--------------------------------------------------------------------------------
Abstract

Arterial tonometry was used to continuously monitor the effects on heart rate and blood pressure of a series of innocuous mechanical stimuli applied to the neck in conscious humans. The stimuli used were derived from procedures commonly employed in clinical examination and physical therapy of the neck. In alert subjects, the stimuli used generally caused small and sometimes statistically significant decreases in heart rate, systolic pressure and diastolic pressure. In alert subjects, statistically significant decreases in systolic and diastolic pressure were particularly associated with stimuli which involved full rotation of the neck. In the course of the prolonged series of stimuli, some subjects slept or reported being on the verge of sleep. In these subjects, the same stimuli produced mixed effects on heart rate and increases in systolic and diastolic pressure which were significantly different from the effects obtained in fully alert subjects.
Key Words: Autonomic Nervous System, Arterial Tonometry, Heart Rate, Blood Pressure, Somatic Stimulation

Introduction

In animal studies, the use of anesthetic to eliminate emotional factors has permitted the resolution of many details of somatoautonomic reflex regulation of cardiovascular function (for review, see Sato et al., 1997). Both noxious and innocuous stimulation of somatic tissues have been demonstrated to produce autonomically-mediated alterations in heart rate and blood pressure. While anesthetic has been essential to the success of animal models used in the investigation of somatoautonomic reflexes, anesthesia effects many aspects of cardiovascular regulation (DeWildt et al., 1983; Faber, 1989; Sun and Reis, 1995). Whereas noxious stimulation characteristically produces marked responses, it has generally been reported, as would be expected in the anesthetized preparation, that innocuous stimulation produces weak and inconsistent responses (Adachi et al., 1990; Kaufmann et al., 1977; Terui et al., 1981). Furthermore, even the responses to noxious stimulation have been shown to be attenuated in a dose-dependent manner by anesthetics (Ogawa et al., 1994; Yanase et al., 1988). Apart from the direct effects of anesthesia on cardiovascular function, respiratory function is altered in the anesthetized animal and this may induce secondary effects on cardiovascular responses (Fukuda et al., 1989). Surgical preparation, such as cannulation and catheterization involve substantial somatic and visceral stimulation, so that any effects induced by experimental stimulation in the anesthetized animal must be presumed to occur against a backdrop of sensory bombardment and a consequent storm of reflex responses. Additionally, in the anesthetized animal, there may be some difficulty in confidently characterizing natural stimulation as either noxious or innocuous.
The current preliminary study was undertaken to determine the effects of innocuous somatic stimulation on heart rate and blood pressure in conscious humans who could, therefore, report on their subjective responses to stimulation. This study was also intended to establish the practicality of continuous measurement of heart rate and blood pressure using arterial tonometry in conscious humans during a prolonged series of somatic stimuli. One of the unanticipated consequences of the design of this pilot study was that a number of subjects fell asleep or were on the verge of sleep before the entire protocol was completed. While the number of trials in subjects who fell asleep or reported being on the verge of sleep was rather small, there were sufficent data to permit a comparison of results to stimulation in the fully alert and the drowsy or sleeping subject.

Methods

2.1 Preparation of subjects and experimental conditions
Volunteers, consisting of 6 males and 4 females aged 27 to 64 years, were recruited from the staff of the Tokyo Metropolitan Institute of Gerontology and the Tokyo Metropolitan Geriatric Hospital. All subjects were familiarized with the experimental design and inherent risks before providing consent. All experiments were conducted between 18:00 and 20:30 hrs, at the end of the subjects` working day, and prior to consumption of an evening meal.
Throughout the experiment, subjects wore loose surgical garb and lay supine on a padded operating table. The room was well lit and ambient temperature was maintained between 24 and 26 degrees Celsius. Investigators and subjects maintained silence throughout the course of the experiment, although subjects were encouraged to report any unpleasant sensation immediately and to make small adjustments in their posture as necessary for comfort. No effort was made to regulate respiratory rate.
2.2 Experimental protocol
As subjects lay supine, a series of mechanical stimuli were applied to the neck. These forms of stimulation were derived from procedures routinely used in the clinical examination of the neck and in physical therapy. Individual subjects received from one to four series of stimuli, with a period of at least one week between consecutive series. Each series of stimuli consisted of i) light, digital pressure applied bilaterally to the suboccipital muscles for 30 seconds, ii) passive rotation of the neck to 45 degrees, both left and right, for 30 seconds, iii) full passive rotation of the neck to the physiological limit (approximately 90 degrees), both left and right, for 30 seconds, iv) rhythmic rotation of the neck as light digital pressure was applied over the intervertebral joints from the top to the bottom of the cervical spine (cervical motion palpation), both left and right, for 30 seconds, v) factitious spinal manipulation consisting of holding the neck close to the physiological limit of rotation, both left and right, and applying a quick, light thrust with the fingers of one hand in the plane of the skin, vi) cervical spinal manipulation, both left and right, consisting of holding the neck at the physiological limit of rotation and applying a quick, light thrust to the intervertebral joints to produce an audible cavitation, vii) full passive rotation of the neck to the physiological limit (approximately 90 degrees), both left and right, for 30 seconds, i.e. procedure (iii) repeated. Each form of stimulation, including consecutive left and right rotational movements, was preceded and followed by a period of 2 minutes during which the patient lay supine with the neck resting in the neutral position. At the end of the series of cervical stimuli, the cold pressor test was applied to each subject, with one hand held in ice water to the limit of endurance.
2.3 Data collection and analysis
Throughout the course of the experiment, heart rate and blood pressure were continuously monitored via radial artery tonometry (Jentow 7700, Colin, Japan). The tonometer was calibrated at the beginning and end of the experimental series. Signals were recorded on a data recorder and displayed on a polygraph.
Changes in heart rate and blood pressure were calculated as the difference between the average measure in the 20 seconds immediately prior to stimulation and the maximum increase or decrease resulting from that stimulation. Values for comparable left and right stimuli (procedures (ii) to (vii) inclusive) were pooled. Pre- and post-stimulation values were compared using the paired t-test. Changes in alert and drowsy/sleeping subjects were compared using the unpaired t-test.

Results

For a single subject, the entire experimental procedure required on the order of 30 minutes. During that time, a number of subjects were observed to fall asleep, or reported diminished wakefulness. Fig. 1 presents representative heart rate and blood pressure recordings for one subject taken from two separate series, one in which the subject was fully alert, and another in which the subject reported diminished wakefulness during the procedures shown. While in the alert subject, heart rate and blood pressure were relatively stable between stimuli, in the subject on the verge of sleep, these same parameters became quite labile. In this particular subject, the cold pressor test produced bradycardia rather than the characteristic tachycardia. This phenomenon of anomalous bradycardia with the cold pressor test has been reported elsewhere and attributed to the intense pain of the test (Shibahara et al., 1996).
Fig. 2 presents the changes in heart rate (fig. 2A), and systolic (fig. 2B) and diastolic (fig.2C) pressure (mean + standard error of the mean) for the various forms of stimulation employed. Data have been separated to distinguish between those from trials in which subjects were fully alert and those from trials in which subjects were observed to fall asleep or reported being on the verge of sleep. All subjects reported that all forms of stimulation, other than the cold pressor test, were painless. There was no discomfort reported due to continuous arterial tonometry. No attempt was made to rouse subjects who slept and so a number of subjects slept through various portions of the experiment.
In all subjects, the cold pressor test resulted in elevation of heart rate and blood pressure, except for the exceptional subject referred to in figure 1 who consistently displayed increased blood pressure with mild bradycardia. For all other forms of stimulation, in the alert subject, heart rate was observed to either not change or decrease on average. These decreases were significant in the cases of procedures i) light touch of suboccipital muscles, iv) cervical motion palpation, and vii) full rotation following spinal manipulation. In subjects who slept or were on the verge of sleep, there were no significant decreases in heart rate, and procedure vi) cervical spinal manipulation was accompanied by a significant increase in heart rate. When comparing responses between alert and drowsy/sleeping subjects (Table 1), significantly different results were only obtained for 2 of the 7 forms of cervical stimulation: procedures (iv) motion palpation and (vi) cervical spinal manipulation
In the alert subject, procedures i) bilateral digital pressure to the suboccipital muscles, ii) passive rotation to 45 degrees, iv) cervical motion palpation and v) factitious cervical spinal manipulation were accompanied by small and insignificant changes in systolic and diastolic pressure. Procedures which required cervical rotation to the physiological limit, iii) full rotation prior to spinal manipulation, vi) cervical spinal manipulation and vii) full rotation after spinal manipulation, were accompanied by small and significant decreases in both systolic and diastolic pressure (fig. 2B and 2C). In subjects who slept or were on the verge of sleep, all forms of stimulation resulted in increases in systolic and diastolic pressure. The increases in systolic pressure were statistically significant in the cases of procedures i) through v) inclusive (fig. 2B). The increases in diastolic pressure were statistically significant for procedures i) bilateral digital pressure to the suboccipital muscles, ii) passive rotation to 45 degrees, and v) factitious cervical spinal manipulation (fig. 2C).
When comparing responses obtained from cervical stimulation in alert and drowsy/sleeping subjects, all responses in systolic and diastolic pressure were significantly different (Table1).

Discussion

The use of anesthetic to eliminate emotional and behavioural factors has been essential to the success of animal models employed in the study of somatoautonomic regulation of cardiovascular function. On the other hand, the use of anesthetic alters the behaviour of reflex centers involved in cardiovascular regulation (DeWildt et al., 1983; Sun and Reis, 1995) and surgical preparation must be presumed to introduce a variety of artifacts. Additionally, in the anesthetized animal, it may be difficult to determine if a particular form of stimulation should be characterized as noxious or innocuous. In conscious humans, numerous studies have been conducted into autonomic regulation of cardiovascular function in response to postural change (visceral stimulation) and exercise (for review, see Joyner and Shepherd, 1997). Relatively few studies have examined cardiovascular responses to noxious somatic stimulation in humans (Hsieh et al., 1995, Malliani et al., 1991). In the present study, we attempted to develop a protocol which would permit the continous measurement of heart rate and blood pressure in conscious humans in response to innocuous mechanical stimulation of somatic cervical tissues. The use of the conscious human permits a correlation between cardiovascular responses to stimuli and the person`s subjective impression of whether the stimuli were noxious or innocuous.
In the anesthetized animal, the adequacy of anesthesia is judged by the stability of physiological parameters such as heart rate and blood pressure. In unanesthetized humans, these parameters are relatively labile. As indicated in figure 1A and 1B, even in the resting supine subject, there were spontaneous rhythmic oscillations in blood pressure, as well as small alterations in heart rate. Rhythmic oscillations in heart rate and blood pressure reflect, either directly or indirectly, oscillations in sympathetic and parasympathetic tone (Akselrod et al., 1987; DeBoer et al., 1987; Grasso et al., 1997; Moltner et al., 1990) and so confound attempts to quantify short term responses in heart rate and blood pressure following brief stimuli. In subjects moving from wakefulness into sleep, the high-frequency oscillations in blood pressure showed a loss of rhythmicity which in part might be attributable to alterations in respiratory rhythm (Koh et al., 1998).
That a number of subjects did fall asleep or reported being on the verge of sleep also implies that over the course of the experiment there might have been significant changes in emotional state associated with, for example, the opportunity to relax at the end of the working day. Conversely, there may have been some anxiety due to the length of the experiment or anticipation of discomfort. Such changes in emotional state could by themselves induce changes in autonomic function and so cardiovascular function (Collet et al., 1997; Ekman et al., 1983).
In the present study, all but one subject displayed brisk increases in heart rate, and systolic and diastolic pressure in response to the cold pressor test. This response pattern is accepted as an indicator of the integrity of sympathetic function, although anomalous results, as shown in figure 1B, have been obtained in some individuals perhaps, due to the pain associated with the test (Shibahara et al., 1996). In this study this form of noxious stimulation, which all subjects characterized as painful, provided heart rate and blood pressure responses which could be compared to those obtained from what were presumed to be innocuous forms of mechanical stimulation.
It has previously been demonstrated that noxious mechanical stimulation of somatic tissues can elicit increases in sympathetic activity in conscious humans (Nordin and Fagius, 1995), an observation which is in accord with daily experience and studies in anesthetized animals (Sato et al., 1997). On the other hand, in anesthetized animals, innocuous mechanical stimulation of somatic tissues has variously been shown to elicit both increases and decreases in sympathetic activity, although the results have often been weak and inconsistent (Ohsawa et al., 1995; Sato and Swenson, 1984).
In the present study, the various forms of cervical mechanical stimuli applied to alert subjects were associated with small but occasionally statistically significant decreases in heart rate, and systolic and diastolic pressure. These findings are consistent with the alert subjects` reports that the stimuli applied were innocuous, and contrast with the tachycardia and pressor responses reported from noxious movements induced in the knee joint of the anesthetized cat (Sato et al., 1984).
Procedures ii) through vii), all involved a degree of cervical rotation. In some exceptional individuals, cervical rotation is associated with a degree of vertebrobasilar insufficiency which often manifests as syncope (Sturzenegger et al., 1994). In asymptomatic volunteers cervical rotation has been shown to result in a decrease in blood flow in the contralateral vertebral artery (Licht et al., 1998). These changes were not shown to correlate with any alteration in blood pressure and, overall, no significant changes in blood pressure were observed. However, the authors did not indicate the time frame over which the rotational procedures were performed, nor is it clear whether the authors recorded the maximum alterations in blood pressure resulting from cervical rotation, or the instantaneous blood pressure at the time of maximum change in vertebral artery blood flow. That is to suggest that the maximum changes in blood pressure may not have been coincident with the maximum changes in vertebral artery flow. In the present study, only those forms of stimulation which produced cervical rotation to the physiological limit were associated with significant changes in both systolic and diastolic pressure in alert subjects. Changes in heart rate did not appear to correlate with degrees of cervical rotation in alert subjects.
In subjects who slept or were on the verge of sleep, all forms of cervical stimulation evoked increases in systolic and diastolic blood pressure, although these increases were statistically significant in only approximately half of the procedures. These results should probably be interpreted as a mild alarm reaction to being roused by the stimulation, although in some instances subjects slept through the procedures. In subjects who slept or were on the verge of sleep, only cervcal spinal manipulation produced a signficant increase in heart rate, and this was not accompanied by significant changes in systolic or diastolic pressure.
In the present study, subjects were indifferent to the continuous application of arterial tonometry over a period of approximately 30 minutes. However, the length of the series of stimuli resulted in several subjects reporting drowsiness or falling asleep, and this change in consciousness significantly effected both the stability of base line measures and the responses to stimulation. The cold pressor test, which all subjects described as painful, produced significant increases in heart rate, and systolic and diastolic pressure. Furthermore, these changes were invariably larger than those obtained with any other form of stimulation. In alert subjects, all forms of cervical somatic stimuli, which all subjects in this study described as painless and even soothing, produced either no significant change or significant decreases in heart rate, and systolic and diastolic pressure.

Acknowledgement: The authors are grateful to Dr. Akio Sato, Emeritus Scientist of the Tokyo Metropolitan Institute of Gerontology, for his encouragement in the performance of this study.

References

Adachi, T., Meguro, K., Sato, A., Sato, Y., 1990. Cutaneous stimulation regulates blood flow in cerebral cortex in anesthetized rats. Neuroreport 1, 41-44.
Akselrod, S., Eliash, S., Oz, O., Cohen, S., 1987. Hemodynamic regulation in SHR: Investigation by spectral analysis. Am. J. Physiol. 253, H176-183.
Collet, C., Vernet-Maury, E., Delhomme, G., Dittmar, A., 1997. Autonomic nervous system response patterns specificity to basic emotions. J. Autonom. Nerv. Syst. 62, 45-57.
DeBoer, R.W., Karemaker, J.M., Strackee J., 1987. Hemodynamic fluctuations and baroreflex sensitivity in humans: A beat-to-beat model. Am. J. Physiol. 253, H680-689.
De Wildt, D., Hillen, F., Rauws, A., Sangster, B., 1983. Etomidate-anesthesia, with and without fentanyl, compared with urethane-anesthesia in the rat. Br. J. Pharmacol., 79, 461-469;
Ekman, P., Levenson, R., Friesen, W., 1983. Autonomic nervous system activity distinguishes among emotions. Science, 221, 1208-1210.
Faber, J., 1989. Effects of althesin and urethan-chloralose on neurohumoral cardiovascular regulation. Am. J. Physiol. 256, R757-765.
Fukuda, Y., Sato, A., Suzuki, A., Trzebski, A., 1989. Autonomic nerve and cardiovascular responses to changing blood oxygen and carbon dioxide levels in the rat. J. Autonom. Nerv. Syst. 28, 61-74.
Grasso, R., Schena, F., Gulli, G., Cevese, A., 1997. Does low-frequency variability of heart period reflect a specific parasympathetic mechanism. J. Autonom. Nerv. Syst. 63, 30-38.
Hsieh, J., Stahle-Backdahl, M., Hagermark, O., Stone-Elander, S., Rosenquist, G., Ingvar, M., 1995. Traumatic nociceptive pain activates the hypothalamus and the periaqueductal gray: a positron emission tomography study. Pain, 64, 303-314
Joyner, M.J.. and Shepherd, J.T., 1997. In: Low, P., ed, Clinical Autonomic Disorders. Lippincott-Raven Publishers, Philadelphia, pp 61-71.
Kaufmann, A., Sato, A., Sato, Y., Sugimoto, H., 1977. Reflex changes in heart rate after mechanical and thermal stimulation of the skin at various segmental levels in cats. Neuroscience 2, 103-109.
Koh, J., Brown, T., Beightol, L., Eckberg, D., 1998. Contributions of tidal lung inflation to human R-R interval and arterial pressure fluctuations. J. Autonom. Nerv. Syst. 68, 89-95.
Licht, P., Christensen, H., Hojgaard, P., Hoilund-Carlsen, P., 1998. Triplex ultrasound of vertebral artery flow during cervical rotation. J. Manipulative Physiol. Ther. 21, 27-31.
Malliani, A., Pagani, M., Lombardi, F., Cerutti, S., 1991. Cardiovascular neural regulation explored in the frequency domain. Circulation 84, 482-492.
Moltner, A., Holzel, R., Strian, F., 1990. Heart rate changes as an autonomic component of the pain response. Pain 43, 81-89.
Nordin, M., Fagius, J., 1995. Effect of noxious stimulation on sympathetic vasoconstrictor outflow to human muscles. J. Physiol. (Lond), 489, 885-894.
Ogawa, S., Saito, H., Saeki, S., Suzuki, H., 1994. Reflex sympathetic activities during inhalation of anaesthetics in cats: nitrous oxide. Neuroscience 168, 16-18.
Ohsawa, H., Okada, K., Nishijo, K., Sato, Y., 1995. Neural mechanism of depressor responses of arterial pressure elicited by acupuncture- like stimulation to a hindlimb in anesthetized rats. J. Autonom. Nerv. Syst. 51, 27-35.
Sato, A., Sato, Y., Schmidt, R., 1984. Changes in heart rate and blood pressure induced by movements of normal and inflamed knee joints. Neurosci. Lett. 52, 55-60.
Sato, A., Sato,Y., Schmidt, R., 1997. The impact of somatosensory input on autonomic functions. Rev. Physiol. Biochem. Pharmacol. 130, 1-328.
Sato, A., Swenson, R., 1984. Sympathetic nervous system response to mechanical stress of the spinal column in rats. J. Manipulative Physiol. Ther. 7, 141-147.
Shibahara, N., Matsuda, H., Umeno, K., Shimada, Y., Itoh, T., Terasawa, K., 1996. The responses of skin blood flow, mean arterial pressure and R-R interval induced by cold stimulation with cold wind and ice water. J. Autonom. Nerv. Syst. 61, 109-115.
Sturzenegger, M., Newell, D., Douville, C., Byrd, S., Schoonover, K., 1994. Dynamic transcranial Doppler assessment of positional vertebrobasilar ischemia. Stroke 25, 1776-1783.
Sun, M., Reis, D., 1995. Urethane directly inhibits chemoreflex excitation of medullary vasomotor neurons in rats. Eur. J. Pharmacol. 293, 237-243.
Terui, N., Numao, Y., Kumada, M., Reis, D., 1981. Indentification of the primary afferent fiber group and adequate stimulus initiating the trigeminal depressor response. J. Autonom. Nerv. Syst. 4, 1-16.
Yanase, K., Meguro, K., Sato, A., Sato, Y., 1988. The effect of sevoflurane on somatically induced sympathetic reflexes. J. Anesth. 2, 272-275.

Legends

Figure 1
Representative recordings of heart rate (HR) and blood pressure (BP) from one subject both when alert (A and B) and when on the verge of sleep (C and D). Duration of stimulation is indicated by horizontal bars beneath recordings. Characteristically, as indicated in these recordings, heart rate and blood pressure were more labile in subjects who slept or reported being on the verge of sleep. When alert, this subject showed an anomalous reaction to the cold pressor test (B), i.e. increased blood pressure, but mild bradycardia. All other subjects in all other trials of the cold pressor test showed increased blood pressure and tachycardia.
Figure 2
A: Heart Rate, B: Systolic Pressure, C: Diastolic Pressure (histograms/vertical bars indicate mean/standard error of the mean). Responses to stimuli in alert subjects (upper panel) and subjects who slept or reported being on the verge of sleep (lower panel). Number of trials/ number of subjects are indicated below the histograms. *: p<0.05, ** :p<0.01 per paired t-test for the null hypothesis that there was no significant difference between prestimulus levels and the maximum change associated with stimulus. Stimuli are indicated by small Roman numerals in brackets as described in detail in the text. Touch - light bilateral suboccipital pressure, 45o - passive cervical rotation of 45 degrees, 90o - passive cervial rotation to physiological limit (approximately 90 degrees), M.P. - motion palpation, F.S.M. - factitious spinal manipulation, S.M. - cervical spinal manipulation, C.P. - cold pressor test.
Table 1
Numerical data from figure 2, i.e. mean plus standard error of the mean for responses in heart rate, systolic blood pressure and diastolic blood pressure. *:p<0.05, **:p<0.01 per unpaired t-test for the null hypothesis that there was no significant difference between responses in alert subjects and subjects who slept or reported being on the verge of sleep.