INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

HYPOTHERMIA IS DIAGNOSED when core temperature (T_core) declines to 35°C or below (18). It is a life-threatening situation, usually resulting from exposure to either a cold outdoor climate or immersion in cold water (12). Certain T_core ranges have been used to categorize different stages of hypothermia and to evaluate the physiological limit humans can tolerate. Pozos et al. (18) defined the signs and the severity of hypothermia (mild, moderate, and profound) by T_core.

T_core as indicated by rectal temperature (T_re) may have been generally accepted as an appropriate physiological measure in the assessment of cold strain (9), although recent studies (16) have used the more rapidly responding esophageal temperature (T_es). However, T_core is not always reduced during cold exposure (18). Unchanged or even elevated T_re is often observed during the initial period of cold exposure (9, 28, 29). An increase in T_core during the initial period of cold exposure is attributed to the sympathetic nervous system mediating peripheral vasoconstriction, which results in redistribution of blood away from the periphery toward the core concomitant with an increase in metabolic heat production (3).

The most commonly used criteria to evaluate the degree of cold stress are ambient temperature (T_a) and wind chill. As T_a decreases, the gradient favoring heat flux from the body to the environment increases. Wind increases heat loss from the body to the environment (26). In 1945, Siple and Passel (22) developed the wind chill index to evaluate the cooling power of the environment by integrating the effects of T_a and wind velocity to assess the convective cooling power. Wind chill index also provides a convenient relative index for comparison between various experimental protocols and field situations. However, T_a alone does not adequately reflect cold strain, and there are also reports published concerning the limitations of the wind chill index (26). Wind chill and T_a only quantify stress to the unprotected body surface area. Further, the wind chill index is only applicable at wind speeds exceeding 20 m/s, overestimates the cooling power for a naked person, and underestimates the cooling power for a clothed person (26). Some corrections were suggested to overcome these limitations. For example, Boutelier (2) suggested inclusion of globe temperature in the index.

In 1950, Scholander et al. (21) described the critical temperature (CT) for exposures in air as the threshold T_a below which energy metabolism increases above the resting level. In 1962, Rennie et al. (19) defined CT for cold-water immersion as the lowest water temperature that an inactive subject could tolerate without exhibiting an increase in O₂. The utility of using CT in air or water to quantify individual resistance or tolerance to cold stress has been challenged (6, 23), and Toner and McArdle (25) discussed this issue and concluded that better indexes are required. In 1987, Bittel (1) suggested using heat debt developed during exposure to cold as an index to assess the extent of the cold adaptation of an individual. However, the complexity of calculating this index [e.g., dry heat exchanges by radiation and convection (R + C), metabolic rate, convective and evaporative heat loss by the respiratory tract, and heat loss by cutaneous perspiration] limits its utility for use to rate cold strain online.

Recently, Moran et al. (15) developed a physiological strain index (PSI) to evaluate heat stress. The PSI is a simple index based on only T_re and heart rate (HR). This index successfully evaluated the heat strain in men who were exposed to a variety of exercise intensities combined with different levels of hypohydration (14) and different combinations of metabolic rate, climate condition, and clothing (15).

The purpose of this study was to develop an analogous, simple PSI to be used in cold environments. This index should be able to differentiate between degrees of cold strain and capable of indicating cold strain in real time as well as able to analyze existing databases on a simple scale of 0-10.

	MATERIAL AND METHODS

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

Three different databases obtained from men exposed to cold were used in this study. The first database, consisting of T_es and HR responses measured during exposure to cold air, served to develop the new index (16). The second database, containing T_re and HR measurements taken from an independent study (28), was used to further validate the new index during cold-air exposures. The third database, taken from another separate study (5), was used to evaluate the applicability of the developed index for men during cold-water immersion and to compare the new index with an independent assessment of cold strain.

Protocol I. Nine healthy young men participated in this study (16). The physical characteristics of the subjects were as follows (means ± SE): age 24 ± 2 yr, height 178 ± 2 cm, body wt 77 ± 4 kg, and maximum rate of O₂ consumption (O_{2 max}) 55 ± 1 ml · kg¹ · min¹. Each subject, dressed in shorts, socks, and shoes, completed three experimental cold-air exposures. The three exposures were completed with the subjects at different levels of hydration [euhydration (Eu), hypertonic hypohydration (HH), and isotonic hypohydration (IH)]. All exposures were completed within 15 days, and subjects served as their own controls. Two methods of dehydration were employed to achieve 4-5% loss of baseline body weight. To induce HH, subjects exercised in the heat (13) on the day before the trial. After 3-4 h of mild-intensity exercise in the heat, which resulted in sweating, fluid replacement was restricted. To achieve IH, subjects ingested a diuretic on the day before the IH trial. Each experimental trial consisted of 30 min of rest in comfortable climatic conditions (25°C), followed by 120 min of rest in cold air (7°C, 40% relative humidity).

Protocol II. This study (28) examined the way in which chronic exertional fatigue and sleep deprivation, coupled with negative energy balance, affected thermoregulation during cold exposure. Eight healthy young men (age 28 ± 2 yr) with body weights of 67 ± 2, 74 ± 2, or 80 ± 2 kg (1st, 2nd, and 3rd trials, respectively) who were dressed in cotton athletic shorts and socks were exposed three times to 4 h of cold air (10°C). The first trial immediately followed completion of 61 days of strenuous outdoor military training (A), the second was after a short recovery of 48 h (SR), and the third trial was after long recovery of 109 days (LR). All three trials were conducted at the same time of the day and were in accordance with the same test protocol (28).

Protocol III. This study (5) examined whether serial cold-water immersions during one day would lead to a blunted response of the thermoregulation system. Eight healthy men (age 24 ± 4 yr, height 178 ± 3 cm, body wt 79 ± 3 kg, and O_{2 max} 50 ± 2 ml · kg¹ · min¹) dressed only in shorts were immersed into cold water (20°C) for 120 min three times (0700, 1100, 1500) during one day ("repeat" group). About 3-4 wk before the experiments, control exposures, in which only a single immersion was employed on one day, were conducted at the same starting times as the three trials during repeat. However, the control trials were performed randomly during different weeks. All cold-water immersion trials employed the same test protocol.

Measurements. The T_re was measured from a thermistor inserted 10 cm past the anal sphincter. In the first protocol, T_es was measured by a thermocouple in a catheter placed at heart level. Skin temperature was measured at nine (protocols I and III) or five (protocol II) skin surface sites, and mean weighted skin temperature (_sk) was calculated according to Gagge and Gonzalez (7). Mean body temperature (_b) was calculated as 0.67 · T_re + 0.33 · _sk. In these three protocols, HR was measured from an electrocardiogram obtained from three chest electrodes and radiotelemetered to an oscilloscope-cardiotachometer. In protocol III, metabolic heat production was estimated from O₂ uptake and respiratory exchange ratio (7). Cumulative body heat debt was expressed as a positive number and was defined as the total negative storage integrated over time. Perception of thermal sensation was rated by using a category rating scale (30). A number of the experiments during the three protocols had to be terminated before the scheduled 120- or 240-min exposure time because subjects voluntarily withdrew or their T_re reached 35.0°C, the medical safety limit.

Calculations. The PSI was calculated as suggested by Moran et al. (15) as follows

(1)

where T_{re 0} and HR₀ are the initial measurements, and T_{re t} and HR_t are simultaneous measurements taken at any time t.

1

2

(2)

(3)

Statistical analysis. Physiological responses for the different experimental exposures were analyzed with SAS version 6.12 software with the use of the general linear models (GLM) procedure (extension of ANOVA) for univariate repeated measurements analysis. Linear models for HR, T_re, T_es, and as dependent variables were fitted by the least squares method (regression procedure), using each group at the different experimental exposure and subject designation as the independent variables. Multiple comparisons were controlled by using Tukey-Kramer tests. For protocol III, a two-way repeated-measures ANOVA was utilized to determine whether significant differences existed between the appropriate control condition and the repeat trial at the same time of day. Significant F ratios were analyzed post hoc by using Newman-Keuls tests. Unless otherwise indicated, significant differences reported herein are at P < 0.05. All data are reported as means ± SE. The materials and methods for the cold-exposure experiments were presented in greater detail elsewhere (5, 16, 28).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

Database I. Generally, the physiological parameter exhibiting the largest relative drop during cold-air exposure was _sk. The fall in _sk was marked from immediately after the start of cold-air exposure throughout the end of the first hour; thereafter, the change was less pronounced. There were no significant differences (P > 0.05) in _sk responses among the trials. The T_es increased during the initial period (~30 min) of the trials (P < 0.05) and then decreased through the end of the experimental exposures (P < 0.05).

(4)

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Heart rate (HR; means ± SE) dynamics (top) and physiological strain index (PSI) obtained from HR and esophageal temperature (Tes) (bottom) of all subjects participating in cold-air experiments at 3 hydration levels: Eu, euhydration; HH, hypertonic hypohydration; IH, isotonic hypohydration (see Ref. 16). bpm, Beats/min.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Cold strain index (CSI) calculated from Tes and mean skin temperature (sk), applied to 3 subjects exposed to same cold-air stress (7°C, 40% relative humidity; see Ref. 16). A, B, and C: subjects I, II, and III, respectively.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Insulation index (top) and CSI (bottom) applied in evaluating 3 hydration levels: Eu, HH, and IH (see Ref. 16). # Insulation was higher during IH than HH (P < 0.03); * CSI was lower during HH than IH or Eu (P < 0.05).

Database II. The most common criteria to assess cold strain are the absolute values of T_re and _b or the changes in these temperatures during time of exposure (T_re or _b). The _b values, calculated from the T_re and _sk of eight subjects exposed to the same 4 h of cold air after different recovery periods from long strenuous military training, are presented in Fig. 4. No significant differences in _b were found between the three trials during the first 120 min of the exposure and during the 240 min between A and LR. However, significant differences in _b were found between A and SR or LR at 150 and 180 min (P < 0.05; Fig. 4, top). Analysis of _b during the three trials revealed no significant difference between A and SR trials. A smaller change was found in LR that was significantly different from A and SR after 150 min, through the end of the exposure (Fig. 4, middle). The cold-strain assessment of the three trials by CSI is presented in Fig. 4 (bottom). Generally, the LR trial was assessed with significantly lower values during the 240-min exposure than the SR and A trials (P < 0.01). The A trial had the highest values (P < 0.05); however, no significant differences were found between A and SR during the second hour (60-120 min) of the exposures.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Mean body temperature (b; top), change in b (b; middle), and CSI (bottom) obtained from 8 subjects participating in 3 cold-air (10°C) experiments at different recovery periods from 61 days of strenuous outdoor military training: A, 2 h after completing training; short recovery (SR), after 48-h recovery; and long recovery (LR), after 109 days recovery (see Ref. 28). Values are means ± SE. # b was lower for A than SR or LR at 150-180 min (P < 0.05); * b was lower during LR than SR (P < 0.01) or A after 150 min through end of exposures (P < 0.05);  CSI was lower during LR than A or SR (P < 0.01) and CSI was higher for A than SR at 10, 30, 150, and 180 min (P < 0.05).

Database III. Cold-water immersion causes greater physiological strain than exposure to air at a similar temperature. In Fig. 5, CSI was applied to T_re and _sk measured in eight men during the six experimental cold-water exposures (5). There were no _sk differences between control and repeat trials. Differences in T_re between the control and the matched repeat trials did not achieve significance until the very end of immersion (120 min) and then only in the 1100 trial (P < 0.05). However, CSI values for repeat immersion exposures at 1100 and 1500 after 80 min through the end of these exposures were higher in comparison with the matched control exposures (Fig. 5, right). Despite these insignificant differences, the CSI differentiated between the repeat trials, assessed with absolute higher strain, and the control trials, demonstrating habituation of the thermoregulatory system by the higher strain assessment.

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Comparison between rectal temperature (Tre; left) and CSI (right) obtained from 8 subjects immersed in 20°C water 3 times in 1 day (repeat) and a single immersion (control) conducted at same time of day (see Ref. 5).

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Heat debt vs. CSI (means ± SE) obtained from 8 subjects at 0700 (top), 1100 (middle), and 1500 (bottom) for control and repeat experiments during cold-water (20°C) immersion (see Ref. 5).

View larger version (17K): [in this window] [in a new window]   Fig. 7.   Rated thermal sensation vs. CSI (means ± SE) obtained from 8 subjects at 0700 (top), 1100 (middle), and 1500 (bottom) for control and repeat experiments during cold-water (20°C) immersion (see Ref. 5). Subjects rated their perception of thermal sensation using a numerical scale; lower numbers correspond to more pronounced sensation of cold.

View larger version (20K): [in this window] [in a new window]   Fig. 8.   Metabolic heat production vs. CSI (means ± SE) obtained from 8 subjects at 0700 (top), 1100 (middle), and 1500 (bottom) for control and repeat experiments during cold-water (20°C) immersion (see Ref. 5).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

The most important new finding from this study is that core and skin temperature measurements can be integrated into a useful new index of cold strain. The CSI, for the three different databases under investigation (5, 16, 28), successfully evaluated the degree of cold strain for men at different levels of hydration status, at different recovery periods from long strenuous outdoor military training, and at different fatigue stages of the thermoregulatory system. This simple-to-use index is based on only two physiological parameters: T_re (or T_es) and _sk, which adequately depict the overall strain reflected by the body when exposed to cold environments. The CSI was applied to cold-air exposures and to cold-water immersion and found to assess quantitatively the cold strain on a universal scale of 0-10.

It is generally accepted that T_re can describe physiological strain during cold exposures (18). However, T_re is independent of environmental temperature over a wide range; therefore, the use of T_re alone as an indication of thermal strain in the cold is of questionable value (9, 25). Minard (11) argued that changes in T_re fail to reflect body heat loss during transient cooling. Greenleaf et al. (8) showed that T_re declined slightly during the first 2 h of immersion in 34.5°C water; after 8 h, the T_re values were higher than the initial T_re. In the present study, cold exposure produced cold strain, as indicated by vigorous shivering during cold-water immersion (5) and in cold air (16, 28), and yet, T_core values often increased, especially during the start of an exposure to these cold environments. Thus T_re alone is often an inadequate quantifier of cold strain. Furthermore, _b and _b were of limited value in assessing cold strain (Fig. 4) for the same reasons, because T_re is the main determinator of _b.

The _sk, on the other hand, is rapidly affected by cold exposure. The skin is the barrier of the body interfacing with the ambient environment for heat-energy exchange by radiation, convection, and conduction. The skin, as stated by Porter and Gates (17), serves as "a transducer of the environment." The _sk can change over a wider range (up to sevenfold more) than T_re. However, although _sk may reflect changes in ongoing dynamic heat exchange, _sk provides relatively little indication of the heat content or the temperature within the body; for that, T_core can be assessed. Thus many investigators calculate _b. We concluded that combining T_re (or T_es), representative of the T_core, with _sk, representative of the heat transfer between the body and the environment, would provide an index of cold strain.

The impact of water immersion on heat-balance mechanisms is greater than that of a similar air temperature because conductivity of water is 25 times greater than that in air (19) and because heat loss in water is particularly sensitive to the skin-water temperature difference. As a consequence, temperature gradients become great, and physiological responses are dramatic (24). Thus, to further the appreciation of the versatility of the CSI, we examined the cold strain from databases obtained from both cold-air and cold-water immersion.

The three indexes (PSI, I index, and CSI) were applied to the first database obtained from cold air (16). The application of the PSI on these data resulted in low positive and negative values across all the trials (Fig. 1). PSI was limited in its ability to evaluate cold stress because of three reasons. First, PSI was originally constructed to evaluate heat stress. As a consequence, it assumes maximal physiological values relevant for heat exposure rather than cold exposure. Second, PSI is based on T_re and HR. However, HR is mainly affected by exercise intensity, whereas, in these cold studies, the subjects were seated at rest, resulting in a relatively small change in HR dynamics during the trials (Fig. 1). In addition, an imbalance between the parasympathetic and the sympathetic nervous systems during cold exposure causes a large variability in HR (10). Third, the PSI lacks any factors reflecting the heat exchange between the environment and the body.

The I index, based on T_es, _sk, and metabolic heat production, successfully discriminated the cold strain during cold exposures at different hypohydration levels. However, to calculate this index, a third parameter, metabolic rate, must be measured. This third parameter is not needed to calculate the simpler CSI. Application of CSI to the same databases clearly evaluated the relative strain with a simple scale ranging from 0 to 10. The use of individual T_es and _sk values for assessment of cold strain revealed that Eu had the lowest strain. However, CSI, which combined these two parameters, rated the HH with the lowest strain. The latter is explained by the smaller change in T_es during the 120-min exposure in the HH trials. The CSI effectively discriminated among cold strain at the different hypohydration levels, at different recovery periods from long outdoor strenuous training, and at different fatigue stages; it was found to be valid for both exposure in cold air and immersion in cold water.

The T_es values are generally lower than the corresponding T_re values (19). The latter values were also found in the cold-air database analyzed in this study (16). However, application of CSI to this database containing T_re (CSI_Tre) or T_es (CSI_Tes) and T_sk measurements revealed no significant differences between CSI_Tre and CSI_Tes for each of the three trials (P > 0.05). The magnitude of this difference suggests that CSI, based on either T_es or T_re, can provide meaningful values for the assessment of cold stress.

Evaluations of different cold strains by either the I index or heat debt and metabolic heat production were all found meaningful. However, the high correlation found between these assessments and the CSI strengthens the ability of CSI to serve as a simple tool constructed only from two parameters (T_core and _sk) for assessment of cold strain. Furthermore, rated thermal sensation during cold exposure as an independent method was found to be highly correlated with CSI.

The ability of an individual to be protected from cold climates and to avoid cold strain can be enhanced by proper clothing (25, 26). An appropriate clothing configuration considering insulation and water permeability can protect an individual from a hostile cold environment. The clothing used during cold exposure establishes a microenvironment at the skin surface and between the skin and the inner side of the clothing that is insulative in effect (26). However, all three databases used in this study had subjects dressed in only shorts or shorts, socks, and shoes. Thus, to apply CSI to different types of clothing, more studies should be conducted for proper validation.

In conclusion, the newly developed index is a complement to the existing literature regarding evaluation and assessment of cold strain. Although CSI might not provide as detailed a description of cold strain as heat debt calculation (1) or CT (26), it is significantly easier to calculate online. Thus CSI may be a useful tool, especially during online data acquisition and when metabolic rate measurements are not available. The CSI may also be useful for explaining discrepancies in experimental findings obtained by different investigators using various ambient conditions by providing a common reference point. This index has the potential to be widely accepted and to serve universally. However, further investigation is required to possibly adjust the CSI for a wider range of cold air and water temperatures and to consider exercise effects.