proper function of a variety of organ systems, such as the kidneys, the heart, and the brain, depends on appropriate perfusion pressure. For example, a reduction in cerebral perfusion pressure below the threshold of autoregulation of cerebral blood flow can cause fainting or transient ischemic attacks (18), whereas an increase in cerebral perfusion pressure above the autoregulatory range can cause hyperperfusion, cerebral edema, or even hemorrhagic stroke (4, 5). To maintain an appropriate perfusion pressure, the body is equipped with several cardiovascular control systems aimed at maintaining blood pressure at a relatively constant level (16). These cardiovascular control systems include the baroreceptor reflex (11), the renin-angiotensin-aldosterone system (7), endothelial control of vascular tone (23), myogenic vascular function (14), and others (16). To study these cardiovascular control systems, analytical techniques have been developed that are based on the analysis of specific patterns of variability introduced to arterial blood pressure and heart rate by these cardiovascular control systems (1, 3, 13, 17, 19, 20). The underlying theory is that the time course of cardiovascular responses to activation or inhibition of individual cardiovascular control systems differs. For example, it has been demonstrated that blood pressure control by the renin-angiotensin system (17) is slower than blood pressure control by the sympathetic nervous system (21). Thus, the renin-angiotensin system affects cardiovascular variability at lower frequencies than the sympathetic nervous system (1). The frequency components of blood pressure and heart rate variability that are affected by different cardiovascular control systems have been well characterized in a variety of species, including humans, dogs, rabbits, cats, and rats (16). However, little is known about the specific frequency components of blood pressure and heart rate variability that are affected by individual cardiovascular control systems in mice. This knowledge is of particular importance, as genetically engineered mouse models are becoming more frequently used to study cardiovascular diseases.
In this issue of The American Journal of Physiology-Regulatory, Integrative and Comparative Physiology, Baudrie, Laude, and Elghozi (2) publish a study aimed at identifying the frequency components of blood pressure and heart rate variability that are affected by sympathetic and parasympathetic nervous system activity in conscious, unrestrained mice (2). Few studies have addressed this issue before. For example, Just et al. (9) performed power spectral analysis of arterial blood pressure and pulse interval in conscious unrestrained mice during application of autonomic receptor blockers and found that α1-adrenergic receptor blockade by prazosin reduced blood pressure variability most prominently at 0.15–0.7 Hz, indicating that this frequency band reflects sympathetic modulation of vascular tone in mice. The results concerning parasympathetic modulation of heart rate were less clear, since muscarinic receptor blockade by atropine reduced heart rate variability at all frequencies with greater effects at frequencies below 1.0 Hz (9). The study by Baudrie et al. (2) differs from that of Just et al. (9) in that these authors used a statistical approach to investigate the effects of autonomic receptor blockers on a total number of 32,896 frequency bands. This huge number of frequency bands results from the permutations of lower and upper frequency limits that can be obtained based on a frequency resolution of 0.0195 Hz and a highest frequency of the power spectra of 5.0 Hz. For each of these 32,896 frequency bands a P value was calculated for the paired t-test comparing spectral powers during control conditions and autonomic receptor blockade. Based on these P values, the frequency bands of blood pressure and pulse interval variability for which autonomic receptor blockade yielded the “most significant” effects were identified. Using this elegant approach, Baudrie et al. (2) identified the frequency bands for sympathetic modulation of vascular tone at 0.15–0.60 Hz [almost identical to the results by Just et al. (9)] and for parasympathetic modulation of heart rate at 2.5–5.0 Hz. The results concerning sympathetic modulation of heart rate were inconclusive because mice were studied during resting conditions where sympathetic modulation of sinus node function may be less prominent.
With this regard, it is important to consider the experimental conditions under which hemodynamic recordings are obtained. For example, recovery from surgery, active vs. inactive states, or day vs. night are well known to affect cardiovascular parameters. It is reasonable to assume that the recordings in the study by Baudrie et al. (2) have been obtained after full recovery from surgical implantation of telemetry probes and during inactive conditions, because heart rate was only 400 beats/min. In addition, heart rate decreased only modestly after β1-adrenergic receptor blockade but increased substantially after muscarinic receptor blockade, indicating that sympathetic tone was low and parasympathetic tone was high under these experimental conditions. To achieve true resting recordings in chronically instrumented mice, the use of telemetric sensors, as utilized in the study by Baudrie et al. (2), may be preferred over the use of exteriorized catheters. Furthermore, it has been demonstrated that a period of ∼2 wk is required for full recovery and to reestablish circadian rhythms for blood pressure and heart rate after implantation of telemetric devices (12). Furthermore, active and inactive states may be separated and analyzed individually. This is particularly important in mice that tend to switch rapidly and spontaneously between active and inactive phases. The activity signal provided by some telemetry devices as well as temporal changes in the blood pressure and heart rate time series may provide some guidance in distinguishing active from inactive states (22).
Because high-frequency variability is defined as respiration-related variability, one may argue that the frequency range for high-frequency variability should be centered around the true respiration rate of each individual animal instead of using a fixed frequency range. However, this approach can prove problematic because determination of the respiration frequency is not always trivial. Under resting and inactive conditions, one typically finds a well-defined respiration peak in the blood pressure and heart rate spectra that can be used to identify the respiration rate. However, during active conditions, the respiration rate can be highly variable, and, thus, the respiration peak can be very difficult, if not impossible, to identify. In this case, a fixed frequency range for high-frequency variability is the only practical option. The data presented by Baudrie et al. (2) demonstrate that atropine affects spectral power of pulse interval “most significantly” in a frequency range from 2.5 to 5.0 Hz. Thus, this frequency range should be used for high-frequency variability in mice, if the experimental conditions do not allow identification of clear respiratory peaks in the power spectra.
It is also important to note, that the frequency ranges suggested by Baudrie et al. (2) were determined in C57BL6 mice obtained from Charles River, L’Arbresle, France. Since no other mouse strain was tested in this study, the question of strain differences remains open. However, the choice of C57BL6 mice is highly relevant, since this mouse strain is frequently used as a genetic background for genetically engineered mouse models (6, 8, 10, 15, 24, 25).
In conclusion, evidence suggests that sympathetic modulation of vascular tone in mice occurs at frequencies above 0.15 Hz (2, 9) and below 0.6 Hz (2) or 0.7 Hz (9). Furthermore, in mice, respiratory sinus arrhythmia affects heart rate variability most in a frequency band between 2.5 and 5.0 Hz. In the absence of clearly detectable respiration peaks in the power spectra, this fixed frequency range appears to be justified for quantification of high-frequency variability. Finally, predominance of sympathetic vs. parasympathetic modulation of cardiovascular function largely depends on experimental conditions. Telemetric recordings can be very useful to standardize these conditions.
- Copyright © 2007 the American Physiological Society