Introduction

Human thermoregulation aims at keeping core temperature at approximately 37 °C and resist thermal fluctuations caused by either endogenous or exogenous factors. Muscular exercise, particularly when performed at vigorous intensity, drastically increases metabolic rate as compared to baseline levels (Jette et al. 1990). Depending on the type of exercise performed, more than 50% of metabolic energy is released as heat (Gonzalez-Alonso et al. 2000; Krustrup et al. 2003) that must be dissipated to the environment to maintain body heat balance. Inadequate heat dissipation results in hyperthermia which limits physical performance especially in endurance-like exercise (Gonzalez-Alonso & Calbet 2003; Nybo 2007, 2009; Tucker et al. 2004). The degree of exercise-induced hyperthermia is largely independent of environmental conditions and directly proportional to the metabolic rate (Nielsen 1938). Sports apparel may also contribute to hyperthermia during physical activity as clothing imposes a barrier for heat exchange with the environment (Gavin 2003). Even in relatively cool environments potentially performance-limiting increases in core temperature may occur if exercise intensity (Ely et al. 2009) and the insulation of the sports apparel worn are sufficiently high (Gavin 2003; Gonzalez et al. 1997).

While systemic increases in body temperature have repeatedly been found to impair endurance performance (Stevens et al. 2016), the functioning of individual skeletal muscles is known to be directly proportional to muscle temperature. Cooling reportedly impairs both the maximal power (Csapo et al. 2017; Drinkwater & Behm 2007) and endurance capacity of skeletal muscles (Bergh & Ekblom 1979; Faulkner et al. 1990; Oksa et al. 2002). In confirmation of the latter, Inoue et al. (2016) recently showed that endurance work performance in a cycle ergometer test was higher when thigh muscle temperature was set to 36 °C as compared to 32 °C.

Sports apparel should therefore meet the competing demands to keep working muscles warm while still facilitating overall heat dissipation to prevent performance-limiting rises in core temperature. In scientific experiments, various attempts have been made to locally influence muscle temperature through warm water immersion (Gray et al. 2006; Sargeant 1987), exposure to hot air (Schlader et al. 2011) or integration of heating/cooling elements (Faulkner et al. 2012; Inoue et al. 2014, 2016). As opposed to these attempts to actively modify muscle temperature, it was decided to reduce heat dissipation through insulation pads that are readily implementable into outdoor sports apparel. Specifically, the goal of this study was to determine the effects of thermal insulation of working muscles on clothing surface temperature and parameters reflecting thermoregulation, physical effort and perceived exertion during endurance exercise in cool environments. Two different combinations of sports apparel were compared: One textile system consisted of a thin shirt and pant to maximize overall heat loss through evaporation of sweat, convection and radiation, whereas the other system featured additional thermal insulation pads placed over the leg muscles with the aim to reduce heat flux from working muscles.

It was hypothesized that the pants containing insulation pads would reduce heat flux, promote local rises in temperature and thereby reduce submaximal exercise responses, as reflected by reduced blood lactate levels and lower heart rates. Furthermore, it was assumed that the insulation pads would not hinder overall dissipation of body heat as they covered only a small part of the whole body surface. Therefore no significant differences in auditory canal and upper body clothing surface temperature were expected between clothing systems.

Method

Study design

To test the influence of insulation pads on clothing surface temperature and parameters reflecting thermoregulation, physical effort and perceived exertion two different running pants with (PINSUL) and without insulation pads (PCOOL) were designed and tested in two sub-studies: Study A was performed with recreationally active sportsmen who were tested while exercising at moderate intensity under temperate climatic conditions. To reflect the demands in competitive sport, additional experiments were conducted with well-trained endurance athletes exercising in a considerably colder environment (study B).

Apparel

Two different, custom-made running pants (PCOOL, PINSUL) were made of identical base fabric and in the same cut. While PCOOL was just made of the base material to facilitate heat dissipation, PINSUL featured additional insulating pads as illustrated in Fig. 1 which were sewn in on the inside. The pads were placed to covered the gluteal, knee extensor, ankle dorsi- and plantar flexor muscles and covered ~ 30% of the surface of PINSUL. The overall insulation of PCOOL (size: S) was 0.055 clo while PINSUL (size: S) featured 2.4-times higher insulation (0.131 clo)Footnote 1 RCT and further material characteristics of the base fabric and insulation pads are evident from Table 1. The pants were combined with thin long-sleeve shirts composed of the same material as the pants. Pants and shirts were provided in different sizes to guarantee a tight fit. Subjects were also provided with identical thin hats and gloves (100% polyester).

Fig. 1
figure 1

Schematic image of PINSUL indicating the position of the additional insulation pads placed on the inside at the front (a) and rear (b) face. The pads were placed to cover the gluteal, knee extensor, ankle dorsi- and plantar flexor muscles and covered ~ 30% of the surface of PINSUL. PCOOL (not pictured) had the same cut and was made of the same base material as PINSUL but was designed without additional insulation pads

Full size image
Table 1 Material characteristics
Full size table

Subjects

Ten male sport students (age: 23 ± 3 y, height: 182 ± 5 cm, weight: 73.4 ± 4.9 kg, body surface areaFootnote 2: 2.3 ± 0.1 m2, Body Mass Index (BMI)Footnote 3: 22 ± 1 kg/m2) and eight well trained endurance athletes (age: 28 ± 2 y, height: 175 ± 5 cm, weight: 72.7 ± 7.3 kg, body surface area: 1.9 ± 0.1 m2, BMI: 24 ± 2 kg/m2) volunteered to participate in studies A and B, respectively. The experiments were approved by the Institutional Review Board of the Department of Sport Science at the University of Innsbruck. Participants were informed about the study purpose and methods involved before giving written consent. Physical readiness to participate was assessed through completion of the Physical Activity Readiness Questionnaire (PARQ) (Adams 1999).

Exercise intervention

To determine their individual maximal running velocity (vmax) an incremental and exhaustive treadmill-based test (5% inclination; pulsar, h/p/cosmos, Germany) was completed by all subjects: After warming up for 5 min at 6 km/h, running speed was increased by 1 km/h every minute until subjects aborted the test due to full exhaustion. If the final stage was not completed for 1 min, vmax was calculated proportionally.

Then, participants were scheduled for two visits to compare the different running pants in a randomized order. Tests were conducted at the same time of the day and interspersed by a minimum of 48 h of passive recovery. On both testing days, subjects warmed up on the treadmill for 5 min at a freely chosen velocity. Then, participants were instructed to run for 45 min at 5% inclination. Running velocity was set to 60% (study A) and 70% (study B) of vmax. All tests were conducted under constant ambient conditions (study A: 7 ± 1 °C and 40 ± 3% relative humidity; study B: 0 ± 1 °C and 40 ± 3% relative humidity) in a climatic chamber (Kältepol, Austria). Light wind (20 km/h) was simulated using a wind machine (TTW 25000 S, Trotec, Germany) positioned next to the treadmill at a 45° angle and facing the subjects. Average running speed was 9.6 ± 0.6 km/h in study A and 11.8 ± 0.8 km/h in study B, respectively.

Measurements

Measurements were taken after warm-up (t0) and after 15 min (t1), 30 min (t2) and 45 min (t3) of running. Heart rate, blood lactate, auditory canal temperature and perceived exertion were measured at all measuring times whereas body mass measurements as well as thermal images were obtained at t0 and t3 only.

Clothing surface temperatures and auditory canal temperature were determined as representative measures of body surface and core temperature. Thermal images were recorded from the upper and lower body (Vario Cam High Resolution, Infratec, Germany) with subjects wearing the test apparel. Average clothing surface temperatures were separately calculated in areas coinciding with the anterior and posterior aspect of legs and upper body using custom-made Labview routines (National Instruments, USA) (Fournet et al. 2015). As both pants (PINSUL, PCOOL) and the long-sleeve shirts were composed of the same material and tight fit, temperature values calculated on the basis of thermal images of the clothing surface can be directly compared without the need to adjust results for differences in material compositions (emission coefficient was assumed to be ε = 0.9) (Maldague 2012; Pastore & Kiekens 2000). Since the temperatures measured on the anterior and posterior side were found to be largely congruent, the average of these two values was calculated for further analysis. According to Burtscher et al. (2012) auditory canal temperature was measured in the left ear using a thermometer, which was kept under ambient temperature conditions between measurements (ThermoScan IRT 4520, Braun, Germany). During the protocol, ears were covered with a cap to prevent potential bias resulting from cooling of the outer auditory canal.

Blood lactate, heart rate and subjects´ loss of body mass were measured as parameters reflecting physical effort. For blood lactate concentration, capillary blood samples were drawn from the earlobe (Double determination; EKF Biosen 5040, Germany). Heart rate was measured via chest strap (Polar Electro, Finland). To estimate the loss of water due to sweating and respiration, the subjects were weighed in underwear before and after running using a high-precision scale (Kern DS 150K1, Kern & Sohn GmbH, Germany), and the average of five consecutive measurements was considered for further analyses (Agache et al. 2004). In addition, Borg’s scale (6 = no exertion; 20 = maximal exertion) was used to inquire perceived exertion (Borg 1982).

Statistical analyses

Factorial ANOVAs with repeated measurements were used to determine the influence of the factors pant and time on blood lactate, heart rate, auditory canal temperature, clothing surface temperature, subjects´ loss of body mass and perceived exertion. Pant × time interaction effects were non-significant for all variables, so the respective results are omitted for improved clarity. In cases where Mauchly’s test indicated a violation of the assumption of sphericity, degrees of freedom were corrected by the Greenhouse–Geisser procedure. Values are reported as mean values ± standard deviation (SD). Differences were considered significant at p ≤ 0.05. SPSS Statistics Version 21 (IBM, USA) was used for all statistical calculations.

Results

Lower body clothing surface temperature

In study A (Fig. 2a) lower body clothing surface temperatures (TLB) (F(1, 9) = 14.138, p = 0.004) were significantly lower for PINSUL compared to PCOOL indicating a lower heat flux with PINSUL. TLB did not change significantly over time (F(1, 9) = 2.855, p = 0.125). Also in study B (Fig. 2b), TLB showed significantly lower values for PINSUL (F(1, 7) = 11,531, p = 0.012). Reflecting the colder conditions in the climatic chamber during study B, TLB dropped significantly during the exercise intervention (F(1, 7) = 9.664, p = 0.017).

Fig. 2
figure 2

Lower body clothing surface temperature (TLB) as measured after warm-up (t0) and after 45 min (t3) of running while subjects were wearing the thermally insulating (PINSUL) and non-insulating (PCOOL) pants in studies A (a) and B (b). Significantly lower TLB were found for PINSUL (p < 0.05) in study A and B indicating a lower heat flux with PINSUL compared to PCOOL. A significant effect of time (p < 0.05) was found in study B. Note: Asterisk indicates significant differences (p < 0.05) between PINSUL and PCOOL

Full size image

Upper body clothing surface temperature

In study A (Fig. 3a), upper body clothing surface temperature (TUB) revealed no significant differences between apparel systems (F(1, 9) = 0.488, p = 0.503) but decreased significantly from t0 to t3 (F(1, 9) = 25.877, p = 0.001). Analogously in study B (Fig. 3b), the influence of pant on TUB (F(1,7) = 1.989, p = 0.201) was non-significant whereas a significant effect of time was found for TUB (F(1,7) = 12,245, p = 0.010).

Fig. 3
figure 3

Upper body clothing surface temperature (TUB) as measured after warm-up (t0) and after 45 min (t3) of running while subjects were wearing the thermally insulating (PINSUL) and non-insulating (PCOOL) pants in studies A (a) and B (b). No significant differences were found between PINSUL and PCOOL. Factor time was significant in study A and B (p < 0.05)

Full size image

Auditory canal temperature

In study A, auditory canal temperature (Fig. 4a) was neither significantly affected by pant F(1, 9) = 0.371, p = 0.558) nor time (F(3, 27) = 1.382, p = 0.270).

Fig. 4
figure 4

Auditory canal temperature as measured after warm-up (t0) and after 15 min (t1), 30 min (t2) and 45 min (t3) of running for the thermally insulating (PINSUL) and non-insulating (PCOOL) pants worn in studies A (a) and B (b). No significant differences were found between PINSUL and PCOOL but factor time was significant in study B (p < 0.05)

Full size image

Just like in study A, the effect of pant (F(1, 7) = 0.192, p = 0.674) on auditory canal temperature was not significant in study B (Fig. 4b). However, a significant drop in auditory canal temperature was observed over time (F(3, 21) = 7.504, p = 0.001).

Parameters reflecting physical effort

In study A (Fig. 5a) no significant difference between blood lactate levels was found between PINSUL and PCOOL (F(1, 9) = 0.612, p = 0.454). In response to the exercise, blood lactate increased at the start of the exercise and then stabilized (F(1.064, 9.579) = 27.564, p < 0.001) at higher levels.

Fig. 5
figure 5

Blood lactate levels as measured after warm-up (t0) and after 15 min (t1), 30 min (t2) and 45 min (t3) of running while wearing the thermally insulating (PINSUL) and non-insulating (PCOOL) pants in studies A (a) and B (b). No significant differences were found between PINSUL and PCOOL but factor time was significant in study A and B (p < 0.05)

Full size image

Just as in study A, the effect of pant (F(1, 7) = 0.050, p = 0.830) on blood lactate levels was not significant in study B (Fig. 5b). Blood lactate increased compared to initial levels and then stabilized during exercise (F(1.292, 9.045) = 29.263, p < 0.001).

In study A (Fig. 6a), no significant effects on heart rate were found for the factor pant (F(1, 9) = 4.896, p = 0.055). Heart rate increased during the running intervention, with all follow-up measures being significantly greater as compared to t0 (F(1.120, 10.076) = 237.294, p < 0.001).

Fig. 6
figure 6

Heart rate as measured after warm-up (t0) and after 15 min (t1), 30 min (t2) and 45 min (t3) of running for the thermally insulating (PINSUL) and non-insulating (PCOOL) pants in studies A (a) and B (b). No significant differences were found between PINSUL and PCOOL but factor time was significant in study A and B (p < 0.05)

Full size image

Also in study B (Fig. 6b), the effect of pant on heart rate (F(1, 7) = 0.661, p = 0.443) was non-significant (F(3, 21) = 0.043, p = 0.988). Heart rate increased from t0 to t1 and then stabilized, revealing a significant effect of time (F(1.118, 8.318) = 53.496, p < 0.001).

Body mass loss

In study A, subjects’ body mass decreased significantly from t0 to t3 by 646 ± 107 g with PINSUL and 645 ± 133 g with PCOOL (F(1, 9) = 330.435, p < 0.001), but no significant differences were found between pants (F(1, 9) = 0.075, p = 0.791).

Comparable losses in body mass (PINSUL: 644 ± 182 g, PCOOL: 680 ± 170 g) were observed in study B. As in study A, no statistical effects were found between pants (F(1, 7) = 0.013, p = 0.911) but the effect of time was significant (F(1, 7) = 104.178, p < 0.001).

Perceived exertion

Perceived exertion as measured with the Borg scale is shown in Fig. 7a, b. As for the parameters reflecting physical effort, the differences between pants (F(1, 9) = 0.438, p = 0.525) failed to reach statistical significance in study A. Perceived exertion increased with both pants during the protocol reflecting a significant influence of time (F(3, 27) = 229.550, p < 0.001).

Fig. 7
figure 7

Perceived exertion as measured after warm-up (t0) and after 15 min (t1), 30 min (t2) and 45 min (t3) of running with the Borg scale for the thermally insulating (PINSUL) and non-insulating (PCOOL) pants in studies A (a) and B (b). No significant differences were found between PINSUL and PCOOL but factor time was significant in study A and B (p < 0.05)

Full size image

A similar time course of perceived exertion was found in study B. No significant differences between PINSUL and PCOOL (F(1, 7) = 0.538, p = 0.487) were found, but the effect of time was significant (F(1.262, 8.832) = 129.561, p < 0.001).

Discussion

The aim of this application-oriented study was to determine the effects of thermal insulation pads placed over working leg muscles on clothing surface temperature and parameters reflecting thermoregulation, physical effort and perceived exertion during endurance exercise in cool environments. In agreement with the hypothesis, significantly lower TLB for PINSUL compared to PCOOL were found, confirming the insulation effect of the additional pads. Furthermore, no significant differences in TUB and auditory canal temperature were found, which confirms the assumption that the additional insulation pads would not impair thermoregulation during exercise. However, no significant differences in parameters reflecting physical effort or perceived exertion between PINSUL and PCOOL were found.

The rationale to test insulation pads placed over working muscles with the aim to improve endurance performance in cool environments was that cold ambient conditions may lead to a drop in muscle temperature (Parkin et al. 1999). This may be of functional importance since cold muscles show impaired performance (Bennett 1985; Bottinelli et al. 1996; Drinkwater & Behm 2007). To counter the expected decreases in muscle temperature, simple insulation pads were used which, unlike active cooling or warming devices (Faulkner et al. 2012; Gray et al. 2006; Inoue et al. 2014, 2016; Sargeant 1987; Schlader et al. 2011), do not require energy supply and could be easily integrated into sports apparel in real-life scenarios.

With PINSUL, TLB were significantly lower as compared to PCOOL. Lower surface temperatures reflect smaller heat loss due to better thermal insulation (Al-Homoud 2005). During exercise, muscles produce heat with heat production being directly proportional to exercise intensity (Gonzalez-Alonso et al. 2000; Jette et al. 1990). The lower heat emission observed with PINSUL is expected to promote higher local temperatures underneath the insulated areas. However, conclusions about the effects on muscle temperature remain speculative since (highly invasive) direct measurements were precluded in this study. No significant differences in TUB and auditory canal temperature between pants were found, implicating that thermoregulation was not negatively influenced by the additional insulation pads. Studies by Fournet et al. (2015) and Gavin et al. (2001) confirm these results. They showed that neither clothing systems featuring the same overall insulation but varying local insulation, nor small differences in overall clothing insulation have a significant influence on core temperature while exercising at intensities comparable to the ones applied in this study.

The expected differences in muscle temperature notwithstanding, parameters reflecting physical effort and perceived exertion were not significantly different between PINSUL and PCOOL. Various factors may explain this result: It is possible that the warming effect of the insulation pads was not large enough to cause significant differences in muscle temperature and, consequently no physiologically relevant differences in physical effort were found. Inoue et al. (2016) induced temperature differences of 4 °C within the muscle to produce a significant effect on endurance performance. It is also possible that, even with the non-insulating PCOOL, muscle temperatures did not drop to an extent where performance deficits may be expected (Noakes 2000). While exercise intensities and climatic conditions were deliberately selected to represent typical conditions encountered by recreationally active sportsmen who exercise in a cool environment (study A) or well-trained athletes exercising in a cold environment (study B), lower exercise intensities in combination with colder environmental conditions are expected to result in more drastic decreases in muscle temperature. In future studies the effect of insulation pads should be evaluated in climatic conditions typical for cold-weather endurance sports like cross country skiing where training and competition are often performed in temperatures below − 15 °C (Larsson et al. 1993).

Several methodical limitations of the current study must be considered. Instead of measuring muscle temperature directly, thermography recordings were used to confirm the insulating effect of the additional pads. In future studies direct measures of muscle temperature should be used to directly determine the differences in muscle temperature induced by insulation pads. Additionally, submaximal exercise responses were investigated, i.e., blood lactate, heart rate, auditory canal temperature, loss in body mass and perceived exertion. While representative of physical effort, such measures may not accurately reflect exercise performance capacity. Thus, in future studies, the effect of insulation pads on the outcomes of competition like events or endurance performance tests (e.g. time trails or time to exhaustion) should be investigated.

Conclusions

Insertion of properly positioned insulation pads into sports apparel is a practical approach to limit heat emission from working muscles during endurance exercise in cool environments without impairing overall body-heat dissipation. However, under the environmental conditions and exercise intensities applied in this study, the insulation of working muscles failed to significantly improve parameters reflecting physical effort or perceived exertion. Two reasons might account for that: Either the warming effect of the insulation pads was not large enough to cause significant differences in muscle temperature or muscle temperatures in both pants did not drop to an extent where performance deficits may be expected. Future studies on the benefit of insulation pads should therefore focus on temperature conditions markedly below 0 °C, where a drop in muscle temperature is likely to be larger. Also muscle temperature should be measured directly.