| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1Heart and Vascular Institute and 2Department of Cellular and Molecular Physiology, General Clinical Research Center, Pennsylvania State University College of Medicine, Hershey; and 3Department of Pharmacology and Physiology, Drexel University College of Medicine, Philadelphia, Pennsylvania
Submitted 13 April 2007 ; accepted in final form 31 July 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Doppler ultrasound; cardiac output; renal vascular conductance; splanchnic vascular conductance
Modulation of renal and splanchnic vascular tone can exert profound effects on systemic arterial blood pressure. Rowell et al. (33, 34) observed increase in splanchnic vascular conductance and mean arterial pressure during skin cooling after whole body heating. It is important to emphasize that in these studies cooling was applied after whole body heating at a time when it is possible that the high internal temperature and skin blood flows are masking skin-surface cooling responses. Additionally, Raven et al. (30) suggested, but did not measure, that prolonged cold stress may increase splanchnic vascular conductance. Thus the effects of cold stress on renal and splanchnic vascular conductance remain unclear.
Many autonomic stressors that elicit systemic pressor responses are mediated in part by vasoconstriction in the splanchnic and renal circulation (17, 22–24). Thus vasoconstriction of the splanchnic and renal vasculatures may contribute to the cooling-induced pressor responses. Besides these implications on systemic blood pressure, cold-induced alterations in splanchnic blood flow may also provide insight into observations of intestinal ischemia during cold stress (28). Accordingly, we tested the hypothesis that skin-surface cooling induces renal and splanchnic vasoconstriction in humans. Such an effect would be important to understand the integrative cardiovascular response to cold stress in humans and the mechanism by which skin-surface cooling elicits a pressor response.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Experimental design. A total of 26 whole body skin-surface cooling trials were completed on the 14 subjects in the postabsorptive state (fasted for a minimum of 6 h). Twelve subjects performed two cold stress trials, and two subjects performed one cold stress trial. Skin-surface cooling consisted of perfusing 15–18°C water through a one-piece high-density tube-lined suit (Med-Eng Systems, Ottawa, ON, Canada) for 20 min. This suit covered the subject's entire body except the left arm, right hand, feet, and head; in addition, the suit contained two-way zippers that allowed access to the chest and abdomen for Doppler measures with minimal skin exposure except directly at the measurement site. The cold stress was designed to be as cool as possible to maximize blood pressure responses without evoking shivering. If preshivering tonus was observed water temperature was increased slightly (1–3°C) to abate this response. Thermoneutral conditions involved perfusing 35°C water through the same suit. Pilot data demonstrated that there were no significant differences in peripheral or visceral blood velocities in thermoneutral conditions over extended durations. Subjects who underwent multiple skin-surface cooling trials repeated the cold stress after 30–60 min of quiet supine rest in thermoneutral conditions to ensure hemodynamic parameters returned to baseline (precooling) levels. Pilot studies indicated that visceral blood velocities returned to baseline values in all subjects tested within this timeframe. Multiple cold-stress trials were utilized to accommodate up to four Doppler ultrasound measures per subject. Doppler measures were not randomized; we employed two Doppler machines and thus could only make four of five measurements on a particular subject (i.e., 2 measures during the first cooling and 2 measures during the second cooling). Initial anatomic scans were performed to identify appropriate vessels and intensity of signals; in some subjects, an adequate signal could not be obtained on a particular vessel and thus these vessels' velocities were not measured. The possible Doppler ultrasound measures on a particular subject included renal, celiac, superior mesenteric, and brachial artery blood velocity and average aortic time velocity integral for the assessment of cardiac output. All subjects were tested in the supine position and maintained this position for a minimum of 30 min before baseline data collection. Experiments were performed in a dimly lit, quiet laboratory maintained at 21–23°C.
Measurements. Heart rate was derived from an electrocardiogram. Arterial blood pressure was measured every 2 min over the brachial artery via an automated sphygmomanometer (Dinamap, General Electric, Waukesha, WI). Four-site mean skin temperature was measured and calculated (0.3 chest + 0.3 arm + 0.2 thigh + 0.2 leg) according to previous reports (29). Oral temperature was measured by securing a thermistor in the sublingual sulcus; subjects were instructed not to converse and to only nasally breathe during the baseline and whole body cooling periods. Under controlled experimental conditions, oral temperature has been demonstrated to responsively track internal temperature (8). Local skin blood flow was indexed via laser-Doppler flowmetry using 2-mm integrating flow probes (FlowLab, Moor Instruments, Wilmington, DE) attached to nonglabrous skin on the dorsal portion of the hand between the first and third metacarpal in an area devoid of large surface veins. In addition to brachial artery measures described below, forearm blood flow was also assessed via venous occlusion plethysmography (39). In brief, measures were made via mercury-in-Silastic strain gauges (Hokanson, Bellevue, WA) secured on the widest aspect of the forearm. Wrist cuff was then inflated to 200 mmHg to remove the hand circulation from the measure, and then an upper arm cuff was inflated to 40 mmHg and cycled every 15 s. Venous occlusion plethysmography was completed on the opposite trial from brachial and cutaneous Doppler measures.
Duplex ultrasound (model HDI 5000, ATL Ultrasound, Bothell, WA) was used to determine blood velocity in the brachial, celiac, superior mesenteric, and renal arteries and in the left ventricular outflow tract. Standard brachial artery and left ventricular outflow track measures were completed as previously described (1, 11). In brief, a 5- to 12-MHz linear transducer with a 6-MHz pulsed Doppler frequency was positioned over the brachial artery just above the antecubital fossa. The insonation angle for measuring mean blood velocity was 
60°. To minimize overestimation of mean blood velocity, the sample volume was adjusted to cover the size of the vessel. Average mean blood velocity was obtained from the Doppler waveform post hoc via software (HDI 5000). To measure arterial diameter a longitudinal view of the artery was taken. Arterial diameter measurements were made in end-diastole (determined by ECG) by measuring the distance between the lumen-intima interface in the near wall and the intima-media interface in the far wall. A 2- to 4-MHz phased array transducer with a pulsed Doppler frequency of 2 MHz was used for the measurement of stroke volume. The left ventricular outflow tract time velocity integral was measured from the apical long-axis view just proximal to the aortic annulus. Average aortic time velocity integral was obtained from the Doppler waveform post hoc via software (HDI 5000). The diameter of the left ventricular outflow tract was measured during systole in the parasternal long-axis view just below the aortic valve.
The visceral arteries were scanned using a curved-array transducer (2–5 MHz) with a 2.5-MHz pulsed Doppler frequency from the anterior abdominal approach. To obtain optimum velocity tracings, the transducer was held in a constant position and data were obtained in the same phase of the respiratory cycle with an insonation angle 60°. Mean blood velocity was obtained from the Doppler waveform post hoc via software (HDI 5000). Velocity measurements were expressed as centimeters per second. It should be noted that blood flow is a function of mean blood velocity and vessel cross-sectional area. Therefore, accurate measurements of vessel diameter are required to determine blood flow. This was possible for brachial artery and left ventricular outflow tract measurements, but due to the limited spatial resolution of the technique for deeper seated structures, accurate measurements of renal, celiac, and superior mesenteric artery diameter are difficult to perform in humans. However, these large conduit vessels show no significant change in diameter during pharmacological flow perturbations or changes in prandial state for the renal artery and celiac and superior mesenteric arteries, respectively (3, 21). Thus, changes in mean blood velocity likely correspond to the blood flow.
Data analysis. All continuous non-Doppler ultrasound data were recorded electronically at 60 Hz (model 16SP Powerlab, ADInstruments, New Castle, Australia). Vascular conductance of the renal artery, celiac artery, superior mesenteric artery, and dorsal hand skin was indexed by dividing Doppler-ultrasound velocity or laser-Doppler flux values by mean arterial blood pressure (MAP). Brachial artery blood flow (brachial blood velocity multiplied by brachial artery cross-sectional area and by 60) and forearm blood flow from venous occlusion plethysmography were expressed as brachial vascular conductance and forearm vascular conductance, respectively, by dividing blood flow by MAP. Stroke volume was calculated by multiplying left ventricular outflow tract time velocity integral and cross-sectional area, and once this value was obtained, multiplying by heart rate to obtain cardiac output. Systemic vascular conductance was calculated from the ratio of cardiac output and MAP. Minute averages were calculated for even minutes of skin-surface cooling for brachial, celiac, cutaneous, superior mesenteric, and renal vascular conductance, while forearm vascular conductance was calculated five times between minutes 2–4, 6–8, 10–12, 14–16, and 18–20. Vascular conductances were normalized to baseline values and expressed as percent change.
To identify possible differences in vascular conductance and temperature across the entire 20-min whole body cold stress, a repeated-measures ANOVA was utilized and Student-Newman-Keuls post hoc analysis was employed if significant main effects were observed. Other study variables (i.e., cardiac output, heart rate, stroke volume, and systemic vascular conductance) were compared between baseline and end cold stress using a paired t-test. No significant differences were noted between men and women in cold-induced changes in arterial blood pressure (P = 0.29) or any of the visceral vascular conductances (range: P = 0.45 to 0.92), and thus these gender data were collapsed for analysis purposes. There were 12 multiple measurements of internal and mean skin temperature and arterial blood pressure within the 14 subjects, but blood velocity parameters were independent (i.e., 1 velocity measurement per subject per vessel). A significance level of P < 0.05 was used for all tests. Results are presented in text and figures as means ± SD.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
This is the first report, to our knowledge, to identify renal vasoconstriction during nonnoxious nonhypothermic cold stress in humans. This renal vasoconstrictor response occurred within the first 4 min and was sustained throughout skin-surface cooling. Previous work in humans and rats noted decreases in renal blood flow during hypothermia (9, 25). However, hypothermia-related changes cannot be dissociated from decreases in cardiac output, MAP, and metabolic rate (37), and thus the role of vasoconstrictor responses during hypothermia is not clear. Noxious cooling of the face or hand also causes increases in arterial blood pressure and decreases in renal vascular conductance (6, 16). Local cooling of the hand to 7°C or below increases MAP, but temperatures above this value do not alter arterial blood pressure (20). The water temperatures used in this study did not approach this threshold, and thus it is unlikely that cold pain elicited the elevations in arterial blood pressure. Although the stimulus in skin-surface cooling is not related to cold pain, it is likely that the autonomic-induced vasoconstriction is mediated through nonnoxious cutaneous thermoreceptors known to be operable within the skin temperatures observed in this study (18, 27).
We also observed decreases in indexes of both celiac and superior mesenteric artery vascular conductance. These decreases in vascular conductance began at 4 min for the celiac artery and at 2 min for the superior mesenteric artery. These decreases persisted throughout the remainder of the 20-min skin-surface cooling trial. Again, we believe this time course is indicative of an autonomic vasoconstrictor reflex possibly mediated via cutaneous thermoreceptors. The celiac artery becomes the common hepatic, left gastric, and splenic arteries, whereas the superior mesenteric artery feeds portions of the small (jejunum and iliac) and large (ascending and transverse) intestine. Although we observed decreases in vascular conductance of the celiac and superior mesenteric arteries, it is not possible to determine whether there was preferential vasoconstriction of various downstream arteries. Previous research has identified splanchnic vasoconstriction during whole body heating and vasodilation during unspecified whole body cooling performed immediately after heating (34). Our data contrast the previous study. However, it is difficult to determine the precise effect of the whole body cold stress because of a previous heat-stress condition, because in many circumstances a previous heat stress can alter responses to a subsequent cold stress (12). Brauer, et al. (7) summarized data indicating that an estimate of splanchnic blood flow decreased during hypothermia (dog colonic temperatures decreased to 30 and 23°C) and that liver vascular conductance decreased during hypothermia (rat liver temperatures decreased to 30 and 20°C). Clearly, there are many differences between these severe cold stresses and this study's mild cold stress, but it is interesting that similar trends can be observed even during a mild cold stress before an observable change in internal temperature. Thus it is possible that skin-surface cooling may be able to be used as a model for testing questions related to such issues as cold-induced intestinal ischemia.
The visceral vasoconstriction observed in this study is of lower magnitude than that observed in peripheral tissues. However, it must be emphasized that during normothermic, resting, and fasting conditions, the renal and splanchnic circulations may account for 
20% and 25% respectively, whereas the cutaneous circulation only receives 4–5% of cardiac output. Thus even small changes in the conductance of the renal and splanchnic vascular beds may exert a significant hemodynamic impact. We observed a 40% decrease in brachial vascular conductance and an over 50% decrease in cutaneous vascular conductance during cold stress. Our use of Doppler ultrasound to assess brachial artery blood flow during whole body cooling confirms the observations in the present study and those of previous studies during cooling via venous occlusion plethysmography methodology (2, 5, 10). Using similar whole body cooling methodology, Durand et al. (15) reported similar magnitude of decreases in cutaneous vascular conductance to the present observations. It is likely that this decrease in cutaneous vascular conductance in dorsal hand skin during cold stress is mediated via increases in skin sympathetic nerve activity during whole body cooling (4, 14, 35). However, it is possible that areas under the tube-lined suit also experienced locally mediated vasoconstriction (19).
Subjects did not experience either overt shivering in our study or changes in cardiac output. If subjects experienced pronounced preshivering tonus, the water temperature perfusing the suits was increased slightly (1–3°C) to abate this response before any measurements were made. Previous whole body cooling studies using similar magnitude and durations of cold stress to this study did not cause an observable increase in multiple-site electromyography indicative of shivering (15). It is likely that previous studies that have observed increases in cardiovascular variables such as cardiac output may have resulted because of the increase in muscle contractions associated with shivering (30, 32). We did not observe significant differences in cardiac output or heart rate with whole body cooling in the supine position. This observation is often reported but is highly variable between reports (36, 38). It is possible that there is a temperature-dependent change in cardiac output and systemic vascular conductance that relates to shivering response during cooling. Shivering is associated with muscle contraction and increases in metabolic rate, which can be as much as a threefold increase in resting metabolic rate (42). Thus it is likely that cold stress without shivering does not increase cardiac output and can maintain decreases in systemic vascular conductance, whereas during cold-induced shivering cardiac output is elevated and systemic vascular conductance may increase from the initial vasoconstricted state to supply blood flow to these working muscles. These observations of skin-surface cooling follow reports from Muza et al. (26), who observed and assimilated that ventilatory and cardiovascular responses to air or water cold stresses are related to the degree that metabolic rate is increased and thus related to the degree to which internal temperature is changed. Skin-surface cooling, which is a milder cold stress, did not significantly alter internal temperature or cardiac output but did result in significant increases in arterial blood pressure and decreases in systemic vascular conductance.
Body composition data were not collected on study subjects. In addition to potential limitations of these measures, we do not believe that the percent body fat would have a profound effect on the data for multiple reasons. First, the vasoconstrictor responses of peripheral and visceral arteries observed in this study occurred within the first few minutes of skin-surface cooling, indicating more of a sympathetic reflex rather than a hypothermic response (or dramatic internal heat loss). Second, cutaneous vasoconstriction (as indexed by laser-Doppler flowmetry) and vasoconstriction to peripheral tissues (as indexed by brachial Doppler and venous occlusion plethysmography) provides the prime indicators of adjustable tissue insulation and thus the ability to change the core to skin temperature gradient (38).
Recently, whole body skin-surface cooling has been utilized as a countermeasure to increase orthostatic tolerance in humans (15, 40). Some of the mechanisms proposed for this preservation of cerebral blood velocity and tolerance include increases in central venous and arterial blood pressure (13, 40). Studies have also observed less of an increase in calf volume with whole body skin cooling (15, 41), indicating less blood pooling during orthostasis. Currently, it is not known whether the splanchnic and renal vasoconstriction observed in this study would contribute to the increase in arterial pressure during orthostatic stress during whole body cooling. Head-up tilt and lower body negative pressure increase splanchnic and renal vascular resistance (24, 33). Thus the data in the present study may provide an additional mechanism whereby skin-surface cooling augments systemic vascular resistance by further constricting both peripheral and visceral vascular beds during orthostatic stress.
Summary. This study identified reductions in indexes of vascular conductance in celiac, superior mesenteric, and renal arteries during skin-surface cooling, indicating vasoconstriction of these vessels. Peripheral vasoconstriction was also observed in brachial artery, total forearm, and cutaneous vasculature during skin-surface cooling. The vasoconstrictor responses occurred in the absence of significant alterations in heart rate, stroke volume, and cardiac output. Therefore, the integrative cardiovascular response to skin-surface cooling includes visceral and peripheral vasoconstriction, which occur independent of changes in cardiac output. Collectively, these hemodynamic responses are likely to contribute importantly to the observed pressor response to skin-surface cooling.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. L. Hess, T. E. Wilson, C. L. Sauder, Z. Gao, C. A. Ray, and K. D. Monahan Aging affects the cardiovascular responses to cold stress in humans J Appl Physiol, October 1, 2009; 107(4): 1076 - 1082. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
) from baseline of brachial vascular conductance (n = 10) and cutaneous vascular conductance of the dorsal hand (n = 12). Significant decreases from time 0 were observed in all subsequent time points, and no significant differences were observed between minutes 4 and 20. Time 0 denotes normothermic baseline; baseline brachial vascular conductance = 32 ± 15 cm·s–1·mmHg–1 and cutaneous vascular conductance = 0.93 ± 0.68 flux units/mmHg. *Significant difference from time 0; P < 0.05. #Significant difference from both time 0 and minute 2, P < 0.05.