Thermoregulatory Responses with Size-matched Simulated Torso or Limb Skin Grafts

Abstract

Skin grafting following a burn injury attenuates/abolishes sweat production within grafted areas. It is presently unknown whether the thermoregulatory consequences of skin grafting depend on anatomical location.

Purpose:

To test the hypothesis that a simulated burn injury on the torso will be no more or less detrimental to core temperature control than on the limbs during uncompensable exercise-heat stress.

Methods:

Nine non-burned individuals (7 males, 2 females) completed the protocol. On separate occasions, burn injuries of identical surface area (0.45 ± 0.08 m² or 24.4% ± 4.4% of total body surface area) were simulated on the torso or the arms/legs using an absorbent, vapor-impermeable material that impedes sweat evaporation in those regions. Participants performed 60 min of treadmill walking at 5.3 km·h⁻¹ and a 4.1% ± 0.8% grade, targeting 6 W·kg⁻¹ of metabolic heat production in 40.1°C ± 0.2°C and 19.6% ± 0.6% relative humidity conditions. Rectal temperature, heart rate, and perceptual responses were measured.

Results:

Rectal temperature increased to a similar extent with simulated injuries on the torso and limbs (condition-by-time interaction: P = 0.86), with a final rectal temperature 0.9 ± 0.3°C above baseline in both conditions. No differences in heart rate, perceived exertion, or thermal sensation were observed between conditions (condition-by-time interactions: P ≥ 0.50).

Conclusion:

During uncompensable exercise-heat stress, sized-matched simulated burn injuries on the torso or limbs evoke comparable core temperature, heart rate, and perceptual responses, suggesting that the risk of exertional heat illness in such environmental conditions is independent of injury location.

INTRODUCTION

Severe burn injuries requiring skin grafts lead to profound thermoregulatory dysfunction (1). Removal of injured tissue and transplantation of grafts impairs/abolishes sweat production in wounded skin regions (2, 3), and the ensuing reduction in evaporative heat loss capacity (4) causes greater heat strain in burn survivors with extensive skin grafts compared to non-injured individuals during exercise-heat stress (5-7). It follows that burn survivors who work or train in hot environments may be more likely to experience decrements in physical performance and may be at greater risk for heat-related illnesses.

U.S. Army Regulation 40-501, Medical Services Standards of Medical Fitness (8), states: “Prior burn injury involving less than 40 percent total body surface area, which results in a loss or degradation of thermoregulatory function does not meet the standard. Examination will focus on the depth of the burn, anatomic location (extensive burns on the torso will most significantly impair heat dissipation), and destruction of sweat glands.” A likely basis for this standard is that local sweat rates are often found to be lower on the limbs than on the torso (9-19). If lower sweat rates on the limbs result in less extensive local sweat coverage relative to the torso, an individual with a torso burn injury (who must rely more on the limbs for sweat production) could have a lower capacity for whole-body evaporative heat loss, leading to a higher elevation in core temperature during physiologically uncompensable exercise-heat stress compared to a burn survivor with a limb injury of equivalent size and intact torso sweat function. Whether this is in fact the case has not been investigated.

The aim of the present study was to determine whether torso burn injuries are more detrimental to core temperature regulation than burn injuries on the limbs during physiologically uncompensable exercise-heat stress. To this end, a simulated burn injury model was used to replicate the impact of a size-matched burn injury on evaporative heat loss within torso and limb skin areas in non-injured individuals.

METHODS

Ethical Approval

The Institutional Reviews Boards of the University of Texas Southwestern Medical Center, Texas Health Presbyterian Hospital Dallas, and the Human Research Protections Office of the Defense Health Agency approved the study protocol (STU 062018-086), which conformed to standards set forth in the Declaration of Helsinki. All study procedures and risks of participation were fully explained to the participants prior to obtaining informed written consent.

Participants

Nine participants (7 males, 2 females) completed the study. All were non-smoking, normotensive, and physically-active individuals (age: 29 ± 6 yr; mass: 72.4 ± 11.38 kg; height: 1.75 ± 0.07 m; body surface area (BSA): 1.86 ± 0.17 m²; body mass index: 23.7 ± 2.4 kg·m⁻²). Female participants were tested during the follicular phase of the menstrual cycle. No medications were being taken at the time of participation, and no known history of cardiovascular, neurological, respiratory, or metabolic diseases was reported. The study was conducted during the winter and spring seasons in Dallas, Texas.

Instrumentation and Measurements

Urine specific gravity was measured via handheld refractometer (Atago Inc., Bellevue, WA). Body mass was measured with a platform scale (Mettler Toledo PBD655-BC120, Toledo, OH), and standing height was measured with a stadiometer (Detecto, Webb City, MO). Mass and height values were used to calculate BSA with the DuBois equation (20). Rectal temperature was measured with an indwelling general-purpose pediatric thermocouple probe inserted 10 cm beyond the anal sphincter (Mon-a-therm, Mallinckrodt Medical, St. Louis, MO). Heart rate was taken from an electrocardiogram (GE Medical Systems, Madison, WI). Rectal temperature and heart rate were sampled at 25 Hz (Biopac MP150, Santa Barbara, CA). Thermal sensation was assessed using a numeric scale that ranged from “unbearably cold” to “unbearably hot” (21). Ratings of perceived exertion were scored using the 6-20 Borg scale (22). Rates of oxygen uptake (VO₂) and CO₂ production (VCO₂) were determined from the analysis of expired gases and volumes using a metabolic cart (Trueone 2400, Parvomedics, Sandy UT), which was calibrated before each trial according to manufacturer instructions. Metabolic rate (M) was calculated from VO_2, the respiratory exchange ratio (RER = VCO₂·VO₂⁻¹), and the caloric equivalents for carbohydrate (e_c, 21.13 kJ·L⁻¹ O₂) and fat (e_f, 19.62 kJ·L⁻¹ O₂) oxidation (23):

$$M={\text{VO}}_2\frac{\left(\left(\frac{\text{RER}-0.7}{0.3}\right)e_c\right)+\left(\left(\frac{1.0-\text{RER}}{0.3}\right)e_f\right)}{60}\times 100(\mathrm{W})$$

External work rate (Wk) was calculated as:

$$\text{Wk}=9.81m_{\mathrm{b}}{v}_{\text{tr}}F(\mathrm{W})$$

where 9.81 represents acceleration due to gravity (m·s⁻²), m_b is total body mass (kg), v_tr is the treadmill belt velocity (m·s⁻¹), and F is the treadmill grade (%grade·100⁻¹). The rate of metabolic heat production was taken as the difference between metabolic rate and the external work rate, and then normalized to total body mass.

To compare the thermoregulatory consequences of torso vs. limb burn injuries, a simulated burn injury model was employed as described previously (24, 25). Absorbent pads with a vapor-impermeable outer layer were cut to cover the entire torso (0.45 ± 0.08 m² or 24.4% ± 4.4% of total BSA) circumferentially between the iliac crest and the lower border of the clavicle, leaving 1.41 ± 0.18 m² of skin surface area available for evaporative heat loss. Absorbent pads for the simulated full torso burn were secured with a stress test elastic vest retainer (Surgilast, Derma Sciences, Princeton, NJ). In female participants, the absorbent material was placed between the skin surface and a sports bra. For simulated limb burns, absorbent pads of identical within-subject surface area (relative to that used for the full torso burn simulation) were split equally between right and left arms and legs, and secured using tubular mesh bandages (Owens & Minor MediChoice, Mechanicsville, VA).

Experimental Protocol

Participants visited the laboratory on two occasions scheduled at the same time of day, but separated by at least 48 h. Prior to each visit, participants were instructed to avoid alcohol and strenuous exercise for 24 h, avoid caffeine for 12 h, and to consume a light meal and ~500 mL of water 2 h prior to the start of their study visit. During the first visit, participants underwent standard screening procedures, including a resting blood pressure measurement and a 12-lead electrocardiogram.

Prior to instrumentation, a urine sample was collected to assess hydration status and to run a pregnancy test in female participants. Euhydration was accepted if urine specific gravity was ≤1.025 (26). Participants then recorded a nude body mass and dressed in a standard clothing ensemble consisting of cotton shorts, socks, running shoes, and a sports bra for females. After instrumentation, the simulated burn injury was applied to either the torso or the limbs. The order in which the “Torso” and “Limbs” trials was random and counterbalanced. Participants then entered the climate chamber with environmental conditions of 40.1°C ± 0.2°C and 19.6% ± 0.6% relative humidity. Baseline measurements were taken, after which the exercise protocol commenced at a treadmill speed of 1.48 m·s⁻¹ (5.3 km·h⁻¹ or 3.3 mph) and an initial grade of 4.0%. If necessary, the treadmill grade was adjusted to target a rate of heat production of 6.0 W·kg⁻¹ of total body mass, which is consistent with moderate-intensity military activities such as foot patrol (27). Metabolic measurements were captured from 0 to10, 25 to 35, and 50 to 60 min of exercise. Ad libitum fluid consumption was permitted between metabolic measurements, with the temperature of the ingested water kept at body temperature using a heated water bath. Ratings of thermal sensation and perceived exertion were collected at 20, 40, and 60 min of exercise.

Data and Statistical Analyses

Rectal temperature and heart rate data were analyzed as average values collected over 2-min time periods ending at 0, 15, 30, 45, and 60 min of exercise. For the change in rectal temperature, heart rate, ratings of perceived exertion, and ratings of thermal sensation, a two-way repeated-measures analysis of variance (ANOVA) was used with the repeated factor of time (0, 15, 30, 45, and 60 min) and simulated burn injury location (Torso, Limbs). The change in rectal temperature from baseline to the end of exercise was further compared between trials using a paired Student’s t-test. A Greenhouse-Geisser correction was applied if the assumption of sphericity had been violated. Statistical analyses were performed using Prism 8.0 (Graphpad, La Jolla, CA). Data are reported as means ± standard deviations. Alpha was set a priori at the 0.05 level.

RESULTS

During the 60-min exercise bout, mean VO₂ (Torso, 18.6 ± 1.2 mL·kg⁻¹·min⁻¹; Limbs, 18.3 ± 1.2 mL·kg⁻¹·min⁻¹; P = 0.11) and the corresponding mean rate of metabolic heat production (Torso, 5.8 ± 0.5 W·kg⁻¹; Limbs, 5.7 ± 0.5 W·kg⁻¹; P = 0.24), did not differ between trials.

Baseline rectal temperature averaged 36.8°C ± 0.5°C and 36.7°C ± 0.5°C in the Torso and Limbs trials, respectively. Rectal temperature increased over time (Fig. 1; main effect: P < 0.01), but no condition-by-time interaction was observed for this variable (P = 0.86). The 60-min change in rectal temperature from baseline was not different between trials (Torso: 0.9°C ± 0.3°C; Limbs: 0.9°C ± 0.3°C; P = 0.55).

FIGURE 1—

Change in rectal temperature during 60 min of exercise in hot-dry conditions with simulated burn injuries of equivalent size on the torso or limbs. Main effects of condition and time, and the condition-by-time interaction, are shown. Data represent means ± standard deviations for nine participants.

Heart rate (Fig. 2) and the ratings of thermal sensation and perceived exertion (Fig. 3) increased over time (main effect: P < 0.01). However, the location of the burn injury did not affect the increase in heart rate (interaction: P = 0.91), ratings of thermal sensation (interaction: P = 0.50), or ratings of perceived exertion (interaction: P = 0.65).

FIGURE 2—

Heart rate response during 60 min of exercise in hot-dry conditions with simulated burn injuries of equivalent size on the torso or limbs. Main effects of condition and time, and the condition-by-time interaction, are shown. Data represent means ± standard deviations for nine participants.

FIGURE 3—

Ratings of thermal sensation (top) and ratings of perceived exertion (bottom) during 60 min of exercise in hot-dry conditions with simulated burn injuries of equivalent size on the torso or limbs. Main effects of condition and time, and the condition-by-time interaction, are shown. Data represent means ± standard deviations for nine participants.

DISCUSSION

The current study is the first to examine whether the anatomical location of a burn injury alters the thermoregulatory response to uncompensable exercise-heat stress. Our results demonstrate that the elevation in core temperature following 60-min of exercise-heat stress did not differ with a simulated burn injury imposed on the entire torso compared to a simulated burn injury on the limbs matched for surface area. Further, no differences in heart rate, perception of exertion, or thermal sensation were evident. Collectively, these results suggest that a full torso burn injury is no more detrimental to thermal, cardiovascular, and perceptual strain than size-matched burn injuries on the limbs.

Studies of thermoregulatory function following a burn injury typically report the percentage of total body surface area injured (%TBSA) without indicating the specific skin regions injured (5-7), making potential effects of burn location on thermoregulatory outcomes impossible to discern from previous studies. A notable exception is the investigation by Austin et al. (28). In that study, two participants were included with similarly-sized third-degree burn injuries (35% TBSA) and skin graft donor sites (27% and 30% TBSA): one with injuries on or near the torso (chest, abdomen, upper arms, back, and head), and the other with injuries only on the limbs (posterior thighs, legs, arms, and buttock). Whole-body sweat rate was 50% lower, and the elevation in core temperature was 0.6°C higher, in the participant with limb burns and, presumably, an intact torso sweat response. While these findings imply superior thermoregulatory control in burn survivors with torso vs. limb burns, these data represent a comparison between only two subjects unmatched for sex, body size, or the rate of metabolic heat generation, each of which influences the core temperature response to exercise-heat stress (29, 30).

In the current study, core temperature responses to moderate-intensity exercise in a hot-dry environment were compared between conditions of simulated torso and limb burn injuries. The simulated burn injury model, which we have employed previously (24, 25), enabled us to impede evaporative heat loss in precisely size-matched areas of the torso and limbs. This approach also allowed us to perform within-subject comparisons of thermoregulatory responses, thereby avoiding potentially confounding effects of differences in sex, heat acclimation status, aerobic capacity, etc. on observed responses with an independent-groups experimental design. Additionally, standardization of the rate of metabolic heat production and environmental conditions ensured that the heat load imposed was nearly identical in each experimental condition. With these experimental controls, any observed differences in core temperature control could be ascribed solely to the impact of burn location. In opposition to the findings of Austin et al. (28), the magnitude by which core temperature increased did not differ between the Torso and Limbs trials after 60 min of exercise in the heat, suggesting that whole-body evaporative heat loss was unaffected by the location of the burn injury (Fig. 1).

The notion that a torso burn injury is more detrimental to heat dissipation is likely based on regional differences in sweat rate. Sweat rates on the torso tend to be greater than those on the limbs (9-19), particularly within the scapular and lumbar regions and along the spinal column (15, 16, 19). This pattern of sweat distribution, and its impact on evaporative capacity, may be more apparent in individuals unacclimatized to heat stress (11, 19) and/or during exposure to a hot-dry environment. Higher humidity levels may diminish differences in torso vs. limb sweat rates (31, 32) due to greater skin wettedness requirements and poorer evaporative efficiency. In the present study, participants were not heat-acclimated or seasonally acclimatized to heat stress, and exercise was performed in a hot-dry, physiologically uncompensable environment, so any differences in evaporative capacity, secondary to differences in regional sweat rate, would manifest through the core temperature response. Unfortunately, the extent to which sweat rate differed between the torso and limbs cannot be determined from the current data. Nonetheless, the absence of any difference in the core temperature response, despite a matched heat load between Torso and Limbs trials, suggests that if a difference in sweat rate occurred between torso and limb sites, it was insufficient to alter evaporative capacity.

Two additional factors could alter the thermoregulatory impact of size-matched torso vs. limb burn injuries. With a more rapid sweat onset and/or higher sweat sensitivity (i.e., rate of change in sweat rate relative to core body temperature) on the torso compared to the limbs, a torso burn would theoretically attenuate whole-body sudomotor responsiveness, contributing to a higher initial rate of body heat storage. Data regarding regional disparities in sweating onset are inconsistent. Delayed sudomotor onset in the thighs and/or arms has been observed (33, 31), yet others have found delayed sudomotor onset in torso vs. non-torso sites (e.g., chest or upper back vs. arms or thighs) (12, 32) or no differences in sudomotor onset between torso and non-torso sites (29, 34, 35). Regional differences in sudomotor sensitivity are similarly inconsistent, with studies demonstrating higher sudomotor gain on the torso relative to the limbs (12, 33) or no differences between torso and limb sites (29, 31, 32). Additionally, variations in local skin temperature could change the average skin vapor pressure and thus the drive for evaporative heat loss, depending on the burn location. However, given the environmental conditions used in the current study (approximately 40°C and 20% relative humidity), regional differences in skin temperature between the torso and limbs were likely negligible (36). Although the extent to which these factors affected the current results cannot be determined from the present data, the net effects of regional sweat rate variability, sudomotor onset and sensitivity, and local skin temperatures were apparently inconsequential towards altering core temperature between trials.

Heart rate, ratings of perceived exertion, and thermal sensation were not different in the Torso and Limbs trials (Fig. 2-3). These parameters were assessed in addition to core temperature because of their well-known effects on physical performance during heat stress (37, 38). Our findings suggest that the effects of a burn injury on physiological and perceptual sources of strain, and possible downstream effects on performance, are unlikely to be dependent on whether the burn injury is located on the torso or limbs during moderate-intensity continuous work in hot conditions.

Perspectives

According to the U.S. Army’s Standards of Medical Fitness (8): “Prior burn injury involving less than 40 percent total body surface area, which results in a loss or degradation of thermoregulatory function does not meet the standard. Examination will focus on the depth of the burn, anatomic location (extensive burns on the torso will most significantly impair heat dissipation), and destruction of sweat glands” (italicized for emphasis). Based on the absence of any difference in the core temperature response to exercise-heat stress between torso and limb simulated burns, U.S. Army personnel performing prolonged work in the heat would be at no greater risk of heat-related illness or injury with a torso burn compared to a burn located on the limbs. Therefore, a revision to the Standard could be made that omits “anatomic location” from the list of burn injury-related factors under examination when determining whether a burned recruit or soldier meets the Standard.

It is also possible that the focus on “anatomic location” in the current Standards of Medical Fitness could in fact increase the risk of heat-related health problems under circumstances in which body armor must be worn. Body armor, which is worn on the torso to protect against ballistic and fragmentation injury to vital organs, impedes evaporative heat loss from covered areas of the torso surface (39). For this reason, the presence of a torso burn injury does not further reduce whole-body heat loss compared to non-injured individuals when body armor is worn (40). In other words, both torso-burned and non-burned individuals rely heavily on non-torso skin (i.e., limbs and head) for heat loss when body armor is worn. However, for soldiers with limb burn injuries, which occur more frequently in combat settings (41), the area of non-torso skin available for evaporative heat loss would be greatly reduced, likely resulting in a more profoundly attenuated capacity for evaporative heat loss when body armor is worn compared to an individual with only torso burns. Whether this is in fact the case needs to be investigated.

Considerations

The use of the simulated burn injury model to replicate the impact of a burn injury/skin grafting on evaporative heat loss has limitations, which have been addressed previously (24, 25). One issue that has not been raised previously is whether the simulated burn model impedes the evaporation of dripping sweat. In actual burn survivors, sweat dripping from anatomically superior skin areas with intact sudomotor function could theoretically evaporate from inferior skin regions with grafts. However, using the simulated burn injury model, this sweat would be absorbed, contributing nothing to whole-body evaporative heat loss. While this may have been the case, the extent to which our results were influenced by the absorption of dripping sweat is unknown.

CONCLUSION

In summary, a simulated burn injury imposed on the torso led to similar elevations in core temperature, heart rate, perception of exertion, and thermal sensation during prolonged uncompensable exercise-heat stress compared to a size-matched simulated burn injury on the limbs. This finding suggests that in uncompensable conditions, a full torso burn injury does not impede evaporative heat loss, and thereby exacerbate heat strain, to a greater extent than an identical size burn injury on the limbs. It follows that torso burns should not be viewed as more detrimental to whole-body thermoregulation when considering whether a US Army recruit or soldier with a burn injury < 40% of TBSA meets the Standards of Medical Fitness.