INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

EACH YEAR IN THE UNITED STATES, the number of individuals older than 65 yr increases (United States Bureau of Census), with the majority of this aging population consisting of women. Women are currently expected to outlive men by ~6 yr. Given that the average age of menopause is 50 yr, a growing number of individuals in our population are postmenopausal women. In recent decades, hormone replacement therapy (HRT) has become an increasingly popular treatment to combat unfavorable symptoms associated with the menopausal period, such as hot flashes, increased risk of heart disease, and osteoporosis.

It has been suggested that tolerance to heat stress diminishes with age (9). For example, people over the age of 65 comprise the majority of victims during severe heat waves (26). The decreased ability of older individuals to tolerate heat stress may be partially due to their reduced capacity to increase skin blood flow (SkBF). Maximal SkBF progressively decreases with age (24) throughout adulthood. In addition, SkBF during submaximal exercise is reduced in older individuals, independently of factors such as maximal oxygen consumption, body composition, and hydration status (21). This diminished capacity to increase SkBF during heat stress is due to both a reduced end-organ sensitivity to active vasodilator system stimulation and to structural changes in skin vessels, rather than to an increased or sustained tonic vasoconstrictor activity (20, 22, 23). Thus any intervention that may enhance an older individual's ability to vasodilate cutaneous vessels may be advantageous under certain environmental conditions.

Acute and chronic estrogen replacement therapy (ERT) alter thermoregulatory and vasomotor function in postmenopausal women, such that core temperature (T_c) is defended at a lower level (6, 31). Specifically, ERT reduces resting T_c by ~0.5°C and shifts the SkBF/T_c curve such that the T_c threshold for increasing SkBF is lower in women receiving ERT compared with women not receiving HRT (our No HRT group). However, addition of progestins to HRT blocks the thermoregulatory effects of estrogens, such that postmenopausal women on a combined estrogen and progesterone therapy (our E+P group) regulate body temperature in a similar manner to women not receiving HRT.

In a previous investigation (6), resting cutaneous vascular conductance (CVC) expressed as a percentage of maximum CVC (%CVC_max) was significantly higher in women receiving ERT compared with a No HRT group. Resting %CVC_max in the E+P group was intermediate to the other two groups of women. A higher absolute resting SkBF in women receiving ERT might contribute to their reduced resting body T_c by enhancing heat loss, since heat loss is directly proportional to the absolute quantity of blood perfusing the skin vasculature. Although resting forearm blood flow (FBF), a quantitative index of SkBF, may be as low as 2-6 ml blood · 100 ml tissue¹ · min¹ in thermoneutral environments, small changes in SkBF can contribute significantly to heat loss (17). In the previous study performed in our laboratory (6), resting SkBF measurements were made in a warm room [dry bulb temperature (T_db) = 36°C] rather than at thermoneutral environmental conditions. Thus it is unclear whether the reduced T_c in postmenopausal women receiving ERT is partially due to a higher SkBF in resting thermoneutral conditions.

Additionally, we recently observed (6) that ERT, E+P, and No HRT groups achieved a similar %CVC_max after 1 h of submaximal exercise in the heat, coupled with similar forearm vascular conductance (FVC) and FBF values. These results provide indirect evidence that maximal CVC (CVC_max) was similar among these three groups of postmenopausal women, but potential effects of HRT on maximal SkBF have not been directly examined. It is known that CVC_max decreases as individuals age. This phenomenon is thought to be due to any of the following: 1) vascular wall thickening, 2) atherosclerotic damage and altered vascular wall structure (4), 3) fewer patent cutaneous capillaries (35), 4) decreased recruitment and response of the vessels to heat (28), and 5) an attenuated response of the vessels to neural vasodilator stimulation (20). Exogenous ERT and estradiol have been shown to cause vasodilation in human coronary and forearm vessels (14, 15). Other reported estrogenic effects include a reduction in collagen accumulation in atherosclerotic rabbit aortas (12), inhibition of vascular smooth muscle cell (VSMC) proliferation in response to injury in rabbits and rats (1, 27), prevention of diet-induced atherosclerosis in primates (8), inhibition of oxidation of low-density lipoproteins in postmenopausal women (29), and reduced intimal thickening in rat femoral arteries (1). One might expect, therefore, that postmenopausal women receiving ERT may have less atherosclerotic and structural damage and preserved vascular wall compliance (3, 16); i.e., ERT may represent an intervention for postmenopausal women to prevent the age-associated decline in CVC_max. However, the effects of HRT on CVC_max have not been directly measured.

Therefore, the purpose of the present investigation was to investigate whether 1) regular exogenous estrogen significantly alters baseline SkBF and CVC in a thermoneutral environment and 2) ERT alters maximal SkBF and CVC_max in postmenopausal women. SkBF was measured, since it is directly proportional to absolute heat loss, and CVC was calculated, since it accounts for differences in mean arterial pressure (MAP) between groups and provides information about the tonic state of the vessel wall. On the basis of previous observations, we hypothesized that resting SkBF and %CVC_max would be higher in women receiving ERT but that maximal SkBF and CVC_max would not be altered by either ERT or E+P.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Subjects

Chronic HRT was defined as continuous therapy for 2 yr. Women were defined as postmenopausal by the following criteria: 1) complete cessation of menses for 1 yr following a history of eumenorrhea and 2) 2-wk repeat serum estradiol of 30 pg/ml. Five women were considered postmenopausal after undergoing both an oophorectomy and hysterectomy. Women included in the study were no longer experiencing symptoms associated with the perimenopausal period (hot flashes, etc.). All but five women on HRT received 0.625 mg of Premarin (Wyeth-Ayerst Laboratories, Philadelphia, PA) on a daily basis. The five exceptions included three women who received 0.625 of Premarin on the first 25 days of the month, a fourth woman who received 0.625 of Premarin on Monday, Wednesday, and Friday, and a fifth woman who received 0.625 mg of Premarin on all odd days of the month and 1.25 mg of Premarin on even days. One of these women also used a vaginal estrogen cream. Progesterone dosages ranged from 2.5 to 10 mg, and, like in estrogen therapy, the pill cycle varied among the women. Progestational agents used included Provera (Upjohn, Kalamazoo, MI) and Cycrin (Esi Lederle, Philadelphia, PA), both of which contain medroxyprogesterone acetate.

Venous blood samples were collected from five No HRT women, six ERT women, and seven E+P women, all of whom had participated in our previous investigation (6). Samples were stored on ice, centrifuged, and later analyzed for 17-estradiol and progesterone concentrations by an ¹²⁵I double-antibody radioimmunoassay procedure (ICN Biomedicals, Costa Mesa, CA). The sensitivity of the assay was 9 pg/ml, and inter- and intra-assay precision coefficients of variation were <12 and <11% for an estradiol range of 28-38 pg/ml, respectively.

Height and weight were measured for each subject before experimental testing. Table 1 presents the mean data for physical characteristics of the three groups of women for each series of studies. Criteria for exclusion included the following: 1) hypertension (resting systolic pressure >140 mmHg or a diastolic pressure >90 mmHg), 2) smoking, 3) any diagnosed metabolic or cardiovascular disease, or 4) taking of any medication with the potential to influence thermoregulatory or cardiovascular variables of interest.

Protocols and Procedures

Experiment 1. In the first series of studies, subjects were randomly tested during the months of December 1995 and January 1996. The study sample included 24 postmenopausal women: 8 from No HRT group, 8 from ERT group, and 8 from E+P group. On arrival at the laboratory, subjects sat upright for 15 min in an environmental chamber (T_db = 24°C) while measurement devices were attached. After an additional 10 min were allowed for responses to stabilize, baseline measurements of FBF and laser-Doppler flux (LDF) were collected for 10 min. In addition, forearm skin temperature (T_fa) was measured at the ventral and dorsal skin surface of the left forearm adjacent to the strain gauge, and MAP was measured before each cycle of FBF recordings. After baseline measurements, the left forearm was heated by using a water-spray device (32), which encircled the entire left forearm, from the wrist to the elbow. Warm water (42-45°C) was pumped through the spray device that would increase T_fa to 42°C and which was adjusted to clamp T_fa at this temperature throughout the remainder of experimental period. Six cycles of FBF measurements and MAP were measured at 0, 15, 30, 45, and 60 min of heating, and FVC was calculated for each time point from these two variables. After 60 min of heating, a reactive hyperemia maneuver was performed on the left arm to verify that the SkBF achieved was maximal (19). This procedure involved inflating a blood pressure cuff to 200 mmHg around the upper left arm for 5 min and recording LDF after release of the occlusion. If the LDF values before the occlusion were lower than the postocclusion LDF values, the arm was heated for an additional 15 min, and FBF and MAP were measured at 75 min. Otherwise, the experiment was terminated at 60 min.

Experiment 2. In the second series of studies, 16 postmenopausal women were recruited: 8 postmenopausal women receiving ERT and 8 postmenopausal women not receiving HRT (No HRT group), and were tested in random order during the months of October and November 1997. This experimental protocol consisted of two 30-min periods of rest in an upright seated position in a thermoneutral [T_db = 24°C, wet bulb temperature (T_wb) = 15°C] environment followed by rest in a warm (T_db = 36°C, T_wb = 25°C) environment. The length of time between these two periods was ~8-10 min. LDF was measured continuously and MAP was measured every 5 min during the protocol. After these two 30-min periods, the room temperature was returned to an intermediate level (T_db = 28°C), and the local temperature of the skin around the laser flow probe was raised to 42°C by a heating device and maintained for 30 min. A reactive-hyperemia procedure was performed after 30 min of local heating to verify that a CVC_max had been attained. LDF was recorded for an additional 10 min after the occlusion.

MAP, FBF, FVC, %CVC_max, and mean skin temperature _sk) data collected from postmenopausal women in resting thermoneutral conditions were compiled with data from experiments 1 and 2, yielding a total of 39 subjects: 14 women not receiving HRT, 13 women receiving chronic ERT, and 12 women receiving chronic E+P. All of these women, except five, participated in two or more of these sets of experiments. Therefore, we were able to examine repeatability of the SkBF measurements within an individual subject.

Measurements

Individual _sk data were collected at a rate of 5 data points/s, averaged over 1-min intervals by using a SuperScope II (GW Instruments, Somerville, MA) data-acquisition system, and stored on a dedicated computer (Macintosh Quadra 650, Apple Computer, Cupertino, CA). Similarly, LDF data were recorded at a rate of 1 data point/s, and a mean was calculated for 1-min intervals.

Analysis of Data

In experiment 1, a repeated-measures ANOVA was performed to examine group differences in MAP, FBF, FVC, and %CVC_max at baseline and during heating. "Time" was the within-subjects factor and "group" was the between-subjects factor. In experiment 2, the %CVC_max measurements from the last 15 min of each 30-min period were averaged for each period. These values were analyzed by a two-way ANOVA by using SAS statistical software. "Group" was the between-subjects factor and "room temperature" was the within-subjects factor. Group differences in MAP were analyzed by a two-way ANOVA. Using SAS statistical software, a repeated-measures ANOVA was performed to examine group and within-subject differences in baseline _sk, MAP, FBF, FVC, and %CVC_max for the compiled data (thermoneutral temperatures only).

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Mean data for subject characteristics for individual experiments and the compiled data are presented in Table 1. There were no significant differences in age, height, or weight for the three groups of postmenopausal women. Estradiol concentrations, when a subset of women from each group was used, were significantly greater in women receiving ERT and in E+P groups compared with the No HRT group (27.6 ± 5.4, 157.4 ± 18.8, and 96.1 ± 35.0 pg/ml for No HRT, ERT, and E+P groups, respectively; P < 0.05). Resting MAP was also similar among No HRT, ERT, and E+P groups (Table 1).

At thermoneutral ambient temperatures, resting FBF, FVC, and %CVC_max were not significantly different among the three groups of women in the present series of studies and in the compiled data (Fig. 1A). Similarly, _sk was similar for No HRT, ERT, and E+P groups at thermoneutral environmental conditions (_sk = 33.0 ± 0.2, 33.1 ± 0.2, and 33.1 ± 0.2°C for No HRT, ERT, and E+P groups, respectively). When ambient temperature and humidity were raised (T_db = 36°C, T_wb = 25°C), resting %CVC_max similarly increased in both No HRT and ERT groups (Fig. 1B).

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Resting forearm blood flow (FBF), forearm vascular conductance (FVC), and cutaneous vascular conductance (CVC), calculated as laser-Doppler flux divided by mean arterial pressure and expressed as a %maximal CVC (%CVCmax), at thermoneutral [dry bulb temperature (Tdb) = 24°C; A] and warm (Tdb = 36°C; B) environmental temperatures are shown for groups of postmenopausal women not receiving hormone replacement therapy (No HRT), receiving estrogen replacement therapy (ERT), or receiving estrogen + progesterone (E+P). Results of maximal FBF (FBFmax) and FVC (FVCmax) in response to prolonged (60-min) local heating of forearm at 42°C are shown for all 3 groups in C. There was no significant group or condition (thermoneutral or warm) effect on resting or maximal skin blood flow. Bars represent SE values.

Maximal FBF and FVC (indexes of SkBF and skin vascular conductance) were not significantly different among No HRT, ERT, and E+P groups of women (Fig. 1C). MAP remained constant during this local heating period.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

This series of studies was undertaken to clarify the effects of HRT on SkBF in postmenopausal women. Direct vasodilatory effects of estrogen on the vasculature are supported by evidence that estrogen increases nitric oxide production, inhibits endothelin action and production, inhibits calcium influx, and increases potassium conductance (10, 25). Similarly, estrogen has been shown to inhibit VSMC proliferation and migration, reduce polyethylene cuff-induced intimal thickening in rat femoral arteries, and inhibit DNA synthesis in fetal rat aortic cell lines independently of nitric oxide, prostacyclin, or cGMP involvement (1). The inhibitory effects of estrogen on the growth of VSMCs have been demonstrated in rabbit aorta and porcine coronary (13, 33). Although most findings support an antiproliferative effect of estrogen, this observation is not consistent. For example, Farhat and co-workers (11) found that estradiol potentiated proliferation of VSMCs in rat pulmonary vessel cultures and canine intrapulmonary artery segments. Similarly, estrogen replacement has been shown to induce c-fos gene transcription in the rat uterus but not in the rat aorta (2). Differences in results are likely due to differences in vessel origin, culture conditions, and methodology. Given these vasodilatory and antiproliferative properties of estrogen (1, 14, 15) on the vasculature wall, we originally speculated that resting CVC and SkBF would be higher in the ERT group compared with the No HRT group. This hypothesis was supported by the significantly higher %CVC_max in the ERT group compared with the No HRT group in our previous investigation (6). Because %CVC_max is defined as CVC/CVC_max, it is much more likely that estrogen increases resting CVC rather than lowers CVC_max. It is clear that estrogen acts centrally to reduce the regulated T_c in humans and animals (6, 7, 30, 31), but the peripheral effects of estrogen on resting SkBF in humans are not as clear.

We did not see a significant difference in resting FBF, FVC, or %CVC_max at thermoneutral conditions among ERT, E+P, or No HRT groups of postmenopausal women. Also, contrary to results from our previous study (6), baseline %CVC_max at rest in the heat was not higher in women receiving ERT compared with the No HRT group. Because measurement of SkBF under controlled conditions was not the primary aim of our initial investigation, the discrepancy between the present study and our previous one (6) may be due to any of several reasons. First, in the previous study, a blood sample was taken from the same arm at which LDF measurements were collected. Although resting SkBF data were consistently collected for 15 min in a warm environment before a 1-h exercise period for each individual subject, the period of time in the warm chamber before these baseline measurements varied between 15 and 40 min due to blood-collection procedures. In certain cases, a heating pad had to be used to collect blood. Also, in the initial study (6), the control skin site was iontophoresed with purified water. This procedure was performed to control for the iontophoresis procedure performed at the bretylium-treated site. Although we do not think that this altered the SkBF response at baseline, it is an inconsistency between the previous study and the present investigation. We feel that these differences in methodology may have caused a spurious difference in baseline %CVC_max in the ERT group.

The lack of a significant difference between resting SkBF among the three groups of postmenopausal women in the present investigation suggests that the lower T_c at rest in thermoneutral conditions in the ERT group is due to other mechanisms besides a higher resting SkBF. For example, there may be small differences in heat production (34) or insensible heat loss among the three groups. However, it is possible that our measurements of SkBF are not sensitive enough to measure subtle differences in SkBF. We do believe that SkBF plays a very important role in the dissipation of heat and maintenance of a lower T_c in postmenopausal women receiving ERT via a centrally mediated reduction in the T_c threshold for the onset of cutaneous vasodilation during periods of heat stress and exercise. This is supported by results from previous studies (6, 31).

Vascular wall thickening, years of exposure to ultraviolet radiation, and atherosclerotic damage of vessels in older individuals may lead to structural alterations of skin vessels (4), fewer patent cutaneous capillaries (35), and an attenuated response of the cutaneous vessels to heat (28) and vasodilator substances (20). Because acute and continuous (3 mo) administration of exogenous estrogens appears to maintain the structural integrity of the vessel wall by inhibiting atherosclerotic occlusion of coronary vessels (16) and oxidation of low-density lipoproteins (29) in postmenopausal women, one might expect maximal SkBF or CVC to be higher in postmenopausal women receiving ERT than in those not receiving HRT (3). However, in the present investigation, maximal FBF and FVC in response to prolonged local heating was not significantly different among the three groups of postmenopausal women. The evidence supporting a vascular-protective role by estrogen originates from studies performed in animals, cell culture, and primarily the coronary vessels rather than the skin. Human skin is unique in that it has an active vasodilator system. We feel that these reasons may be responsible for the discrepancy between the in vitro antiproliferative, vascular-protective effects of estrogen on VSMCs and the in vivo results from the present investigation, which showed a lack of an effect by ERT on maximal skin vascular conductance. In addition to the aforementioned mechanisms, it has been proposed (5) that two populations of vessels exist in the skin: one for nutritive tissue perfusion and one responsible for temperature regulation. If this situation exists, perhaps estrogen exerts its vascular-protective effects on the nutritive vessels rather than on thermoregulatory vessels. Although estrogen may enhance vascular conductance in other vascular beds, such as coronary vessels, it does not appear to do so in the skin. In coronary arteries, where blood flow is critical for oxygen delivery, prevention of intimal thickening and maintenance of optimal conductance by estrogen has important implications. Except in situations of severe heat stress, a high capacitance by the cutaneous vasculature is of lesser importance in the daily life of an individual.

In summary, resting FBF, FVC, and %CVC_max were not significantly different among No HRT, ERT, and E+P groups in a thermoneutral environment, and HRT did not alter maximal SkBF or CVC in postmenopausal women. From these findings, HRT does not appear to significantly alter resting SkBF in thermoneutral temperatures nor does it prevent the reduction in maximal skin vascular conductance associated with the aging process.