HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Wallin, B. G. Articles by Elam, M. Search for Related Content PubMed PubMed Citation Articles by Wallin, B. G. Articles by Elam, M.

Two neural mechanisms for respiration-induced cutaneous vasodilatation in humans?

B. Gunnar Wallin, Karin Båtelsson, Peter Kienbaum, Tomas Karlsson, Bertil Gazelius * and Mikael Elam

Received 24 June 1998; accepted after revision 18 August 1998.

	ABSTRACT

Top Abstract Introduction Methods Results Discussion References

1. In humans, a deep breath is known to induce cutaneous vasoconstriction in the warm state, and vasodilatation in the cold state. To investigate whether vasodilatation in the cold state is related to reduction of sympathetic vasoconstrictor nerve traffic, we studied the effect of a deep breath on vascular resistance in a skin area on the dorsum of the hand, in which release of noradrenaline from sympathetic nerves was blocked by iontophoretic pretreatment with bretylium tosylate. Simultaneous measurements were made in two control areas. In eight healthy subjects, data were obtained from deep breaths taken before bretylium in the warm state, after general cooling to a finger skin temperature below 25 °C and after rewarming to above 32 °C.

2. In the warm state before bretylium pretreatment, the deep breath evoked short-lasting vasoconstrictions at all sites. In the cold state there was no change of vascular resistance in the bretylium-pretreated area, whereas in the control areas an initial tendency towards vasoconstriction was followed by a significant transient vasodilatation. After rewarming, transient vasoconstrictions reappeared at the control sites, whereas only a transient vasodilatation occurred at the bretylium-pretreated site.

3. In six healthy subjects we also monitored the effects of a deep breath on skin sympathetic nerve activity (recorded by microneurography in the peroneal nerve), and skin vascular resistance within the innervation zone of the impaled nerve fascicle in the foot. Data from thirty deep breaths per subject were averaged.

4. In the cold state, the deep breath induced a strong increase in neural discharge, followed by a transient reduction of nerve traffic lasting approximately 15 s and associated with a subsequent reduction of vascular resistance.

5. We conclude that the deep breath-induced vasodilatation in the cold state is due to reduction of sympathetic vasoconstrictor nerve traffic. The vasodilatation after bretylium treatment in the warm state raises the possibility that a deep breath induces two simultaneous neural reactions, a vasoconstrictor and an active vasodilator component, the latter being weaker and normally masked by the strong vasoconstrictor component.

	INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

Thermoregulatory reflexes are important for the control of skin blood flow. In cold human subjects, sympathetic vasoconstrictor impulses induce generalized skin vasoconstriction, in thermoneutral subjects there is an intermediate blood flow to skin and no vasoconstrictor traffic, and in warm subjects skin blood flow is high because of active neurogenic vasodilatation (Grant & Holling, 1938; Rowell, 1977). Skin blood flow is also influenced by arousal, mental stress and respiratory stimuli (such as a deep breath), and surprisingly, these manoeuvres induce transient vasoconstriction in warm subjects and mainly vasodilatation in cold subjects (Oberle et al. 1988). The vasoconstriction in warm subjects is due to reflex activation of vasoconstrictor impulses, but whether the vasodilatation in cold subjects is caused by activation of vasodilator nerve fibres or by inhibition of the background vasoconstrictor activity is not known.

The aim of the present study was to test the hypothesis that the vasodilatation in cold subjects is due to a reduction of sympathetic vasoconstrictor nerve traffic. To this end we performed two types of experiments using a deep breath as the test stimulus. In the first study, we compared the stimulus-induced skin blood flow responses in thermoneutral and cold conditions between three skin areas, two serving as control, the third being pretreated with bretylium tosylate. Since bretylium is known to block blood flow responses induced by noradrenergic nerve traffic (Boura & Green, 1959; Haeusler et al. 1969), the difference in flow responses between the treated and untreated areas should be a measure of the vasoconstrictor component of the response. In the second study, which was carried out in cold subjects, we made microneurographic recordings of multi-unit skin nerve sympathetic activity (SSA, which is dominated by vasoconstrictor impulses; Bini et al. 1980a), and analysed the effect of repeated deep breaths on the nerve traffic and on the blood flow within the innervation zone of the impaled nerve fascicle.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Subjects

After approval by the human ethics committee at the University of Göteborg, and with the written, informed consent of the subjects, experiments were performed on sixteen healthy volunteers (fourteen men, two women) aged 21-33 years (mean, 26 years). Ten males participated in experiments with bretylium tosylate, but two were excluded because of incomplete vasoconstrictor block (see below) and therefore the final analysis is based on data from eight men. Six subjects (four men, two women) participated in the microneurographic recordings.

Bretylium experiments

Measurements. Skin perfusion (i.e. flux of red blood cells) was measured with the laser Doppler technique. Two drug delivery probes (see below) (PF 481), which were not heated, were attached to a dual channel Periflux 4001 Master (Perimed AB, Järfälla, Sweden) that transmitted a laser beam at a wavelength of 780 nm. These probes were placed between the first and second metacarpal bone on the dorsal surface of the left and right hands. A third probe (PF 308) was attached to the skin between the third and fourth metacarpal bone on the left hand and connected to a PF2b monitor (Perimed AB), which used a wavelength of 632·8 nm. All probes had a fibre separation of 250 µm. Electrical calibration for zero flow was made in all recordings. The output of Doppler meters does not yield absolute values for blood flow, but shows relative changes in the flux of red blood cells expressed in perfusion units (PU; 1 PU = 10 mV). Skin temperature was measured on the skin on the dorsal side of the left hand and the volar tip of the second left finger with small temperature sensors (PF 442) connected to a PeriTemp (PF4005, Perimed AB). Blood pressure was monitored with the volume clamp method on the third left finger (Finapres, Ohmeda, Louisville, CO, USA), and respiratory movements were monitored by a strain gauge strapped around the chest with a rubber band. Analog signals of all parameters were stored on a personal computer (sampling rate, 16 s^-1).

Iontophoresis. Bretylium tosylate (10 mM; Research Biochemicals Inc.) solution was added to a iontophoretic drug delivery electrode chamber (PF383, Perimed AB), which had a laser Doppler probe (PF481, Perimed AB) inserted through a hole in the centre of the electrode chamber. In this way perfusion could be monitored continously during the iontophoresis. Electrodes were connected to a battery-driven PeriIont power supply (PF382, Perimed AB) working with constant current. The cathode was at the site of the perfusion measurement probe between the first and second metacarpal bone on the dorsal side of the left hand. The iontophoresis was performed for 10 min with a current density of 0·4 mA cm^-2. Immediately afterwards, iontophoresis of NaCl (isotonic solution) was performed at the corresponding site on the right hand and with the same time and current parameters. In both cases the anode was placed on the ulnar dorsal side of the left hand.

Changes in ambient temperature. The subject was dressed in a thermal suit lined with plastic tubing, through which water of different temperatures could be perfused so as to cool or warm the subject. The hands and feet were not covered by the suit.

Experimental procedure. Subjects were supine. All recording probes were arranged as described above, and the subjects practised taking single maximal deep breaths before the recordings were initiated. If necessary, subjects were warmed until finger tip temperature was above 32°C and then the experiment started. On three occasions, at approximately 5 min intervals, the subject took a single deep breath. Following this, bretylium tosylate and NaCl were applied iontophoretically to the dorsal side of the left and right hand, respectively, for 10 min. This led to a transient increase in perfusion lasting about 50 min (see Results). When the perfusion had returned to the control level, the cooling of the subject began. When the finger tip temperature was below 25°C, a single deep breath was taken on three occasions at intervals of approximately 5 min. Finally the subject was rewarmed and when the finger tip temperature was above 32°C, another three breaths were taken. The protocol is illustrated in Fig. 1, which shows perfusion, skin temperature and blood pressure data for one subject.

Analysis. In two subjects, the deep breaths taken after rewarming evoked vasoconstrictor responses at the bretylium-treated site, indicating incomplete neural block. Data from these subjects were excluded from further analysis.

To calculate changes in perfusion induced by the deep breath, the averaged maximal reduction of perfusion over a period of 3·0 s within 2-12 s after the start of the deep breath was compared with the mean perfusion during a 30 s control period before the breath. The last 10 s immediately prior to the start of the breath were not included in the control period. In a given individual the perfusion responses was found to vary between individual breaths. The coefficient of variation (mean ± S.D.) in the control situation before application of bretylium was 23 ± 10, 12 ± 6 and 21 ± 10 % in untreated, NaCl and bretylium sites, respectively. In the cold state, corresponding figures for the untreated and NaCl sites were 17 ± 12 and 18 ± 19 %, respectively. The coefficient of variation for the perfusion increase in the cold state was 22 ± 9 and 15 ± 11 % in the untreated and NaCl sites, respectively. To reduce the effects of such random variability, perfusion records were averaged. To this end, the computer used individual perfusion records from the three breaths under each temperature condition to calculate the average perfusion in each subject at each of the three recording sites. For each temperature condition, an average blood pressure curve was also calculated. All subsequent calculations were made using each subject's averaged records. The vascular resistance at each recording site was then calculated as blood pressure divided by perfusion.

In a given individual and recording site, the vasoconstrictor response was calculated as the ratio (expressed as a percentage) between the average vascular resistance during a 30 s control period before the breath, and the average resistance over 3·0 s of the most pronounced increase of resistance occurring within 2-12 s after the start of the deep breath in the untreated control site on the left hand. The same measurement period was then also used for the bretylium - and the NaCl control - sites. The last 10 s immediately before the start of the breath were not included in the control period. Vasodilator responses were expressed as the ratio (as a percentage) between the average vascular resistance during the same control period as above, and the average resistance during a 10 s period, beginning 20 s after the start of the inspiration (which corresponded approximately to the most pronounced reduction of vascular resistance in the cold state, see below). After rewarming, vascular resistance at the bretylium-treated site was also quantified for 10 s, beginning 7 s after the start of the deep breath. The time frames for analysis of vasomotor responses were chosen after inspection of the averaged response curves for the whole study group.

Microneurography experiments

Measurements. For nerve recordings, multi-unit postganglionic sympathetic nerve activity was recorded in the peroneal nerve at the fibular head by using tungsten microelectrodes with a shaft diameter of 0·2 mm and a tip of a few micrometres. A reference electrode with a larger uninsulated tip was inserted subcutaneously a few centimetres away. The signals were amplified (gain 50 000), filtered (band-width, 0·7-2 kHz) and passed through an amplitude discriminator to improve the signal-to-noise ratio. The original neurogram was passed through a resistance-capacitance integrating network (time constant, 0·1 s) to obtain a mean voltage display of the multi-unit nerve activity. The receptive field of the impaled nerve fascicle was mapped with touch stimuli and then small adjustments of the electrode position were made until a recording site was obtained in which multi-unit sympathetic activity could be recorded. Technical details and evidence that the recorded activity was of sympathetic origin have been published previously (Vallbo et al. 1979; Bini et al. 1980a).

Skin perfusion was measured at two sites with the laser Doppler meter (Periflux 4001). One probe was placed on the dorsum of the foot within the receptive field of the impaled nerve fascicle, and one on the plantar side of the ipsilateral big toe. Skin temperature was recorded on the plantar side of the ipsilateral big toe and on the volar side of the tip of the third finger with thermocouple probes. Skin electrical resistance was recorded by van Gough GSR modules (type IGSR/7A) using Ag-AgCl electrodes (Medicotest). At least one and, if possible, both electrodes were positioned within the receptive field of the impaled nerve fascicle. Skin resistance was also recorded by two electrodes placed on the sole of the ipsilateral foot. The measuring current was 12 µA, and filter settings were 0·7-100 Hz. An ECG was recorded via surface electrodes on the chest and respiratory movements were monitored by a strain gauge strapped around the chest with a rubber band.

Analog signals of all parameters were stored on a personal computer (sampling rate, 200 s^-1) and magnetic tape (V-store, Racal, Southampton, UK). During the experiment, selected signals were monitored on an ink-jet recorder (Mingograph 800, Siemens-Elema, Solna, Sweden).

Experimental procedure. Subjects lay supine dressed in the thermal suit. After arrangement of the electrodes and probes for monitoring ECG, respiration and blood pressure, body cooling was started and the tungsten microelectrode was inserted in the peroneal nerve at the fibular head. When an adequate recording of skin sympathetic nerve activity (SSA) had been obtained, the innervation zone of the impaled nerve fascicle was defined and probes for measurements of perfusion, skin resistance and skin temperature were attached. Subjects were cooled until skin temperature on the dorsum of the foot was below 25°C (average skin temperature was 23·3°C in the foot and 24·7°C in the hand). During the cooling procedure, single rapid deep inspirations were performed at approximately 2 min intervals, the subject being instructed to take the deep breath when the experimenter said 'now'. In this way, approximately thirty deep breaths were obtained and included in the analysis.

Analysis. For each deep breath, signals of respiration, integrated nerve activity, blood pressure and perfusion were quantified in 5 s epochs over a total of 105 s. For each epoch, vascular resistance was calculated as mean blood pressure divided by perfusion. The observation period comprised 40 s before the start of the breath, 5 s corresponding to the breath and 60 s following the breath. The control period was defined as the initial 30 s (six epochs) of the observation period, the 10 s immediately prior to the start of the breath being excluded in order to avoid contamination from possible arousal effects of the instruction to take the breath. To quantify the nerve activity, a baseline was defined manually in the mean voltage neurogram and for each 5 s epoch, the computer calculated the mean voltage area above the baseline. In each subject, data from all breaths were averaged and further calculations were based on such averaged records. For nerve activity and perfusion, data from each 5 s epoch were expressed as a percentage of the respective mean control value. Respiration was quantified by measuring the number of peaks and their amplitudes in the respiratory curve that occurred within each 5 s epoch. Due to the method of measuring respiration, the first and the last 5 s epochs of the 105 s analysis period had to be excluded.

SSA consists of a mixture of vasomotor and sudomotor fibre discharges that cannot be separated from each other in the neurogram (Bini et al. 1980a). The sweating induced by sudomotor fibre activation can be detected as a change in skin electrical resistance. Therefore, skin electrical resistance reductions were recorded in order to assess the extent to which the stimulus-induced SSA bursts included sudomotor nerve impulses. Since previous studies from our laboratory have shown that the amplitude of a skin resistance change is poorly related to the strength of the underlying nerve traffic (Kirnö et al. 1991; Kunimoto et al. 1992), the analysis was limited to counting the number of resistance changes during the 30 s control period and the 60 s period following the deep breath (expressed as the respective number of deflections min^-1).

Statistics

Data are presented as means ± S.E.M. In the bretylium experiments the hypothesis to be tested was based on the occurrence of a significant vasodilatation after a deep breath at the control sites in the cold state. To this end the period of minimum resistance in the untreated site after the breath (20-30 s after the start of the breath) was chosen as the period to be tested. Application of Shapiro-Wilk's W test showed that blood flow and vascular resistance data were not normally distributed, and therefore a two-way analysis of variance by ranks with one repeated measure, followed by Duncan's multiple range test, was used to test the significance of changes over time. A similar procedure was used in the microneurography experiments. Student's t test was used to test the significance of the fall in vascular resistance induced by a deep breath at the bretylium site after rewarming. P values less than 0·05 are considered significant.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Bretylium experiments

Resting perfusion. Before cooling and application of bretylium, the resting skin perfusion levels were similar at all sites. The iontophoretic applications of bretylium and sodium chloride both gave rise to marked increases of perfusion, lasting 14-75 min (mean, 51 min) (Fig. 1). Cooling led to decreases of perfusion at all sites. The mean reduction was less in the bretylium-treated area, but when compared with the reductions at the control sites, the difference did not reach statistical significance (P = 0·12, ANOVA). With rewarming, perfusion increased to, or above, control levels at all three sites (Fig. 2, Table 1).

		View larger version [in this window] [in a new window]
Figure 1. Experimental procedure Records of laser Doppler perfusion at the three sites, skin temperature at a finger tip and finger blood pressure during a whole experiment. Periods of whole body cooling and warming are indicated by the horizontal bar in the top panel. DB, deep breaths. Periods of iontophoretic application (Ionto) of bretylium and NaCl are indicated by the horizontal bars below the respective perfusion records.

		View larger version [in this window] [in a new window]
Figure 2. Laser Doppler perfusion and blood pressure responses to a deep breath in the control situation (finger skin temperature > 32 °C), in the cold state (finger skin temperature < 25 °C) and after rewarming to a finger skin temperature > 32 °C Averaged data from eight subjects. Iontophoretic application (Ionto) of bretylium and NaCl was made after the control period (arrow). The NaCl perfusion record is not shown. The dotted vertical lines indicate the start of deep breath. Note the virtual absence of perfusion changes at the bretylium site in the cold state.

Table 1. Skin perfusion and blood pressure (BP) at rest (n = 8)

Responses to deep breaths. The amplitude and duration of the deep inspirations did not differ significantly between the control state, the cold state and after rewarming. All data are given in Tables 2 and 3. Figure 2 shows averaged, deep breath-induced blood pressure and perfusion changes at the bretylium-treated and the control sites in the left hand (not shown are the perfusion changes in the NaCl control site on the right hand, which were similar to those in the left hand control site).

In the control situation (before cooling and before application of bretylium), a single deep breath evoked a distinct reduction in perfusion at all three recording sites with durations of 10-15 s, with minimum perfusions varying between 31 and 47 % of the resting level and occurring 6·5 ± 0·7 s (mean ± S.D.) after the start of the deep breath. Beginning at approximately the same time, there was also a transient decrease in blood pressure, but the increase in skin vascular resistance was nevertheless highly significant (P < 0·001) at all three skin areas (Fig. 3).

		View larger version [in this window] [in a new window]
Figure 3. Vascular resistance responses induced by a deep breath in the control situation, in the cold state, and after rewarming (note logarithmic scales) Averaged data from eight subjects. Iontophoretic application (Ionto) of bretylium and NaCl was made after the control period (arrow). Note the absence of resistance changes in the cold state and reduction of resistance after rewarming at the bretylium site.

In the cold state, the deep breath evoked initial significant reductions of perfusion at the two control sites down to 58 and 62 % of the resting levels, respectively, occurring 7·8 ± 2·2 s after the start of the breath (Fig. 2, Table 2). These reductions were followed by increases of perfusion (to 155 and 129 % of resting perfusion, respectively). Again there was a transient blood pressure reduction that started with the reduction of perfusion, but was also maintained during the subsequent increase in perfusion. When expressed as a change in vascular resistance, the initial increase in resistance (vasoconstriction) was significant compared with the resting level in the NaCl site, but not in the untreated site, whereas the subsequent decrease in resistance (vasodilatation) was significant at both sites (Table 3 and Fig. 3). At the bretylium-treated site, no significant changes in perfusion or vascular resistance occurred (Figs 2 and 3, Tables 2 and 3).

After rewarming, the deep breath evoked only transient vasoconstrictions (reductions of perfusion) at the control sites, similar to those seen before cooling (P < 0·01) (Figs 2 and 3, Tables 2 and 3). At the bretylium-treated site, there was no vasoconstrictor response and 20-30 s after the start of the breath, there was no vasodilator response. However, there was a transient increase of perfusion (Fig. 2, Table 2) and a decrease of vascular resistance (P < 0·05) (Fig. 3, Table 3) occurring between 7 and 17 s after the beginning of the breath.

Table 2. Effects of deep breath on skin perfusion and blood pressure (BP)

Table 3. Effects of deep breath on cutaneous vascular resistance

	Control	Cooled	Rewarmed
(1) Vasoconstrictor response
Bretylium site	219 ± 35 ²	97 ± 2	86 ± 4
Untreated site	244 ± 58 ²	184 ± 44	213 ± 44 ²
NaCl site	191 ± 19 ²	190 ± 32 ²	161 ± 24
(2) Vasodilator response (20-30 s)
Bretylium site	87 ± 6	93 ± 2	96 ± 6
Untreated site	93 ± 12	65 ± 5 ²	104 ± 9
NaCl site	83 ± 7 *	76 ± 5	100 ± 4
(3) Vasodilator response (7-17 s) (calculated only after rewarming for the bretylium site)
Bretylium site	-	-	80 ± 5 *

Microneurography experiments

During the cooling period there was a progressive increase in spontaneous nerve traffic. To judge from the skin resistance and Doppler records, the resting activity contained only vasoconstrictor activity in two subjects. However, in four subjects some skin resistance changes still occurred, indicating that an admixture of sudomotor impulses was still present, despite the subjects being cold.

The amplitude and the duration of the deep breaths varied slightly within subjects; four showed a weak trend towards increasing size of the inspirations, whereas the other two subjects showed a trend towards a decrease. On a group basis the amplitude of the deep breath was unchanged throughout the course of the experiments. The respiratory frequency and amplitude between deep breaths was stable throughout the study (Fig. 4). Although in some subjects the respiratory frequency and amplitude tended to decrease after each deep breath, on a group basis the effect was not statistically significant.

		View larger version [in this window] [in a new window]
Figure 4. Effects of a deep breath on respiration amplitude (logarithmic scale) and frequency, skin sympathetic activity recorded in the peroneal nerve (logarithmic scale), laser Doppler perfusion within the innervation territory of the impaled nerve fascicle in the foot, mean blood pressure and calculated vascular resistance Averaged data from six subjects, each taking approximately thirty deep breaths. Note that the deep breath (starting at vertical dotted line) caused an initial increase followed by a reduction of sympathetic activity. Significance levels: * P < 0·05, ** P < 0·01, *** P < 0·001.

Figure 4 summarizes deep breath-induced mean changes of nerve traffic, skin perfusion, blood pressure and vascular resistance from all six subjects, and Fig. 5 shows examples from two subjects. When the subjects were asked to take a deep breath, a sympathetic neural discharge often occurred before the beginning of the breath, presumably as an arousal response (Fig. 5). This is seen in the average record as an increase in the 5 s epoch immediately prior to the breath. In conjunction with the breath there was always a single strong neural discharge, seen in Fig. 4 as a further increase of activity during the 5 s epoch after the start of the breath (note logarithmic scale in Fig. 4, top trace). Immediately after the respiration-related increase there was a reduction of nerve traffic for about 15 s, followed by a gradual return to the control level.

The deep breath was followed by an initial reduction in skin perfusion with a corresponding (non-significant) increase in vascular resistance. Subsequently there was an increase in perfusion and a reduction of vascular resistance (P < 0·01), which reached its maximum 30-40 s after the start of the breath. The deep breath was also followed by a transient reduction of skin electrical resistance (Fig. 5A and B). When comparing the number of transient skin resistance reductions before and after deep breaths, there was an increase after the breaths in two, a decrease in two and no change in two subjects, the mean change being an increase of 131 % (range 71-248 %).

		View larger version [in this window] [in a new window]
Figure 5. Effects of a single deep breath on skin sympathetic activity (mean voltage record, arbitrary units), skin resistance changes (reduction upwards) and laser Doppler perfusion (PU, perfusion units) within the neural innervation territory, and blood pressure in two subjects denoted A and B Note that in both cases the deep breath-induced neural burst was preceded by a large, presumably arousal-induced discharge. The vertical dotted line indicates the start of the deep breath.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

There were three main findings in the present study: (1) bretylium treatment eliminated the cutaneous vasoconstriction and subsequent vasodilatation that were induced by a deep breath in cold subjects, (2) multi-unit skin sympathetic activity first increased and then decreased after a deep breath in the cold state, and (3) after bretylium treatment, a deep breath in warm subjects led to a short-lasting increase in cutaneous perfusion and reduction in vascular resistance.

Effects of a deep breath in cold subjects

Vascular resistance. In agreement with a previous report (Oberle et al. 1988), the effect of a deep breath on blood flow in untreated skin differed depending on the thermoregulatory state of the subject: in the warm state, a short-lasting vasoconstriction occurred, whereas in the cold state an initial vasoconstriction was followed by a transient vasodilatation. In contrast, in bretylium-treated skin in the cold state, the deep breath had no effect on vascular resistance. Since the perfusion changes in NaCl-treated skin were similar to those in the untreated control area, the elimination of changes of perfusion and vascular resistance in bretylium-treated skin must have been bretylium dependent, and not non-specific effects of the electrical current.

Bretylium is known to eliminate effects of -adrenergic nerve activity by blocking noradrenaline release (Boura & Green, 1959; Haeusler et al. 1969) and iontophoretic application has been validated as a way of blocking adrenergic vasoconstrictions in the human skin (Kellogg et al. 1989). To verify the completeness of the bretylium block, we rewarmed the subjects after the cooling procedure, the requirement for inclusion being that after rewarming, a deep breath did not evoke a vasoconstrictor response. The fact that general cooling induced a marked reduction of baseline perfusion even in bretylium-pretreated skin does not invalidate our supposition regarding the completeness of the block. Prolonged general cooling leads to vasoconstriction, not only in small diameter vessels in the skin, but also of larger supply vessels, and this will lead to an overall downstream flow reduction. Such an effect would not be influenced by elimination of neurally mediated vasoconstriction in a small local area of skin and therefore a marked flow reduction is to be expected in the bretylium-treated area.

We interpret the lack of change of perfusion and vascular resistance, after a deep breath in bretylium-treated skin in cold subjects, as a strong indication that both the vasoconstrictor and the vasodilator components of the response in untreated skin were due to changes in sympathetic vasoconstrictor activity. The inference is that the initial increase of vascular resistance was due to an increase, and the subsequent decrease of resistance to a decrease of vasoconstrictor nerve traffic.

Sympathetic nerve traffic. Multi-unit skin sympathetic activity may contain both vasoconstrictor and sudomotor impulses, which can be recorded in the same electrode site (Hagbarth et al. 1972; Bini et al. 1980a). Although the bretylium and nerve recording experiments of the present study were made in different extremities, it is probably valid to compare the data since simultaneous double nerve recordings indicate a high degree of similarity between sympathetic traffic to the skin of hand and foot (Bini et al. 1980b).

In our microneurography experiments, the deep breath evoked an initial short-lasting increase of nerve traffic to the foot, followed by a more long-lasting decrease. These neural effects were associated with similar perfusion and resistance changes, as occurred in the control areas in the hand in the bretylium experiment. Although in cold subjects, spontaneous skin sympathetic activity is dominated by vasoconstrictor impulses (Bini et al. 1980a), two findings indicated the presence of sudomotor impulses in our neurograms. First, the deep breath evoked a neural discharge that was followed by a reduction in skin electrical resistance in all subjects. Second, in four of our subjects there was evidence of occasional spontaneous sudomotor discharges at rest. The first finding suggests that the strong neural discharge evoked by the deep breath contained both sudomotor and vasoconstrictor impulses. This is in agreement with previous results (Hagbarth et al. 1972). The second finding raises the possibility that the reduction of nerve traffic after the deep breath was due to a reduction of sudomotor impulses. However, this is unlikely: no evidence of sudomotor traffic was present in two subjects and in the remaining four the nerve activity decreased regardless of whether the number of skin resistance deflections increased or decreased. Thus we conclude that the transient reduction of sympathetic activity after the deep breath was due, in all probability, primarily to a reduction of vasoconstrictor nerve traffic.

To our knowledge, a reduction of cutaneous vasoconstrictor activity after a deep breath has not been described previously. However, in direct nerve recordings in anaesthetized cats innocuous stimuli applied to the skin led to an initial activation followed by a reduction of skin vasoconstrictor activity (Horeyseck & Jänig, 1974), i.e. a pattern similar to that found in the present study. In the same species, noxious stimulation also led to a reduction of the vasoconstrictor drive, but this was usually not preceded by activation (Jänig, 1975; Grosse & Jänig, 1976). In the anaesthetized rat, a biphasic excitation-inhibition response to noxious stimuli has been recorded from sympathetic fibres in the splanchnic nerve (Elam et al. 1986). The mechanism underlying our present finding of a reduced vasoconstrictor activity after a deep breath is unclear, but a non-specific post-excitatory depression is a possibility. The physiological role of the increased blood flow after the deep breath is also unclear.

Effect of a deep breath in warm subjects

After rewarming, there was a transient fall in cutaneous vascular resistance in bretylium-treated skin 7-17 s after the start of the breath. In contrast, vascular resistance increased in both control areas. The reason for the earlier appearance of the vasodilatation in the cold state than after rewarming may be a more sluggish vascular contractile mechanism in the cold state. However, it may be that different mechanisms were involved in the two types of vasodilatation.

The mechanism underlying the vasodilatation after rewarming is unclear. A passive redistribution of blood into the bretylium-treated vessels from neighbouring vessels that constricted is conceivable. A more likely possibility is that in the warm state a deep breath activates both a vasoconstrictor and an active vasodilator reflex mechanism in the skin of the human hand and foot. If so, the vasodilator component has not been observed in previous studies, presumably because the vasoconstrictor component normally dominates. Such opposing neurally mediated influences that vary in relative strength would, however, be a possible mechanism contributing to the often marked variability of response amplitudes between different breaths.

In several animal species there is evidence of sympathetic vasodilator fibres innervating cutaneous blood vessels (including arterio-venous shunts) in the extremities (for references, see Bell & Robbins, 1997). Physiological data suggest that the neurotransmitters may include acetylcholine, histamine and dopamine (Brody & Shaffer, 1970; Bell & Lang, 1979), but immunohistochemical support is less convincing (Gibbins, 1997). In humans, electrical stimulation of the sympathetic chain may evoke cutaneous vasodilatation in the foot (Lundberg et al. 1989), but whether this was due to activation of vasodilator fibres or secondary to sudomotor fibre activation could not be determined. The same uncertainty also applies to the present study: in addition to the reduction in skin vascular resistance, the deep breath also evoked a transient reduction of skin electrical resistance, indicating sudomotor fibre activation.

Limitations of the study

In part, the blood pressure increase seen in the cold state may be an effect of the increases of skin (Bini et al. 1980a) and muscle (Fagius & Kay, 1991) sympathetic vasoconstrictor activity, which are known to occur during generalized cooling. However, we cannot exclude the possibility that the increase in blood pressure level was, in part, a methodological artifact. Blood pressure was recorded by the volume clamp method by using a cuff around a finger: this method provides reliable measures of changes of blood pressure, but data on the absolute level of blood pressure is more uncertain (Parati et al. 1989; Imholz et al. 1990). In cold subjects, finger blood flow is sometimes reduced to such an extent that finger blood pressure recordings cannot be obtained. Although this did not occur in the present study, the increase in blood pressure level in the cold state may in part be related to the large cooling-induced reduction of finger blood flow. On the other hand, there is no reason to suspect that the short-lasting deep breath-induced changes in blood pressure found in the cold state were contaminated by major artifacts, such as a change in the direction of stimulus-induced blood pressure changes.

Clinical diagnostic consequences

The amplitude of the cutaneous blood flow reduction induced by a deep breath has been used clinically as an indirect measure of the strength of the sympathetic vasoconstrictor discharge (Low et al. 1983). The present results suggest that this test has substantial limitations. In comfortably warm subjects there is no spontaneous cutaneous vasoconstrictor nerve traffic (Bini et al. 1980a), blood flow is fairly high and therefore the vasoconstrictor response to a deep breath will be large. In cool subjects with weak to moderate spontaneous vasoconstrictor nerve traffic and a correspondingly reduced basal skin blood flow, one would still expect the deep breath test to evoke an initial vasoconstrictor response (but with reduced amplitude), probably followed by some vasodilatation (due to the inhibition of vasoconstrictor nerve traffic shown here). However, if the test is used in subjects with skin temperatures below 25°C, basal skin blood flow will be low and then the vasoconstrictor response may be difficult to recognize (as in the untreated control site in the hand in the bretylium experiments, and in the foot in the microneurography experiments of the present study). In this situation, the deep breath may induce only a transient vasodilatation. Thus the deep breath test may be of diagnostic value only if used in comfortably warm subjects without spontaneous skin vasoconstrictor nerve traffic. As soon as such traffic is present, the amplitude of the vasoconstrictor response will become an unreliable measure of the strength of the neural drive.

To summarize, the present study indicates that a deep breath may lead to vasodilatation by two different mechanisms, passively in cold subjects (reduction of vasoconstrictor activity), and actively in warm subjects (recruitment of impulses causing vasodilatation). The complexity of the vascular resistance changes highlights the need for rigorous temperature control if cutaneous vascular reflexes are to be used to evaluate sympathetic neural function.

	REFERENCES

Top Abstract Introduction Methods Results Discussion References

Bell, C. & Lang, W. J. (1979). Evidence for dopaminergic vasodilator innervation of the canine paw pad. British Journal of Pharmacology 67, 337-343	[Medline]
Bell, C. & Robbins, S. (1997). Autonomic vasodilatation in the skin. In Autonomic Innervation of the Skin, ed. Morris, J. L. & Gibbins, I. L., pp. 87-110. Harwood Academic Publishers, The Netherlands.
Bini, G., Hagbarth, K.-E., Hynninen, P. & Wallin, B. G. (1980a). Thermoregulatory and rhythm-generating mechanisms governing the sudomotor and vasoconstrictor outflow in human cutaneous nerves. The Journal of Physiology 306, 537-552	[Abstract]
Bini, G., Hagbarth, K.-E., Hynninen, P. & Wallin, B. G. (1980b). Regional similarities and differences in thermoregulatory vaso- and sudomotor tone. The Journal of Physiology 306, 553-565	[Abstract]
Boura, A. L. A. & Green, A. F. (1959). The actions of bretylium: Adrenergic neurone blocking and other effects. British Journal of Pharmacology 14, 536-548.
Brody, M. J. & Shaffer, R. A. (1970). Distribution of vasodilator nerves in the canine hindlimb. American Journal of Physiology 218, 470-474	[Medline]
Elam, M., Svensson, T. & Thorén, P. (1986). Locus coeruleus neurones and sympathetic nerves: Activation by cutaneous sensory afferents. Brain Research 366, 254-261	[Medline]
Fagius, J. & Kay, R. (1991). Low ambient temperature increases baroreflex-governed sympathetic outflow to muscle vessels in humans. Acta Physiologica Scandinavica 142, 201-209	[Medline]
Gibbins, I. L. (1997). Autonomic pathways to cutaneous effectors. In Autonomic Innervation of the Skin, ed. Morris, J. L. & Gibbins, I. L., pp. 1-56. Harwood Academic Publishers, The Netherlands.
Grant, R. T. & Holling, H. E. (1938). Further observations on the vascular responses of the human limb to body warming; evidence for sympathetic vasodilator nerves in the normal subject. Clinical Science 3, 273-285.
Grosse, M. & Jänig, W. (1976). Vasoconstrictor and pilomotor fibres in skin nerves to the cat's tail. Pflügers Archiv 361, 221-229	[Medline]
Haeusler, G., Haefely, W. & Huerlimann, A. (1969). On the mechanism of adrenergic nerve blocking action of bretylium. Naunyn-Schmiedeberg's Archives of Pharmacology 265, 260-277.	[Medline]
Hagbarth, K.-E., Hallin, R. G., Hongell, A., Torebjörk, H. E. & Wallin, B. G. (1972). General characteristics of sympathetic activity in human skin nerves. Acta Physiologica Scandinavica 84, 164-176	[Medline]
Horeyseck, G. & Jänig, W. (1974). Reflexes within postganglionic fibres within skin and muscle nerves after noxious stimulation of skin. Experimental Brain Research 20, 125-134	[Medline]
Imholz, B. P. M., Settels, J. J., Van Der Meiracker, A. H., Wesselin, K. H. & Wieling, W. (1990). Non-invasive continuous finger blood pressure measurement during orthostatic stress compared to intra-arterial pressure. Cardiovascular Research 24, 214-221	[Medline]
Jänig, W. (1975). Central organization of somatosympathetic reflexes in vasoconstrictor neurones. Brain Research 85, 305-312.
Kellogg, D. L., Johnson, J. M. & Kosiba, W. A. (1989). Selective abolition of adrenergic vasoconstrictor responses in skin by local iontophoreses of bretylium. American Journal of Physiology 257, H1599-1606	[Medline]
Kirnö, K., Kunimoto, M., Lundin, S., Elam, M. & Wallin, B. G. (1991). Can galvanic skin response be used as a quantitative estimate of sympathetic nerve activity in regional anesthesia? Anesthesia and Analgesia 73, 138-142	[Abstract]
Kunimoto, M., Kirnö, K., Elam, M., Karlsson, T. & Wallin, B. G. (1992). Non-linearity of skin resistance response to intraneural electrical stimulation of sudomotor nerves. Acta Physiologica Scandinavica 146, 385-392	[Medline]
Low, P. A., Neumann, C., Dyck, P. J., Fealey, R. D. & Tuck, R. R. (1983). Evaluation of skin vasomotor reflexes using Doppler velocimetry. Mayo Clinic Proceedings 58, 583-586	[Medline]
Lundberg, J., Norgren, L, Ribbe, E., Rosén, I., Steen, S., Thörne, J. & Wallin, B. G. (1989). Direct evidence of active sympathetic vasodilatation in the skin of the human foot. The Journal of Physiology 417, 437-446	[Abstract]
Oberle, J., Elam, M., Karlsson, T. & Wallin, B. G. (1988). Temperature dependent interaction between vasoconstrictor and vasodilator mechanisms in human skin. Acta Physiologica Scandinavica 132, 459-469	[Medline]
Parati, G., Casadei, R., Groppeli, A., Di Rienzo, M. & Mancia, G. (1989). Comparison of finger and intra-arterial blood pressure monitoring at rest and during laboratory testing. Hypertension 13, 647-655	[Abstract]
Rowell, L. B. (1977). Reflex control of the cutaneous vasculature. Journal of Investigative Dermatology 69, 154-166	[Medline]
Vallbo, Å. B., Hagbarth, K.-E., Torebjörk, H. E. & Wallin, B. G. (1979). Somatosensory, proprioceptive and sympathetic activity in human peripheral nerves. Physiological Reviews 59, 919-957	[Medline]

Acknowledgements

Supported by the Swedish Medical Research Council Grant 3546. P. Kienbaum was supported by a grant from the Deutsche Forschungsgemeinschaft (DFG Ki 629/1.1).

Corresponding author

B. G. Wallin: Department of Clinical Neurophysiology, Sahlgren University Hospital, S-413 45 Göteborg, Sweden.

Email: gunnar.wallin{at}nfys.gu.se

Author's present address

P. Kienbaum: Department of Anaesthesiology and Intensive Care, University Clinic, Essen, Germany.

This Article Abstract Full Text (PDF) Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Wallin, B. G. Articles by Elam, M. Search for Related Content PubMed PubMed Citation Articles by Wallin, B. G. Articles by Elam, M.