Heat acclimation enhances the cold-induced vasodilation response

Apart from the exercise training incorporated in all three acclimation protocols, two additional environmental stressors (heat and hypoxia) were added to the protocols to observe how different combinations would interact and affect cardiorespiratory and thermoregulatory responses (Fig. 1).

All three acclimation protocols were aiming at a central heat adaptation as a result of training and/or prolonged exposure to heat. Primarily, the idea was to observe how this central adaptation would further reflect peripherally in the CIVD response. Additionally, the study aimed to investigate the signs of local heat acclimation by immersing a hand, that was also tested for CIVD response, daily in warm water for 30 min. Since local cold exposure of the hands has been shown to augment the CIVD response (Krog et al. 1960; Nelms and Soper 1962), the hypothesis was that such benefits could also be observed after local heat exposure.

In all three studies, measurements of the CIVD response were conducted before and after each 10-day intervention (2 days before and 2 days after the protocol). Participants did not perform any exercise before the CIVD measurements were conducted and they remained seated for 20 min after entering the laboratory to adapt to the controlled ambient temperature (25 °C) and to allow us to ensure stable blood flow in the hands. Thereafter, a temperature sensor (MSR 147WD, MSR-Electronic GmbH, Switzerland) was placed on the side of each finger of the right hand attached with thin air-permeable tape (Tegaderm, 3M, USA). Once instrumented, the forearm was placed on a flat surface at the hip level for 3 min, providing the baseline measurements. Immediately after, a thin plastic bag was placed on the hand. Excess air was removed, and the bag was sealed with micropore tape (3M, USA) approximately 10 cm above the wrist. Next, the instrumented right hand was immersed up to the ulnar and radial styloid processes sequentially in tanks of warm (35 °C) and cool water (8 °C) for 5 and 30 min, respectively. During the hand immersion, participants were seated on a padded chair, with the elbow supported by a flat surface and thick towel. After the 30-min cold-water immersion, the hand was taken out of the water, the plastic bag removed, and the forearm again placed on a flat surface at the hip level for another 15 min (recovery phase). Auditory canal temperature (T_ac) was measured every 5 min, using a commercial infrared thermometer (ThermoScan, Braun, Germany). Heart rate (HR) was monitored with a heart rate monitor (Polar T34, Finland) throughout the trial. CIVD measurements were conducted at the same time of the day with similar daily activities being reported pre- and post-acclimation for each participant.

To distinguish any potential benefits of the daily warm-water hand immersion, both hands were tested before and after the NorEx protocol. Both hands were immersed during the same visit, however, the immersion of each hand was separated by 30 min. Half of the participants commenced the immersion with their left hand, and the other half with the right hand. The same order of hand immersion was repeated for each participant before and after the 10-day acclimation protocol. The CIVD measurement was performed in the same manner as described above.

During the CIVD measurements, participants provided ratings of thermal comfort and thermal sensation of the immersed hand every 10 min, using the following descriptors: 0–0.5: comfortable, 1–1.5: slightly uncomfortable, 2–2.5: uncomfortable, 3: very uncomfortable for thermal comfort and 3: hot, 2: warm, 1: slightly warm, 0: neutral, − 1: slightly cool, − 2: cool and − 3: cold for thermal sensation (ANSI/ASHRAE 2013).

The area under the curve (AUC) as an incremental area under the curve above 5 °C (Gorjanc et al. 2018), was defined and determined from the 2nd to the 30th min of the cold-water immersion, since the first 2 min were transitional with finger temperature decreasing rapidly. It was calculated from a downloadable spreadsheet—the Time Series Response Analyser (Narang et al. 2020). For other CIVD parameters, identified in previous studies (Daanen 2003; Keramidas et al. 2010; Mekjavic et al. 2008), a computer program written in LabVIEW (National Instruments, Austin, Texas) was developed. Briefly, these parameters included:

Following the HeA protocol, the average temperature of all five fingers was higher both during cold-water immersion (from 13.9 ± 2.4 to 15.5 ± 2.5 °C; p = 0.049) and during the recovery phase (from 22.2 ± 4.0 °C, to 25.9 ± 4.9 °C; p = 0.023). Observing the responses of individual fingers during immersion, only the thumb displayed a higher temperature post- (16.0 ± 2.8 °C) compared to pre-HeA protocol (14.2 ± 2.2 °C; p = 0.032), whereas the temperature of the other fingers remained similar between the trials (Fig. 2). The rewarming rate was improved in all fingers (Fig. 2), except the little finger (p = 0.067), although a tendency towards improved rewarming was shown. Accordingly, the AUC after the HeA protocol was greater in the thumb during cold-water immersion (from 230 ± 74 to 278 ± 81 [arbitrary units]; p = 0.044) and, with the exception of the little finger (from 246 ± 65 to 288 ± 66; p = 0.069), in all remaining fingers during the recovery phase (thumb: from 243 ± 42 to 291 ± 74, p = 0.026; index: from 245 ± 57 to 303 ± 73, p = 0.016; middle: from 239 ± 63 to 304 ± 69, p = 0.014; ring finger: from 242 ± 71 to 297 ± 77, p = 0.044; [arbitrary units]).

The analysis of CIVD parameters indicated greater number of waves after the HeA protocol, observed in the thumb (from 1.4 ± 1.1 to 2.3 ± 1.3; p = 0.025), index (from 1.5 ± 0.9 to 2.1 ± 1.3; p = 0.046) and middle finger (from 1.5 ± 0.9 to 2.3 ± 1.2; p ˂ 0.001). T_max in the thumb increased from 14.1 ± °C to 17.8 ± 3.3 °C after the HeA protocol (p = 0.010). The onset time from immersion to the first T_min and T_max in the little finger was reduced by ~ 4 min after the HeA protocol (p ˂ 0.05).

The average temperature of all five fingers remained unaltered during the cold-water immersion (before: 13.4 ± 2.1 °C, after: 13.9 ± 2.6 °C; p = 0.491) as well as the recovery phase (before: 21.8 ± 4.5 °C, after: 23.2 ± 4.7 °C; p = 0.491) after the combined HeA/HypA protocol. None of the individual fingers displayed pre- to post-HeA/HypA difference (Fig. 3). AUC remained unaltered in all of the fingers.

After the HeA/HypA protocol, the number of waves increased in the thumb (from 1.0 ± 0.9 to 1.8 ± 0.7; p = 0.048) and middle finger (from 1.1 ± 1.2 to 2.0 ± 0.8; p = 0.041). No other differences in any of the recorded CIVD parameters were noted.

During the cold-water immersion, the average finger temperature of both hands did not differ between the pre- (left hand: 14.2 ± 1.2 °C, right hand: 14.1 ± 1.6 °C) and post-NorEx protocol (left hand: 14.2 ± 1.4 °C, right hand: 14.2 ± 1.2 °C; p = 0.933). Furthermore, during the recovery phase, the average finger temperature of both hands remained similar before (left hand: 18.7 ± 4.3 °C, right hand: 18.6 ± 3.9 °C) and after the applied protocol (left hand: 17.1 ± 2.8 °C, right hand: 17.5 ± 4.0 °C; p = 0.125). Individual fingers did not display any pre- to post-acclimation differences in the measured temperature (Fig. 4). AUC remained unaltered in all of the fingers.

The number of waves did not differ in any of the left and right-hand fingers when comparing pre- to post-NorEx data. The average number of waves was less than one per finger (more precisely, the average was 0.7 ± 0.6 for the left hand and 0.7 ± 0.6 for the right hand), with several participants not displaying any CIVD response before and/or after the NorEx protocol. The analysis considered only the participants, who displayed CIVD response both before and after the protocol.

Left hand: the analysis was limited to 7, 8 and 9 participants with CIVD response in the thumb, little finger and remaining fingers, respectively. Based on the available data, it was shown that the temperature amplitude (ΔT: the difference between T_min and T_max) was actually reduced in some of the fingers following the application of the protocol, including the middle (from 2.9 ± 1.5 to 1.7 ± 1.0 °C; p = 0.003), ring (from 2.4 ± 1.3 to 1.7 ± 1.0 °C; p = 0.041) and little finger (from 2.8 ± 1.6 to 1.5 ± 0.6 °C; p = 0.021). The onset time from immersion to the first T_min was shorter after NorEx, observed in the index (from 16 ± 4 to 11 ± 4 min; p = 0.013) and ring finger (from 16 ± 3 to 11 ± 4 min; p = 0.021). The latter also displayed shorter time to T_max after NorEx (from 22 ± 5 to 17 ± 5 min; p = 0.035).

Right hand: there were only 2 participants with an evident thumb CIVD response. The data for this finger were, therefore, not considered for analysis. There were 8 participants with CIVD responses in the index and middle finger, whereas the ring and little finger displayed CIVD response in 10 and 9 participants, respectively. Further analysis revealed no differences in the measured CIVD parameters before and after NorEx in any of the fingers.

As evident from the present results, the CIVD response was improved following the HeA protocol, whereas no such modifications were noted after the HeA/HypA protocol. Both studies used an identical exercise training protocol in a warm environment with the difference being that the participants in the HeA protocol were not confined and lived at an altitude of 295 m (in Ljubljana, Slovenia). The participants in the HeA/HypA protocol were confined in normobaric hypoxic conditions (a simulated altitude of ~ 4000 m) for the duration of the 10-day protocol with the 90-min exercise training protocol performed in normobaric normoxic conditions (the hypoxic facility is located at a natural altitude of 940 m). As participants in the HeA/HypA were confined to normobaric hypoxia, the lower FiO₂ (0.141 ± 0.005) in that protocol might have inhibited heat training-anticipated benefits on the CIVD response. Indeed, these findings are in line with thermoregulatory outcomes following steady-state and incremental exercise to exhaustion (Sotiridis et al. 2019a,2019b). After the HeA protocol, our participants demonstrated both lower resting and exercise T_re as well as lower T_re thresholds for sweating and vasodilation (Sotiridis et al. 2019a), exhibiting adaptations pertinent to a classic heat adaptation phenotype (Patterson et al. 2004; Sawka et al. 2011). After the HeA/HypA protocol, only partial adaptations to heat acclimation were observed (Sotiridis et al. 2019b). It has, however, previously been reported that the CIVD response can be enhanced after a sleep high–train low regimen. The study by Amon et al. (2012) assessed the CIVD response before and after 28 days of aerobic training with daily 1-h exercise (50% of W_peak) in thermoneutral conditions with the experimental group sleeping (9–12 h per night) at a simulated altitude of 2800 m (week 1) to 3400 m (week 4) and the control group sleeping close to sea level. Despite an enhanced aerobic performance post-intervention observed in both groups, the enhancement in the CIVD response was more profound in the experimental group. Some field studies have also demonstrated enhanced CIVD response in alpinists following high-altitude Himalayan expeditions (Felicijan et al. 2008; Gorjanc et al. 2018). While the study by Amon et al. (2012) was a controlled laboratory study with the hypoxic exposure limited to night time, the studies by Felicijan et al. (2008) and Gorjanc et al. (2018) were field studies involving continuous hypoxic exposure. Common to these studies is the long-term exposure (≥ 3 weeks) to simulated (normobaric hypoxia)/actual (hypobaric hypoxia) altitude and its acclimatisation effect on CIVD, which was initially observed by Mathew et al. (1977) and later explored by Daanen and van Rujten (2000). The authors observed a reduction in the CIVD response during the first weeks of high-altitude exposure, which was restored following the acclimatisation (~ 3 weeks). The acclimatisation to altitude could further explain the augmented CIVD response after three or more weeks of altitude exposure (Amon et al. 2012; Felicijan et al. 2008; Gorjanc et al. 2018). Although in the study by Amon et al. (2012) the simulated altitude was limited only to night time with no cold acclimatisation commonly experienced in the high-altitude field studies, the hypoxic exposure alone seemed to be sufficient to induce changes in the CIVD response. On the contrary, the hypoxic exposure in our HeA/HypA study might have been too short to affect CIVD, but might also have blunted the thermoregulatory benefits of the HeA protocol (Sotiridis et al. 2019a).

The exercise training itself, performed without any environmental stressors (NorEx) elicited minor benefits in the CIVD response. Mean finger temperature and the number of waves remained unaltered following the applied supervised exercise training protocol. The average number of waves was less than one per finger before and after the experimental intervention with a great number of participants displaying no CIVD response in one or more fingers. Some of the fingers of the left hand, in fact, exhibited reduced temperature amplitude after acclimation. The index and ring fingers were the only fingers displaying some improvements, evident in a shorter time interval required to reach T_min and/or T_max. Previous research (Keramidas et al. 2010) has shown an increase in average finger skin temperature and the number of waves after 4 weeks of exercise training in thermoneutral conditions (1 h daily, 5 days per week, at 50% of W_peak). Exercise training significantly increased VO_2peak, whereas the potential thermoregulatory gains were not reported. Similarly, our NorEx protocol with 10-day exercise training exhibited an improvement in V̇O_2peak in less fit individuals. While the peak sweat rate increased in both groups when performing submaximal aerobic exercise in the heat, sweating was initiated at lower T_re only in the more fit participants. No other thermoregulatory benefits of the acclimation protocol were noted (Sotiridis et al. 2020). As such, the improvements in performance and minor thermoregulatory gains were evidently not sufficient to provoke CIVD enhancement in the present study. While in the study by Keramidas et al. (2010) participants trained for 20 days with intermittent breaks during weekends, our participants trained for 10 consecutive days. Additionally to the enhanced aerobic capacity, the longer training protocol might have further provided thermoregulatory benefits reflected in an augmented CIVD response as in the study by Keramidas et al. (2010).

During the 10 days of HeA/HypA and NorEx protocols, participants immersed their right hand in warm water (35 °C) before attending daily exercise training. This concomitant local hand-heat acclimation was introduced to observe potential effects of such local exposure on the CIVD response. To distinguish between local and central acclimation, CIVD measurements were performed on both hands before and after the NorEx protocol. The results indicated no differences in the temperature responses between the left and right hand, meaning that the daily hand immersion in warm water did not induce any local thermoregulatory adaptations. The CIVD response also remained unchanged following the HeA/HypA protocol, confirming no effect of the local HeA protocol. It is well documented that the CIVD response is enhanced in people whose hands are regularly exposed to cold (Krog et al. 1960; Nelms and Soper 1962; Purkayastha et al. 1992). On the contrary, some laboratory studies show no such trainability of the response (Daanen et al. 2012; Mekjavic et al. 2008). A valid argument might be that studies, targeting specific groups and occupations regularly exposed to local cold stress, are actually recruiting people demonstrating a higher level of local cold tolerance, whereas individuals who experience severe negative physiological or psychological reactions to local cold exposure will likely not engage in such occupations. As such, the observed differences may not be due to an acute or chronic acclimatisation response, but rather due to pre-existing innate physiological differences (Cheung and Daanen 2012). Nonetheless, the study on Korean female divers, who daily harvest the sea bed in cold waters, provides some intriguing findings. The divers’ CIVD response was improved once they started wearing thick wetsuits, which reduced the cold stress experienced by the divers from the whole-body to local cold stress (Lee et al. 2017). Conversely, the present study observed an enhanced CIVD response as a consequence of a whole-body heat stress (training in a warm environment at 35 °C), whereas the daily hand immersion in warm water (35 °C) might not have provided the participants with a stimulus strong enough to induce local heat acclimation. Warmer water, longer daily exposures (more than 30 min per day) and longer protocols (higher number of total acclimation sessions) might modify the outcome in future investigations.

When conducting CIVD measurements, interest is usually focussed on the finger skin temperature during cold-water immersion (Daanen et al. 2012; Keramidas et al. 2010; Mekjavic et al. 2008). However, some studies also consider the recovery phase that is initiated with the rewarming of the hand once it is removed from the cold water (Kingma et al. 2019; Ko et al. 2020; Lee et al. 2013; Keramidas et al. 2019). The present study employed a 15-min recovery phase after cold-water immersion to observe potential alterations in rewarming of the hand as a result of the acclimation protocol. Following the HeA protocol, a significant increase in finger temperature was observed during both cold-water immersion and the recovery phase. Conversely, no enhancement in the CIVD response during cold-water immersion might further explain our inability to observe improvements in rewarming of the fingers following both the HeA/HypA and NorEx protocols. Similarly, Ko et al. (2020) observed that higher minimum finger temperature during cold-water immersion is associated with a faster recovery during the rewarming phase. Cold injury may not only be attributed to tissue cooling but also to the lack of recovery following the cold exposure. Thus, future studies aiming to investigate the vasomotor response to cold should also include an assessment of that response during the rewarming/recovery phase.

In the present study, a variety of CIVD patterns was observed among different participants as well as individual fingers. The CIVD response is often presented as a typical skin cooling response interspersed with slight elevations in skin temperature. The present study confirms the atypical nature of the CIVD, as previously emphasised by Mekjavic et al. (2008). Throughout the three acclimation protocols of the present study, several types of responses were observed, such as no CIVD, small wave, big wave, early plateau, late plateau, a quick first wave and double wave. Interestingly, following the HeA and HeA/HypA protocols, four participants exhibited a profound increase in finger skin temperature during cold-water immersion or recovery. In contrast to HeA, the HeA/HypA protocol exhibited no overall significant increase in the CIVD response. Still, half of the participants notably increased the average finger skin temperature following the protocol. These individual cases suggest that combined HeA/HypA protocol was beneficial for some participants, but might require a longer adaptation protocol to achieve more consistent responses. Due to the small sample size, the current study may have been underpowered. The outcome might have been different if the number of participants was greater. Following the NorEx protocol, no such extreme cases were noted. During the recovery phase, a substantial number of participants exhibited a reduction in the mean finger skin temperature by more than ~ 2 °C (six participants in the fingers of the left hand, and five participants in the fingers of the right hand) whereas only a small number of participants improved the rewarming of the fingers after the NorEx protocol (three participants in the left and two participants in the right-hand fingers). This outcome was somewhat surprising as exercise training alone has been shown to enhance the CIVD response (Keramidas et al. 2010). As explained earlier, a longer exercise training protocol with intermittent breaks might have provided a different, more beneficial outcome.

The CIVD measurements were performed on all fingers of either the right (HeA, HeA/HypA) or both hands (NorEx). In general, fingers displayed different temperatures, number of waves as well as individual CIVD parameters. These different responses, observed in individual fingers, suggest that CIVD should be measured in all fingers, as opposed to past studies conducting the measurements on a single finger (Daanen and Van Ruiten 2000; Lee et al. 2013; O'Brien et al. 2015). It has been demonstrated that the CIVD response is more pronounced following finger immersion and reduced after hand or forearm immersion (Ducharme et al. 2001; Sendowski et al. 1997). Nonetheless, a single-digit cold-water immersion might not reflect the responses of the remaining fingers. Based on previous work (Mekjavic et al. 2013), we concluded that the CIVD responses observed during immersion in 8 °C water were not statistically different from those observed during immersion in 5 °C water. Similarly, Tyler et al. (2015) observed the CIVD response of a finger to 0 and 8 °C water immersion and concluded that both temperatures induced a CIVD response with the latter more suitable in optimising the number of CIVD waves. The participants also reported the immersion in 8 °C water as being much less uncomfortable than immersion in 5 °C water (Mekjavic et al. 2013). There being no further gain anticipated by immersions in water temperatures lower than 8 °C, this water temperature was used in the present study, and has also been used in previous studies investigating hand cold-water immersion (Gorjanc et al. 2018; Cheung and Mekjavic 2007; Daanen et al. 1997).

The present study conducted acclimation protocols (HeA, HeA/HypA, NorEx) in different seasons—the summer 2017, spring 2018 and autumn 2018. Some (Ihzuka et al. 1986; Kakitsuba et al. 2011; van Ooijen et al. 2004), but not all (Ciuha et al. 2020) thermoregulatory responses can vary seasonally. Moreover, it has been shown that CIVD can vary greatly between summer and winter (Tanaka 1971). Since none of the acclimation protocols were conducted in the winter, the seasonal differences were at least partly accounted for. Furthermore, the CIVD measurements were initiated at similar core and skin temperatures to minimise any potential seasonal effects. The daily hand immersion in warm water, investigating potential local heat acclimation, was introduced only during the HeA/HypA and NorEx protocols. Unfortunately, due to logistical reasons, we were not able to conduct the local heating in the HeA protocol. Nonetheless, to distinguish between the local and central heat acclimation, the NorEx protocol was the crucial one, performing exercise training in thermoneutral conditions with no external thermal stressor, showing no changes in the CIVD response, and therefore, no benefits of this local heating. The CIVD response was evaluated by measuring the skin temperature of our participants’ fingers. Whether human feet would display a similar response remains unclear. Considering previous research work, a different outcome might be expected (Cheung and Mekjavic 2007; Daanen et al. 2012). Therefore, the future work should also consider evaluating the CIVD response of the human feet.

The present study demonstrated that whole-body heat acclimation increased the vasodilatory response during cold-water immersion and enhanced the rewarming rate of the hand, thus potentially contributing to improved cold tolerance. The increased vasodilatory response was a result of a heat acclimation protocol, which has previously been shown to enhance the vasodilatory response. Whether the present results can be attributed to cross-adaptation warrants further investigation. The relative contributions of central and peripheral factors to CIVD remain unresolved. The results of the present study would suggest that whole-body heat acclimation increased the CIVD response, whereas daily immersion of the hand in warm water did not. This would favour the contribution of central over peripheral factors to the observed response.