Acute Finnish sauna heating and cold water immersion effects on cardiovascular dynamic response in normotensive women

Abstract

The study aims to evaluate the acute effects of sauna heating and cold immersion on cardiovascular dynamic response in normotensive women. Twenty-eight healthy females underwent a sauna protocol comprising three consecutive 10-min exposures, each separated by a 10-min cooling interval. Blood pressure and heart rate (HR) were measured immediately after leaving the sauna room and in the last minute of the cooling period. Three acute responses after heating in the sauna and three cooling responses had a statistically significant effect on SBP (systolic blood pressure) (p < 0.001), DBP (diastolic blood pressure) (p < 0.001), and HR (p < 0.001). SBP was significantly higher in the first heating session compared to the baseline measurement, while DBP was significantly lower. HR was significantly higher in all three heating sessions compared to the baseline measurement (p < 0.001). Progression analysis revealed a decreasing trend in SBP across heating sessions, whereas no significant trend was observed during cooling sessions. DBP and HR remained stable across heating and cooling cycles. SBP shows high sensitivity to repeated sauna stress, suggesting adaptive cardiovascular effects. However, it is still a preliminary study in young, healthy women, and in the future, more longitudinal studies are needed to identify cardiovascular responses in different age and sex groups, as well as the impact of sauna cycles on individuals with coexisting cardiovascular diseases.

Introduction

Bathing in a Finnish sauna is a form of passive heat therapy and is commonly used for relaxation, wellness, and pleasure¹. It has also been shown that regular sauna bathing can be beneficial to health², reducing the risk of sudden cardiac death, fatal coronary heart disease, fatal cardiovascular disease, all-cause mortality³, dementia, and Alzheimer’s disease⁴. In sports, sauna exposures are used as one of the passive heat acclimation strategies to prepare the athlete for competing in a high-temperature environment⁵ and may lead to enhanced athletic performance⁶. The typical Finnish sauna is characterised by low humidity levels (10–20%) and elevated temperatures, generally ranging from 80 to 100 °C⁷. The Finnish sauna session consists of one to three cycles alternating between heat exposure and subsequent cooling⁸. Each heat exposure within the sauna chamber lasts between 5 and 20 min and is followed by a cooling phase, which may include passive rest at ambient temperature, a cold shower, or immersion in cold water^7–10. Alternating heat and cold has a bidirectional effect on the human body, depending on the heat-to-cold ratio and the intensity of heat and cold exposure¹¹. There is a strong physiological relationship between the performance of the human cardiovascular system and the ambient temperature, as vasoconstriction and vasodilation are regulated by temperature¹². Heat stimulation leads to a diversion of blood, and therefore heat, to the skin by opening multiple arteriovenous anastomoses¹³. Blood flow through the skin can increase to 7–8 l/min in response to acute heat stress. In addition, it has been proven that during heat exposure, blood flow also increases in skeletal muscle and bone marrow^13,14. On the other hand, cold environments also force the body to make acute adaptive responses to maintain thermal balance, primarily by reducing heat loss. Reflexively and via local mechanisms, cutaneous blood vessels constrict, which reduces heat transfer from the core to the skin and other distal parts of the body¹⁵. In summary, cooling the body immediately after heat exposure in a sauna rapidly lowers the skin temperature and facilitates the normalisation of the core body temperature¹⁶, thereby preventing overheating and enabling the safe repetition of sauna cycles. Alternating exposure to heat and cold, which elicits contrasting cardiovascular responses, can thus be regarded as a form of conditioning for the cardiovascular system.

Thermal sensitivity differs between men and women^17,18. Women report greater discomfort or thermal sensations at the same absolute temperature compared to men and often detect these changes earlier^17,19,20. In addition, women have an increased perception of skin wetness, which modifies thermal sensations²¹. Differences in skin temperature between men and women are likely due to differences in body composition and blood flow regulation²². Women have a greater proportion of body fat²², and fat may have an insulating effect during passive exposure to high external temperatures, which may lead to higher skin temperatures. Estrogens, including 17β-estradiol, promote vasodilation and enhance blood flow by increasing nitric oxide production, which in turn dilates blood vessels and elevates skin temperature¹⁹. Women were selected for this study due to evidence suggesting greater cardiovascular and autonomic reactivity to thermal stress compared to men. This heightened sensitivity may allow for more pronounced and detectable changes in response to repeated sauna and cooling exposure, making female participants a particularly relevant group for investigating dynamic cardiovascular adaptations. Moreover, women still constitute a very small proportion of participants in thermoregulation research (e.g., 12–18% of the total over the last decade)²³.

Most previous studies regarding the effect of Finnish sauna on haemodynamic responses focused solely on the heating phase, neglecting the cooling phase^1,10,11. In a study involving 4 cycles of alternating heating in a Finnish sauna and cooling, a significant decrease in diastolic blood pressure (DBP) and no significant changes in systolic blood pressure (SBP) and heart rate (HR) were observed in healthy women⁹. A single cycle involved a 12-minute stay in the sauna at a temperature of 90–91 °C, followed by 6 min of cooling, which included a 2-min immersion in cold water at 9–11 °C. Studies involving a single 16-min sauna session and a 2-min cooling phase showed a significant decrease in HR, SBP, and DBP after the procedure when the cooling phase included a 2-min cold water immersion²⁴. In turn, when participants took a cold shower for 30 s and then rested for 90 s in a temperature-neutral room after sauna exposure, an increase in HR and no statistically significant changes in blood pressure were observed. These studies were conducted in a group of young men leading a sedentary lifestyle.

The above-mentioned studies explained the simple effects of heating and cooling before and after complete sauna bathing. However, to date, and to the best knowledge of the authors, there is a lack of results for the course of dynamic cardiovascular responses in blood pressure and heart rate to the several cycles of the alternate heating and cooling down sauna bathing phases. Moreover, to minimise physiological variability associated with sex-based differences in thermal and cardiovascular responses, the present study was conducted exclusively in women. Therefore, the purpose of this study was to assess acute sauna heating and cold water immersion effects on cardiovascular dynamic response in normotensive women. Specifically, we aimed to (1) assess the acute and cumulative effects of repeated sauna exposure (H1–H3) and subsequent cold water immersion (C1–C3) on cardiovascular parameters regarding mean values and (2) model individual cardiovascular trajectories (SBP, DBP, and HR) across repeated exposures using linear regression while accounting for baseline physiological state. This approach enables the assessment of temporal response patterns and the influence of baseline values on the magnitude and direction of changes over time.

Materials and methods

Sample size

The G*Power software (version 3.1) was employed to determine the required sample size a priori. For the one-way repeated measures analysis of variance (RM ANOVA) with one group and seven repeated measures (baseline, three overheating and three cooling sessions), an effect size of 0.2, a significance level of 0.05, and a power of 0.80, a total sample size of 26 was recommended as optimal for results credibility. In addition, 10% (two persons) of dropout was assumed (in case of necessity of excluding individuals during the experiment). Finally, our sample size was 28 individuals.

Participants

The study group consisted of twenty-eight female participants (age 22.28 ± 0.63 years; body height 168.79 ± 6.07 cm; body mass 60.11 ± 6.91 kg; BMI 21.07 ± 1.87 kg/m²) with no experience in sauna bathing. Exclusion criteria included hypertension (> 140 mmHg), tachycardia (> 100 bpm), and contraindications to sauna bathing. All participants were informed about the purpose of the study and measurement procedures and voluntarily signed a consent form. All participants were instructed to refrain from smoking, caffeine consumption, alcohol, and intense physical activity on the day of the study⁹. The research protocol and study design were approved by the Research Ethics Committee of Wroclaw University of Health and Sport Sciences (28/02/2011) in accordance with the Declaration of Helsinki.

Procedure

Before the sauna bath, anthropometric measurements, blood pressure, and heart rate were taken. Participants then underwent a single sauna session, during which they entered the sauna room three times for 10 min. Each exposure to high sauna temperature was interspersed with a 10-min cooling period. Blood pressure and heart rate were measured immediately after leaving the sauna room and in the last minute of the cooling period.

Anthropometry

Anthropometric measurements involved using anthropometers (GPM Anthropological Instruments) to record two body height measurements with an accuracy of 0.1 cm. Body weight was determined utilising an InBody230 body composition analyser bioelectric impedance method with an accuracy of 0.1 kg (InBody Co. Ltd., Cerritos, CA, USA).

The above data were used to calculate body mass index (BMI):

Blood pressure and heart rate

Blood pressure (systolic blood pressure—SBP, diastolic blood pressure—DBP) and heart rate (HR) were measured with an automatic digital blood pressure monitor equipped with an upper arm cuff (M7 Intelli IT, Omron, Tokyo, Japan). All measurements were performed in the sitting position on the left arm. Before the baseline measurement, the subject sat for 10 min. Then, two consecutive blood pressure and heart rate measurements were taken. The average value of the two measurements was used for further analysis.

Sauna bathing intervention

The exposure to high temperature in the sauna room lasted 10 min, followed by a 10-min cooling-off period. This cycle was repeated three times. The temperature in the sauna room at the height of the middle bench was on average 90 ± 3 °C, and the humidity was in the range of 6–10%. The temperature was measured to an accuracy of 1 °C and the humidity to 1% using a thermometer and hygrometer designed by Harvia Oy (Finland). During the first 2 min in the sauna room, the participants sat on the lowest bench for adaptation, and the next 8 min were spent sitting on the second (middle) bench. After leaving the sauna, the participants sat on a bench in the cooling-off room (average temperature 20 °C), where blood pressure and heart rate were measured. Participants spent 6 min in this room, then took a cold shower (temperature 15 °C) to rinse off sweat, and then immersed their bodies (up to the neck) in a pool of cold water (temperature 15 °C) for 2 min. Immediately after cooling off in the pool, blood pressure and heart rate were measured, after which another exposure to the sauna was performed. Participants were supervised by a physiotherapist and were allowed to leave the sauna room at any time they felt discomfort.

Statistics

The data distribution was tested with the Shapiro–Wilk test of normality of the distribution. All quantitative variables were presented as mean, standard deviation, and 95% confidence interval (95% CI). To test pre-post changes in subsequent measurements (heating phases and cooling phases), a one-way repeated measures analysis of variance (RM ANOVA) was conducted. Before running ANOVA, homoscedasticity and sphericity assumptions were tested with Levene’s and Mauchly’s tests (with Greenhouse-Geisser correction when the sphericity assumption was violated). In case of significant omnibus ANOVA results, post-hoc tests were conducted with Bonferroni corrections applied. To complement null-hypothesis significance testing, we reported standardised mean differences for baseline-referenced contrasts at each phase (H1–H3; C1–C3) for SBP, DBP, and HR. Because measurements were repeated within participants, we calculated Cohen’s d_z statistic²⁵:

, where M_diff is the mean of the paired differences (i.e., how much values changed from baseline to the phase of interest, averaged across participants), SD_diff is the standard deviation of those paired differences (i.e., how much individual participants vary in their changes).

The sign indicates the direction of change relative to baseline. Conventional benchmarks were used to aid interpretation (≈ 0.2 small, ≈ 0.5 medium, ≈ 0.8 large). Effect sizes are reported to contextualise the magnitude and potential physiological relevance of observed differences; primary inference still relies on RM ANOVA and within-subject regression models (β, SE, p, R², AIC/BIC). Additionally, for clinical context, absolute changes (mmHg/bpm) and percentages vs. baseline are described in the results.

To capture personalised dynamic trends, individual regression slopes were computed per participant. A grand (group-level) regression line was calculated to illustrate the overall direction of response across time. Baseline measurements (e.g., SBP₀) were included as independent variables to assess their predictive value on the trajectory of change. Linear regression models, applied across three time points (H1, H2, and H3, or C1, C2, and C3) for each cardiovascular parameter (SBP, DBP, and HR), lead to evaluating the direction and strength of cardiovascular response trends over time. A result at the level of p < 0.05 was considered significant. Calculations were carried out using Statistica 13.3 (StatSoft Poland 2024, Cracow, Poland) and R software with RStudio (PBC, Boston, MA, USA URL http://www.rstudio.com/ (accessed on 15 March 2025)).

Results

Descriptive statistics of direct responses of heating (H) and cooling (C) phases (baseline, after 1st, 2nd, and 3rd phases), as well as p-values derived from post-hoc tests after the omnibus test of repeated measures (RM ANOVA), are presented in Table 1. Three acute responses after heating in the sauna and three cooling responses had a statistically significant effect on SBP levels (F(6, 27) = 127.42, p < 0.001), DBP levels (F(6, 27) = 82.29, p < 0.001), and HR (F(6, 27) = 72.94, p < 0.001) (Table 1). The highest effect size, defined by mean square (MS), was observed in SBP (MS = 13510.16) and DBP (MS = 11904.08), while it was less in HR (MS = 11329.19).

Table 1.

Descriptive statistics of heating and cooling effects in systolic blood pressure (SBP), diastolic blood pressure (DBP), and heart rate (HR). P-values are derived from post-hoc tests conducted after repeated measures analysis of variance. SBP, systolic blood pressure; DBP, diastolic blood pressure; HR–, heart rate; H1, the first heating session; H2, the second heating session; H3, the third heating session; C1, the first cooling session; C2, the second cooling session; C3, the third cooling session.

Variable	Phase	Mean ± SD	P (vs. baseline)	Cohen’s dz
SBP	Baseline	111.50 ± 6.38	–	–
	H1	120.21 ± 11.47	p < 0.05	0.67
	H2	117.32 ± 9.32	n.s.	0.66
	H3	112.03 ± 13.85	n.s.	0.04
	C1	116.50 ± 10.93	n.s.	0.44
	C2	117.21 ± 9.65	n.s.	0.51
	C3	115.32 ± 10.17	n.s.	0.29
DBP	Baseline	71.25 ± 3.85	–	–
	H1	63.64 ± 8.50	p < 0.05	− 0.94
	H2	64.89 ± 11.57	n.s.	− 0.63
	H3	64.50 ± 8.26	n.s.	− 0.81
	C1	72.89 ± 9.23	n.s.	0.18
	C2	73.50 ± 11.19	n.s.	0.22
	C3	73.93 ± 10.21	n.s.	0.26
HR	Baseline	77.14 ± 8.08	–	–
	H1	116.61 ± 17.11	p < 0.001	2.06
	H2	111.43 ± 24.92	p < 0.001	1.40
	H3	116.93 ± 25.13	p < 0.001	1.43
	C1	72.28 ± 13.43	n.s.	− 0.42
	C2	77.46 ± 13.65	n.s.	0.03
	C3	77.11 ± 11.06	n.s.	0.00

Systolic blood pressure was significantly higher in the first heating session compared to the measurement before entering the sauna (p < 0.05). This increase amounted to about + 9 mmHg (≈ 8% of baseline SBP), which can be considered physiologically meaningful. Diastolic blood pressure was significantly lower in the first heating session compared to the measurement before entering the sauna (p < 0.05). The decrease reached nearly − 8 mmHg, which represents about 11% of baseline DBP values. Heart rate was significantly higher in all three heating sessions compared to the measurement before entering the sauna (p < 0.001), as summarised in Table 1 and illustrated by individual slopes in Fig. 1 (SBP), Fig. 2 (DBP), and Fig. 3 (HR). On average, HR increased by ~ 40 bpm during heating, reflecting a robust cardiovascular activation. For transparency, standardised mean differences (Cohen’s d) vs. baseline were: SBP-H1 = 0.67, H2 = 0.66, H3 = 0.04; cooling: C1 = 0.44, C2 = 0.51, C3 = 0.29. DBP-H1 = − 0.94, H2 = − 0.63, H3 = − 0.81; cooling: C1 = 0.18, C2 = 0.22, C3 = 0.26. HR-H1 = 2.06, H2 = 1.40, H3 = 1.43; cooling: C1 = -0.42, C2 = 0.03, C3 = 0.00.

Fig. 1.

SBP regression slopes–heating and cooling phases. Individual regression slopes of systolic blood pressure (SBP) across repeated sauna heating (H1–H3) and cooling (C1–C3) sessions. Grey lines represent individual participant slopes, and the red line represents the grand regression line for the group. A statistically significant downward trend was observed during the heating phase, suggesting a potential adaptive cardiovascular response. No significant trend was detected during the cooling phase.

Fig. 2.

DBP regression slopes–heating and cooling phases. Individual regression slopes of diastolic blood pressure (DBP) across repeated sauna heating (H1–H3) and cooling (C1–C3) sessions. Grey lines represent individual trajectories, and the red line indicates the grand regression line. No significant trends were observed in either phase, suggesting DBP remained stable across thermal exposures.

Fig. 3.

HR regression slopes–heating and cooling phases. Individual regression slopes of heart rates (HR) across repeated sauna heating (H1–H3) and cooling (C1–C3) sessions. Grey lines represent individual trajectories, and the red line indicates the grand regression line. No significant trends were observed in either phase, suggesting DBP remained stable across thermal exposures.

In the present analysis, systolic (SBP) and diastolic (DBP) blood pressure responses were assessed across three repeated sauna exposures (heating phase) and three subsequent cooling phases (cooling phase) using linear regression. Individual regression slopes were computed per participant. A grand (group-level) regression line was calculated to illustrate the overall direction of response across time.

In heating phases, SBP showed a statistically significant decreasing trend across sessions (β = − 4.09 mmHg/session, p = 0.010) (Fig. 1; Table 2), where individual participant trajectories are displayed, suggesting an adaptive response of the cardiovascular system. This corresponds to a cumulative decrease of ~ 8 mmHg from H1 to H3, which may be clinically relevant in terms of blood pressure regulation. The model explained approximately 7.8% of the variance in SBP (R² = 0.078). While, in cooling phases, SBP did not show significant trends. The course decreased slightly (β = − 0.59, p = 0.667) (Fig. 1; Table 2). The absolute changes across C1–C3 were small (< 2 mmHg), supporting the interpretation of SBP stability during cooling.

Table 2.

Summary of linear regression models for SBP and DBP across heating and cooling Phases. Estimated regression coefficients (β) and standard error of beta (SE β), p-values, coefficients of determination (R²), and model fit indices (AIC and BIC) for systolic (SBP) and diastolic (DBP) blood pressure and heart rates (HR) across repeated sauna heating (H1–H3) and cooling (C1–C3) sessions.

Variable	Phase	β	SE β	p-value	R-squared	AIC	BIC
SBP	Heating	-4.09	1.56	0.010	0.078	655	662
	Cooling	-0.59	1.37	0.667	0.002	633	640
DBP	Heating	0.37	1.27	0.737	0.001	621	628
	Cooling	0.52	1.36	0.704	0.001	632	640
HR	Heating	0.16	3.03	0.958	≈ 0.000	767	774
	Cooling	2.41	1.71	0.161	0.024	670	677

In contrast, in heating phases, DBP exhibited a non-significant, slight increase (β = +0.37 mmHg/session, p = 0.737), with virtually no explained variance (R² = 0.001) (Fig. 2; Table 2). Across H1–H3, DBP values fluctuated within ± 2 mmHg of baseline, indicating that DBP remained stable. This suggests that DBP remained stable across heating cycles and did not show evidence of adaptation. Similarly, in cooling phases, DBP increased slightly (β = +0.52, p = 0.704), but these changes were also not statistically significant, and the explained variance was near zero (R² = 0.001). Again, changes were marginal (< 3 mmHg), reinforcing the interpretation of stability. Consistently, paired d_z values were negative and large during heating (H1 = − 0.94, H2 = − 0.63, H3 = − 0.81) and small and positive during cooling (C1 = 0.18, C2 = 0.22, C3 = 0.26).

During the heating phase (H1–H3), heart rate showed no meaningful change over time. The estimated regression slope was β = 0.16 bpm per session, with a p-value of 0.958 (Fig. 3), indicating a completely non-significant effect (Table 2). The R² was nearly zero (R² = 0.00003), suggesting that session number did not explain any variability in heart rate during heating. This means that HR values fluctuated randomly around ~ 115–117 bpm, confirming stability rather than adaptation. These results indicate that repeated sauna exposure did not elicit a cumulative cardiovascular activation as reflected by heart rate (Fig. 3), consistent with the non-significant regression model (Table 2). In the cooling phase (C1–C3), the regression slope was β = 2.41 bpm per session, which suggests a slight increase in heart rate across cooling cycles. However, this effect did not reach statistical significance (p = 0.161), and the explained variance remained low (R² = 0.024). The observed rise was only ~ 5 bpm across three cooling phases, thus of minor practical importance. While the upward trend could point to accumulating cardiovascular stress or incomplete recovery, it cannot be confirmed based on this analysis alone (Fig. 3; Table 2). The paired d_z estimates corroborate this pattern: negligible during cooling (C1 = − 0.42, C2 = 0.03, C3 = 0.00).

To compare the relative fit of linear regression models across the measured cardiovascular parameters (SBP, DBP, HR), model fit indices, including the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC), were examined.

Although the DBP heating model yielded the lowest AIC (621) and BIC (628), indicating the best mathematical fit, its explanatory power was minimal (R² = 0.0014) (Table 2), and the regression coefficient was not statistically significant (p = 0.737) (Table 2). This supports the interpretation of DBP as physiologically stable during heating. In contrast, the SBP heating model had a higher AIC (655) but showed a significant downward trend (p = 0.010), with meaningful variance explained (R² = 0.078) (Table 2). This effect (− 4 mmHg per session) is both statistically significant and physiologically meaningful, given the role of SBP in cardiovascular regulation. Therefore, while the AIC suggests slightly better residual distribution in the DBP model, the SBP regression better reflects a physiologically relevant and statistically robust adaptation effect. On the other hand, this discrepancy suggests the presence of substantial inter-individual variability in SBP responses. Such variability—potentially due to distinct responder and non-responder subgroups—may reduce the overall fit of the group-level model despite a meaningful average effect. In line with this interpretation, paired d_z values for SBP ranged from medium during heating (H1 = 0.67, H2 = 0.66) to negligible (H3 = 0.04) and remained small during cooling (C1 = 0.44, C2 = 0.51, C3 = 0.29), indicating heterogeneous responses at the participant level.

Discussion

This study assessed the dynamic cardiovascular responses—specifically systolic blood pressure (SBP), diastolic blood pressure (DBP), and heart rate (HR)—across three repeated cycles of sauna heating and cold-water immersion in normotensive women. Significant differences in mean values were observed in all three parameters, mainly in the first heating phase compared to the baseline values. There were no significant differences between heating phases or cooling phases. However, SBP showed a general statistically significant downward trend during the heating phase, indicating a potential adaptive response to repeated thermal stress. No significant changes in SBP were observed during the cooling phase, suggesting stable post-exposure recovery. DBP remained largely stable across both heating and cooling phases. This indicates that DBP is less responsive to short-term thermal stress or adaptation compared to SBP. HR also showed no significant changes across heating sessions, and although a mild increasing trend was observed during cooling, it did not reach statistical significance. Overall, heart rate remained stable across sessions, suggesting limited cumulative cardiovascular activation in response to alternating heat and cold exposure.

As shown in Fig. 1, individual SBP trajectories display considerable variability, although the overall group trend is downward. Figure 2 demonstrates that DBP remained stable at both the individual and group levels. As seen in Fig. 3, HR slopes were heterogeneous but did not indicate systematic changes. These participant-level plots enhance the transparency of our findings by illustrating both individual differences and the group-level trend. Different results from ours were obtained by Podstawski et al.²⁶, who found a continuous increase in HR, SBP, and DBP during four consecutive Finnish sauna exposures. In our study, HR and DBP behaved stably during successive heats, while SBP showed a decreasing trend. The discrepancy in the results may be due to different cooling sessions. In the aforementioned study, cooling was 2 times shorter than heating (5 min of cooling vs. 10 min of heating) and consisted of resting in a room at 18 °C, a shower at 14–15 °C, and an optional cold water immersion (10 °C). The authors did not provide information on the duration of the shower and the number of subjects using the cold water immersion. Another reason for the discrepancy in the results may have been the selection of the study group, which consisted of overweight men²⁶. The authors of the study conclude that 5-minute cooling after 10-minute heating is too short for a complete physiological recovery²⁶. Therefore, for individuals planning multiple sauna exposures during a single treatment, the protocol used in our study, in which the cooling time is equal to the heating time, seems more beneficial. Additionally, when studies are based on populations with overweight men, it should be stressed that in this group we may observe impaired thermoregulation in hot environments due to the insulation properties of fat tissue hindering heat loss²⁷.

Other studies included 4 cycles of alternating heating and cooling, and a single cycle consisted of a 12-min exposure in a Finnish sauna at 90–91 °C, followed by 6-min cooling, including a 2-min immersion in cold water (9–11 °C)⁹. After the 72-min treatment, SBP and HR were not significantly different compared to the values measured before the treatment, while a significant decrease in DBP was observed. In our study, there were no statistically significant differences in SBP, DBP, and HR when comparing pre-treatment values with the last measurement taken after the third cooling session. The different DBP responses in our study and those mentioned earlier⁹ may be due to some differences in study protocols. Our protocol included three heating and cooling cycles with a total time of 60 min. We used identical heating and cooling times (10 min), while in the aforementioned work, the cooling time was twice as short as the heating time. Additionally, our study concentrated on the responses of SBP, DBP, and HR throughout the entire procedure, specifically during the successive heating and cooling sessions. In contrast, the authors of the previously mentioned paper assessed only the overall effect of the procedure by measuring haemodynamic parameters before and immediately after⁹.

The procedure used during cooling can significantly affect haemodynamic responses. In a study involving a single 16-min heating session in a Finnish sauna, there was a significant decrease in HR, SBP, and DBP after a 2-min cold water immersion, while when a 30-second cold shower followed by a 90-s rest in a temperature-neutral room was used, there was an increase in HR and no statistically significant changes in SBP and DBP²⁴.

During our investigation, we may have noted the temporary blood pressure increase, and the reason for the observation may be complex. Some authors, for instance, Podstawski et al., also found the temporary and simultaneous slight increase in systolic blood pressure²⁶. They worked with a group of sedentary and overweight participants, which cannot be fully comparable with our study group; however, some ideas seem to be reasonable. First, participants were not frequent sauna users, and the psychological effect could be present in this case due to the unfamiliar, new circumstances. Young participants in our study could initially react in a stressful manner, and this response was normalised after some time. Pilch et al. obtained similar results, characterising a study group as lacking both sport training and regular sauna use²⁸. Furthermore, the authors suggest that variations in blood pressure may be influenced by a patient’s habits, indicating that regular sauna bathing is generally better tolerated and results in more significant and healthier changes in blood pressure. In summary, new studies should evaluate this problem, defining the details and including the habits of the participants.

In our study we selected young and healthy women in order to avoid the confounding factors in the cardiovascular system, as with age the number of health complications increases, and at the same time various pharmacological strategies may be used to cure hypertension and other medical conditions. The observed moderate changes in systolic blood pressure and relatively stable diastolic blood pressure and heart rate may constitute the proof for the safety of sauna therapy, especially for those who suffer from non-severe cardiovascular diseases. There are some studies suggesting that long-term sauna bathing may help lower blood pressure in patients with hypertension; however, additional studies are needed to confirm these findings in the future²⁹. Hannuksela and Ellehham²⁹ claim that very few acute myocardial infarctions and sudden deaths occur in saunas, but alcohol consumption during sauna bathing increases the risk of hypotension, arrhythmia, and sudden death and should be avoided. It is possible that one of the main mechanisms is via blood pressure and heart rate increase in this situation, promoting arrhythmia and myocardial ischaemia²⁹. Currently, cardiovascular contraindications to sauna bathing include unstable angina pectoris, recent myocardial infarction, and severe aortic stenosis²⁹. Sauna bathing is thought to be safe for most people with coronary heart disease with stable angina pectoris or old myocardial infarction⁸. Laukkanen and Kunutsor³⁰ ask the question if sauna bathing may be protective of sudden cardiac death, and the authors suggest that there is emerging evidence that sauna bathing is associated with reduced risk of adverse cardiovascular disease (CVD) and non-CVD outcomes as well as mortality; however, the evidence is uncertain, and the studies should be continued. According to the authors, few reports have linked sauna baths with sudden cardiac deaths (SCDs), but these few cases have been attributed to the effects of dehydration, hypotension, and cardiac arrhythmias due to a combination of sauna exposure and alcohol consumption, as it was previously mentioned³⁰. Sauna bathing is safe for most healthy people and even among patients with stable cardiovascular disease if used sensibly and with caution^3,8. In a comprehensive review, Laukkanen et al. have analysed the plausible pathways underlying the protective effect of sauna bathing, summing up that they may be linked to reduced arterial stiffness, decreases in inflammation and oxidative stress, stabilisation of the autonomic nervous system, beneficial changes in circulating lipid profiles and other CVD risk markers, and lowering of systemic blood pressure⁷. In 2015, in JAMA, the same author evaluated the association between sauna bathing and fatal cardiovascular and all-cause mortality events, and it has been revealed that increased frequency of sauna bathing is associated with a reduced risk of SCD, coronary heart disease (CHD), CVD, and all-cause mortality; however, the authors comment that further studies are warranted to establish the potential mechanism that links sauna bathing and cardiovascular health³.

The aforementioned reports indicate the majority of mechanisms by which sauna has its beneficial effects are not fully known and should be elucidated. This is why the observational studies like ours are essential to analyse each step of sauna bathing in regard to cardiovascular response. The other aspect, even more complex, is to define inter-individual variability in SBP responses, as we have mentioned in the Results section. In larger studies, the subgroups of responders and non-responders should be identified, but in the current paper we deliberately restricted our analysis to descriptive trajectories to avoid overextension.

Our study has some limitations, which include the lack of information about the menstrual cycle time of the women participating in the experiment; it would be beneficial if both phases before and after ovulation could be compared. Resting body temperature, measured immediately after waking, is 0.3 °C to 0.7 °C higher in the luteal phase compared to the follicular phase, which is mediated by a higher concentration of progesterone³¹. During heat exposure, the higher core body temperature observed in the luteal phase is linked to an elevated core temperature threshold for heat loss mechanisms, such as cutaneous vasodilation and sweating²². In addition, some studies have shown in women a higher rate of sweating and skin blood flow during the luteal phase when exposed to heat³². There are also other potential confounding factors, including individual thermal sensitivity, the thickness of subcutaneous adipose tissue layers, fitness and physical activity levels, various diet habits, and hydration status, which could be analysed. Additionally, results refer to a group of young and healthy women, and more longitudinal studies are needed to verify the impact of sauna cycles on men, as well as in different ethnic groups. Moreover, the evaluation of different age groups should be carried out in the future. It is known saunas are safe for the majority of young healthy persons; however, as interest in them increases in spa centres and resorts all over the world, additionally, older people with coexisting cardiovascular diseases should be taken under study, including the most common diseases: arterial hypertension and coronary artery disease in their stable phases. Taking into account the limitations, our results should be treated as preliminary and promising for future research, as safety and efficacy have been evaluated for young and healthy women.

Conclusions

Our findings suggest that systolic blood pressure is the most sensitive parameter of the repeated heating stress in sauna conditions, reflecting possible cardiovascular adaptation. The diastolic blood pressure and heart rate remained stable, showing no clear signs of adaptation or cumulative stress response under the conditions tested. The observed reduction in SBP across repeated sauna exposures could have potential implications for applied settings; however, additional future studies are still necessary in this field. The present study fills the gap in this field by focusing on female participants, but there is the need for more detailed research on diverse populations. Repeated thermal stress via sauna bathing may serve as a non-pharmacological intervention to support cardiovascular regulation, particularly in normotensive or pre-hypertensive individuals; however, at present, the safety and efficacy are proven for young and healthy women. Novel longitudinal studies addressing older participants with coexisting health problems are needed to estimate their adaptation abilities and safety. Furthermore, besides the preventive health role, saunas may have potential in athletic recovery and stress management contexts where cardiovascular modulation is desirable. Further studies should address inter-individual variability in response patterns, possibly by applying responder classification frameworks to better understand who benefits most from repeated thermal stress.

Author contributions

R.S. designed and supervised the study. R.P. and J.D. analysed the data. R.S., R.P., and J.D. wrote the main manuscript text. All authors read and approved the final version of the manuscript.

Funding

This study received no external funding.

Data availability

Data for the current study will be available upon reasonable request from the corresponding author.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval

The research protocol and study design were approved by the Research Ethics Committee of Wroclaw University of Health and Sport Sciences (28/02/2011) in accordance with the Declaration of Helsinki. Informed consent was obtained from all subjects involved in the study.