Factors influencing acute tympanic temperature changes during cold water exposure Authors: Anna Karafová, Marian Karafa, Bartłomiej Kopciński, Jakub Kortas, Federica Verme, Paolo Capodaglio Published: 2026 ABSTRACT AbstractExposure to ice water, despite its growing popularity and documented health benefits, is limited by the risk of distressing physiological reactions. The aim of this study was to evaluate the relationships between individual, environmental factors and acute tympanic temperature changes during cold exposure. A total of 90 individuals either practicing cold water immersion (CWI) or winter swimming (WS) were enrolled in the study. They were divided into 4 subgroups according to the distance swum (200 m or 450 m) and immersion time (5 or 9 min). The tympanic temperature of all participants was measured, and BMI was calculated. This study found significant pre-post exposure changes in temperature [°C] (p>0.05) in all the subgroups, except for the group immersed for 5 min. In winter swimmers, a significant negative correlation between BMI and tympanic membrane temperature was found p = 0.006) for both females and males. The greatest change in tympanic temperature occurred in participants with 3 years of experience in practicing those activities. A five-minute immersion in water below 1°C did not allow for the observation of a significant body response to ice water. The results indicate that a higher BMI was associated with a smaller decrease in tympanic membrane temperature in the winter swimming subgroup.Keywords: Cold water immersion, ice water swimming, body heat loss, tympanic membrane temperature, body mass index Introduction Cold-water immersion (CWI) has become a popular form of recreation, while ice water swimming (WS) was formalised as a competitive discipline in 2009 by the International Ice Swimming Association (IISA) and until recently was only reserved for a few extreme athletes from northern countries [1–3]. CWI can include various forms, such as cold showers and ice bathing and can be carried out with complete or partial immersion of the body in cold water, generally understood as water <15°C [4,5]. In order to strictly define swimming rules, the IISA in their guidelines differentiate between swimming in ice, freezing and cold water, providing water temperature ranges: −2°C to +2°C, +2.1°C to 5°C and +5.1°C to +9°C, respectively [1]. Various health benefits of CWI and WS have been reported: time-dependent inflammation, sleep quality, stress and general well-being [4]. In a randomised trial, Buijze et al. [6] found that practicing CWI reduced by 29% the absence from work because of sickness on 3018 participants. Despite the potential benefits associated with CWI and WS, it should be emphasised that such activities also carry certain risks [3]. Exposure to ice water most often leads to cardiorespiratory disorders due to excessive stimulation of the sympathetic nervous system, which can result in hyperventilation, arrhythmia and even death in healthy and young people [7]. Proper priming and acclimatisation, including routine immersion under the supervision of an experienced instructor, allows extension of the trunk to reduce risks resulting from sudden and unreasonable contact with cold water [3]. However, it should be reminded that, despite experience, prolonged immersion in ice water leads to a drop in core body temperature below 35°C, which can result in hypothermia [3,5,8]. It is reported that to prevent life-threatening complications, contact with ice water during CWI should not exceed 30 min [9]. The scale of the reaction caused by immersion in ice water varies and depends on many factors, among which the water temperature is important [5]. In current reports it has been observed that higher body mass index (BMI) in ice swimmers was associated with a higher chance to successfully complete the competition and with a decreased risk of core temperature loss [2,10]. According to data, there is a correlation not only between adiposity and core cooling but also body mass and size should be taken into consideration, while in larger individuals, the cooling process is slower [11]. Body mass index is a commonly used indicator for assessing health and body habitus in general, including in sports medicine [11]. This indicator has certain limitations, as it can be overestimated in situations with both increased muscle mass and increased fat mass. Despite this, it has been shown to be protective against core cooling in swimmers, regardless of their body composition [11]. The aim of our study was to assess whether such a relationship would differ between static immersion and swimming as well as the influence of other individual and environmental factors, such as the number of years of immersion or swimming practice and the duration of exposure to core cooling, expressed as a decrease in tympanic membrane temperature. In line with our expectations, a higher BMI, greater experience and shorter exposure time would be associated with a smaller decrease in tympanic membrane temperature in both groups. Methods Participants All the participants were informed about the anonymity of the study and they have given written informed consent to publish all the needed details. A total of 90 healthy individuals, including 48 winter swimmers (WS) registered for the Gdynia Winter Swimming Cup (GWSC) 2021 competition (17 female and 31 male) and 42 practicing CWI (22 female and 20 male) from all over Poland, were enrolled in the study. All of them provided informed consent to participate in the study. The age of the participants was 21–69 years (39.01±11.99), the mean body weight (46.0–115.1) was 78.51±15.68 kg and an average height of 1.76±0.1 m (1.60–1.98 m). A few days before the study began, all the participants were examined by a physician. All the participants received regular medical screenings and check up every three months. Experimental procedures The data were collected during the Gdynia Winter Swimming Cup (GWSC) 2021 (Feb 19–21, 2021) and during morning winter swimming (22.02.2021–01.03.2021), which took place in the Polish Baltic Sea. All participants were asked to provide information such as gender, age and years of experience in ice–water immersion (CWI) or winter swimming (WS). Before the study began, measurements such as body height and weight were taken, which allowed the calculation of the BMI index based on the World Health Organization (WHO) classification for people over 20 years of age (Table 1). The tympanic temperature was measured using a tympanic thermometer Braun PRO6000 ThermoScan® as a proxy of core body temperature. The temperature was measured before starting CWI or WS and immediately after their completion (in 1 min after exposure). The probe was placed into the ear canal tightly, directed in the collateral temple and kept stationary. According to the manufacturer, the final temperature calculation includes data on the direction and depth of the probe, while a new special sensor system allows the reduction of measurement errors resulting from different ear canal structures and different measurement techniques. For each participant, the measurement was performed in the same ear, and an additional green light indicated that the probe was positioned correctly. This thermometer has a 60 s times which remains solid at 15 s and it is repeated for each 15 s interval. A new disposable probe cover was used for each participant. The clinical repeatability given by the manufacturer was 0.19°C. Based on the International Winter Swimming Association (IWSA) requirements, the water temperature was measured using the DT-1 electronic thermometer reference HAACP. During the days of data collection, the water temperature for the winter swimmers and CWI practitioners was 0.5°C, while the ambient temperature was −2°C. The distance for WS was either 200 or 450 m, while the immersion time for CWI was either 5 or 9 min. The immersion time for CWI was calculated based on the arithmetic mean swimming times of WS performing the 200 and 450 m competitions. The study on the 4 groups of participants (WS 200, WS 400, CWI 5 and CWI 9) was observational in nature, and the participants were not randomised. During the competition, the participants were supervised by a doctor and paramedics. WS were required to wear a standard competition outfit, which consisted of a swimsuit, two silicone caps, swimming goggles and earplugs. However, it was forbidden to the use of substances that heated the body or materials that had an insulating function (neoprene shoes and gloves). During ice water immersion, CWI participants were in a static position in the water with their hands out of water behind their heads, with the water level reaching the nipple line. The outfit consisted of a swimsuit and a silicone cup. Table 1.Anthropometric measurements of all participants included into the study. Immersed 5 min(CWI 5) (n=27)Immersed 9 min(CWI 9) (n=15)Swimmers 200 m(WS 200) (n=31)Swimmers 450 m(WS 450) (n=17)pBody mass (kg)73.9±13.679.2±18.380.9±14.880.8±17.20.32Body height (m)1.74±0.11.76±0.11.77±0.11.77±0.010.56BMI (kg/m2)24.3±3.225.5±4.525.7±4.025.6±4.30.57Age (years)36.4±13.636.5±9.740.0±11.943.6±10.40.2Note: values are presented as mean±standard deviation; BMI—body mass index. *BMI calculated using the formula: BMI = kg(body weight)/m2 (height)2. Anthropometric measurements of all participants included into the study. Note: values are presented as mean±standard deviation; BMI—body mass index. *BMI calculated using the formula: BMI = kg(body weight)/m2 (height)2. Statistical analysis Statistical analysis was performed using Statistica 13.1 software. Graphs were created with GraphPrism 7 software. Before starting the final statistical analyses, a preliminary assessment of the data distribution was performed. For visualising this distribution, box plots were used, while for identifying outliers, the Grubbs test was performed. The normality of the distribution of individual variables was assessed using the Shapiro–Wilk test. Depending on the fulfilment of the assumptions of normality and sphericity (verified by the Mauchly test), appropriate parametric or nonparametric tests were selected. Repeated measures ANOVA was used to assess the significance of changes in tympanic temperature before and after cold exposure in the four groups (CWI 5 and CWI 9; WS 200 and WS 450), with repeated measurements. Additionally, sex was included as a between-subjects factor to evaluate potential differences between males and females. Comparisons between independent groups (e.g. different exposure lengths, winter swimming/swimming experience) were made using one-way ANOVA, with Tukey's post hoc test used to correct for multiple comparisons. The relationship between body mass index (BMI) and the extent of hypothermia (delta temp.) was analysed using Pearson's correlation coefficient (for normally distributed data) or Spearman's correlation coefficient (for non-normally distributed data). All analyses were performed at a significance level of p≤0.05. Results Anthropometric data and BMI data for the 4 study groups (CWI 5, CWI 9 and WS 200, WS 450) are summarised in Table 1. The CWI 5 group consisted of 13 men and 14 women and the CWI 9 group of 7 men and 8 women. The WS 200 group consisted of 21 men and 10 women, and the WS 450 group of 10 men and 7 women. The majority of male swimmers at both distances (n=13 in WS 200, n=8 in WS 450) were classified as individuals with overweight or with obesity according to BMI (BMI 25–29.99 kg/m2; 30–34.99 kg/m2, while women in most cases (n=7 in WS 200; n=5 in WS 450) were within the normal weight range (BMI 18.5–24.99 kg/m2). The majority of male in the CWI 5 group (n=7) and in the CWI 9 group (n=5) were also individuals with overweight or obesity, while female participants mostly (n=13) were in the normal weight range. The analysis began with an examination of the significance of changes in tympanic temperature before and after exposure to ice water in each subgroup (Table 2). No significant changes were found in CWI 5 (p>0.05), whereas a statistically significant difference in tympanic temperature was found in the other three groups studied (Figure 1). Table 2.Pre-post exposure changes in tympanic membrane temperature in all subgroups [°C].SubgroupPre-exposure temperaturePost-exposure temperatureCWI 536.44±0.4336.27±0.63CWI 936.41±0.5435.91±0.88WS 20036.91±0.4234.79±0.60WS 45036.82±0.5134.57±0.50Note: values are presented as mean±standard deviation. Pre-post exposure changes in tympanic membrane temperature in all subgroups [°C]. Note: values are presented as mean±standard deviation. Changes in tympanic temperature in the four study groups (CWI 5, CWI 9, WS 200, WS 450). Correlation analysis for all individuals exposed to ice water did not show a significant relationship between BMI and a decrease in tympanic temperature (ρ=−0.049, p=0.646). However, for all WS (Figure 2), a significant negative correlation was found (r=−0.391, p=0.006), which was confirmed for both women (r=−0.498, p=0.042) and men (r=−0.459, p=0.009). In CWI, this relationship was not statistically significant. Relationship between body temperature and BMI in a group of all swimmers. The data, including the experience of all subgroups with immersion or swimming in ice water expressed as the number of seasons in years, are summarised in Table 3. Table 3.Years of experience of all subgroups with immersion or swimming in ice water.SubgroupSwimming or immersion season 1234Over 5TotalCWI 518220527CWI 95232315WS 2006774731WS 4501445317Total301516111890 Years of experience of all subgroups with immersion or swimming in ice water. Analyses showed that participants with an experience of 3 years of practicing ice water had a significantly greater decrease in tympanic temperature than those with 1 year of experience (p=0.013, Figure 3). No other significant differences were found between the groups with longer practicing experience. The results obtained did not confirm any statistically significant gender differences in relation to all parameters of our study. Change in tympanic temperature vs years of experience in exposure to ice water. Discussion Exposure to ice water during winter periods, including ice swimming and CWI, despite its growing popularity and health benefits, is limited by the risk of hypothermia [1,5,7]. The aim of this study was to evaluate the relationships between individual (BMI, years of practicing with immersion or swimming in ice water), environmental (duration of exposure to ice water) and tympanic body temperature as a proxy of core temperature. Owing to the restrictions of the pandemic period, we measured tympanic membrane temperature, instead of using rectal probes or ingestible sensors, as a proxy for core temperature. The tympanic membrane temperature is considered the gold standard in clinical conditions; however, when it is not possible to perform such invasive methods, measuring the temperature of the tympanic membrane is reliable and accepted as a suitable alternative [12]. Brinnel and Cabanac in their study showed that during 30 min of exposure to cold water at 0°C, the changes in tympanic temperature were comparable to those in oesophageal temperature, while the influence of the air temperature was equal only 0.0025°C per °C drop in air temperature [13]. The authors suggested that tympanic temperature measured accurately can be considered as an index of core temperature, while its changes may manifest fluctuations in brain temperature [13]. Since 90% of heat is lost through the skin, physical activity consequently contributes to a drop in body temperature, while heat loss during arm movements is more pronounced than that during lower limb exercises [9,14]. Such a phenomenon could explain our results, which showed a significantly greater decrease in body temperature in the WS compared to both CWI groups (p<0.001). According to Wallingford et al. [15], triceps skinfold thickness is a statistically stronger predictor of distance covered compared to body adiposity or the decrease in core body temperature. In both CWI groups, a slight increase in tympanic body temperature was observed immediately after leaving the ice water, while in both WS groups, a decrease in body temperature was noted. Janský et al [16] in their study have obtained similar results, showing that body temperature tends to increase during the first 20 min of head-out ice water immersion and then to decrease about 0.75°C. In our study, the duration of static immersion was significantly shorter, being 5 and 9 min in the first and second CWI groups, respectively. Despite this, in some cases (n=23 for 42), an immediate drop in body temperature was noted, suggesting that there are factors that determine different individual response to cold. It has been widely reported in the literature that repeated exposure to ice water, age, gender, BMI or individual experience may have adaptational characteristics and influence an individual's rate of cooling [16,17]. One of the forms of human adaptation to cold is acclimatisation, which seems to decrease the risk of hypothermia [9,18]. Several adaptations, including their metabolic, insulative and hypothermic types, can be exhibited by winter swimmers [19]. It was well documented that in cold acclimated sportsmen, muscle shivering was reduced and cold thermogenesis has occurred in lower body temperature, saving of about 20% of total body heat production during a single immersion in cold water [16]. In our study, it was shown that individuals with 3 years of experience (n=16) (Table 3) in ice water practicing, both in CWI and WS, had a significantly greater decrease in tympanic temperature than those with 1 year of experience (p=0.013). This finding is difficult to interpret because of the limited number of participants in the subgroups and should be taken into consideration in further studies. The authors have noticed a decreased cold sensation and changes in cold thermogenesis regulation after four sessions of CWI and observed an increased fat body content in acclimated individuals [16]. This could explain the fact that in our study, in the subgroups with four years (n=11) and more than five years of experience (n=18) in ice water, no statistically significant differences in the decrease in tympanic temperature were found. Many authors have reported that increased BMI was a significant protective factor against hypothermia during open winter swimming [2,20]. It is known that BMI is commonly used in research, but it is not a direct measure of body fat content, however it is widely used worldwide as an indicator of obesity and adiposity [11,21]. To obtain more accurate results, skinfold thickness would need to be measured, which was not included in our study [20]. It is commonly known that the measurement of body fat can pose difficulties and is more complex than determining body mass index [11]. Although in some cases, lean individuals with high muscle mass may have an elevated BMI. Crow et al showed in their study a positive correlation between % body fat and BMI in winter swimmers [11]. Subcutaneous fat plays an important role in insulating against cold, while its thermal resistivity is greater compared to skin and muscles [1,14]. However, owing to their different distributions within the trunk (approximately 50% of insulation), limbs (~30%) and hands and feet (~3%), the ability to stabilise temperature in these parts of the body is different [1]. The results of our study did not show any difference between women and men in the rate of heat loss, which was dependent on BMI, in any of the studied groups. Therefore, further analyses were conducted based on a comparison of the CWI and WS groups as a whole, without differentiating them by gender. Solianik et al. [22] examined the effect of gender on the response to cold in a group of 18 men and 14 women, showed that the insulating response in women was statistically greater, while metabolic heat production and shivering were more pronounced in men. However, despite such differences, the cooling rate was similar regardless of gender. In our study, correlation analysis for all individuals (n=90) exposed to cold did not show a significant relationship between BMI and a decrease in tympanic body temperature (ρ=−0.049, p=0.646). However, in the WS group, independent from the distance swimmed, a significant negative correlation was found (r=−0.391, p=0.006, Figure 2), which was confirmed in both women (r=−0.498, p=0.042) and men (r=−0.459, p=0.009). In the CWI group, this relationship was not statistically significant. In our study, no significant changes in tympanic body temperature were found in the CWI 5 group in ice water (p>0.05) (Figure 1). A five-minute immersion in water below 1°C did not allow for the observation of a significant body response to ice water, probably due to the influence of measurement sensitivity and variability. The results showed that only in the CWI 9 group a significant change in tympanic membrane temperature was found (p=0.034). The duration of immersion was calculated based on the time needed to cover distances of 200 m and 450 m in WS. In this group, significant changes in temperature were found in both subgroups (200 m and 450 m; p<0.001). The time to swim the 200-m distance was about 5 min, similar to the immersion duration of the CWI 5 practitioners. Tympanic membrane temperature changes were noticeable in WS, while they were not statistically significant in the CWI groups, indicating that the body cools down more during physical activity. According to the IWSA, the recommended swimming distance in such icy waters as in our study should be limited to 200 m; however, in our study, any interaction between immersion time in the two CWI groups and distance length was not statistically significant (p=0.152; p=0.497). Conclusions The results of our study showed that during immersion in water below 1°C, no relationships between individual (BMI), environmental (exposure duration to ice water—5 and 9 min) and tympanic temperature as a proxy of core temperature were observed. In WS, the changes in tympanic temperature were statistically significant because of physical activity and faster heat loss. The results indicate that higher BMI was associated with a smaller decrease in tympanic membrane temperature in this group, regardless of gender. The lack of differences between women and men suggests that the body's response to cold has many dimensions and does not depend only on the amount of fat tissue and its insulating function. The lack of significant differences in tympanic temperature among individuals with over 3 years of experience with immersion or swimming in ice water is difficult to interpret due to small size of the sample and further studies are needed. It should be emphasised that the study is not free from limitations. The small size of the subgroups, as well as the lack of collected data on the participants' body composition or on the menstrual cycle of the females, could have significantly impacted the interpretation of the results. Acknowledgements None. Disclosure statement No potential conflict of interest was reported by the author(s). Funding This research submitted to the journal received no specific grant from any funding agency in the public, commercial or not-for-profit sectors. Confirmation of ethical compliance The study was approved by the Bioethics Committee of the Medical University of Gdansk, Poland (approval no. KB–1/18). All methods used in this study were performed in accordance with the Declaration of Helsinki. Practical applications The results of this study indicate that the group of winter swimmers may have longer contact with icy water because their body cools down more slowly compared to people practicing cold water immersion. References