Journal of Applied Physiology
Vol. 81, No. 5, pp. 2192-2198, November 1996
CONTROL OF BREATHING, CIRCULATION, AND TEMPERATURE

Baroreceptor control of the cutaneous active vasodilator system

C. G. Crandall, J. M. Johnson, W. A. Kosiba, and D. L. Kellogg Jr.

Departments of Physiology and Medicine, University of Texas Health Science Center at San Antonio, San Antonio, Texas 78284

ABSTRACT

INTRODUCTION

METHODS

RESULTS

DISCUSSION

ACKNOWLEDGEMENTS

FOOTNOTES

REFERENCES

ABSTRACT

Crandall, C. G., J. M. Johnson, W. A. Kosiba, and D. L. Kellogg, Jr. Baroreceptor control of the cutaneous active vasodilator system. J. Appl. Physiol. 81(5): 2192-2198, 1996.We sought to identify whether reductions in cutaneous active vasodilation during simulated orthostasis could be assigned solely to cardiopulmonary or to carotid baroreflexes by unloading cardiopulmonary baroreceptors with low levels of lower body negative pressure (LBNP) or unloading carotid baroreceptors with external pressure applied over the carotid sinus area [carotid pressure (CP)]. Skin blood flow was measured at a site at which adrenergic function was blocked via bretylium tosylate iontophoresis and at an unblocked site. During LBNP of 5 and 10 mmHg in hyperthermia, neither heart rate (HR) nor cutaneous vascular conductance (CVC) at either site changed (P > 0.05 for both), whereas forearm vascular conductance (FVC) was reduced (5 mmHg: from 21.6 ± 4.8 to 19.8 ± 4.1 FVC units, P = 0.05; 10 mmHg: from 22.3 ± 4.0 to 19.3 ± 3.7 FVC units, P = 0.002). LBNP of 30 mmHg in hyperthermia reduced CVC at both sites (untreated: from 51.9 ± 5.7 to 43.2 ± 5.1% maximum, P = 0.02; bretylium tosylate: from 60.9 ± 5.4 to 53.2 ± 4.4% maximum, P = 0.02), reduced FVC (from 23.2 ± 3.6 to 18.1 ± 3.3 FVC units; P = 0.002), and increased HR (from 83 ± 4 to 101 ± 3 beats/min; P = 0.003). Pulsatile CP (45 mmHg) did not affect FVC or CVC during normothermia or hyperthermia (P > 0.05). However, HR and mean arterial pressure were elevated during CP in both thermal conditions (both P < 0.05). These results suggest that neither selective low levels of cardiopulmonary baroreceptor unloading nor selective carotid baroreceptor unloading can account for the inhibition of cutaneous active vasodilator activity seen with simulated orthostasis.

skin blood flow; baroreflex; human; cardiopulmonary baroreceptors; carotid; temperature regulation; peripheral circulation; bretylium

INTRODUCTION

IN THERMALLY NEUTRAL ENVIRONMENTS, skin receives ~5-10% of cardiac output, whereas in conditions of heat stress, skin blood flow (SkBF) can reach 50-70% of cardiac output, approaching 8 l/min (12). This increase in SkBF is accomplished initially through withdrawal of sympathetic vasoconstrictor activity and, in nonglabrous areas, by increased sympathetic vasodilator activity. The active vasodilator component mediates the majority of the reflex thermoregulatory increase in SkBF (23). In addition to these thermoregulatory reflexes, SkBF is also controlled by nonthermoregulatory reflexes, including baroreflexes (12, 15, 25, 30). For example, Kellogg et al. (15) used bretylium iontophoresis to block adrenergic vasoconstrictor activity and measured cutaneous vascular conductance (CVC) at unblocked and bretylium-treated sites during 40 mmHg lower body negative pressure (LBNP) in normothermia and hyperthermia. They found reductions in CVC induced by this level of LBNP in normothermia to be due to enhanced vasoconstrictor activity, whereas in hyperthermia, reductions in CVC were due primarily to withdrawal of cutaneous active vasodilator activity. Thus baroreceptor-mediated control of CVC can be important in the maintenance of blood pressure, particularly in hyperthermia, when the skin represents a significant fraction of the total vascular conductance.

The mechanisms controlling SkBF in response to a challenge to blood pressure may depend on the degree of unloading of the individual baroreceptor populations (i.e., cardiopulmonary and sinoaortic baroreceptors). Moreover, the control by individual baroreceptor populations of the cutaneous vasoconstrictor and active vasodilator systems may differ. Cardiopulmonary baroreceptor unloading via low levels of LBNP (13, 19) in normothermia yielded results suggesting cutaneous vasoconstriction in some (30), but not all (2, 32), studies. When greater levels of LBNP are applied, unloading a combination of cardiopulmonary and sinoaortic baroreceptors, substantial reductions in SkBF are more consistently observed (12, 15, 25, 30). On the other hand, muscle blood flow shows a much more consistent vasoconstriction in response to low levels of LBNP. A question that naturally arises from these observations is which baroreceptor population is mediating these changes in SkBF and whether this is accomplished through modulation of the cutaneous vasoconstrictor and/or active vasodilator systems. The purpose of the present study was to identify whether selective unloading of cardiopulmonary or of carotid baroreceptors is capable of reducing cutaneous active vasodilator activity.

METHODS

Subjects

Experimental Procedures

Internal temperature was monitored with a thermocouple placed sublingually (T_sl). T_sk was monitored from the electrical average of six thermocouples placed at representative sites on the body surface (28). The subject wore a two-piece tube-lined suit that permitted alterations in T_sk of both the upper and lower body by changing the temperature of the water perfusing the suit. The suit covered the entire body surface with the exception of the head, feet, and arms.

Data Collection and Statistics

RESULTS

Protocol 1

Table 1. Hemodynamic and skin temperatures before and during LBNP LBNP Stage  30 mmHg P Value  5 mmHg (HS) P Value  10 mmHg (HS) P Value  30 mmHg (HS) P Value MAP, mmHg   Pre-LBNP 79 ± 2  71 ± 2  74 ± 3  74 ± 3    LBNP 77 ± 2  0.12 73 ± 2  0.31 74 ± 3  0.52 71 ± 5  0.39 HR, beats/min   Pre-LBNP 59 ± 3  81 ± 6  82 ± 5  83 ± 4    LBNP 68 ± 2  <0.01 83 ± 6  0.13 85 ± 6  0.26 101 ± 3  <0.01 FVC, units   Pre-LBNP 6.1 ± 1.1  21.6 ± 4.8  22.3 ± 4.0  23.2 ± 3.6    LBNP 3.6 ± 0.8  <0.01 19.8 ± 4.1  0.05 19.3 ± 3.7  <0.01 18.1 ± 3.3  <0.01 Tsk, °C   Pre-LBNP 34.67 ± 0.20  38.10 ± 0.17  38.11 ± 0.19  38.05 ± 0.18    LBNP 34.61 ± 0.19  0.12 38.03 ± 0.17  0.15 38.10 ± 0.19  0.6 37.97 ± 0.21  0.3 Values are means ± SE. LBNP, lower-body negative pressure; HS, heat stress; MAP, mean arterial pressure; HR, heart rate; FVC, forearm vascular conductance; Tsk, skin temperature. P values are for comparison between pre-LBNP and LBNP periods.

Increasing T_sk from 34.6 ± 0.1 to 38.1 ± 0.2°C led to an increase in T_sl from 36.59 ± 0.06 to 37.12 ± 0.14°C and engaged the cutaneous active vasodilator system because CVC at bretylium-treated sites was significantly increased. During hyperthermia, LBNP at 5 mmHg did not significantly change CVC at the untreated site (from 52.6 ± 6.4 to 51.5 ± 6.4% maximum CVC; P = 0.15) or at the bretylium-treated site (from 63.8 ± 7.1 to 64.4 ± 6.5% maximum CVC; P = 0.58; see Fig. 1). Similarly, LBNP at 10 mmHg did not change CVC at either site (untreated: from 51.6 ± 6.3 to 50.4 ± 5.6% maximum CVC, P = 0.38; bretylium-treated: from 60.7 ± 6.4 to 60.9 ± 6.4% maximum CVC, P = 0.79). LBNP at 30 mmHg significantly reduced CVC both at untreated sites (from 51.9 ± 5.7 to 43.2 ± 5.1% maximum CVC; P = 0.02) and at bretylium-treated sites (from 60.9 ± 5.4 to 53.2 ± 4.4% maximum CVC; P = 0.02). These reductions were not different between sites (P = 0.63), suggesting that the reduction in CVC at the untreated site was primarily due to a reduction in active vasodilator activity. There were no significant changes in HR (Table 1) during LBNP trials at 5 or 10 mmHg. HR significantly increased during LBNP at 30 mmHg in hyperthermia. FVC (Table 1) was significantly reduced by all levels of LBNP. No significant change in T_sk occurred during LBNP in either thermal condition (P  0.12; see Table 1).

Protocol 2

Application of CP during hyperthermia raised both HR (from 86.9 ± 6.7 to 95.4 ± 6.2 beats/min; P = 0.002) and MAP (from 78.2 ± 3.5 to 89.1 ± 4.1 mmHg; P = 0.005), whereas CVC at untreated sites (from 77.0 ± 6.1 to 75.4 ± 5.9% maximum CVC; P = 0.6) and at bretylium-treated sites (from 62.8 ± 5.2 to 65.2 ± 5.6% maximum CVC; P = 0.5) as well as FVC (from 19.7 ± 4.7 to 19.2 ± 4.2 FVC units; P = 0.6) were unaffected (Figs. 2 and 3).

DISCUSSION

A number of studies have shown that the cutaneous circulation participates in the baroreflex-mediated responses to actual or simulated orthostasis (e.g., Refs. 12, 15, 25, 30). In normothermia, vasoconstrictor activity is increased, whereas in hyperthermia cutaneous active vasodilation is withdrawn during simulated orthostasis (15). However, the baroreceptor population(s) initiating these responses is not known. The primary purpose of this study was to test whether the mechanism for the withdrawal of active vasodilator activity is mediated solely by cardiopulmonary or carotid baroreflexes. Our findings do not support these singular mechanisms. Cardiopulmonary baroreceptor unloading in hyperthermia with 5 and 10 mmHg LBNP did not evoke reductions in CVC. Similarly, selectively decreasing carotid sinus transmural pressure in either normothermia or hyperthermia did not elicit responses in CVC despite increases in HR and MAP, reflective of a carotid baroreflex. When lower extremity pooling of blood was further increased with 30 mmHg LBNP, significant and equal reductions in CVC were observed at both the untreated and bretylium-treated sites.

Application of low levels of LBNP was selected to unload specifically the cardiopulmonary baroreceptors, whereas 30 mmHg LBNP was chosen to unload both the cardiopulmonary and sinoaortic baroreceptors. In normothermia, graded levels of LBNP up to about 20 mmHg cause limb and splanchnic vasoconstriction, but they do so in the absence of measurable changes in aortic mean pressure, pulse pressure, or rate of pressure change, signals important to arterial baroreceptors (13, 19). There is also no increase in HR with these levels of LBNP (13, 19). At more severe levels (beyond 20 LBNP), aortic pulse pressure narrows and HR increases (13). For this reason, it is generally accepted that increases in HR indicate a change in the signal to the arterial baroreceptors (19), an assumption adopted in this study. However, the possibility that the aortic and carotid baroreceptors were affected by 5 and 10 mmHg LBNP cannot be ruled out because these perturbations have been shown to change aortic pulse area (27) and carotid arterial diameter (16). Whether such changes are sufficient to evoke sinoaortic baroreflexes is unknown. Nevertheless, neither reductions in CVC nor increases in HR with these low levels of LBNP were observed.

Despite the lack of changes in CVC or HR at 5 and 10 mmHg LBNP in hyperthermia, significant reductions in FVC occurred. Therefore, we are confident that cardiopulmonary baroreceptors were sufficiently unloaded to evoke efferent responses, presumably mediating vasoconstriction of noncutaneous vasculature in the forearm. Such an observation of a preferential efferent response during cardiopulmonary baroreceptor unloading is not new because Tripathi and Nadel (30) made similar observations during graded LBNP in normothermia.

In contrast to results from 5 and 10 mmHg LBNP in hyperthermia, CVC decreased and HR increased during 30 mmHg LBNP in both normothermia and hyperthermia. In normothermia, decreases in CVC during LBNP were solely attributable to increased sympathetic adrenergic activity because there were no significant changes in CVC at the site lacking a functional sympathetic adrenergic vasoconstrictor system (i.e., the bretylium-treated site). This lack of change in CVC at the bretylium-treated site during LBNP suggests that changes in the control of SkBF occur within the area of blockade. Therefore, reductions in SkBF at the untreated site are not due to vasoconstriction of larger vessels upstream. In hyperthermia, LBNP at 30 mmHg caused equal reductions in CVC at the untreated and bretylium-treated sites, suggesting this reduction in CVC in both cases was mediated largely or entirely by withdrawal of cutaneous active vasodilator activity, in keeping with earlier findings by Kellogg et al. (15).

Vissing et al. (32) suggested the reductions in SkBF during LBNP in hyperthermia were due to LBNP-induced reductions in T_sk and not to baroreceptor reflexes because in their studies when T_sk was held constant, no changes in SkBF were observed during LBNP. In the present study, however, T_sk was maintained constant throughout all LBNP procedures (see Table 1) and significant decreases in CVC were observed during the LBNP trials at 30 mmHg. However, in agreement with Vissing et al., our findings suggest that low levels of LBNP in hyperthermia do not affect CVC.

Previously, it was not known whether the carotid baroreflex contributed to the reduction in CVC during general baroreceptor unloading in hyperthermia. To address this question, we applied CP to the subjects in both normothermia and hyperthermia to identify whether the carotid baroreflex has an efferent limb governing SkBF because this stimulus has been shown to selectively unload the carotid baroreceptor and evoke appropriate baroreflex-mediated HR and MAP responses (7). We did not observe changes in CVC during the CP trials. Because unloading of carotid baroreceptors during LBNP at 30 mmHg should be far more subtle than during the CP procedure, the reduction in CVC observed during higher levels of LBNP was not, independently, the result of carotid baroreceptor unloading.

Our data suggesting a lack of significant control of SkBF by the carotid baroreflex are consistent with the results of others. For example, unlike muscle sympathetic nerve activity, skin sympathetic nerve activity often does not have the detectable cardiac rhythm suggestive of modulation by the arterial baroreceptors (11). Furthermore, carotid sinus nerve stimulation did not measurably change skin sympathetic nerve activity (34), suggesting that neural control of SkBF is not under significant carotid baroreflex control. In contrast, Beiser et al. (3) evoked reductions in estimated skin vascular resistance with increased carotid baroreceptor transmural pressure. Beiser et al. estimated forearm SkBF via its elimination by epinephrine iontophoresis in one arm. We are unsure how to explain the differences in results between the present study (or those involving measurements of skin sympathetic nerve activity) and those reported by Beiser et al. They may relate to the direction of application of the stimulus (i.e., carotid distention rather than compression) or the method of assessing SkBF. Nevertheless, we are confident that we evoked a carotid baroreflex, as evidenced by changes in HR and MAP. We do not rule out the possibility that more prolonged or more marked carotid unloading might have measurable effects on the skin circulation and the active vasodilator system.

Unloading the carotid sinus with 45 mmHg CP in normothermia and hyperthermia also did not significantly reduce FVC. These results are surprising in light of data suggesting muscle sympathetic nerve activity to increase during this procedure (10, 22, 33). Previous findings regarding the effects of carotid sinus baroreceptor perturbations on reflex control of the forearm vasculature are mixed, with some studies reporting responses in forearm vasculature (3, 6, 17, 31) but others finding no change (1, 4, 8, 24). These differences in results may relate to the time during the carotid stimulation in which the data were obtained because some studies show evanescent efferent responses to neck pressure and/or suction (6, 17, 33). Differences may also be attributed to neck collar design, neck pressures applied, and mode of pressure delivery. Regardless, in the present study a carotid baroreflex was evoked, as evidenced by the changes in MAP and HR, but this same stimulus did not cause a change in CVC.

It is possible that the relative contribution of the aortic baroreflex in modulating CVC is greater than that of the carotid baroreflex and that the aortic baroreceptors are the primary baroreceptor population governing CVC during conditions such as orthostasis. Disproportionate control of an efferent response by the carotid and aortic baroreceptors has previously been identified. For example, in humans it has been shown that the aortic baroreceptors have a greater role than the carotid baroreceptors in controlling HR (9, 18) and muscle sympathetic nerve activity (26). Unfortunately, procedures used to selectively activate and/or deactivate the aortic baroreflex in humans (5, 9, 26) cannot address the questions asked in this study because vasoactive drugs are administered, which would confound our SkBF measurements. Although this problem might be addressed by using skin sympathetic nerve activity as the dependent variable, it could prove difficult to distinguish active vasodilator activity from the recording of mixed nerve activity.

Although we did not see active vasodilator withdrawal at lower levels of LBNP, we did confirm that response at a more severe level. In humans it is difficult to unload the cardiopulmonary baroreceptors to a comparable level to that observed with 30 mmHg LBNP in hyperthermia without also unloading the sinoaortic baroreceptors. Nevertheless, if substantial cardiopulmonary baroreceptor unloading was required to evoke a cutaneous vascular response, then this threshold might not be reached until sinoaortic baroreceptors were also unloaded. This hypothesis suggests the cardiopulmonary baroreceptors mediate the change in CVC during high levels of LBNP, whereas concomitant unloading of sinoaortic baroreceptors has no affect on the cutaneous vasculature. Such a hypothesis would be difficult if not impossible to test in healthy humans.

Alternatively, it might be that a combination of cardiopulmonary and sinoaortic baroreceptor unloading is required to evoke a withdrawal of cutaneous active vasodilator activity. Previous studies suggest that unloading the cardiopulmonary baroreceptors increases the sensitivity of the sinoaortic baroreceptor reflex (20, 31). Thus the sinoaortic baroreceptors may contribute to the reduction in CVC during higher levels of LBNP, but the sensitivity of these baroreceptors must first be increased through reduced cardiopulmonary afferent firing.

In conclusion, data from the present study suggest that low levels of LBNP in hyperthermia, which primarily unload the cardiopulmonary baroreceptors, do not affect cutaneous vasoconstrictor or active vasodilator activities. Despite the absence of a cutaneous vascular response, these levels of LBNP significantly reduced FVC, suggesting that cardiopulmonary baroreflexes were engaged. Similarly, carotid baroreflex unloading initiated a carotid sinus baroreflex but did not evoke changes in CVC in normothermia or hyperthermia. Reductions in CVC during LBNP at 30 mmHg in both normothermia and hyperthermia must therefore be due to greater cardiopulmonary baroreceptor unloading, aortic baroreceptor unloading, or an interaction between cardiopulmonary and sinoaortic baroreceptor unloading.

ACKNOWLEDGEMENTS

The authors express their appreciation to the subjects for their willing participation.

FOOTNOTES

Address for reprint requests: J. M. Johnson, Dept. of Physiology, Univ. of Texas Health Science Center at San Antonio, San Antonio, TX 78284-7756.

Received 11 December 1995; accepted in final form 10 June 1996.

REFERENCES