Cooling Effectiveness of a Modified Cold-Water Immersion Method After Exercise-Induced Hyperthermia

Exertional heat stroke (EHS) is one of the most common causes of sudden death among athletes and military personnel.^1–4 The risk increases during the summer months, when ambient temperature and relative humidity increase.³ Adding protective equipment further compromises normal thermoregulation.⁵ The diagnostic criteria for EHS are a core temperature of more than 40.5°C plus alterations in central nervous system function.^1,6 Once a patient is diagnosed, it is critical that core body temperature be reduced to below 38.9°C as quickly as possible to prevent sequelae and to limit the chance of mortality.^6,7

The National Athletic Trainers' Association recommends whole-body cold-water immersion (CWI) as the best method for treating EHS.⁶ Colder water temperatures optimize CWI (ie, provide faster cooling), particularly with continuous water circulation to prevent a barrier of warm water from forming adjacent to the patient.^6,8–12 Oscillating the water supplies a massaging effect to the skin, which increases vasodilation of the peripheral vasculature.¹² Therefore, circulated cold water allows more blood to be cooled at the periphery through conduction and convection and subsequently transported to the core.

Cold-water immersion typically requires a large tub filled with water and ice to be on-site and ready for use during activities. However, Mazerolle et al¹³ found that many athletic trainers do not implement this method due to lack of resources and insufficient daily time to maintain the tub. Also, transporting a large tub to certain venues, such as military drills and off-road races, is problematic. Although some¹² contend that patient comfort and cardiovascular status are compromised by CWI, the method consistently demonstrates successful outcomes.^{6,8–12,14,15}

A recent review¹¹ of cooling rates for a variety of modalities identified categories for cooling rates (ie, how long it would take an EHS patient to be cooled appropriately). For efficient cooling to take place in less than 30 minutes, whole-body cooling must be ≥0.078°C·min⁻¹. Typical cooling rates for CWI range between 0.129°C·min⁻¹ and 0.350°C·min⁻¹ and produce 100% survival in EHS patients.^8,9,14 In some military and athletic situations, continual rotation of ice-soaked towels or sheets has been successful as an alternative to CWI.¹⁶ However, limited evidence to date supports the use of CWI alternatives for the treatment of exercise-induced hyperthermia.

The tarp-assisted cooling with oscillation (TACO) method was developed to address the challenges of CWI. A plastic tarp held by staff members serves as the container for cold water while the patient sits or lies in the middle. The method is portable and inexpensive (<$20) and has been used successfully in the military and during the 2012 Boston Marathon to treat patients with EHS (C. Troyanos, oral communication, May 2012; M. B. Smith, oral communication, June 2011). Despite anecdotal reports of success with TACO, no studies have been conducted to determine its effectiveness. Therefore, the purpose of our investigation was to determine the cooling rate achievable through the TACO method. We hypothesized that TACO would cool participants at an acceptable rate (≥0.078°C·min⁻¹) after exercise-induced hyperthermia.

METHODS

Instrumentation

Participants were provided with a heart-rate (HR) strap (model T31; Polar, Lake Success, NY) to wear for the duration of the exercise and treatment. A blood pressure cuff was placed on the upper arm and 3 electrocardiograph electrodes were attached (right and left subclavicular fossa and right anterior axillary line) to allow arterial blood pressure to be measured by auscultation of the brachial artery via electrosphygmomanometry (model Tango+; SunTech Medical, Inc, Raleigh, NC). Participants inserted a rectal thermistor (model RET-1; Physitemp Instruments, Inc, Clifton, NJ) 15 cm beyond the anal sphincter for rectal temperature (T_re) measures throughout trials. Participants were educated on perceptual measures (rating of perceived exertion,¹⁷ thirst,¹⁸ thermal sensation,¹⁹ and perceived muscle pain²⁰) that were assessed throughout exercise and treatment.

Exercise Protocol

After the instrumentation was applied, the participant entered an environmental chamber (model ETC 01; Can-Trol Environmental Systems, Ltd, Markham, ON, Canada; temperature = 33.4°C ± 0.8°C, relative humidity = 55.7% ± 1.9%). He or she then sat on a chair for 10 minutes, after which baseline values for T_re, blood pressure, HR, and perceptual measures were obtained. The blood pressure cuff and electrode leads were removed immediately, and the participant moved to and began exercise on either a bicycle or treadmill. The leads and cuff were removed for participant safety, to prevent their catching on the exercise equipment. During the first trial, each participant self-selected exercise intensity, mode, and duration; these values were recorded by the research staff as sufficient to elevate T_re to at least 39°C. This exercise protocol was repeated for the second trial, ensuring a match in metabolic heat production during exercise. The T_re and HR were recorded continuously throughout exercise and cooling via computer software (model LabChart7; ADInstruments Inc, Colorado Springs, CO) at 50 Hz during exercise and treatment. Perceptual measures (rating of perceived exertion, thirst sensation, thermal perception, and muscle pain) were taken every 10 minutes during exercise. Exercise was ceased immediately if a participant began to experience signs or symptoms of exertional heat illness or body temperature reached 41°C. During exercise, participants were encouraged to consume ad libitum water heated to 38°C. Once a T_re of at least 39°C was reached, they ceased exercise and moved to the cooling portion of the trial.

Treatment Protocol

A standardized transition of 5 minutes was allowed to simulate effective movement of a patient with EHS to a treatment area. Participants moved to the treatment area inside the environmental chamber and assumed a semirecumbent position on a standard tarp (8 × 10 feet [2.5 × 3.1 m]) held at the corners and edges by researchers (minimum of 6). Before starting treatment, the blood pressure cuff and leads were reapplied, and precooling perceptual measures were taken. To begin treatment, the researchers elevated their tarp corners, forming a taco shape.

During the TACO treatment, 151 L of ice water (temperature = 2.1°C ± 0.8°C) from four 37-L water coolers were poured on and around the participant, ensuring that the head and upper chest were above the water. The participant remained immersed for 15 minutes or until T_re had returned to 38.1°C (approximately 101°F). After immersion, the participant was assisted in exiting the tarp, toweled dry, and returned to sitting on a chair inside the environmental chamber for 30 minutes. During this latter portion of recovery, T_re and HR were recorded every 10 minutes. The participant then exited the chamber to towel completely dry and remove the instrumentation.

During the CON trial, the participant remained semirecumbent in the dry tarp inside the environmental chamber for 15 minutes. He or she then moved to the chair and sat inside the environmental chamber for 30 minutes before toweling completely dry and removing the instrumentation. For both treatments, T_re and HR were monitored continuously during and after treatment. Blood pressure was measured immediately precooling and postcooling. Perceptual measures were taken at the beginning and end of the treatment as well.

Cooling rate was calculated using the following equation:

Blood pressure values were used to calculate mean arterial pressure (MAP) as follows:

RESULTS

The 24-hour diet logs demonstrated no differences between trials (kilocalories: t₁₅ = −0.97; P = .35; ES = 0.24; 95% confidence interval [CI] = −834, 312; protein: t₁₅ = −1.40; P = .18; ES = 0.35; 95% CI = −38, 8; carbohydrates: t₁₅ = −0.85; P = .93; ES = 0.02; 95% CI = −41, 37; fat: t₁₅ = −0.88; P = .39; ES = 0.22; 95% CI = −58, 24, data not shown). Hydration status was not different between trials when measured using either body mass (t₁₅ = 0.03; P = .98; ES = 0.01; 95% CI = −0.4, 0.4) or urine osmolality (t₁₅ = −1.91; P = .08; ES = 0.48; 95% CI = −168, 9). Fluid consumption during exercise was not different between the CON (0.26 ± 0.29 L) and TACO (0.19 ± 0.26 L; t₁₂ =1.73; P = .11; ES = 0.48; 95% CI = −0.02, 0.14) treatments.

An interaction of time and treatment was demonstrated for T_re (F_1,15 = 50.40, P < .001; partial η² = 0.77), given that it was similar at the pretreatment time point for both conditions (CON: 39.27°C ± 0.26°C; TACO: 39.30°C ± 0.39°C; P = .62; ES = −0.09; 95% CI = −0.17, 0.10) but was lower posttreatment during the TACO (38.10°C ± 0.16°C) compared with the CON (38.74°C ± 0.38°C, P < .001; ES = 2.27; 95% CI = 0.43, 0.86) condition. Cooling rate was faster during TACO than during CON (t₁₅ = −8.84, P < .001; ES = 2.21; 95% CI = −0.13, −0.08; Figure 1). The HR was not different between conditions (F_1,15 = 1.15, P = .30; partial η² = 0.07; 95% CI = −3, 10), and no interaction was evident (F_1,15 = 3.29, P = .09; partial η² = 0.18). However, HR decreased independent of treatment (F_1,15 = 182.52, P < .001; partial η² = 0.92; 95% CI = 47, 65) from pretreatment (158.8 ± 17 beats/min) to the end of cooling (102.2 ± 11 beats/min). Change in HR from pretreatment to posttreatment was not different between CON (−51 ± 19 beats/min) and TACO (−62 ± 22 beats/min; t₁₅ = −1.81, P = .09; ES = 0.45; 95% CI = −22, 2). An interaction of time and treatment was noted for MAP (F_1,13 = 12.51, P = .004; partial η² = 0.49; Figure 2), which was greater at the end of treatment in TACO than in CON (P < .001; 95% CI = 13, 21).

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

An interaction of time and treatment was shown for thirst (F_1,13 = 19.12, P < .001; partial η² = 0.60), with lower thirst ratings posttreatment in TACO than in CON (P < .001; 95% CI = 0.9, 2.3; Figure 3). Thermal sensation also exhibited an interaction of time and treatment (F_1,13 = 140.17, P < .001; partial η² = 0.92), with responses lower posttreatment in the TACO compared with the CON (P < .001; 95% CI = 2.6, 4.1; Figure 3) condition. In addition, muscle pain demonstrated an interaction effect (F_1,13 = 14.48, P = .002; partial η² = 0.53) and increased posttreatment in TACO versus CON (P = .007; 95% CI = 0.7, 3.3; Figure 3).

View Figure

• Download as Powerpoint
• Download Figure

During posttreatment recovery, a significant interaction of time and treatment was present for T_re (F_3,42 = 5.72; P = .02; partial η² = 0.29; Figure 4). The HR during posttreatment recovery was lower in the TACO than the CON condition regardless of time point (F_1,14 = 58.93, P < .001; partial η² = 0.81; 95% CI = 17, 30) and decreased independent of treatment (F_3,42 = 16.55, P < .001; partial η² = 0.54), with no interaction of treatment and time (F_3,42 = 2.40, P = .08; partial η² = 0.15; Figure 4).

View Figure

• Download as Powerpoint
• Download Figure

DISCUSSION

We found that using TACO for whole-body cooling of individuals with hyperthermia was more effective than the CON condition. Furthermore, our data established that TACO provided a cooling rate previously deemed acceptable for the treatment of patients with EHS.¹¹ Decreased body temperature, safe cardiovascular responses, and enhanced perceptual outcomes were evident with TACO after exertional hyperthermia. Our data are important for athletic trainers who work with military and athletic personnel and develop emergency action plans that must be executed in remote or restricted areas where traditional CWI using a tub may be impossible.

Though our water temperature (2.1°C ± 0.8°C) was comparable with that of the coldest water used in other studies, our cooling rate was slightly less than the CWI cooling seen elsewhere (up to 0.35°C·min⁻¹).⁹ This difference is likely due to the fact that our participants were immersed only to the iliac crest, whereas the Proulx et al⁹ participants were immersed in 2°C water up to their clavicles. Even with the limitation that TACO is partial-body CWI, our cooling rate of 0.14°C·min⁻¹ would allow for safe cooling of EHS patients within 30 minutes, making this an acceptable (>0.078°C·min⁻¹) modality for emergency treatment.¹¹

For ethical reasons, we could not allow individuals to reach core temperatures indicative of EHS (>40.5°C). Strong evidence indicates that the rates at which hyperthermic study participants and patients with EHS cool when using CWI are similar. This point was supported by data²¹ from 274 patients with EHS who were cooled with CWI at a mean rate of 0.22°C·min⁻¹ ± 0.11°C·min⁻¹. Reports¹⁰ indicate that cooling rates using CWI for hyperthermic participants ranged from 0.129°C·min⁻¹ to 0.35°C·min⁻¹. Cooling rates were similar in the literature between patients with EHS and hyperthermic research participants.^10,11,15

For field use, TACO has potential benefits compared with a traditional CWI tub. Tarps can be easily folded and stored in medical kits and can cost as little as $15, whereas a stock tank costs approximately $200. We found that 151 L of ice water was sufficient to submerge participants at least up to the iliac crest with little variation in water temperature. This amount of water was chosen during pilot testing because most venues have at least 4 coolers of ice water available in case of emergency. To use this method in the field, we recommend having at least 6 people to assist in holding the tarp because 151 L of water weighs approximately 150 kg, independent of the patient's body weight. As with all emergency action plans, medical staff and all individuals who may be involved in emergency procedures at the venue should practice this method before an actual emergency occurs. Furthermore, practice should be conducted with the recognition that many patients with EHS become combative due to central nervous system dysfunction. Standing close to the patient helps to maintain the water level and allows the tarp to be oscillated efficiently with repeated knee bends to maintain water circulation. Although we did not quantify the amount of water circulation in the current study, having the same researchers for each trial ensured consistent oscillations. In addition, given our acceptable cooling rate, this method provided sufficient movement to facilitate cooling.

Our cooling rates using TACO were slightly better than those previously identified using constantly rotating towels or sheets for heat-illness treatment.¹⁶ When ice-soaked towels were continually rotated, Armstrong et al¹⁶ identified a mean cooling rate of 0.11°C·min⁻¹ in 7 patients with EHS. Other anecdotal reports have described constant rotation of ice sheets as successful in treating heat casualties in the military. However, this treatment option requires further research documentation to be considered evidence based. One advantage of rotating ice-soaked towels or sheets is that the treatment requires less cold water than our TACO trial (approximately 38 L versus 151 L, respectively). This factor may decrease the feasibility of TACO when cold water is in short supply, yet it may also be a limiting factor in the cooling rate. The more water in contact with the patient's skin, the better the cooling, and less water used in treatment may compromise cooling efficiency.^10,12 On the basis of our data, TACO is recommended over ice-sheet rotations. Future researchers should investigate and document the efficacy of ice-sheet rotations.

Our participants had similar body types, so we suggest future researchers use this method on a more heterogeneous sample population to ensure effectiveness for all body types and provide the best recommendations for use in the field. For example, previous researchers^10,21 have shown that differences in body surface area (BSA) and the ratio of BSA to lean body mass affect cooling rates in hyperthermic individuals. Because CWI relies on conductive and convective heat loss, it follows that a greater BSA allows more contact with the cold water and, therefore, a greater amount of heat loss can be achieved. Even though individuals with a larger BSA relative to body mass index may have greater adiposity, this had a limited effect on cooling rate.²¹

The HR responses to water immersion after exercise vary from nonsignificant decreases in thermoneutral water to significant decreases in cold water (<16°C).^22,23 Although our change in HR was not different between the CON and TACO conditions, the greater MAP (Figure 2) in TACO postcooling indicates better maintenance of cardiac function during TACO than during the CON condition. During CON, the participants were semirecumbent and not moving for 15 minutes; blood likely pooled in the lower extremities and delayed recovery. Furthermore, Kenny et al²⁴ demonstrated reductions in sweating and skin blood flow while esophageal temperature was still elevated after exercise. These reductions occur from nonthermal contributions leading to impaired heat loss. Given that our participants' MAP decreased similarly at the end of treatment, it is possible that these perturbations in heat-loss mechanisms may have led to the slower cooling rates in our CON trial.

Perceptual measures provided insight into how our participants felt during cooling. Thirst was rated lower postcooling in TACO than in CON, which is likely due to the maintenance of blood pressure and circulation during immersion. Because we used an extreme water temperature, we expected to find decreased thermal sensation with TACO compared with CON. Our identified increase in muscle pain with cooling may reflect participants' difficulty differentiating between muscular and cutaneous pain sensation caused by the water temperature. Some of our participants began to shiver during cooling as well, which may have increased muscular pain.

CONCLUSIONS

The purpose of our study was to examine the cooling rate of TACO after exercise-induced hyperthermia. Given the resulting cooling rate of 0.14°C·min⁻¹ ± 0.06°C·min⁻¹, TACO is an acceptable method of reducing an individual's core temperature in an appropriate amount of time for emergency treatment of EHS. Though CWI in a traditional tub remains the criterion standard for treatment of patients with EHS, TACO could be included in emergency action plans for remote venues or in situations where the use of a CWI tub is not immediately feasible.