Learn more: PMC Disclaimer | PMC Copyright Notice
Temperate-Water Immersion as a Treatment for Hyperthermic Humans Wearing American Football Uniforms
Abstract
Context:
Cold-water immersion (CWI; 10°C) can effectively reduce body core temperature even if a hyperthermic human is wearing a full American football uniform (PADS) during treatment. Temperate-water immersion (TWI; 21°C) may be an effective alternative to CWI if resources for the latter (eg, ice) are unavailable.
Objective:
To measure rectal temperature (Trec) cooling rates, thermal sensation, and Environmental Symptoms Questionnaire (ESQ) scores of participants wearing PADS or shorts, undergarments, and socks (NOpads) before, during, and after TWI.
Design:
Crossover study.
Setting:
Laboratory.
Patients or Other Participants:
Thirteen physically active, unacclimatized men (age = 22 ± 2 years, height = 182.3 ± 5.2 cm, mass = 82.5 ± 13.4 kg, body fat = 10% ± 4%, body surface area = 2.04 ± 0.16 m2).
Intervention(s):
Participants exercised in the heat (40°C, 50% relative humidity) on 2 days while wearing PADS until Trec reached 39.5°C. Participants then underwent TWI while wearing either NOpads or PADS until Trec reached 38°C. Thermal sensation and ESQ responses were collected at various times before and after exercise.
Main Outcome Measure(s):
Temperate-water immersion duration (minutes), Trec cooling rates (°C/min), thermal sensation, and ESQ scores.
Results:
Participants had similar exercise times (NOpads = 38.1 ± 8.1 minutes, PADS = 38.1 ± 8.5 minutes), hypohydration levels (NOpads = 1.1% ± 0.2%, PADS = 1.2% ± 0.2%), and thermal sensation ratings (NOpads = 7.1 ± 0.4, PADS = 7.3 ± 0.4) before TWI. Rectal temperature cooling rates were similar between conditions (NOpads = 0.12°C/min ± 0.05°C/min, PADS = 0.13°C/min ± 0.05°C/min; t12 = 0.82, P = .79). Thermal sensation and ESQ scores were unremarkable between conditions over time.
Conclusions:
Temperate-water immersion produced acceptable (ie, >0.08°C/min), though not ideal, cooling rates regardless of whether PADS or NOpads were worn. If a football uniform is difficult to remove or the patient is noncompliant, clinicians should begin water-immersion treatment with the athlete fully equipped. Clinicians should strive to use CWI to treat severe hyperthermia, but when CWI is not feasible, TWI should be the next treatment option because its cooling rate was higher than the rates of other common modalities (eg, ice packs, fanning).
American football players are a population that is especially at risk for exertional heat illnesses such as heat exhaustion and exertional heatstroke (EHS). American football players have an 11 times higher risk of developing exertional heat illnesses than athletes in all other sports combined.1 In the last 84 years, 146 American football players have died from EHS.2 The greatest risk for developing severe hyperthermia occurs when the heat index risk is high or extreme.3
The current standard of care for EHS patients consists of rectal temperature (Trec) assessment followed by whole-body cold-water immersion (CWI, 1.7°C–15°C).4 When this standard of care is followed within 30 minutes of collapse, survival rates are 100%.5 Acceptable Trec cooling rates for EHS victims are 0.08°C/min to 0.15°C/min, whereas ideal cooling rates exceed 0.16°C/min.6
Some authors7–11 have argued that temperate-water immersion (TWI; ≥20°C) may be a suitable alternative to CWI for treating severe hyperthermia. Temperate-water immersion Trec cooling rates in minimally clothed hyperthermic participants varied from 0.06°C/min to 0.19°C/min in water baths with temperatures between 20°C and 26°C.10,12,13 Consequently, some scientists advocated for using TWI over CWI in EHS situations because it minimizes potential cold-shock responses and keeps the patient more comfortable during treatment, yet still reduces Trec at acceptable rates.8
Many experts14,15 recommend removing football equipment before EHS treatment, but we observed16,17 that hyperthermic humans wearing a full American football uniform (PADS) during CWI had excellent Trec cooling rates (>0.21°C/min) and similar thermal sensation and Environmental Symptoms Questionnaire (ESQ) responses as minimally clothed participants. However, it may be difficult to maintain water baths at recommended cold temperatures (1.7°C–15°C) because of limited access to ice or the bath being exposed to hot ambient conditions for extended periods of time. Therefore, the purpose of our study was 3-fold. First, we compared Trec cooling rates of participants wearing PADS or minimal clothing (NOpads) during TWI. Second, we questioned if wearing PADS during TWI would result in acceptable6 Trec cooling rates. Finally, we wondered if wearing PADS during TWI would affect thermal sensation18 or ESQ responses.19 Based on the results of preceding studies,16,17 we hypothesized that NOpads would not result in higher Trec cooling rates, lower thermal-sensation scores, or fewer environmental symptoms.
METHODS
Participants
A convenience sample of 13 healthy, recreationally active, unacclimatized men (Table 1) volunteered. Participants were excluded if they reported any of the following: (1) an injury that impaired the ability to exercise, (2) any neurologic, metabolic, gastrointestinal, respiratory, or cardiovascular disease, (3) taking any medication that could affect fluid balance or temperature regulation, (4) a sedentary lifestyle (defined as exercising <30 minutes, 3 times per week),20 (5) a history of heat-related illness in the 6 months preceding data collection, (6) illness at the time of data collection, or (7) any recent diarrheal illness, anal surgery, anal fistula, hemorrhoid, or anal fissure. All procedures were approved by our institutional review board, and participants provided written consent before testing.
Table 1.
Participant Demographics and Descriptive Information
| Variable | Group | ||
| Total (n = 13) | NOpads | PADS | |
| Mean ± SD | |||
| Age, y | 22 ± 2 | ||
| Height, cm | 182.3 ± 5.2 | ||
| Body mass index | 24.8 ± 3.8 | ||
| Body density | 1.08 ± 0.01 | ||
| Body fat, % | 10 ± 4 | ||
| Body surface area, m2 | 2.0 ± 0.2 | ||
| Body mass pre-exercise, kg | 82.5 ± 13.4 | 82.5 ± 13.4 | |
| Body mass postexercise, kg | 81.6 ± 13.2 | 81.5 ± 13.1 | |
| Pre-exercise urine specific gravity | 1.007 ± 0.005 | 1.005 ± 0.004 | |
| Pre-immersion water temperature, °C | 21.03 ± 0.01 | 21.02 ± 0.02 | |
| Hypohydration level, % | 1.1 ± 0.2 | 1.2 ± 0.2 | |
| Shivering onset, min | 8.8 ± 3.1 | 14.2 ± 6.1 | |
| No. | |||
| Participants who reported shivering | 8 | 3 | |
Abbreviations: NOpads, shorts, undergarments, and socks; PADS, shorts, undergarments, socks, T-shirt, practice jersey, shoulder pads, helmet, game pants, and padding over the thighs, knees, hips, and tailbone.
Procedures
Participants reported to the laboratory on 2 days separated by at least 48 hours. They were instructed to avoid exercise, caffeine, and alcohol for at least 24 hours before testing; maintain a consistent diet; drink water consistently the day before and the day of testing; and fast for 2 hours before testing. They self-reported compliance before each day's testing.
On testing days, participants emptied their bladders completely so we could determine hydration status (model SUR-Ne refractometer; Atago USA Inc, Bellevue, WA). If they were euhydrated (urine specific gravity < 1.02),21 they were weighed nude. If they were hypohydrated, testing was rescheduled for at least 48 hours later. Skinfolds were measured in triplicate and averaged from the thigh, abdomen, and chest (model Baseline skinfold caliper #12-1110; Fabricated Enterprises, Inc, White Plains, NY).22 Skinfolds were summed and used to estimate body density23 and percentage of body fat.22 Body surface area was also estimated.24
Each participant donned a heart-rate monitor, self-inserted a thermistor (model 401; Advanced Industrial Systems, Prospect, KY) 15 cm past the anal sphincter,25 and put on PADS. For a complete description of the PADS worn, we direct the reader to previous work from our laboratory.16,17 The participant entered an environmental chamber (temperature = 40°C, relative humidity = 50%) and stood on a treadmill for 10 minutes to acclimate to the hot and humid environment. After the rest period, Trec, thermal sensation, and ESQ score were recorded. Participants performed successive 5-minute exercise bouts consisting of walking at 4.83 km/h (3 mph, 0% incline) for 3 minutes and then 2 minutes of running at 90% of their age-predicted heart-rate maximum. Heart rates were not recorded; we monitored heart rate during exercise only to ensure that participants were exercising at the correct intensity. Rectal temperature was recorded every 5 minutes during exercise; exercise continued without rest until Trec was 39.5°C.
Once Trec reached 39.5°C, participants stopped exercising and rated their thermal sensation. Then they performed 1 of 2 procedures, depending on which condition was being tested. For NOpads (first testing day), we helped participants remove all equipment and clothing except for their shorts, undergarments, and socks (uniform removal time = 1.3 ± 0.3 minutes). They immersed themselves up to the neck in temperate water (21°C; model 4247 noncirculating water tub with 1135.6-L capacity; Rubbermaid, Atlanta, GA). We defined TWI as 21°C water because other authors12 used similar water-bath temperatures, and CWI was defined as water with a temperature <15°C.4 Water temperature was constantly monitored and maintained at 21°C while participants exercised by adding ice or hot water as necessary.
For PADS (second testing day), participants removed only their shoes and waited to enter the tub for the same amount of time it took to remove their uniform during the NOpads condition. Once the participant's foot touched the water, we started a standard stopwatch to determine immersion duration. When Trec was 38.75°C (half finished with cooling), the participant rated his thermal sensation. Participants self-reported shivering (if any occurred) during TWI. The water bath was stirred every 2 minutes and Trec was recorded every 30 seconds. Participants remained immersed until Trec decreased to 38°C.
On reaching a Trec of 38°C, participants exited the water bath, rated their thermal sensation a fourth time, and completed the ESQ a second time. They sat in the heat for 15 minutes so we could observe any potential afterdrop. Rectal temperature was recorded every 5 minutes during recovery. After recovery, participants exited the environmental chamber, dried themselves, removed the rectal thermistor, were weighed nude a second time, and were then excused. No fluids were given to participants at any time during testing.
Statistical Analysis
Pre-exercise urine specific gravity, preimmersion water-bath temperature, hypohydration level, shivering onset, and exercise duration on each testing day were not statistically analyzed but are reported for descriptive purposes (Table 1 and Figure 1). Separate dependent t tests were used to determine if differences existed between equipment conditions for TWI duration and Trec cooling rates. Repeated-measures analyses of variance were used to determine if differences in thermal sensation or ESQ scores existed between uniform conditions over time. For the ESQ data, we summed the scores from the 16-item questionnaire and created a new cumulative score.19 Sphericity was assessed with the Mauchly test. Geisser-Greenhouse adjustments to P values and degrees of freedom were made when sphericity was violated. When significant interactions or main effects were observed, we used Tukey-Kramer post hoc tests to identify differences between uniform conditions at each time point. Significance was accepted when P < .05 (version 2007; Number Cruncher Statistical Software, Kaysville, UT).
Rectal temperatures during exercise (left) and temperate-water immersion (right) for participants wearing a full American football uniform or minimal clothing (mean ± standard deviation, n = 13). Data are shown until the shortest exercise and water-immersion durations common to at least 10 of the 13 participants. X-axis error bars indicate the standard deviation for the final times to reach 39.5°C during exercise and 38°C during temperate-water immersion. Abbreviations: NOpads, shorts, undergarments, and socks; PADS, shorts, undergarments, socks, T-shirt, practice jersey, shoulder pads, helmet, game pants, and padding over the thighs, knees, hips, and tailbone.
RESULTS
Data provided are means ± standard deviations with 95% confidence intervals (CIs). Temperate-water immersion durations (NOpads = 14.5 ± 5.7 minutes [95% CI = 11.1, 17.9 minutes], PADS = 13.6 ± 6.7 minutes [95% CI = 9.6, 17.7 minutes]; t12 = −0.56, P = .71; Figure 1) and Trec cooling rates (NOpads = 0.12°C/min ± 0.05°C/min [95% CI = 0.09°C/min, 0.15°C/min]; PADS = 0.13°C/min ± 0.05°C/min [95% CI = 0.10°C/min, 0.16°C/min]; t12 = 0.82, P = .79) were similar for each condition. Seven of 13 participants (54%) cooled more quickly in NOpads.
We did not observe an interaction between uniform condition and time for thermal-sensation scores (F2,22 = 0.6, P = .61), nor did we observe a difference between uniform conditions (F1,12 = 0.7, P = .43). However, thermal sensation changed over time (F3,36 = 171.2, P < .001; Figure 2). Pre-exercise thermal sensation was different from all other times (P < .05). Postexercise thermal sensation was higher than the scores reported when Trec was 38.75°C or post-TWI (P < .05).
Thermal-sensation scores before, during, and after temperate-water immersion while wearing a full American football uniform or minimal clothing (mean ± standard deviation, n = 13). Scale ratings ranged from 0 (unbearably cold) to 8 (unbearably hot). A score of 4 indicated participants were comfortable. a The pre-exercise score was different than scores at all other times (P < .05). b Postexercise > Trec at 38.75°C and postimmersion (P < .05). Abbreviations: NOpads, shorts, undergarments, and socks; PADS, shorts, undergarments, socks, T-shirt, practice jersey, shoulder pads, helmet, game pants, and padding over the thighs, knees, hips, and tailbone.
We did not observe an interaction between uniform condition and time (F1,12 = 1.5, P = .25) or a main effect of uniform condition for ESQ score (F1,12 = 0.33, P = .57; Table 2). However, fewer environmental symptoms were reported before exercise than immediately after TWI (main effect of time; F1,12 = 8.9, P = .01).
Table 2.
Environmental Symptoms Questionnaire Responsesa (n = 13)
| Statement | Condition, Mean ± SD | |||
| NOpads | PADS | |||
| Pre-Exercise | Post-TWI | Pre-Exercise | Post-TWI | |
| 1: I feel lightheaded. | 0 ± 0 | 1 ± 1 | 0 ± 0 | 1 ± 1 |
| 2: I have a headache. | 0 ± 0 | 0 ± 1 | 0 ± 0 | 0 ± 1 |
| 3: I feel dizzy. | 0 ± 0 | 1 ± 1 | 0 ± 0 | 1 ± 1 |
| 4: I feel faint. | 0 ± 0 | 0 ± 1 | 0 ± 0 | 0 ± 1 |
| 5: My coordination is off. | 0 ± 0 | 1 ± 2 | 0 ± 0 | 1 ± 1 |
| 6: It is hard to breathe. | 0 ± 0 | 0 ± 0 | 0 ± 0 | 0 ± 1 |
| 7: I have a chest pain. | 0 ± 0 | 0 ± 0 | 0 ± 0 | 0 ± 0 |
| 8: I have a muscle cramp. | 0 ± 0 | 0 ± 1 | 0 ± 0 | 0 ± 0 |
| 9: I feel weak. | 0 ± 0 | 1 ± 1 | 0 ± 1 | 1 ± 1 |
| 10: I feel sick/nauseated. | 0 ± 0 | 0 ± 1 | 0 ± 0 | 0 ± 1 |
| 11: I feel irritable. | 0 ± 0 | 0 ± 1 | 0 ± 0 | 0 ± 1 |
| 12: My heart is pounding. | 0 ± 0 | 0 ± 1 | 0 ± 0 | 1 ± 1 |
| 13: I feel feverish. | 0 ± 1 | 1 ± 1 | 0 ± 1 | 0 ± 1 |
| 14: I feel warm. | 1 ± 1 | 1 ± 1 | 1 ± 1 | 1 ± 1 |
| 15: My vision is blurry. | 0 ± 0 | 0 ± 1 | 0 ± 0 | 0 ± 0 |
| 16: I feel goosebumps. | 0 ± 0 | 0 ± 1 | 0 ± 0 | 0 ± 1 |
| Total (95% confidence interval) | 2 ± 3b (1, 4) | 9 ± 10 (3, 15) | 1 ± 1b (1, 5) | 8 ± 11 (2,14) |
Abbreviations: NOpads, shorts, undergarments, and socks worn during TWI; PADS, shorts, undergarments, T-shirt, socks, practice jersey, shoulder pads, helmet, game pants, and padding over the thighs, knees, hips, and tailbone worn during TWI; TWI, temperate-water immersion.
DISCUSSION
Some experts recommend removing PADS before CWI if an athlete is diagnosed with EHS.14,15 In contrast, the National Athletic Trainers' Association recommends initiating equipment removal during CWI because complete PADS removal can be time consuming.4 Delays in initiating CWI may also occur because the PADS are difficult to remove or patients are noncompliant. We undertook this study because there may be situations when CWI is not possible (eg, no access to ice, long exposure of water to hot ambient conditions). The current data suggest that (1) wearing PADS during TWI did not impede cooling and (2) TWI produced acceptable, though not ideal, cooling rates when hyperthermic humans wore PADS during immersion. These results are consistent with research16,17 demonstrating that PADS did not negatively influence CWI Trec cooling rates.
To date, this is the third experimental study demonstrating that PADS did not impair Trec cooling rates.16,17 The reasons for this finding are that the body still has access to water, even with equipment donned, and water is much more effective than air in removing heat.26,27 Thus, conductive and convective cooling can still occur. Not having to remove PADS before CWI or TWI simplifies the treatment paradigm, especially for lay responders; most secondary schools do not employ full-time medical professionals.1 Moreover, not having to remove PADS can save time if EHS treatment has been delayed or patients are noncompliant. Saving this time could be crucial given that EHS morbidity and mortality increase the longer an individual's Trec stays above 40.5°C.15,28 In the majority of cases, we recommend equipment removal after starting immersion4 although there may be situations when (1) a clinician suspects the athlete may have a respiratory or cardiac emergency or (2) the cooling tub is not large enough to accommodate a fully equipped athlete.
We observed an average Trec cooling rate of 0.125°C/min between groups. Proulx et al12 noted a cooling rate of 0.19°C/min ± 0.10°C/min when hyperthermic humans (Trec = 40°C) were immersed in 20°C water postexercise. Their12 faster cooling rates were likely due to their participants experiencing more convective cooling from the continuous circulation of water, whereas we stirred the bath only every 2 minutes. Rectal temperature cooling rates depend on a multitude of factors, including water-bath temperature, stirring rate, Trec at the onset of cooling, and patient anthropometrics.13,29 However, our Trec cooling rates were consistent with those of other authors10,13 who used water temperatures above 20°C to cool participants with exercise-induced hyperthermia in laboratory settings. In fact, the authors of a recent critically appraised topic11 reported an average Trec cooling rate of 0.11°C/min ± 0.06°C/min when water-bath temperature ranged from 20°C to 26°C.
Although TWI produced acceptable cooling rates, it is not as effective as CWI for reducing body temperature quickly. Using a similar exercise and treatment protocol, hyperthermic participants who wore PADS during CWI displayed Trec cooling rates of 0.28°C/min ± 0.12°C/min16 and 0.21°C/min ± 0.11°C/min.17 The clinical implication of all of these studies is that treatment duration will be longer if TWI is used instead of CWI and if PADS are not removed. Assuming an average cooling rate of 0.245°C/min from the CWI studies16,17 and the participants observed in this study, reducing an equipment-laden athlete's Trec from 42°C to 38.6°C approximately doubles the immersion duration if TWI is used (26.2 minutes) instead of CWI (13.9 minutes). Theoretically, a similar prognosis for the athletes treated with TWI and CWI would be likely because Trec would still be reduced within 30 minutes of EHS symptom development.15 Debate exists regarding TWI or CWI for EHS,8 but Trec cooling rates for water immersion are higher than for all other cooling modalities (eg, ice packs, fanning, chilled intravenous fluid infusion),6 and water immersion should be implemented whenever EHS is diagnosed. However, clinicians may consider taking practical steps to ensure that CWI is available in the event of an EHS emergency by keeping ice in coolers next to treatment tubs and keeping tubs in shaded areas.
Even though CWI is the criterion-standard treatment for EHS, there may be times when it is unavailable or impractical. In these situations, TWI may offer benefits. First, TWI requires fewer resources (eg, ice) to maintain water-bath temperature. Second, some authors8 have argued that patient discomfort, shivering, and the risk for potentially dangerous cold-shock responses (eg, increased sympathetic nervous system activity, reduced cerebral blood flow) may be lessened if hyperthermic patients undergo TWI. Like other authors,12,17 we noted fewer instances of shivering and a later onset of shivering when TWI was used. This is likely because more of the skin surface was exposed to convective cooling in NOpads, resulting in lower skin temperatures and greater afferent signaling to the preoptic area of the anterior hypothalamus.27 However, patient discomfort, shivering, and cold-shock responses are likely minor concerns given that EHS patients rarely shiver during water-immersion treatment30; the duration of treatment is often too short to elicit shivering responses12; participants report similar levels of discomfort with CWI as with TW12; and EHS survival rates are 100% when CWI is implemented within 30 minutes of symptom development.5 Third, hypothermic afterdrop is reduced with TWI. Afterdrop is a potential concern because only 10%–25% of clinicians surveyed used Trec in EHS situations.31 In the current study, participants' average afterdrop 15 minutes postimmersion was 0.81°C with NOpads and 0.68°C with PADS. With CWI, afterdrop can exceed 1.2°C.12,16 Thus, the extent to which patients may need to be rewarmed postimmersion is reduced if TWI is used to cool the patient. Yet if clinicians follow recommendations to use Trec to monitor patients' progress while cooling and remove EHS patients from cooling tubs when Trec drops below 38.9°C (102°F),4 afterdrop should not be a concern. Fourth, although PADS did not affect thermal sensation or ESQ scores in this study, the former scores were 1 point higher during treatment with TWI than with CWI.17 Thus, TWI provides minor improvements in patient comfort.
We acknowledge the possible limitations of our study. First, we did not quantify heat flux because we did not measure skin temperature during TWI. However, participants reported similar thermal-sensation scores postexercise, and thermal sensation is closely related to skin temperature.18 Similarly, our results cannot be explained by other well-known contributors to heat storage, including hypohydration level,32 intensity and duration of exercise,33 and environmental temperature and relative humidity,34 because these were similar for each condition. Second, shivering onset was self-reported and not quantified using oxygen consumption or heart rate as other authors12 did. Third, a potential order effect may have occurred because all participants removed PADS before TWI on the first day of testing. We did not counterbalance the trials because (1) we felt it was more important that participants experienced the same delay in treatment and (2) any physiological adaptation to heat (eg, increased skin blood flow)4 or cold (eg, shivering threshold, thermal comfort)27,35 requires more than 1 day of exposure. Finally, similar to the results of all laboratory hyperthermia studies in humans, our results may not apply to EHS situations or all equipment-laden athletes.
In summary, removing PADS before TWI did not increase Trec cooling rates or affect thermal-sensation or ESQ scores. Clinicians should strive to use CWI in the field, but if this is not possible, TWI produces acceptable6 (though not ideal) Trec cooling rates, regardless of how much equipment is worn during treatment. Because PADS do not significantly interfere with body cooling, it is imperative to start whole-body immersion as quickly as possible and then remove equipment for further lifesaving measures and medical transport as necessary.
ACKNOWLEDGMENTS
We thank Michael McPike, MS; Greg McGillvary, MA, ATC; and Joseph Fox, MA, ATC, from the Central Michigan University Athletics Department for donating the equipment for this study; Ms Leah Komisak and Ms Shelby Schneider for their help with data collection; and the Central Michigan University Office of Research and Graduate Studies for funding this project.