INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

DURING HYPERTHERMIA, humans dissipate heat through evaporation of sweat and an increase in skin blood flow (SkBF), which convects heat from the body core to the surface. Reflex thermoregulatory control of SkBF is accomplished primarily through two opposing arms of the sympathetic nervous system: a noradrenergic vasoconstrictor system and an active vasodilator system, the latter of which is less well understood. As core temperature rises, vasoconstriction is withdrawn and the vasodilator system is activated. The active vasodilator system is responsible for most of the vasodilatory response to heat stress and can increase SkBF >10-fold.

During passive heating and dynamic exercise, men and women >60 yr of age have an attenuated SkBF response compared with younger individuals (2, 6, 20, 21, 28, 39). To assess the respective roles of vasoconstrictor withdrawal vs. active vasodilation in this attenuated SkBF response to heat stress, researchers (7, 15-17, 22, 26, 35) have used the iontophoresis of bretylium tosylate to locally inhibit the release of norepinephrine (12) and other constrictor cotransmitters (33) from cutaneous sympathetic nerves without influencing active vasodilation (16). The attenuated SkBF response to heat stress in older subjects is attributable to decreased active cutaneous vasodilation, rather than augmented or sustained vasoconstriction (22). Although older and young individuals can increase SkBF to a similar percentage of maximal local SkBF during hyperthermia (22), maximal SkBF decreases with age (27, 31), resulting in a decreased absolute SkBF response in aged skin. Furthermore, under these experimental conditions (2, 20-22, 28), the decrement is characterized by a reduced slope or sensitivity of the SkBF-core temperature relationship, rather than a difference in the threshold for the initiation of active vasodilation.

It is unknown whether the attenuated SkBF response of older subjects during passive heat stress is a function of the older men having 1) a similar area of skin containing vasodilated microvessels that is less well perfused, 2) a reduced area of skin containing dilated microvessels that is similarly perfused, or 3) a reduced area of skin containing dilated microvessels that is also less well perfused.

In contrast to those studies relying on standard laser-Doppler flowmetry, the present study employed laser-Doppler imaging (LDI) to examine larger regions of forearm skin to test the hypothesis that the attenuated vasodilator response of aged skin to hyperthermia is due to decreased cutaneous area associated with vasodilation and reduced vascular conductance within that vasodilated area at a given change in core temperature. Therefore, at incremental increases in core body temperature, LDI produced a "snapshot" map of SkBF composed of color-coded pixels, each representing a flux value. The mean cutaneous vascular conductance (CVC) was calculated for each map. Then pixels were classified as representing vasodilated or nonvasodilated areas, and the amount of vasodilated area, the increase in CVC of that vasodilated area, and the distribution of the pixels exhibiting defined ranges of CVC in the vasodilated area were compared between young and older men. Local iontophoresis of bretylium isolated the role of the active vasodilator system in the SkBF response.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Subjects. The Biomedical Committee of the Institutional Review Board for the Protection of Human Subjects of the Pennsylvania State University approved the project. Each subject was given a detailed description of the experiment and provided informed consent before participation. The screening included a medical history, physical examination, and resting electrocardiogram. Adiposity was estimated from skinfold thickness measured at seven sites (13, 32) (Table 1). Subjects performed a graded exercise test to volitional fatigue on a motor-driven treadmill to determine maximal O₂ uptake (O_2 max). During the test, the respiratory exchange ratio exceeded 1.1 in all but one older subject. Each participant's O_2 max was within the middle 60th percentile for his respective population (1). Exclusion criteria included 1) body mass index 30 or 20 kg/m², 2) smoking, 3) medications that could alter cardiovascular or thermoregulatory control or response, 4) a history of heat intolerance, 5) dermatological conditions or diseases, 6) hypertension (resting systolic pressure >140 mmHg or diastolic pressure >90 mmHg), 7) any diagnosed cardiovascular or metabolic disease, or 8) a positive electrocardiogram during the graded exercise test.

Experimental procedures. Twelve older (64- to 75-yr-old) and 12 younger (19- to 29-yr-old) noninstitutionalized, healthy men participated in the study. The protocols were conducted between 0700 and 1200. Older and younger subjects participated in experiments each month between June and December.

Measurements. Urine osmolarity was determined by freezing-point depression (DigiMatic osmometer model 3D2, Advanced Instruments, Needham Heights, MA), and urine specific gravity was measured by refractometry (A300 CL clinical refractometer, Atago, Tokyo, Japan). T_es was measured by using a thermistor inserted nasally into the esophagus to a distance one-fourth of the subject's height (30). Mean skin temperature was calculated as the unweighted average of temperatures measured by type T thermocouples attached to the calf, thigh, abdomen, chest, upper arm, and upper back. Mean arterial blood pressure and heart rate were monitored by using a cuff placed on the right middle finger (Finapres blood pressure monitor model 2300, Ohmeda, Madison, WI). T_es, mean skin temperature, mean arterial pressure, and heart rate data were acquired with a Macintosh computer (Quadra 650, Apple Computer, Cupertino, CA) at 5 points/s, and 1-min averages were computed by using Superscope II (GW Instruments, Somerville, MA). Venous occlusion plethysmography (model EC4 plethysmograph, Hokanson, Belleview, WA) determined FBF of the left arm (8, 22, 38).

Data analysis. The pixels of the LDI were color coded as follows: blue/purple represented low, green/yellow/orange signified medium, and red identified high SkBF (Fig. 1). CVC was calculated for each pixel in the image by dividing the laser-Doppler flux (LDF) by mean arterial pressure. The distribution of CVC within each image was determined by grouping the pixels into seven bins corresponding to their respective CVC values as follows: 0-500, 501-1,000, 1,001-1,500, 1,501-2,000, 2,001-2,500, 2,501-3,000, and >3,001 units. The average number of pixels for young and older men within each binned range (i.e., magnitude of CVC) was plotted to yield a histogram of the distribution. A mean CVC per pixel (mean CVC) for the 30 × 30-pixel image was determined algebraically.

View larger version (84K): [in this window] [in a new window]   Fig. 1.   Laser-Doppler images (LDI) from a representative young and older subject mapped at baseline and at 0.4 and 1.4°C increases in esophageal temperature (Tes). Respective absolute Tes is shown above each image. There is a distinct age difference in the percentage of scanned area that is dilated as well as in magnitude of mean vasodilation during passive heat stress; however, age differences decrease as Tes increases. Line along the left side of images from the younger subject is due to a pen mark on the skin. CVC, cutaneous vascular conductance.

1

1

1

Statistical procedures. Descriptive plots of FVC, CVC, CVC, and %P_VD vs. T_es were analyzed as follows. To determine whether the FVC, CVC, and CVC vs. T_es responses were significantly different between young and older men, an analysis of variance with repeated measures was performed by using a linear model (SAS version 8, SPSS). Group differences in %P_VD vs. T_es were determined by using a repeated-measures analysis of variance (Proc Mixed) fitting a cubic model, because this model described the best-fitting curve through the data. The differences between the responses of older and younger men were compared at each 0.2°C increase in T_es. Critical  level was set at 0.05 for significance of factors (age and T_es) and their interaction. A Student's t-test determined age-group differences in the characteristics presented in Table 1 and in the percent pixels associated with the binned CVC after local heating. Values are means ± SE.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Subject characteristics. Participants were chosen so that their O_2 max was within the middle 60th percentile for their respective populations; therefore, O_2 max was significantly lower in the older than in the young men (Table 1). The mean baseline T_es was lower in the older than in the young men. However, the baseline T_es was strikingly lower in one older man than in the other subjects in his group. If he is excluded, there is no difference in baseline T_es between age groups, but all other results are unchanged. Sweating rate was similar between groups, and all individuals were considered normally hydrated.

Mean arterial pressure and FVC. The mean arterial pressure of older men was higher than that of young men at baseline (P < 0.001), with age-related differences lessening as T_es increased (Fig. 2A). FVC was not different between age groups at baseline (Fig. 2B), but throughout heating, the FVC of the young men was greater than that of the older men at all T_es intervals (P  0.001 at T_es = 0.2°C and P < 0.0001 at T_es  0.4°C).

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Mean arterial pressure (MAP) and forearm vascular conductance (FVC = FBF/MAP, where FBF is forearm blood flow) plotted against Tes for young and older men at baseline and at every subsequent 0.2°C increase in Tes (n = 12 for both groups at all levels of Tes except Tes = 1.4°C, where n = 11 for the young men). A: older men exhibited a higher MAP than young men at 0°C  Tes  0.6°C, but age difference diminishes over time as Tes increases. B: FVC was greater for young than for older men throughout passive heating. Values are means ± SE. ** P < 0.01;  P  0.001; P < 0.0001.

Bretylium tosylate-treated vs. control skin sites. There were no differences in LDF between control and bretylium-treated sites in either age group. The origin of the regression, LDF bretylium vs. LDF control (Fig. 3), was not different from zero, and the regression line (r² = 0.94) resided near the line of identity (P = 0.58). Visual examination of the images of bretylium-pretreated and control sites within subjects revealed no apparent differences in the spatial pattern of CVC or the onset of vasodilation. Likewise, the distributions, "percentage of total pixels vs. CVC," were unaltered by bretylium pretreatment.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Regression comparing mean laser-Doppler flux (LDF) from control sites with LDF from sites iontophoresed with bretylium tosylate to prevent noradrenergic vasoconstriction. Data are from 5 young and 6 older subjects. Regression is close to line of identity, and origin is not significantly different from 0 (P > 0.58). Data suggest no significant involvement of the constrictor system in age differences described in the present study. Rather, differences in skin blood flow are due to age-related changes in the active vasodilator system.

CVC. The mean CVC was not different between age groups at baseline (Fig. 4, A-C). Also, CVC_baseline + 2 SD and SD were not different between groups. In Fig. 4A, the mean CVC was greater for the young than for the older men at T_es = 0.2°C to 0.6°C (P < 0.01) and 0.8°C (P < 0.05). At maximum vasodilation, the mean CVC was significantly lower (P < 0.01) for the older than for the young men (2,259 and 3,380 flux units/mmHg, respectively). The change in CVC from baseline (CVC) (Fig. 4B) within the vasodilated area was greater in the young than in the older men at T_es = 0.2°C (P < 0.01), 0.4°C (P < 0.001), and 0.6°C (P < 0.01). When CVC was expressed as percent maximum CVC, age effects were minimized. The percent maximum CVC of young men was higher only at T_es = 0.4°C (P < 0.05). At T_es = 1.4°C, the CVC of older men reached a greater percentage of their maximum CVC (P < 0.05).

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Mean CVC responses of young and older men to increasing Tes. Data were obtained from a 1.82-cm2 scanned skin region on the right ventral forearm by LDI (n = 12 for both groups at all levels of Tes, except Tes = 1.4°C, where n = 11 for the young men). A: young men exhibited a significantly higher mean CVC at Tes = 0.2-0.8°C than did older subjects. B: change in CVC (CVC) of the vasodilated pixels from baseline was defined as CVC = CVCTes  (CVCbaseline + 2 SD). Similar to A, young men had a higher CVC at Tes = 0.2-0.6°C than their older counterparts. C: when expressed as a percentage of the maximum CVC (%maxCVC), young men had a significantly higher %maxCVC only at Tes = 0.4°C. Older subjects reached a higher %maxCVC at Tes = 1.4°C. D: percentage of pixels that were vasodilated as a function of total pixels (%PVD). Vasodilation was defined as CVC > CVCbaseline + 2 SD. %PVD for young men at Tes = 0.2-0.6°C was greater than that for older men, whereas %PVD at baseline and Tes  0.8°C were similar between the 2 age groups. Values are means ± SE. * P < 0.05; ** P < 0.01.

Vasodilated area. The number of pixels that corresponded to vasodilated area (%P_VD) was lower in the older than in the young men during early (0.2°C  T_es  0.6°C) passive heating (P  0.05). The LDI (Fig. 1) illustrate this difference in %P_VD at T_es = 0.4°C between representative young and older subjects, chosen for their similar baseline T_es. The dark blue, blue, and some of the purple pixels correspond to nonvasodilated areas. At T_es = 0.4°C, the %P_VD of the older man was lower, despite a slightly greater absolute T_es. At maximum vasodilation, there was no difference in %P_VD between young and older men (99.96 ± 0.04 and 99.95 ± 0.05%, respectively).

View larger version (27K): [in this window] [in a new window]   Fig. 5.   Distribution of CVC in older (A) and young (B) men expressed as percentage of total number of pixels lying within specified ranges of CVC. Tes increases in subsequent horizontal pairs of graphs. At baseline and Tes = 0.2, 0.4, and 0.6°C, older men had more pixels corresponding to lower levels of CVC than did young men. At Tes = 1.2°C, difference between groups becomes less evident. A spline was applied to each distribution to aid in visualization. Values are means ± SE.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The present study employed LDI to explore whether decreased cutaneous area associated with vasodilation as well as reduced vascular conductance within that vasodilated area is responsible for the attenuated SkBF response of aged skin to hyperthermia. This study yielded the following main results: 1) as T_es increased ~0.6°C during early passive heating, vasodilated area within the mapped region of the ventral forearm was smaller in the older than in the young men; 2) CVC per unit area was lower in this vasodilated area in the older than in the young men; and 3) when T_es rose 0.8°C, the vasodilated area was similar in both groups, although CVC remained reduced in the older men.

In addition, local heating elicited reduced CVC during maximum vasodilation in the older men, in agreement with previous observations (27, 31). The impaired conductance was characterized by reduced CVC per vasodilated area, rather than a smaller vasodilated area. Also, bretylium tosylate iontophoresis and observations by LDI confirmed that attenuated active vasodilation, rather than sustained vasoconstriction, caused the decrement in SkBF response to heat stress in older men (23). The comparison of the control and bretylium-pretreated areas also indicated that vasoconstriction did not cause quantitative or apparent spatial differences in the cutaneous area associated with vasodilation or the amount of CVC in the vasodilated areas.

The T_es threshold for the onset of vasodilation in the present study occurred between T_es = 0 and 0.2°C in both age groups. Earlier studies had shown little or no difference between the thresholds in older and young men during heat stress (20, 23, 24, 27, 35), which suggested that, under these conditions, the differences in the SkBF response were more likely due to peripheral changes associated with aging than to a central (hypothalamic) effect.

Although active vasodilation was initiated by T_es = 0.2°C, both age groups exhibited pixels corresponding to nonvasodilated areas (non-P_VD) that appeared to "dilate" at higher T_es. The non-P_VD persisted in multiple images of a series and at increasing T_es, suggesting that the non-P_VD were not due to cyclic temporal variations. Braverman and Schechner (5) described a subpopulation of microvessels unresponsive to local thermal stimuli. They proposed two subclasses of microvessels in the skin: one responsible for tissue perfusion and one responsible for thermoregulation. Although vasodilation induced through local warming was under different control from that induced by reflex through increasing core temperature (10, 18, 19), it was possible that the non-P_VD in the present study reflected functionally diverse microvessels that eventually dilated through different mechanisms. Likewise, Mack (25) suggested that the two subclasses of microvessels could explain why baroreceptor unloading failed to elicit a uniform reduction in SkBF in images of the forearm. However, although Mack reported a universal onset of thermoregulatory vasodilation, our definition for P_VD [CVC_pixel > (mean CVC_baseline + 2 SD)] was more conservative and could explain the nonuniform onset of vasodilation we observed. Also, because the arterioles were probably 1.5-1.7 mm apart (4), some of the 0.45 × 0.45-mm pixels that comprised the images in the present study may have contained only capillaries and venules. Because of the inherently low CVC of capillaries and venules (4), some of these pixels may have required more advanced vasodilation of parent arterioles before CVC increased enough to satisfy our definition of "vasodilation."

The older men had twice the non-P_VD (100%  44% = 56%) of the young men (100%  72% = 28%) at T_es = 0.2°C (Fig. 4D). Also, more non-P_VD persisted at T_es = 0.6°C, reflecting a slower recruitment and filling of skin microvessels in older men. Because both groups were normally hydrated (Table 1), differences between groups were not related to dehydration in the older men. Research has suggested the possible impairment of the control and function of the microvessels through age-related changes in vessel anatomy, ultrastructure, and surrounding tissue, as well as alterations in cutaneous innervation (3, 9, 29) that could retard the vasodilatory response in older individuals. Also the increased non-P_VD and reduced CVC in older men could be due, in part, to fewer thermoregulatory microvessels (5). Interestingly, the loss of thermoregulatory arterioles may not contribute to the apparent delayed recruitment of skin microvessels. Vollmar et al. (36) observed that cutaneous arteriolar and venular number and branching pattern are conserved throughout the life span of hairless mice. On the other hand, they reported a 40% decrease in capillarity. Likewise, others have observed a decrease of capillary loops in the papillary dermis in humans (29) but have also recorded a decrease in venular cross-sectional area (11). Decreasing capillarity with age (29, 36), as well as the low conductance of capillaries and regions containing primarily capillaries (4), could have caused reduced SkBF in some areas. The greater persistence of non-P_VD in older men could have resulted when regions that contained only capillaries and venules in reduced number required larger relative increases in CVC per microvessel before being defined as vasodilated. However, despite the small area represented by each pixel, no pixels were devoid of microvessels, because vasodilation occurred throughout at higher T_es and during local heating.

During early passive heating, the reduced mean CVC for the 30 × 30-pixel area in older men was characterized not only by reduced vasodilated area but also by impaired CVC in that vasodilated area. Because the baseline mean CVC was the same between age groups, this impairment in CVC resulted in reduced mean CVC in the vasodilated area. At T_es = 0.8°C, P_VD was the same between age groups, and mean CVC for the 30 × 30-pixel area was reduced in older men; therefore, one might expect the CVC to be reduced. However, the large standard error for the older men may have contributed to the difference in CVC, narrowly missing statistical significance.

Because the P_VD was not different between age groups at T_es  0.8°C but the mean CVC of older men remained lower than that of the young men at T_es = 0.8°C, the age-related decrement in SkBF response at T_es = 0.8°C was due to a reduced CVC per dilated area, rather than a reduction in the vasodilated area. This could be due to a decrease in the number of microvessels per pixel and/or a decreased CVC per microvessel. As a result, we saw a progression in the mechanisms for the difference in the SkBF response to heat stress between the young and older men. In early hyperthermia, a reduced vasodilated area and a lower CVC in that vasodilated area contributed to the impaired SkBF response of older men. However, at higher core temperatures, reduced CVC per dilated area alone was responsible for the attenuated SkBF response. At the highest T_es, the differences between CVC became smaller as the dilation of individual vessels in the older men approached that of the young men. However, to achieve the same CVC at T_es = 1.4°C as that of the young men, the older group must operate at a higher percent maximum CVC, leaving less reserve vasodilatory capacity (Fig. 4C).

The similar mean CVC and CVC in both age groups at higher T_es, despite the decreased sensitivity of the SkBF response in older men, suggested that the body has a targeted, upper limit for SkBF (15). With reduced maximal SkBF observed in the present study and others (27, 31), one would expect a reduced CVC in older men at higher T_es if they were to increase SkBF to the same percent maximum CVC as do young men. The upper limit to the CVC response may be imposed by a baroreceptor-mediated limitation to active vasodilation designed to maintain central blood pressure (14, 17).

Technical comments. The scanning rate of the LDI can affect the resolution and integrity of the images. Wardell et al. (37) noted that regions of perfused skin may best be discerned at 1,000 ms/pixel. Mack (25) suggested using 50 ms/pixel as the optimum rate for a 6.25-cm² area of skin. In the present study, we used 4 ms/pixel to reduce the scanning time to ~2 min for the largest areas (32-47 cm²) and 1 min for the smallest areas. The shorter scanning time promoted consistent physiological conditions throughout the scan, enabled us to perform successive scans quickly, and reduced the chance of movement artifacts.

Summary. In conclusion, the present study demonstrated that during early passive heating the reduced SkBF response of older men to is due to a decreased area corresponding to vasodilated microvessels as well as a decreased CVC within those vasodilated areas. This decrement is due to a reduced sensitivity of the %P_VD vs. T_es, CVC vs. T_es, and CVC vs. T_es responses in older men. At a higher T_es, the decrement in the SkBF response in older men was due solely to a reduced CVC within the vasodilated areas. Eventually, these differences in vasodilated area and CVC between age groups became smaller at the greatest values of T_es. However, the older group must maintain a greater percent maximum CVC to achieve the same absolute SkBF at the greatest T_es.