN has such a major influence over a very broad range of time scales (i.e., not only at a time scale close to 24 h), the SCN must be one of the principal nodes of the network controlling heart rate fluctuations.
The preservation of a scale-invariant pattern in heart rate fluctuations in SCNx rats at time scales <3.6 h indicates that another neurophysiological source other than the SCN must be responsible for much of the scale-invariant patterns of heart rate over this shorter time range. The observation of similar correlations across the time scale from minutes to 3.6 h and across the time scale from 3.6 to 24 h in intact animals also implies that the SCN and the non-SCN mechanisms of cardiac regulation are coupled. Thus, our results strongly support the hypothesis of the existence of coupled intrinsic cardiac control nodes operating at different time scales (). The precise anatomical sites of the other components of such a multiscale regulator are unknown, and the nature of the interactions between the SCN and these components remains to be elucidated. It is also possible that the SCN itself contains interacting nodes that together are responsible for the scale invariance of heart rate fluctuations in the range of time scales from 3.6 to 24 h.
Figure 5 Schematic diagram of the contributions of the suprachiasmatic nucleus (SCN) to the complex statistical features in heart rate fluctuations. Control rats exhibit virtually identical scale-invariant patterns of heart rate fluctuations across a time scale (more ...)
Establishing a physiologically meaningful model and understanding the origin of scale-invariant behavior in cardiac dynamics poses stimulating challenges for physiology, physics, and mathematics. The first crucial step is to map the anatomical architecture of a neurophysiological network responsible for scale invariance. Prior to this study no key elements of the cardiac control network contributing to scale-invariant behavior have been identified. We report a neural site (SCN) that is entirely responsible for scale-invariant cardiac regulation over a range of specific time scales (>~4 h). This finding is surprising because the SCN, serving as the endogenous circadian pacemaker, has been thought to function mainly at a specific time scale, that is, generating and coordinating rhythms close to 24 h in many physiological systems. It is still unclear how the same neural site can generate a relatively stable rhythm at a fixed time scale (~24 h) and simultaneously display scale-invariant fluctuations over a wide range of time scales (~3.6–24 h). Generally, such coexistence of rhythms and scale-invariant patterns in a system indicates nonlinear feedback coupling among rhythms and fluctuations at different time scales. However, specific mathematical models are needed to explain the complex functions of the SCN at multiple time scales.
Separate Networks Responsible for Cardiac and Activity Scale Invariance
In our previous studies, we found similar scale-invariant patterns of activity fluctuations in rats and humans, and that such patterns were independent of mean activity levels and environmental influences (Hu et al., 2004a
). Those findings indicated that, similar to heart rate fluctuations, there exists a complex control network influencing locomotor activity. It seems possible that the same network is responsible for scale invariance in the two physiological outputs such that the observed cardiac scale invariance is secondary to feedback control of locomotion. On the other hand, although mean activity can obviously influence mean heart rate, it is not necessarily the case that the fluctuations around the mean levels in two physiological signals are related—and it is the fluctuations that are being assessed in the “detrended” fluctuation analysis (because changes in the mean levels are subtracted). The ideal way to test whether cardiac scale invariance is independent of locomotion feedback is to control activity. This cannot easily be done in rats, as immobilization could lead to stress-related effects on heart rate. But it is relatively easy to achieve in humans. Indeed, we recently discovered a circadian rhythm of cardiac scale invariance in humans and that this rhythm persists throughout 38 h of voluntary inactivity (Ivanov et al., 2007
). These findings indicate that activity and cardiac scale invariance in humans can be uncoupled. Moreover, we showed in this study that, as with humans, in rats (1) the cardiac scale invariance is essentially unchanged by changes in mean activity level (), and (2) activity scale invariance does not predict scale invariance in cardiac dynamics (). Therefore, we conclude that there are likely to be separate feedback networks responsible for cardiac and activity scale invariance.
Our results clearly indicate that the SCN is a major node in both control networks of heart rate and activity fluctuations. However, by ablating the entire SCN and looking only at the downstream variables, we cannot yet distinguish different effects from intervening parts of the control pathways. For example, the SCN could influence heart rate in a unique way (e.g., the locomotory and cardiac influences from the SCN may emerge from distinctly different parts of the SCN and have different patterns), or the SCN could influence many variables in the same scale-invariant manner. Even if the latter occurs, the feedback interactions for the different variables are likely to result in unique downstream patterns. Indeed, there are numerous possible sites where SCN’s influences and integration of information and pattern can occur, as there are multisynaptic pathways for control of both locomotion and heart rate (Scheer et al., 2001
). Because autonomic impairment or blockade significantly alters cardiac scale invariance (Penttila et al., 2003
; Beckers et al., 2006
; Merati et al., 2006
; Aoyagi et al., 2007
), the SCN-related autonomic regulation may potentially provide an explanation of the SCN’s contribution to cardiac scale invariance.
At the current stage, we still cannot identify neuronal pathways that are responsible for the SCN influence on cardiac and activity scale invariance. Nevertheless, our studies indicate clearly that (1) scale invariance is a universal characteristic of physiological fluctuations, and (2) the SCN is a major node in both scale-invariant control networks of activity and cardiac dynamics, imparting both activity and heart rate fluctuations over a wide range of time scales, especially at large time scales (>~4 h). To confirm that the SCN is a major node in the network responsible for scale-invariant control of physiology, future studies could test whether a similar scale invariant pattern exists in temperature regulation.
Impact of Scale Invariant Cardiac Control on Physiology
The finding of scale-invariant heart rate fluctuations challenges the classical principle of homeostasis, which postulates that physiological systems return to equilibrium after perturbation and that linear causality controls the pathways of physiological interaction. The cardiac scale invariance has been linked to situations found in physical dynamic systems far away from equilibrium. Such systems are comprised of a network of nonlinear feedback interactions and never settle down to constant output, but rather exhibit complex fluctuations (Stanley, 1971
; Kurths et al., 1995
; Shlesinger and West, 1998
). Models of these physical systems imply that cardiac scale invariance may derive from a network of controlling factors operating at different time scales with feedback interactions, which lead to an overall organization of fluctuations (or rhythms) in heart rate signals at all time scales. For instance, factors that influence heart rate at varied time scales can include temperature changes, activity related influences, endocrine and autonomic influences, and influences internal to the heart itself. Alteration in cardiac scale invariance is associated with cardiovascular disease and predicts reduced survival (Bigger et al., 1996
; Perkiomaki et al., 2001
; Makikallio et al., 2004
). Thus, analogous to some known physical systems, loss of the scale invariance of cardiac control may indicate a simpler physiological control system that is less adaptive to perturbations and is more vulnerable to disturbing events.
In this study we showed that the SCN is a major node in the network of cardiac control, and that the SCN lesion imparts temporal organization of heart rate fluctuations over a wide range of time scales. However, as noted above, our experiment was not designed to determine whether the SCN influence on cardiac scale invariance derives from neuronal interactions within the SCN, or the influence is generated by feedback interactions between the SCN and other neural sites influencing heart rate. Regardless of the underlying mechanisms, our results reveal previously unrecognized SCN contribution to scale-invariant behavior in cardiac dynamics, which can perhaps be used for the assessment of SCN-related pathological alterations. To fully understand scale-invariant cardiac control and to ultimately build a physiological network model, future studies could be directed at determining the anatomical sites of other components of such a multiscale cardiac regulator, and elucidating the nature of the interactions between the SCN and these components.