Effect of cold and warm water ingestion on physiological and performance parameters during simulated hot burn surgeries

ABSTRACT

This study assessed the effects of hydration and hydration/cooling on various psycho-physiological and cognitive responses in staff during a simulated burns surgery. Twelve participants completed three 2.5-h trials in the heat (33.6°C, 36.4% RH) whilst walking on a treadmill at a rating of perceived exertion of 12 on the Borg scale. Trials consisted of: i) ingestion of 37°C water (HYD); ii) ingestion of 5°C water (COLD); and iii) a no cooling/hydration control (CON). Water ingestion (0.9% of body-mass) was based on fluid loss calculated during a previous 2.5-h burn surgery. Results demonstrated that while treadmill distance was similar between trials (p > 0.05), cold water ingestion resulted in improved manual dexterity (p = 0.03), better thermal comfort (p < 0.01) and lower core and skin temperatures (p < 0.01), compared to CON. Skin temperature was also lower in COLD vs HYD (p < 0.01). Moderate to large effect sizes (ES, g = 0.38–0.77) were observed in favor of COLD versus CON and/or HYD for manual dexterity, counting span, grammatical reasoning and several perceived workload subsets at various time points, however associated 95% confidence intervals were wide and crossed zero, suggesting statistical uncertainty. Similarly, moderate to large ES (g = 0.45–0.77) favored HYD over CON for counting span (120 min) and various perceived workload outcomes, though again confidence intervals suggest that these effects were not statistically conclusive. No differences were observed between trials for sweat loss, thermal sensation, or heart-rate (p > 0.05). Overall, cold water ingestion resulted in benefit to numerous variables assessed here. Small boluses of cold water ingestion are recommended during hot burn surgeries.

Introduction

Burn patients with injuries covering ≥20% of their total body surface area often experience compromised thermoregulation, leading to excessive heat loss [1]. To counteract this, operating theaters (OT) are usually kept between 30–40°C [1] to prevent infections and organ damage resulting from hypothermia [2]. While a hot OT can benefit the burn patient, it can cause excessive heat accumulation in surgical staff. Heat accumulation increases sweating, which can result in hypohydration, cardiovascular strain (elevated core temperature [T_c] and heart-rate), as well as thermal strain [3]. Notably a T_c ≥38.5°C (hyperthermia) [3] has been associated with heat illnesses and impaired exercise and cognitive performance [4–7]. Further, the wearing of personal protective equipment (PPE) by surgical staff leaves minimal skin exposed for heat dissipation [8], which can further exacerbate cardiovascular and thermal strain. Moreover, burn surgery staff typically avoid drinking water during surgeries to prevent bathroom breaks and the need to scrub back into surgery [9]. This can result in greater dehydration, which can hinder heat dissipation due to reduced blood flow to the skin required for cooling [10]. Even mild dehydration can impede thermoregulation during physical activity in hot conditions [11,12].

To date, only three studies have assessed the effects of working in a hot (31–34°C) compared to a cooler OT (23–24°C) during ~2.5 h of either simulated [13,14] or real-world burn surgeries [9]. Results demonstrated that staff working in the heat during simulated conditions had significantly higher T_c, heart-rates, and sweat loss, as well as a tendency for impaired manual dexterity [13] and complex cognitive performance [14], compared to working in cooler OT. Additionally, 13 of 17 staff members in the hot OT were found to be mildly to significantly hypohydrated prior to the simulation [13]. In real-world burn surgeries, staff working in hot OT had higher heart-rates and T_c, as well as higher levels of perceived fatigue, perceived workload demands, thermal sensation, thermal discomfort and ratings of perceived exertion, compared to when working in cooler OT [9]. Sweat loss was also significantly higher in staff working in hot OT compared to cooler surgeries, with urinary specific gravity (USG) indicating greater hypohydration levels post hot burn surgeries [9]. Overall, these results highlight the need to explore hydration and cooling strategies to mitigate thermal-related issues in burn surgery staff.

An obvious hydration strategy for burn surgery staff is water ingestion, which can mitigate plasma volume loss from sweating, thus assisting in the maintenance of peripheral blood flow and potentially preventing an excessive rise in T_c and hence cardiovascular and thermal strain [15]. Of importance to staff is that water ingestion does not result in the need for bathroom breaks. Therefore, the strategy of attempting to match water intake (using small regular doses over time) with sweat loss calculated from previous burn surgeries of similar durations [13] may address this concern.

In respect to cooling strategies, traditional cooling methods used in exercise and occupational settings are not suitable for the sterile environment of a hot OT. For example, fans are generally discouraged due to concerns about surgical site infections [16]. Notably, cooling vests, tested during real-world orthopedic surgeries (90–150 min in duration) performed in an ambient temperature of 20°C [17], were found to improve thermal comfort, reduce fatigue, and decrease perceived exertion in medical staff. However, these findings are not directly applicable to hot burn surgeries, where ambient OT temperatures range between 30–40°C and surgeries can last longer than 150 min [18]. Additionally, commercially available cooling vests typically provide cooling for only ~120 min and weigh ~1 kg, which could be restrictive for some staff. A promising alternative that addresses both heat and dehydration concerns is cold water ingestion. Cold water ingestion has been found to decrease T_c and improve endurance (cycling) performance, as well as improve perceptions of thermal sensation and thermal comfort [15], all of which could positively impact cognitive and physical performance of burn surgery staff.

Therefore, this study aimed to simulate a burn surgery environment and assess the effects of consuming either body-temperature water (37°C: hydration intervention) or cold water (5°C: hydration and cooling intervention) in regular doses designed to match sweat loss, compared to standard procedures (no water intake/control), on physical and cognitive performance and psycho-physiological measures in participants with similar demographics to burn surgery staff. It was hypothesized that ingestion of body-temperature or cold water during a 2.5-hour surgery simulation would improve work done (treadmill distance covered), preserve cognitive and manual dexterity performance, and reduce perceived workload compared to a control trial. It was also expected that cold water ingestion would offer greater benefits due to its combined hydration and cooling effects, leading to greater reductions in thermal and cardiovascular strain. Finally, it was hypothesized that matching water intake with sweat loss calculated from previous burn surgeries of similar durations would not result in bathroom breaks.

Materials and methods

Twelve participants (4 female and 8 males) were recruited for this study. This sample size was determined from an a priori power analysis using G*Power (Version 3.1.9.3) based on our primary variable: physical performance (total distance covered) (α = 0.05, 1–β = 0.8). The effect size (ES) used in this analysis was 0.47, based on reported average ES for differences in physical performance in prolonged self-pacing running laboratory studies [19,20]. Demographics for the twelve participants were: mean age 50.9 ± 8.9 y, height 176.8 ± 7.0 cm, and body-mass 75.3 ± 12.6 kg. Participants had comparable demographics (age range and gender balance) to nurses and doctors described in the studies by Palejwala et al. [9,13], and were all employed in professional occupations. Participants gave informed consent prior to commencement of trials, and Ethics approval was granted by the Human Research Ethics Committee of the University of Western Australia (UWA: 2023/ET000310). All mandatory laboratory health and safety procedure as determined by UWA were complied with during this study.

This study used a randomized, counterbalanced, crossover design as determined by a computer-generated program designed to ensure the equal distribution of interventions. No imputations were made for missing data. Burn surgery environmental simulations were conducted in a climate chamber set to 34°C and 40% relative humidity (RH) to replicate the environmental conditions of a previous burn surgery simulation [13]. During the experimental trials, participants walked on a treadmill at a rating of perceived exertion (RPE) of 12 (between “light” and “somewhat hard”) on the 6–20 Borg scale [21]. This RPE value was based on average values reported by surgery staff after ∼2.5 h of work in recent real-world hot burn surgeries [9]. This approach is similar to that of Uchiyama et al. [22], who used treadmill walking at a target RPE of 11 to simulate working on a hot mine site. Walking was chosen as the task as it is a common physical activity undertaken by all surgical staff, regardless of their specific duties during surgery. Participants wore PPE typically worn by burn surgery staff [9,13], that included scrubs, surgical aprons, surgical caps, masks, gloves and enclosed shoes. Participants visited the laboratory on four occasions approximately one week apart: one familiarization session followed by three separate surgical environmental simulation trials. The three trials consisted of: i) ingestion of 37°C (body-temperature) water (HYD trial); ii) ingestion of 5°C water (COLD trial), representing internal cooling and hydration intervention; and iii) a no cooling/hydration control (CON trial), representing standard practice in hot burn surgeries. In the two water trials, participants consumed 0.9% of their body-mass in water, divided into four equal portions every 30 min, commencing at the 20 min timepoint. This amount was based on average sweat loss measured in a previous hot burn surgery simulation [13]. Water was ingested from a paper cup via a bendable straw that could be inserted under the face mask, allowing participants to drink as quickly as desired at each timepoint. Male participants completed trials approximately one week apart. To minimize confounding effects of the menstrual cycle, females completed trials within the follicular phase of a single or two consecutive cycles.

Participants attended a familiarization session approximately one week prior to their first experimental trial. Anthropometric measurements including height (cm) and body-mass (kg) were recorded, and a questionnaire was administered to determine recent heat exposure and potential heat acclimation/acclimatization (with none found). Non-menopausal female participants provided information regarding their menstrual cycles (i.e. typical cycle length and the first day of their last menstrual period) to ensure that their trials occurred during their follicular phase. Participants also practiced the cognitive and manual dexterity tests (detailed below) and engaged in 5 minutes of treadmill walking at an RPE of 12. Finally, participants were asked to complete a food and activity diary 24 h prior to their first trial and to replicate this diary for the 24 h period preceding the subsequent two trials.

Experimental protocol

Prior to each experimental session, participants ingested a T_c capsule (CorTemp, HQ Inc., Palmetto, United States of America; USA) approximately 8 h before the trial as per the manufacturer’s instructions. Additionally, female participants were required to complete an ovulation urine strip test to confirm that they were in their follicular phase. Upon arrival to each trial, participants provided a ∼20 mL mid-stream urine sample to determine USG. Nude body-mass (NBM) was measured to the nearest 0.1 kg using a digital platform scale (SOEHNLE, Style sense comfort 100, Germany) to determine sweat lost. Participants were fitted with a heart-rate monitor (Polar T31, Finland), and four wireless skin thermistors (iButton DS1922L, USA) were taped to the left side of the chest (T_chest), left mid-posterior calf (T_leg), left mid-anterior thigh (T_thigh), and left mid-anterior forearm (T_arm). Participants then entered the climate chamber, where baseline measurements of T_c, skin temperature (T_skin), heart-rate, manual dexterity, cognitive tests, thermal sensation and thermal comfort (described below) were made. The exercise protocol involved treadmill walking consisting of 10 intervals of 15 min each. Assessments of manual dexterity, cognitive performance, thermal sensation and thermal comfort were performed at various time points when the participant was off the treadmill (described below). After completing the specified tasks, treadmill walking resumed, and this process continued throughout the trial. Importantly, participants could alter the treadmill speed, using an unlabeled dial located on the treadmill in front of them, to maintain an overall RPE of 12. Participants were reminded of this RPE target every 10 min during the walking protocol. Workload demands, as measured by the NASA Task Load Index (NASA-TLX: described later), were assessed immediately on completion of the walking task. Participants were advised that they could take bathroom breaks if needed. A commercially available purpose designed container was provided (if required) to measure urine output to assist in determining body-mass loss.

Physical performance

Physical work completed was assessed by the total treadmill distance covered (m) over the 2.5 h duration of treadmill walking, with distance recorded using a computerized program (Treadmill Measuring System 2.0, UWA, Australia). Participants were blinded to the treadmill speed and distance throughout all trials.

Cognitive and manual dexterity performance

Cognitive function was assessed using three tests: two computerized tests on a laptop which consisted of: (i) a counting span task [23]; (ii) Baddeley’s 3-minute grammatical reasoning test [24]; and (iii) a verbal test: the serial sevens task [25].

The counting span task (millisecond software) assessed working memory and executive function. This test was performed at 0 min, 60 min, 120 min and 150 min time points of the exercise protocol. Participants were shown cards featuring 3–9 green target dots (which they were required to count), and 3–9 yellow distractor dots. After a given number of cards (commencing with a span size of three and increasing to seven, with two trials per counting span), participants were required to recall the number of target dots in a serial manner starting with the first card they were presented with. Baddeley’s grammatical reasoning (millisecond software) assessed fluid reasoning through sentence comprehension and decision making (“true” or “false”) according to grammatical rules [24]. This task was performed at 0 min, 30 min, 90 min and 150 min time points, with a higher score representing higher accuracy.

The serial sevens test assessed working memory and concentration and was performed at 0 min, 15 min, 45 min, 75 min, 105 min, 135 min and 150 min marks. In this verbal task, participants were required to count backwards by sevens from a randomly generated number between 900 and 1000 for 60 s. Responses were recorded and scored based on the number of correct subtractions. Errors were noted, and subsequent subtractions were scored based on the new number after the mistake. It is important to note that these cognitive tests are not specific to cognitive skills required in burn surgeries but rather reflect tasks that could be considered as transferable to other settings.

Manual dexterity, which measures the ability to make skilled hand movements under time constraints, was assessed using the Perdue pegboard test [26]. The test involved placing as many pegs as possible in a straight line with the dominant hand within 30 s [26]. Manual dexterity was assessed at 0 min, 30 min, 60 min, 90 min, 120 min and 150 min time points. Test-retest reliability assessment of the Purdue Pegboard task was performed in a previous study, yielding a typical error score of ±0.5 and a coefficient of variation of 3.1% for the dominant hand (12). This “general skills” task was chosen due to its ability to be easily mastered by burn surgery staff (nurses and surgeons) who possessed a range of specialized skills.

Workload demands

The NASA-TLX assessed workload demands across multiple dimensions including mental, physical and temporal demands, performance, effort and frustration after the completion of a trial [27]. Participants rated their perceived workload in these categories on a 20-point Vermin Scale ranging from “very low” to “very high.”

Thermal sensation and thermal comfort

Thermal sensation and thermal comfort were measured using color-coded scales ranging from “very cold” to ‘very hot; (green to red scale) and from “very comfortable” to “very uncomfortable” (white to black graduation) [28], respectively. Measurements were taken at 0 min, 15 min, 45 min, 75 min, 105 min, 135 min and 150 min. Scores ranged from 0 to 20 on a color-coded scale visible to the researcher only, with higher scores representing feeling hotter and more uncomfortable for thermal sensation and thermal comfort, respectively.

Physiological responses

Heart-rate, T_c and T_skin were recorded at baseline and after each 15-min bout of treadmill walking throughout the experimental session. Mean skin temperature was estimated using the following equation by Ramanathan [29]: T_skin = (0.3× T_chest)+ (0.3× T_arm)+(0.2× T_thigh)+ (0.2× T_leg). Nude body-mass was measured pre- and post-trial and sweat loss was calculated using the following equation: sweat loss = Pre-trial NBM minus post-trial NBM plus fluid intake minus urine output. Urine specific gravity was used to assess hydration status pre-trial, and was classified according to Kavouras [30] where USG > 1.030, 1.010–1.020 and < 1.010 represented “serious dehydration,” “minimal dehydration,” and “well hydrated,” respectively.

Statistical analysis

The data were analyzed using R Studio (Version 1.4.1717). A linear model (obtained using the lmer function) was used to assess all dependent variables across all time points and trials through the analysis of variance test function. Post-hoc comparisons were conducted using the Tukey test. Significance was accepted at p ≤0.05. Where appropriate, post hoc comparisons using Bonferroni adjustments were conducted. All results in tables are expressed as mean ± SD, while all figures are presented as mean ± SEM. Due to our sample size, Hedges’s g effect sizes (ES) were calculated to assess differences in primary variables between trials. Only large (g=≥0.76) and moderate (g = 0.38–0.75) ES and associated ±95% confidence intervals (CI) are reported [31].

Results

Average environmental conditions in the climate chamber were 33.6 ± 0.9°C and 34.9 ± 5.54% RH across the three trials. There were no significant differences between trials for ambient temperature (p = 0.06) or RH (p = 0.49). In HYD and COLD, average total water intake was 675 ± 110 mL.

Physical performance

Total treadmill distance completed was not significantly different between the CON (10059 ± 2663 m), HYD (10479 ± 2297 m) and COLD (10522 ± 2473 m; p = 0.88; Figure 1).

Figure 1.

Total treadmill distance completed in the control (CON), hydration (HYD) and cold water (COLD) trials (n = 12). Horizontal lines = average distance; circles represent individual participants.

Cognitive performance

No significant main effects of trial (p = 0.49), time (p = 0.20), or interaction (p = 1.00) were found for the serial sevens task (Table 1). No significant main effects were found for the grammatical reasoning task for trial (p = 0.76) or time (p = 0.15). Nor was there a significant interaction effect (p = 0.95). Although a moderate effect size was observed at 30 min (g = 0.38 [−0.43, 1.18]), the wide confidence interval crossing zero suggests the result is inconclusive. Mean scores were higher in COLD (36 ± 12) than CON (32 ± 7), but the uncertainty in the estimate limits interpretation. For the counting span task, no significant main effects of time (p = 0.35), trial (p = 0.48), or interaction (p = 0.86) were found. A moderate ES was observed for HYD (6.3 ± 0.8) compared to CON (5.9 ± 0.8; g = 0.45 [−0.36, 1.26]) at 120 min and for COLD (6.5 ± 0.7) versus CON (vs 6.0 ± 1.1; g = 0.55 [−0.27, 1.36]) and HYD (6.2 ± 0.7; g = 0.43 [−0.38, 1.24]) at 150 min, however again associated CI values indicated inconclusive results.

Table 1.

Serial sevens, grammatical reasoning, and counting span scores (mean ± SD) in the control (CON), hydration (HYD) and cold water (COLD) trials (n = 12).

	0 min	15 min	30 min	45 min	60 min	75 min	90 min	105 min	120 min	135 min	150 min
SERIAL SEVENS
CON	19 ± 6	22 ± 7	–	22 ± 7	–	22 ± 7	–	22 ± 8	–	24 ± 8	24 ± 7
HYD	18 ± 4	21 ± 4	–	22 ± 5	–	21 ± 6	–	21 ± 6	–	22 ± 4	23 ± 6
COLD	20 ± 6a	21 ± 6	–	21 ± 6	–	22 ± 5	–	21 ± 6	–	22 ± 5	21 ± 5
GRAMMATICAL REASONING
CON	32 ± 10	–	32 ± 7	–	–	–	37 ± 10	–	–	–	36 ± 10
HYD	30 ± 11	–	33 ± 9	–	–	–	35 ± 10	–	–	–	36 ± 12
COLD	33 ± 10	–	36 ± 12b	–	–	–	36 ± 10	–	–	–	36 ± 12
COUNTING SPAN
CON	5.9 ± 1.0	–	–	–	6.0 ± 0.8	–	–	–	5.9 ± 0.8c	–	6.0 ± 1.1
HYD	5.8 ± 1.1	–	–	–	6.1 ± 0.9	–	–	–	6.3 ± 0.8	–	6.2 ± 0.7
COLD	6.0 ± 0.8	–	–	–	6.0 ± 0.8	–	–	–	6.1 ± 0.7	–	6.5 ± 0.7de

^aindicates moderate effect size between COLD and HYD (g = 0.48 [−0.33, 1.29])

^bindicates moderate effect size between COLD and CON (g = 0.38[−0.43, 1.18])

^cindicates moderate effect size between CON and HYD (g = 0.45[−0.36, 1.26])

^dindicates moderate effect size between COLD and HYD (g = 0.43[−0.38, 1.24])

^eindicates moderate effect size between COLD and CON (g = 0.55[−0.27, 1.36])

Manual dexterity performance

No significant main effect of time (p = 0.86) or interaction effect (p = 0.71) were found for the Purdue pegboard task. However, a significant main effect of trial showed better performance in COLD compared to CON (p = 0.03, Table 2). Although moderate ES were observed between COLD and CON at 30 (g = 0.43 [−0.38, 1.24]), 60 (g = 0.50 [−0.31, 1.31]), 90 (g = 0.40 [−0.41, 1.21]) and 150 min (g = 0.43 [−0.38, 1.24]) time points, the associated 95% CI crossed zero indicating statistical uncertainty. Similarly, moderate ES were observed between COLD and HYD at 30 min (g = 0.51 [−0.31, 1.32]) and 60 min (g = 0.63 [−0.19, 1.45]), suggesting potential but inconclusive benefits of COLD.

Table 2.

Dominant hand Perdue pegboard scores (mean ± SD) after completion of the task for 30 s in the control (CON), hydration (HYD) and cold water (COLD) trials (n = 12).

MANUAL DEXTERITY
	0 min	30 min	60 min	90 min	120 min	150 min
CON	15 ± 2	15 ± 2	15 ± 3	15 ± 2	15 ± 3	15 ± 3
HYD	15 ± 2	15 ± 1	15 ± 2	15 ± 2	15 ± 2	16 ± 2
COLD*	15 ± 2	16 ± 2bc	16 ± 2bc	16 ± 2b	16 ± 2	16 ± 2b

*significant main effect for trial with scores for COLD being higher than CON.

^bindicates moderate effect size between COLD and CON at 30 min (g = 0.43[−0.38, 1.24]), 60 min (g = 0.50[−0.31, 1.31]), 90 min (g = 0.40[−0.41, 1.21]) and 150 min (g = 0.43[−0.38, 1.24]).

^cindicates moderate effect size between COLD and HYD at 30 min (g = 0.51[−0.31, 1.32]) and 60 min (g = 0.63[−0.19, 1.45]).

Workload demands

For the NASA-TLX, no significant differences were observed between trials for mental (p = 0.36), physical (p = 0.10) and temporal (p = 0.09) demands, nor for performance (p = 0.41), effort (p = 0.44) or frustration (p = 0.13; Figure 2). While moderate to large ES were found between CON versus HYD and/or COLD for numerous variables, associated CI suggested uncertainty in interpreting these results (see Figure 2). Similarly, a moderate ES found between COLD and HYD scores for perceived performance needs to be interpreted with caution based on wide ranging CI values (see Figure 2).

Figure 2.

Sub-dimensions of task load (mean ± SEM) from the NASA task load index (NASA-TLX), administered post-trial in the control (CON), hydration (HYD) and cold water (COLD) trials (n = 12).

Thermal sensation and thermal comfort

No significant differences were found between trials for thermal sensation (p = 0.29), nor was there a significant interaction effect (p = 1.00, Table 3). Thermal sensation scores significantly increased over time, with participants feeling hotter as the trial progressed (p < 0.01). While there was no significant interaction effect for thermal comfort (p = 0.83), a main effect of trial (p < 0.01) showed that participants felt more uncomfortable in CON compared to COLD (p <0.01). Thermal comfort scores significantly increased over time (p < 0.01).

Table 3.

Thermal sensation and thermal comfort scores (mean ± SD) at 15-minute intervals in the control (CON), hydration (HYD) and cold water (COLD) trials (n = 12).

THERMAL SENSATION*
	0 min	15 min	45 min	75 min	105 min	135 min	150 min
CON	12.0 ± 2.6	14.8 ± 1.8	16.0 ± 1.4	16.3 ± 1.8	16.2 ± 2.2	16.0 ± 2.1	16.7 ± 2.1
HYD	12.4 ± 2.7	14.3 ± 1.8	15.7 ± 1.6	15.6 ± 1.8	15.8 ± 1.7	16.0 ± 1.8	15.8 ± 1.9
COLD	12.0 ± 3.2	14.4 ± 2.4	15.3 ± 1.8	15.7 ± 1.6	16.0 ± 1.5	16.0 ± 1.7	15.3 ± 2.0
THERMAL COMFORT*
CON#	11.1 ± 1.5	13.2 ± 2.3	14.6 ± 1.7	14.8 ± 1.8	15.4 ± 1.6	15.5 ± 1.9	15.8 ± 2.3
HYD	11.1 ± 2.0	12.8 ± 1.2	14.2 ± 1.4	14.4 ± 1.2	14.5 ± 1.4	14.5 ± 1.5	14.7 ± 1.5
COLD	9.2 ± 3.7	12.7 ± 1.5	13.7 ± 1.2	14.4 ± 1.4	14.6 ± 1.2	14.7 ± 1.6	14.2 ± 1.9

*indicates significant main effect for time (p < 0.05).

^#indicates significant main effect for condition between CON and COLD (p < 0.05).

Core temperature

Core temperature was significantly higher in the CON trial compared to COLD (p < 0.01) and HYD (p < 0.01; Figure 3), however there was no significant difference between COLD and HYD (p = 0.91). Additionally, a significant effect of time was found, with T_c increasing over time in all trials (p < 0.01). No significant interaction effect was found for T_c (p = 1.00).

Figure 3.

Mean ± SEM core temperature (°C) at 15-minute intervals in the control (CON) (n = 9), hydration HYD (n = 12) and cold water (COLD) trials (n = 11).

Skin temperature

There was a significant effect for trial (p < 0.01) with mean T_skin being significantly lower in COLD compared to both CON (p < 0.01) and HYD (p < 0.01), with no significant differences between CON and HYD (p = 0.98) (Figure 4). There was no significant main effect of time (p = 0.09) or interaction effect (p = 1.00) for T_skin.

Figure 4.

Mean ± SEM skin temperature (°C) at 15-minute intervals in the control (CON), hydration (HYD) and cold water (COLD) trials (n = 12). Skin temperature was estimated using the following equation: T_skin = (0.3× T_chest)+ (0.3× T_arm)+(0.2× T_thigh)+ (0.2× T _leg) [28].

Heart-rate

Heart-rate was not significantly different between trials (p = 0.58) but increased significantly over time during all trials (p < 0.01; Figure 5). There was no significant interaction effect found for this variable (p = 1.00).

Figure 5.

Heart-rate (bpm; mean ± SEM) at 15-minute intervals in the control (CON), hydration (HYD) and cold water (COLD) trials (n = 12).

Sweat loss and hydration status

No significant differences in sweat losses were found between trials for CON (1.20 ± 0.45 L), HYD (1.24 ± 0.52 L) and COLD (1.22 ± 0.49 L) post-trial (p = 0.98). Changes in body-mass (adjustments made for water intake and urine output) post- versus pre-trial were significantly less in the two water ingestion trials (HYD: −0.52 ± 0.46 kg; COLD: −0.55 ± 0.43 kg) compared to CON (no water intake: −1.2 ± 0.45 kg; p < 0.01). These changes represented a body-mass loss of 0.70% (HYD), 0.72% (COLD) and 1.63% (CON) when compared to pre-trial body-mass values. Pre-trial average USG values were similar (1.016 ± 0.006 in CON, 1.013 ± 0.007 in HYD and 1.013 ± 0.009 in COLD (p = 0.72)). In CON, one participant was well hydrated, nine were minimally dehydrated, and two were significantly dehydrated. In HYD, five were well hydrated, five minimally dehydrated, and two significantly dehydrated. In COLD, four participants were well hydrated, five were minimally dehydrated, and three were significantly dehydrated.

Discussion

To our knowledge, this is the first study to assess the effects of thermoneutral (37°C; hydration) and cold water (5°C hydration and cooling) ingestion on physical and cognitive performance and psycho-physiological parameters in participants during a simulated burn surgery. Cold water ingestion resulted in significantly better manual dexterity performance and thermal comfort as well as significantly lower T_c and T_skin, compared to CON. Skin temperature was also significantly lower in COLD versus HYD. Further, moderate and large ES demonstrated a greater magnitude of effect favoring COLD and/or HYD versus CON, as well as for COLD versus HYD at several time points for manual dexterity, various cognitive tests and perceived workload demands, however associated CI values suggested uncertainty in the interpretation of these outcomes. While it was hypothesized that water ingestion here would not result in bathroom breaks, there was one incident over 36 trials where a 60-y male participant in the HYD trial temporarily left the laboratory to urinate (320 ml). This need may have been due to his age, as older individuals are reported to urinate more frequently [32].

Treadmill distance covered

Similar distances covered on the treadmill between trials was unexpected. Earlier studies have reported impaired exercise performance in temperate to hot ambient conditions (21–32°C 43–80% RH), when participants did not drink water compared to water (14–16°C) ingestion trials, with this detriment associated with a greater degree of dehydration [33,34] and/or a higher T_c ( > 38.5°C, hyperthermia) [34,35]. It was also expected that exercise performance would be superior in the COLD trial due to the creation of an internal heat sink that absorbed metabolically produced heat, which in turn would reduce T_c, T_skin, and thermal discomfort [14]. Previous studies have reported improved prolonged cycling and running performance in the heat (35°C, 60% RH [15], 34°C 42% RH [36];), after cold water (4–10°C) ingestion compared to warm water ingestion (37°C).

Comparable physical performance across all trials may have been due to hypohydration not reaching levels associated with impaired exercise performance [3,37] and/or T_c not reaching hyperthermic levels [38]. Specifically, a peak T_c of 38.0°C was recorded for the CON trial, with this occurring at the end of the exercise protocol (150 min). A longer trial may have resulted in hyperthermia in the CON trial as T_c values in this trial significantly increased over time. In respect to hypohydration, losses in body-mass of 1.7–5.6% due to dehydration have been reported to impair aerobic exercise performance in ambient temperatures ranging from 19–40°C [3,37]. In our study, body-mass losses did not reach these levels, with body-mass changes pre- to post-trial being significantly less in both water ingestion trials (HYD: 0.70% loss COLD: 0.72% loss) compared to CON (1.63%).

The similar treadmill walking results in the two water ingestion trials found here partially align with those of Uchiyama et al. [22], who performed a mining simulation in the heat (37°C, 40% RH), where participants wore PPE and walked on a treadmill at an RPE of 11 [19] for 225 min, interspersed with rest periods that varied in frequency and duration (Experimental versus Current Practices trials). During both trials, participants ingested warm water (37°C) whilst walking in the heat chamber and cold water (5°C) during rest breaks in the laboratory (22°C, 35% RH) ad libitum. Despite participants drinking significantly more cold water during the Experimental trial, there was no significant difference in treadmill distance between trials. These researchers attributed this finding to peak T_c not reaching hyperthermic levels in either trial, with T_c and sweat loss being similar between trials. It is possible that cold water ingestion may not have been sufficient in the current study to reduce T_c and possibly improve exercise performance.

Cognitive performance

No significant differences were found over time or between trials for any cognitive task. Importantly, T_c did not reach 38.5°C (hyperthermia) at any time point in any trial, a threshold reported to impair performance on some complex cognitive tests [5,39,40]. While the moderate ES found between COLD and/or HYD versus the CON trials for grammatical reasoning and counting span need to be interpreted with caution due to inconclusive CI values, it is possible that these results may relate to the 1.6% loss in body-mass over the course of the trial due to dehydration in the CON trial, compared to only a 0.72% and 0.70% body-mass loss in COLD and HYD, respectively. Specifically, Ganio et al. [41] reported adverse effects of a body-mass loss of 1.59% due to dehydration on cognitive performance (vigilance and working memory), despite T_c not reaching 38.5°C. This suggests that dehydration may have contributed to the cognitive scores recorded here.

While not supported by definitive CI values, a moderate ES was also found for counting span scores between the COLD and HYD trials at the end of the trial, with mean scores favoring the COLD trial. This result is difficult to explain as the only secondary variable that significantly differed between these two trials was T_skin, which was lower in COLD. While lower T_skin during exercise has been associated with lower cardiovascular strain and hence better physical performance, as well as better thermal comfort and a lower sense of effort compared to higher T_skin values [3], these variables were not different between the COLD and HYD trials. A longer trial may provide more insight into the mechanisms driving this trend.

Manual dexterity performance

In the current study, manual dexterity scores were significantly better in the COLD trial compared to the CON trial. To date, no previous studies have assessed manual dexterity performance in the heat associated with water ingestion compared to a no-water/cooling trial. However, Maroni et al. [42] assessed manual dexterity during cycling bouts in the heat (35.0°C, 52.5% RH) using a variety of external cooling methods (cooling glove, cooling jacket and combination of glove and jacket), compared to a no-cooling trial. These researchers reported no significant differences in manual dexterity between trials and suggested that these results may be due to similar T_c and sweat loss values recorded between all trials. Consequently, results found in the current study may relate to the significantly higher T_c and T_skin values recorded, along with a greater degree of dehydration occurring (as represented by body-mass loss) in the CON versus the COLD trial, which may have impacted physical/manual dexterity performance. Further, significantly improved thermal comfort levels in COLD, compared to CON, may also have played a role here. Of relevance, Gonzalez-Alonso et al. [38] reported that elevated T_skin and higher perceptions of effort prompted athletes to reduce pace or intensity so to limit thermal strain, noting the importance of these factors on physical performance. Consequently, participants in the COLD trial may have felt more comfortable in the heat performing this task. While moderate ES were observed between the COLD and HYD trials at various time points, the wide range in associated CI values suggests caution when interpreting these results. Furthermore, these results are difficult to explain due to similar physiological and psychological variables recorded between the two water trials, apart from T_skin (lower in COLD).

Workload demands

There were no significant differences found between trials for any sub-dimensions for perceived workload demands. Moderate to large ES, which need to be interpreted with caution due to inconclusive CI values, were found between COLD and/or HYD versus CON for physical, mental and temporal demands, effort and frustration where scores favored the water ingestion trials. It is possible that these results could relate to the significantly greater thermal comfort, as well as the significantly lower T_c and T_skin values recorded in one or both water ingestion trials compared to CON, as described earlier. Similarly, a moderate ES (with inconclusive CI values) was found between the COLD and HYD trial for perceived performance. As most secondary variables were similar between trials, further research is needed to explore these outcomes. Importantly, possible reductions in perceived workload with water ingestion could have implications for the overall wellbeing of burn surgery staff, as a high perceived workload has been linked to job burn out in healthcare workers [43].

Thermoregulatory responses

While T_c increased over time in all trials, T_c was significantly lower in COLD and HYD compared to CON. This finding is supported by Haseawa et al. [35], who reported that water (14–16°C) ingestion attenuated a rise in rectal temperature compared to a no-fluid control during prolonged cycling in hot ambient conditions (32°C, 80% RH). Interestingly, T_c values in the current study were not different between the COLD and HYD trials (p > 0.05), suggesting that hydration alone played a major role in attenuating T_c. This finding contradicts our hypothesis that cold water would reduce physiological thermal strain more effectively than thermoneutral water due to its cooling capabilities. These results differ from Lee et al. [15] who found that cold (4°C) water ingestion lowered T_c during cycling to exhaustion in heat (35°C, 60% RH), compared to warm (37°C) water ingestion. The differences between studies may be due to the frequency and the amount of fluid ingested by participants during the protocol. Specifically, Lee et al. [15] provided participants with 300 mL of flavored cold or warm water every 10 min during a 30-min rest period, followed by 100 mL every 10 min during 52–63 min of cycling. In comparison, participants in the current study ingested 0.9% of their body-mass of warm or cold water, that was divided into four equal portions and provided every 30 min, commencing at the 20 min timepoint of the protocol. The more frequent and larger intake of cold water in the study by Lee et al. [15] most likely contributed to mitigating the increase in T_c compared to the warm water trial. Therefore, the smaller, less frequent ingestion of cold water in the current study may explain why no significant difference in T_c was observed between the water ingestion trials.

Thermal comfort was significantly improved in the COLD compared to the CON trial. This result may relate to cold water ingestion stimulating thermoreceptors in the mouth and gut that in turn enhanced overall comfort [34,44]. Additionally, thermal comfort has been found to be strongly correlated with T_skin [15], with T_skin being significantly lower in COLD compared to CON in the current study. As noted earlier, higher T_skin and perceptions of thermal strain have been associated with impaired physical performance [38]. The improvement in thermal comfort in the COLD trial is noteworthy due to a study by Davey et al. [45] that reported thermal discomfort associated with wearing PPE in thermoneutral conditions. In contrast, no significant differences were found between trials for thermal sensation. Nonetheless, thermal sensation scores were consistently higher in the CON group compared to the water ingestion trials at each assessment point after the trial began.

Fluid replacement and sweat loss

Fluid ingestion in the current study was carefully considered so to prevent the need for bathroom breaks and, theoretically, subsequent scrubbing in during the simulated surgery. This issue is often noted anecdotally by burn staff as the main reason for not drinking during surgery and is supported by previous studies where no water was ingested by staff during either real [9] or simulated burn surgeries [13]. The total amount of fluid ingested here was based on mean sweat loss calculated from staff participating in a previous burn surgery simulation [13]. This equated to 0.9% of pre-trial body-mass, but ultimately proved inadequate for total sweat replacement, as reflected by body-mass losses of −0.52 ± 0.46 kg (0.70% loss) for HYD and −0.55 ± 0.43 kg (0.72% loss) for COLD. This discrepancy most likely relates to using a different cohort to determine fluid replacement. Importantly, previous studies have reported benefits of water/fluid ingestion during prolonged exercise in the heat when fluid intake matched 100% sweat loss [35,36,46]. Results here suggest that fluid ingestion during future burn surgery research should be > 0.9% so to better maintain fluid balance and hence hydration status.

Limitations

A key limitation to this study was that it was conducted as a simulation in a university laboratory setting that used a walking protocol to assess physical performance. This approach was necessary due to restrictions to access to hospital facilities imposed by the COVID-19 pandemic, while reasons for choosing a walking protocol were provided earlier. While the laboratory simulation allowed for numerous assessments of cognitive, manual, and psychological variables – something not feasible in a real burn surgery – it is acknowledged that the results may not fully reflect the complexities of real surgical conditions that involve varying distractions (i.e. noises from machines and discussions between staff), as well as operating on critically ill patients. Furthermore, while participants were not informed of any possible benefits associated with the interventions, it is possible that some participants may have altered their walking speed due to preconceived notions about the effects of a particular intervention. This would not have been detected by researchers given the subjective nature of the walking intensity requirement (RPE of 12).

Conclusion

Cold water ingestion resulted in significant improvement in manual dexterity and thermal comfort, as well as significantly attenuated a rise in T_c and T_skin over the course of a simulated burn surgery in the heat, compared to no-water ingestion. Warm water ingestion also significantly tempered a rise in T_c compared to no-water intake. While moderate to large ES were calculated in favor of COLD versus CON and HYD trials for numerous variables, these results need to be interpretated with caution due to inconclusive CI values. Further, the strategy of consuming small volumes of water throughout the 2.5 h simulation successfully addressed concerns regarding bathroom breaks, commonly reported by burn surgery staff. Results here may encourage staff to participate in controlled hydration strategies during future hot surgeries. Notably, larger and more frequent boluses of cold water may further enhance physical and cognitive performance without increasing the need for bathroom breaks.

Funding Statement

This study was funded by the University of Western Australia.

Abbreviations

CI

Confidence intervals

CONT trial

Control trial

ES

Effect sizes

h

Hour

HYD trial

Hydration trial

Kg

Kilogram

min

Minute

NASA Task Load

NASA-TLX

NBM

Nude body mass

OT

Operating theaters

PPE

Personal protective equipment

RPE

Rating of perceived exertion

RH

Relative humidity

T_arm

Arm temperature

T_chest

Chest temperature

T_leg

Leg temperature

T_skin

Skin temperature

T_thigh

Thigh temperature

T_c

Core temperature

UWA

University of Western Australia

USG

Urine specific gravity

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available from the corresponding author [KW] upon reasonable request.