The perceptual and physiological responses of high and low fat women exposed to SoC air for 120 minutes

Transcription

1 Wilderness and Environmental Medicine. 9, (1998) ORIGINAL ARTICLE The perceptual and physiological responses of high and low fat women exposed to SoC air for 120 minutes E.L. GLICKMAN-WEISS; C.M. HEARON; R. PRISBY; N. CAINE From the Department of Biological and Exercise Sciences. Northeastern Illinois University, Kent State University, Kent, OH (Dr Glickman-Weiss, Ms Prisby, Ms Caine), and the Exercise Sciences Laboratory, Chicago, IL (Dr Hearon). The present investigation dichotomized female subjects by percentage of fat (low fat [LF] == 19.2 ± 3% [n = 4] vs high fat [HF] = 29.9 ± 3% [n == 4]) to elucidate the perceptual and physiological responses during acute exposure to 5 C air for 120 min. Correlational analyses were used to assess the relationship between thermal sensection (TS) and differences between LF and HF groups. Significant (p < 0.01) differences existed between the groups for TS when values were pooled over time, where HF had a lower TS (5.9 ± 0.6) than LF (7.9 ± 0.5). Additionally, a significant (p < 0.01) main effect for time existed, where TS, when pooled over groups T5 (3.8 ± 0.7) and T30 (5.4 ± 0.9), was less than T60 (7.4 ± 0.6), T90 (8.6 ± 0.5), and Tl20 (9.1 ± 0.4). No significant fat X time interaction was noted despite apparent perceptual differences between the groups at most of the time points. Further, the relationships between TS and percentage of fat and fat mass were either significant or exhibited strong trends toward significance at time points T5, T30, and T60. When rectal temperature (Tre) was pooled over the four time points, the fat groups differed significantly (p < 0.05); LF (-0.21 ± 0.04 C) exhibited a greater ATre than HF (-0.03 ± 0.04 C). When Tre was pooled across the fat groups, significant (p < 0.01) differences over time were detected. From these data, it appears that the modified TS scale tested may be a good marker of thermal discomfort in females exposed to the cold because perceptual differences were noted between the groups, accompanied by apparent differences in Tre between the LF and HF groups. Key words: thermal perception, thermal sensation, temperature regulation, body composition. females Introduction The relation between cold perception (ie, an individual's ability to sense and scale extreme thermal discomfort) during acute cold exposure and selected physiological parameters is an area that has considerable practical application to those individuals who may work in cold environments or even individuals that recreationally are placed in the wilderness setting. For example, Fine [1] examined the use of subjective rating scales as substitutes for physiological measurements (mean skin temperature and metabolic rate) in six enlisted men exposed as a group to different environmental conditions on eight successive days. Fine [1) reported that the subjective scale might be used instead of the physiological measurements for the purpose of comparing conditions with regard to relative warmth or coldness of groups of in- Reprint requests to School of Exercise. Leisure and Sport. Kent State University. 163E Gym Annex. Kent. OH (Dr Glickman Weiss). dividuals within the conditions. In addition, Gagge et al [2] (Table 1) quantified thermal sensation on three unclothed subjects who sat quietly in a range of ambient temperatures. However, when relations between thermal perception and a physiologic response to a thermal challenge are examined, manipulating one stressor, ie, ambient temperature, and then measuring the physiological, thermal, and perceptual responses to the stressor is important [3]. If this is not done (as in the case when one would exercise in the cold), then when using the thermal scale, the individual would be unlikely to be able to discriminate between the two opposing challenges (ie, exercise that presumably would increase heat production and thereby reduce the cold stressor), which may affect the actual thermal sensation. Therefore, the purpose of this study was to revisit the Glickman-Weiss et al [3] thermal sensation scale that was developed as a valid and reliable instrument for resting cold exposure protocols (using a

2 Perceptual and physiological responses of high and low fat cold-exposed women 205 Table 1. Thermal sensation scale* Table 2. New rating scale of thermal sensation* Scale Thermal sensation Scale Thermal sensation 0.0 Unbearably cold Very cold Cold Cool Neutral (comfortable) Warm Hot Very hot Unbearably hot *Gagge et al [2]. sample of males, n == 8) with a population of low vs high fat females. The Glickman-Weiss et al [3] scale (Table 2) uses similar semantics and was designed to be similar to the universally utilized Borg lo-point scale [4,5]. The Borg [4,5] scale is often used as a tool to rate the perception of exertion or exercise intensity during exercise testing or exercise training. Alternatively, the Gagge et al scale [2] does not use a range ofnumerical values that increase as thermal discomfort increases. Instead, the Gagge et al [2] scale presents environmental extremes differently (ie, cold, warm, and neutral) on one scale. This results in both a very small continuum for scaling and an improbable physiological continuum when one is exposed for a protracted period of time to cold as an environmental stressor. Further, the majority of the environmental physiology literature, or even most exercise physiology literature, has focused on a population of males. However, we [6] recently demonstrated that females, more specifically high vs low fat females, physiologically and thermally respond to an environmental stressor differently from their male counterparts. As we know, body fat provides significant protection during cold air exposure by providing resistance to heat transfer from the body's core to the environment (7]. Although body fat appears to play a role in reducing core cooling, Glickman-Weiss et al [8] demonstrated that rectal temperature (Tre) was maintained similarly in low (8%-12% fat) and high fat males (18%-22% fat) via a >10.0 *Glickman-Weiss et al [3]. significantly greater aerobic metabolic rate. Alternatively, high fat (HF: 28%-35%) females demonstrated higher core temperatures and tissue insulation values than their low fat (LF: 15%-22%) counterparts that were not accompanied by a differential response in aerobic metabolic rate. Thus, the impact of body composition upon energy expenditure to maintain Tre during cold exposure appears to differ between LF and HF males and females. Therefore, we believe that the thermal and physiological responses to acute cold exposure differ between males and females. The purpose of this investigation was to explore in low and high fat females the differences and look at the relation between thermal, physiological, and perceptual parameters. Methods SUBJECT SELECTION Nothing at all Moderately cool Cool Very cool Very, very cool Somewhat cold Cold Very cold Very, very cold Unbearably cold Height, weight, body surface area, and body composition were obtained for 27 female volunteers. From this pool of potential subjects, four LF (15%-22%) and four HF (28%-35%) normal healthy active women aged yr of age were identified. Descriptive data on the HF and LF subjects are presented in Table 3. Prior to participation in the investigation, subjects signed informed consent. The protocol was reviewed and approved by the institutional human ethics committee. As required by the human ethics committee, prior to experimental testing, all subjects underwent a pregnancy test that was assessed via urine analysis. However, no subject was excluded on the basis of testing positive for pregnancy. Subjects that were excluded (n = 19) did not meet the body fat criterion for this study as outlined above. Subjects who

3 206 Glickman-Weiss et at Table 3. Subject characteristics (mean ::!:: SD) Mass (kg) Stature (cm) Percent fat (%) VOz max (ml kg- l min- l ) Body surface area (m 2 ) *Means are different, p < were taking oral contraceptives were not excluded from the study. BODY COMPOSITION ASSESSMENT EXPERIMENTAL TRIALS Low fat 58.5 ::!:: ::!:: ::!:: 2.9* ::!:: ::!:: 0.1 High fat 68.1 ::!:: ::!:: ::!:: 3.3* 38.6 ::!:: ::!:: 0.1 A hydrostatic weighing system was employed to assess body density. Underwater densitometry was conducted in a X 152A-cm tanle During the body composition analysis, subjects sat on a specially designed chair suspended by a Chatillon@ scale. In this process, the subject was asked to exhale as completely as possible while suspended underwater for approximately 5-10 seconds. In doing so, the subject's mass underwater was measured. Relative fat content (percentage of fat) was then derived with the Brozek et al [9] equation by obtaining body density (ie, density = mass/volume; where mass is the mass in air and volume is the individual's mass in air minus her mass in water minus her residual lung volume). Residual lung volume was determined prior to the hydrostatic weighing procedure above by a modification of the standard oxygen-dilution technique [10]. During the measurement of residual lung volume, subjects were seated, in air, in the same anatomical position as they assumed during the underwater weighing procedure. Surface-area-to-mass ratio was determined with the Du Bois equation [11] for body surface area, body mass was determined with a balance beam scale, and stature was determined with a measuring rod. Subjects were divided into two groups with respect to body fat and exposed to cold (5 C) air for 120 min. Subjects reported to the laboratory at approximately 0900 h. Approximately 120 min prior to the experimental trial, subjects ate either a bagel or two slices of toast. Core temperature was assessed via a rectal thermistor (Yellow Springs Instruments Model 410 probe) inserted 130 mm into the rectum. Thermocouples (Yellow Springs Instruments Model 427) for the measurement of mean skin temperature (Tsk) were taped (Hytape@) to Differences between fat groups over time were examined by an analysis of variance with repeated measures (one between subjects effect [fat group], one within-subject effect [time]). In those situations where the repeated observations violated the ANOVA assumption of sphericthe right triceps, chest, forearm, thigh, and calf. Mean skin temperature was calculated with the following equation: Tsk = 0.22(calf temperature) (thigh temperature) (chest temperature) (forearm temperature) (tricep temperature). Both Tre and skin temperature (Tsk) were monitored with a tele-thermometer interfaced to a 12-channel switch box (Yellow Springs Instruments 2100 tele-thermometer, YSI Model 4002 switch box). Exposure air temperature was maintained within ±OSC of the criterion temperature. Oxygen uptake (V0 2 ) was obtained with respiratory metabolic system consisting of the following: 1) gas analysis equipment, ie, CO 2 analyzer (Applied Electrochemistry, CD3A), flowmeter (Applied Electrochemistry, R-1), and an O 2 analyzer (Applied Electrochemistry, S-3A1); 2) 5-L mixing chamber; and 3) pneumotachometer. The gas analyzers were calibrated immediately prior to testing, every 15 min during testing, and at the completion of testing with standard reference gases. Metabolism, M, was determined from VOz measurements and expressed in W m- 2. The change in M (M) was calculated as the difference from one-.!ime pot to the previous time point. In addition, the Tsk and Tre were calculated similarly. After baseline parameters, ie, preexposure Tsk, Tre, and Vo 2, were assessed, the subject, wearing only cotton shorts and jog bra, entered the chamber and sat with limbs separated and exnded on a mesh lawn chair. Oxygen uptake, Tre, and Tsk were determined after 5 and 30 min of exposure. Measurements were then made every 30 min until the trial was terminated. All experiments were 120 min in length. However, as a medical safeguard, trials would be terminated if Tre decreased to 35 C (however, no test was terminated early). At the completion of the trials, subjects exited the chamber and consumed warm fluids. All tests were conducted within a 12-week period. Because core temperature is regulated at a higher temperature during the luteal phase compared with the follicular phase of the menstrual cycle [12], the cold exposure had to be consistent between all subjects. Each subject, therefore, was tested within 10 days after the onset of menses, during the follicular phase. Subjects were given the opportunity to acquaint themselves with the protocol during an initial familiarization visit. STATISTICAL ANALYSIS

4 Perceptual and physiological responses of high and low fat cold-exposed women T-;:======:; _, 10 to be perceptual differences between HF and LF at earlier time points, no significant (p > 0.05) fat X time interaction (Fig 1) was noted. RELATIONSHIP BETWEEN TS AND BODY COMPOSITION AT EACH TIME POINT 2 o Fig 1. Thermal sensation (mean ± SD): fat X time interaction, p = The relationships between TS and both percentage of fat and fat mass were either significant (p ::5 0.05) or exhibited a strong trend toward significance up through 60 min of exposure. However, at later time points, no significance or trends toward such were noted. Generally, no significant (p > 0.05) relationship existed between TS and stature, body mass, body surface area, or fat free mass at any time point (Table 4). ity, the subsequent p-values for the within-subject effects were adjusted with the Hunyh-Feldt estimate. Post ANOVA analysis entailed a series of 1 df contrasts chosen a priori. The level of significance was set a priori (p ::5 0.05) and was maintained throughout all post-ano VA tests. Relationships between variables were described with a coefficient of correlation obtained from simple linear regression where the level of significance was set a priori (p ::5 0.05). Results THERMAL SENSATION (TS) DIFFERENCES BETWEEN HF AND LF SUBJECTS When the data were pooled over the five time points, the LF group (7.9 ± 0.5) had a significantly (p < 0.05) higher mean TS than the HF group (5.9 ± 0.6). As was expected, when TS was pooled across both fat groups, a significant (p < 0.01) increase in TS was reported as exposure time increased. However, despite what appear DIFFERENCES IN TRE BETWEEN HF AND LF SUBJECTS When Tre was pooled over time, LF subjects (-0.21 ± OA C) exhibited a greater (p < 0.05) change than HF subjects (-0.03 ± 0.04 C). Additionally, when Tre was pooled across both fat groups, the change was significantly (p < 0.01) greater at the later time points than at the earlier time points. Again, despite apparent differences in Tre between LF and HF subjects at the earlier time points, no significant (p > 0.05) fat X time interaction (Fig 2) was found, and the small sample size relative to variation magnitude is believed to be the reason for the lack of significance of this interaction. DIFFERENCES TSK BETWEEN HF AND LF SUBJECTS The Tsk did not differ (p > 0.05) between HF and LF subjects whee pooled over time, but a significantly (p < 0.01) lower Tsk was noted at the later time points com- Table 4. Relationship between thermal sensation and body composition variables across time Coefficient of correlation with thermal sensation 5 min 30 min 60 min 90 min 120 min Body composition variable r P r P r P r P r P Body mass (kg) Body stature (cm) Body fat (%)* ** Fat mass (kg) ** Fat-free mass (kg) Body surface area (m 2 ) * Percentage of fat determined via hydrostatic weighing. **p :

5 208 Glickman-Weiss et al 0.3 IT!- 0.2 l!! 0.1 II) <l r ;::====::::;-, = ' Fig 2. Rectal temperature differences between low and high fat subjects (mean ::!:: SD): fat X time interation p = o e: -1 l!! 2 II) Q. -3 t e -4 e :e -5 <l -6 L-=- ===== Fig 3. Skin temperature differences between low and high fat subjects (mean ::!:: SD): fat X time interaction, p = pared with the earlier time points when Tsk was pooled across fat groups. The fat X time interaction (Fig 3) was not significant (p > 0.05). DIFFERENCES IN AM BETWEEN HF AND LF SUBJECTS Metabolism did not differ significantly (p > 0.05) between HF and LF subjects when pooled over time and did not change significantly (p > 0.05) over time when pooled across the two fat groups. Despite what appear to be perceptual differences between the HF and LF groups at the earlier time points, no significant (p > 0.05) fat X time interaction (Fig 4) was noted. Small sample size relative to variation magnitude might account for the lack of statistical significance of this interaction. Discussion Body fatness is a significant factor in the relationship between human thermal equilibrium and environmental cold stress [13]. Additionally, LeBlanc [14] established positive relationships between skinfold thickness and insulation index (resistance of heat transfer from the core to the skin), and these relationships became stronger with decreased air temperature. Buskirk et al [15] similarly identified a linear positive relationship between Tre response and percentage of fat in looc air. In attempting to understand the data on low and high fat women in the present investigation, we examined the perceptual response. The HF subjects demonstrated an apparently, though nonsignificant, lower TS as compared with their LF counterparts (Fig 1). This perceptual rating may be linked to the physiologic and thermal responses and is believed to differ between low and high fat subjects. However, due in part to the statistical power of the data, several valuable trends and apparent differences in the data may be discussed. For example, when Tre was pooled over time, the LF group exhibited a greater change than the HF group, demonstrating a greater physiological strain in the LF than the HF group. Despite apparent differences between Tres in the LF and HF groups, which were greater in the earlier time points (Fig 2), no significant fat X time interaction was demonstrated. This thermal marker is similar to the perceptual marker, ie, TS. When pooled over time, TS was significantly higher in the LF group than in the HF group. Further, though perceptual differences between the LF and HF groups appeared to be greater at the earlier time points (Fig 1), no significant fat X time interaction was noted. The small sample size relative to variation magnitude might account for the lack of statististical significance of both of the aforementioned interactions. However, it is certainly worth noting, 200 I ;:::====:::::; E.r! 0 "5.Q.I! -50 <l l ' Fig 4. Metabolism differences between low and high fat subjects (mean ::!:: SD): fat X time interaction, p =

6 Perceptual and physiological responses of high and low fat cold-exposed women 209 from the standpoint of wilderness and survivability, that an apparent difference exists between LF and HF women in how they may sense and scale cold sensation, and this sensation is most likely linked to a core temperature response that differs according to one's percentage of fat, ie, passive tissue insulation. We found in previous work in our laboratory that low fat males were able to maintain similar core temperature to their high fat counterparts through a significantly higher metabolic rate, ie, shivering thermogenesis. This trend was not found in the present investigation, perhaps for several reasons. Buskirk and colleagues [15] suggested that the greater the individual's total body fat the greater the individual's total fat distribution to the subcutaneous tissue. Consequently, the amount of subcutaneous fat is apparently a more important factor than total body fat in the body's defense against cold air. Because females, on average, differ from their male counterparts and are typically fatter than males and distribute fat in both the upper (chest) and lower body (hip, buttocks, thigh region), the group variations in body fatness and, thereby, subcutaneous fatness may be considerable. This difference may, in part, explain the lack of metabolic adjustment that the males demonstrated to maintain Tre. However, the metabolic-type adjustment may be on average replaced by the passive tissue insulation differences. Additionally, available recent research indicates that the amount of cold stress, the individual's percentage of fat and level of fitness, and exposure time all have an impact on the physiological response to cold. The present investigation used a cold air stressor, the same stressor that was used to derive the Glickman-Weiss et al [8] thermal sensation scale. However, the low fat and high fat Glickman-Weiss et al [8] study utilized cold water as a thermal cold stressor, so perhaps the greater thermal stressor required the low fat individuals to maintain a hypermetabolic rate as opposed to a vasomotor adjustment to maintain Tre. Therefore, both the variance in the females' vs the males' percentage of fat and its distribution as well as the reduced thermal stressor imposed in air vs water may have contributed to different results between the present investigation and previous work in this area. Again, a limitation of the present investigation is the small sample size and, therefore, the reduced statistical power inherent in this study. However, the majority of the environmental/hypothermia literature does involve very small sample sizes; therefore, the statistical power of those investigations must be considered as well as the difficulty in obtaining subjects to volunteer to participate in this area of research. It is clear when perusing the available literature in this area that more research is needed. Additionally, because women physiologically respond to an environmental stressor and may attempt to maintain core temperature with a different physiological strain and, thereby, a different perceptual strain than their male counterparts, research in this dimension is also warranted. From a survival standpoint, more women today are involved in demanding physical activities and recreational pursuits. Therefore, the need to explore the physiological, perceptual, and metabolic responses of women is both critical and timely. Additionally, caution should be taken in interpreting the preponderance of available data that utilizes a sample of men and in attempting to generalize these data to women. Conclusion From these data, it appears that the modified TS scale tested may be a good marker of thermal discomfort in females exposed to the cold because perceptual differences were noted between LF and HF subjects that were accompanied by a lower Tre response (ie, the hallmark measure of hypothermia) during protracted resting cold exposure. References 1. Fine BJ. The comparative effectiveness of some psychological and physiological measures in ranking the impact of diverse environmental conditions. J Appl Psychol. 1958;42: Gagge Ap, Stolwijk AJ, Hardy JD. Comfort and thermal sensation associated with physiological responses at various ambient temperatures. Environ Res. 1967;1: Glickman-Weiss EL, Hearon CM, Nelson AG. A thermal perception scale for use during resting exposure to cold air. Percept Motor Skills. 1994;79: Borg G. A category scale with ratio properties for intermodal and interindividual and intermodal comparisons. In: Geissler HG, Petzolds P, eds. Psychosocial Judgement and the Process of Perception. Berlin, Germany: Deutscher Verlag Wissenchaften; 1982: Borg G, Ljunggren G, Ceci R. The increase of perceived exertion, aches and pains in the legs, heart rate and blood lactate during exercise on the bicycle ergometer. Eur J Appl Physiol. 1985;54: Glickman-Weiss EL, Nelson AG, Hearon CM, Prisby R, Caine N. Thermal and metabolic responses of high and low fat women during exposure to 5 and 27 C for 120 minutes Aviat Space Environ Med. In press. 7. Toner MM, McArdle WD. Physiological adjustments of man to the cold. In: Pandolf KB, Sawka MN, Gonzalez RG, eds. Human Performance Physiology and Environmental Medicine at Terrestrial Extremes. Brown and Benchmark; 1985:361.

7 Glickman-Weiss EL, Goss FL, Robertson RJ, Metz KF, Cassinelli DA. Physiological and thermal responses of males with varying body compositions during immersion in moderately cold water. Aviat Space Environ Med. 1991;62: Brozek J, Grande F, Anderson JT, Keys A. Analysis of body composition: Revision of some quantitative assumptions. Ann NY Acad Sci. 1963;110: Wilmore JA, Vodak PA, Girandola RN, Billing IE. Further simplification of a method for determination of residuallung volume. Med Sci Sports Exerc. 1980;12: Glickman-Weiss et al 11. Du Bois D, Du Bois EE A formula to estimate the approximate surface area if height and weight be known. Arch Int Med. 1915;17: Stephenson LA, Kolka MA. Thermoregulation in women. Exerc Sports Sci Rev. 1993;21: Toner MM, McArdle WD. Human thermoregulatory responses to acute cold stress. Handb Physiol-Environ Physiol. 1997;17: LeBlanc J. Subcutaneous fat and skin temperature. Can J Biochem Physiol. 1954;32: Buskirk ER, Thompson RH, Whedon GD. Metabolic responses to cold air in men and women in relation to total body fat content. J Appl Physiol. 1963;18: