Acetylsalicylic acid does not alter thermo-effector responses during mild whole-body passive heat stress in young men

Abstract

Acetylsalicylic acid (ASA), aspirin, exerts potent systemic effects that may interfere with normal thermo-effector responses. We investigated the influence of commonly ingested ASA doses on measures of skin blood flow (SkBF) and local sweat rate (SR) during whole-body, passive heat stress. Seven male participants completed counter-balanced trials to compare ASA treatments (single dose 325 mg or 4 consecutive days 81 mg (4-d 81 mg)) to control (no ASA). Laser-Doppler flowmetry provided an index of SkBF. A ventilated capsule measured local sweat rate via capacitance hygrometry. Mean body temperature () was increased by 1 °C above baseline using a water-perfused suit. was similar at the onset of cutaneous vasodilation among trials. Cutaneous vascular conductance, expressed as a percentage change from baseline, was not different among trials. Additionally, at the onset of local SR and SR sensitivity did not differ among trials. While ASA has previously been shown to influence SkBF during heat stress, it is possible our cohort’s relatively young age may have contributed to our dissimilar results.

• Aspirin
• cyclooxygenase
• sweating
• thermoregulation
• vasodilation

Introduction

Prolonged exposure to hot ambient conditions leads to a rise in core body temperature. Upon reaching a relative ‘threshold’, increased skin blood flow (SkBF) and eccrine sweating are initiated to defend against a rise in body temperature [1]. Accordingly, hypothalamic-mediated temperature regulation relies on central and peripheral signalling to preserve cellular and enzymatic activity [2]. Since normal physiological function necessitates a relatively narrow core thermo-balance (37 ± 1 °C) [3], any environmental or pharmacological challenges that threaten such balance may pose a health risk. One such challenge may be the use of acetylsalicylic acid (ASA), a commonly used over-the-counter medication for treating fever and pain.

Through the inhibition of cyclooxygenase (COX), ASA can elicit potent systemic effects through the suppression of central and peripheral signalling [4]. While the efficacy of ASA to treat fever is widely recognised, more recently a daily regimen of low-dose (81 mg) ASA has been prescribed for its antithrombotic properties [5].

With regard to temperature regulation, current understanding suggests nitric oxide synthase- [6,7] and cyclooxygenase- (COX-) [8] mediated pathways are obligatory for complete reflexive cutaneous vasodilator expression, especially in the young. However, attenuated SkBF responses during whole-body, passive heat stress have been repeatedly observed in middle-aged participants taking low-dose ASA daily [9,10]. Researchers postulated that diminished SkBF effects may have resulted from altered vessel wall signalling, inhibition of vascular COX, and/or altered vascular shear stress due to a change in blood viscosity [9].

Additionally, acute ASA (1000 mg) ingestion has been shown to inhibit current-induced cutaneous vasodilation (i.e. a model for neurogenic inflammation) [11]. Researchers reported that ASA treatment significantly altered vasodilator responses and theorised that prostaglandins of endothelial or smooth muscle origin were unlikely to be involved. Instead, it is possible that platelet activity may have been modified, which serves to highlight the importance of peripheral signalling for proper homeostatic function. While previous work has demonstrated ASA use significantly alters SkBF responses, research investigating the influence of ASA on local sweat rate is limited.

During heat stress, the release of acetylcholine initiates reflexive cutaneous vasodilation, and additionally stimulates sweating through muscarinic receptor activation [12]. Accordingly, increased SkBF and sweating often occur sequentially as body temperature rises, although research has demonstrated that decreased SkBF and decreased local temperature independently alter sweat rate (SR) sensitivity [13]. More recently, Smith and colleagues [14] determined that the onset and rise of regional sweating is dependent on increased SkBF during passive heat stress. Due to the role of COX-mediated pathways during reflexive cutaneous vasodilation [8], ASA-induced modifications in SkBF may indirectly affect local sweating. Indeed, any impairment in heat dissipation via SkBF and/or sweating may increase heat storage, possibly leading to heat injury.

The purpose of this study was to determine how a commonly ingested acute ASA dose (e.g. 325 mg) and four consecutive days of 81 mg low-dose ASA (4-d 81 mg) affected measures of temperature regulation during whole-body passive heat stress. A four-day low-dose ASA loading period was selected as it has been described to achieve effective blood platelet inhibition [5]. It was hypothesised that SkBF responses would be attenuated during ASA trials during passive whole-body heating, and additionally eccrine sweating responses would be weakened as evidenced by lowered SR sensitivity and lowered peak SR.

Methods

Ethical approval

Prior to inclusion, all participants were informed of the methods and risks associated with this study. Verbal and written consent were acquired from each participant during each visit. The consent and experimental procedures were approved by the local institutional review board. All procedures conformed to the standards set by the Declaration of Helsinki.

Participants

Seven healthy, male participants (mean ± SD, 28 ± 3 years) were recruited to participate in this study. A power analysis [15] revealed this sample size was adequate to detect a moderate effect size (d = 0.60) [16] for peak local SR, assuming a power of∼ 0.85 and correlation among repeated levels of each factor at∼ 0.90. All participants were normotensive and self-reported non-smokers, maintaining moderately active lifestyles in accordance with American College of Sports Medicine guidelines [17]. Participants were not taking any medications, including herbal supplements, and did not report an allergy to ASA. During the 12 h prior to each experimental session, participants abstained from caffeine, alcohol, and vigorous physical activity. A randomly assigned counterbalanced, crossover design was used to reduce an ordering effect among the following trials: control, single 325 mg ASA, and 4 days of 81 mg ASA. Experimental trials were conducted at a similar time of day (within 1 h) to minimise the influence of circadian rhythms on body temperature.

Instrumentation and measurements

Following arrival at the laboratory, participants provided their consent and completed a 24-h health history questionnaire to confirm pre-instruction adherence. Hydration status was assessed by urine specific gravity (USG) through refractometry (SUR-NE, Atago, Higham, MA). Adequate hydration was defined as values < 1.020 [18]. Once hydration status was confirmed, participants had their standing height and nude body mass measured (Tanita, Tokyo, Japan). Lange calipers (Beta Technology, Santa Cruz, CA) were used to estimate body fat percentage from the sum of three skinfolds [19]. To measure rectal temperature (T_re), participants self-inserted a rectal thermocouple (RET-1, Physitemp, Clifton, NJ) 8–10 cm beyond the anal sphincter [20]. Mean skin temperature () was assessed by weighted measures of six thermocouples placed on the upper back, upper chest, lower back, abdomen, anterior thigh, and calf. Mean body temperature () was calculated from weighted measures of T_re and from the following equation [21]: Participants were then instrumented with heart rate monitors (Polar, Stamford, CT).

An index of SkBF was obtained via laser-Doppler flowmetry (LDF) (MoorVMS-LDF2, Moor Instruments, Axminster, UK) with a single probe positioned on the dorsal aspect of the forearm. Cutaneous vascular conductance (CVC) was calculated from arbitrary LDF units divided by mean arterial pressure. Mean arterial blood pressure was determined by automated oscillometry and calculated from the following equation:

Local SR was continuously measured from a ventilated capsule (3.98 cm²) placed near the LDF probe. Compressed nitrogen gas perfused through the capsule at a fixed rate of 300 mL/min. Humidity from the effluent air was measured via capacitance hygrometry. Measurements from anatomical landmarks of the forearm were taken to facilitate consistency between trials. Body temperature was manipulated with a water-perfused suit covering the whole body apart from the face, hands feet, and instrumented forearm. An impermeable suit (i.e. rain suit) was worn over the perfusion suit to maximise heat storage and encourage sweating. To mitigate the risk of room air affecting local skin temperature of the experimental forearm, the sleeve of the rain suit was extended. This permitted the area to be enclosed while not obstructing instrumentation. Total body sweating was estimated from the difference in nude body mass before and after the experimental protocol.

Experimental protocol

Standard USA full-dose (325 mg) ASA was selected, as it is commonly used for pain relief and required no additional preparation. Since peak plasma levels occur within 30–40 min and altered platelet function is detectable within 1 h of ingestion [5], a non-enteric coated pill was used to evaluate the acute effects of ASA. Once hydration was confirmed, a single dose of 325 mg ASA was administered with a small snack to mitigate gastrointestinal distress. Participants were then instrumented to ensure the passive heating protocol began within 30–40 min after ASA ingestion.

Standard USA low-dose (81 mg) ASA was selected due to its widespread use as an anti-thrombotic. Accordingly, a 4-day loading period of 81 mg ASA has been shown to be a sufficient time-course to thoroughly inhibit platelet activity [5]. The cumulative dose of 81 mg over 4 days equates to 324 mg. Because the heating protocol was consistently repeatable, this study compared the effects of ASA based on timing (i.e. acute versus daily) since the absolute doses were similar (i.e. single 325 mg versus 4-d 81 mg (324 mg)). It should be noted that during the loading period, no experimental procedures were conducted and participants were encouraged to resume normal free-living activities, apart from any vigorous exercise. Following the 4-day loading period, participants returned to the laboratory within∼ 18–22 h following their last ASA dose. All experimental ASA trials were separated by a minimum of 10 days to permit normal platelet re-synthesis [5]. Due to counter-balancing and to ensure adequate recovery, control trials which took place before or between ASA trials were given at least 72 h before another experimental session was conducted.

During the trials, instrumented participants entered an environmentally controlled chamber set to temperate conditions (22 °C, 30–35% relative humidity). Participants were aided onto a gurney and remained in the supine position throughout the duration of the experiment. All instruments were connected and values verified. To clamp at∼ 34 °C (thermoneutral), water (∼ 31.3–31.5 °C) was perfused through the suit. A 10-min period was permitted to allow to stabilise. During the final minute of this period, baseline data were collected. Next, hot water (∼ 49 °C) was circulated through the suit to increase 1.0 °C above baseline temperature. Upon reaching the target and during the final minute of heat stress, peak data values were collected. Cool water was then perfused through the suit to lower within∼ 0.4 °C of baseline. Throughout the heating protocol, heart rate was continuously monitored, while blood pressure was measured every 5 min via automated oscillometry. At the cessation of the experiment, participants were aided off the gurney to remove instrumentation. After towel drying, participants reported their nude body mass and were given fluids to consume ad libitum. All experimental trials were conducted in this manner.

Statistical analyses

Data were continuously sampled at a rate of 50 Hz using a data acquisition system (Biopac, MP150, Goleta, CA) and stored for offline analysis. Changes in temperature variables (i.e. , T_re, ) before heating were compared with peak heat stress values and analysed using a repeated-measures analysis of variance (ANOVA). CVC responses during heat stress were expressed as a percentage change from baseline ((peak CVC − baseline CVC)/baseline CVC) × 100). The onset of cutaneous vasodilation and sweating was determined by an experienced investigator, blinded to the treatment, by visually examining arbitrary LDF units and local SR graphed as a function of time. The at the indicated time was then used to identify the onset of cutaneous vasodilation or sweating. The at the plateau of local sweating or at the final if a plateau did not occur was used for calculation [13]. A linear regression was used to calculate the slope of the local SR: relationship from all data points between the onset of sweating and peak of heat stress (or plateau if appropriate). One-way repeated measures ANOVA was used to compare the slope and peak local SR among control and ASA treatments. In instances where the assumption of sphericity was violated, subsequent degrees of freedom (df values) for within-subject effects were adjusted using the Greenhouse-Geisser correction. Data were analysed using SPSS version 22 (IBM, New York). Data are presented as means ± SD. Significance was accepted at p values ≤ 0.05.

Results

Participant characteristics are shown in Table 1. At baseline was not different between trials (control, 36.5 ± 0.1 °C; 325 mg, 36.4 ± 0.2 °C; 4-d 81 mg, 36.5 ± 0.2 °C; p = 0.14). Prior to heating, baseline measures revealed CVC responses were not different among trials (control, 0.14 ± 0.2; 325 mg, 0.11 ± 0.3; 4-d 81 mg, 0.15 ± 0.3; p = 0.11). Cardiovascular and thermal responses during the final minute of heat stress are presented in Table 2. ASA treatments did not produce any significant change in measured dependent variables. Additionally, total body water losses (mL) were not different among trials (p = 0.19).

Table 1. Participant characteristics (n = 7).

Age (years)	28 ± 3
Body mass (kg)	84.7 ± 15.1
Height (m)	1.77 ± 0.04
Body fat (%)	19.9 ± 7.8
Systolic BP (mm Hg)	109 ± 6
Diastolic BP (mm Hg)	70 ± 7
MAP (mm Hg)	83 ± 6
Resting HR (beats*min^−1)	59 ± 4

Values are presented as means ± SD; BP, blood pressure; MAP, mean.

Table 2. Cardiovascular and thermal responses to 1 °C increase in mean body temperature above baseline (n = 7).

	Control	325 mg	4-d 81 mg	p value
Rectal temperature (°C)	+ 0.6 ± 0.1	+0.6 ± 0.1	+0.6 ± 0.1	0.30
Mean skin temperature (°C)	+5.1 ± 1.0	+4.7 ± 0.8	+4.6 ± 0.6	0.23
HR (beats*min^−1)	+38 ± 4	+37 ± 8	+36 ± 5	0.79
Systolic BP (mm HG)	+15 ± 10	+7 ± 10	+8 ± 6	0.09
Diastolic BP (mm HG)	−1 ± 5	−4 ± 9	−1 ± 7	0.50
MAP (mm HG)	+4 ± 5	0 ± 8	+2 ± 5	0.36
Total body sweat loss (mL)	740 ± 220	830 ± 110	940 ± 340	0.19
Duration of heat stress (min)	65 ± 11	68 ± 15	71 ± 14	0.39

Values are presented as means ± SD; HR, heart rate; BP, blood pressure; MAP, mean arterial pressure; p values indicate omnibus ANOVA.

Thermo-effector responses

At the onset of cutaneous vasodilation, mean body temperature was similar among trials (control, 36.7 ± 0.2 °C; 325 mg, 36.7 ± 0.2 °C; 4-d 81 mg, 36.7 ± 0.3 °C; p = 0.91) (Figure 1). During the final minute of heat stress, the magnitude of change from baseline (CVC) was not different among trials (control, 523 ± 219%; 325 mg, 714 ± 372%; 4-d 81 mg, 545 ± 173%; p = 0.42) (Figure 2).

Figure 1. Mean body temperature (means ± SEM) at the onset of cutaneous vasodilation during passive, whole-body heat stress among control and acetylsalicylic acid trials (325 mg; 4-d 81 mg), (n = 7); p value indicates omnibus ANOVA.

Figure 2. The percentage change (%Δ) of cutaneous vascular conductance (CVC) (means ± SEM) from baseline to the end of passive, whole-body heat stress among control and acetylsalicylic acid trials (325 mg; 4-d 81 mg), (n = 7); p value indicates omnibus ANOVA.

Neither ASA treatment influenced any measures of local sweating responses (Table 3). Changes in local SR were measured as a function of , and thus, provided insight into the sensitivity of this thermo-effector response. Consequently, differences in local SR sensitivity ((mgċcm⁻²ċmin⁻¹)/ °C) were not seen among trials (control, 1.23 ± 0.26; 325 mg, 1.24 ± 0.34 °C; 4-d 81 mg, 1.20 ± 0.36 °C; p = 0.87). Furthermore, peak local SR was similar among trials and did not appear to be influenced by either ASA treatment (p = 0.87).

Table 3. Sweat responses during whole-body, passive heat stress among control and acetylsalicylic acid trials (325 mg; 4-d 81 mg) (n = 7).

	Control	325 mg	4-d 81 mg	p value
at onset of sweating (°C)	36.9 ± 0.1	36.8 ± 0.1	36.8 ± 0.2	0.43
SR sensitivity [(mgċcm^−2ċmin^−1)/ °C]	1.23 ± 0.26	1.25 ± 0.34	1.20 ± 0.36	0.87
Peak SR (mgċcm^−2ċmin^−1)	0.89 ± 0.24	0.88 ± 0.25	0.87 ± 0.19	0.87

Values are presented as means ± SD; , mean body temperature; SR, sweat rate; SR sensitivity, expressed as a function of . P values indicate omnibus ANOVA.

Discussion

Previous work has demonstrated that daily ingestion of low-dose ASA significantly attenuated SkBF responses during passive heat stress [9,10]. More recently, Smith and colleagues [14] determined that the onset and rise of regional sweating is dependent on increased SkBF during passive heat stress. Within this context we reasoned that ASA ingestion could alter local sweat responses indirectly through an attenuation of SkBF. Accordingly, the purpose of this study was to determine how a commonly ingested acute ASA dose (e.g. 325 mg) and a 4-day loading of low-dose (81 mg) ASA affected measures of temperature regulation during passive heating in young male participants. Contrary to our initial hypotheses and independent of ASA dosing strategy (i.e. acute or daily), the principal outcome of this investigation suggested ASA ingestion did not influence measures of thermoregulation within the framework of this study.

Given that ASA readily permeates the blood–brain barrier and irreversibly inhibits blood platelet activity, it exerts both central and peripheral effects. In the present study, measures of temperature and cardiovascular-related variables were similar among control and ASA trials at baseline and peak heat stress (Table 2). Recently though, Bruning et al. [22] observed higher oesophageal temperatures in middle-aged participants taking low-dose ASA (7–10 days) compared to controls following 40 min of exposure to warm (30 °C) ambient conditions. This was especially interesting, since ASA and control trials began with similar core temperatures prior to heating. Despite the higher core temperature reached during the ASA trials, no differences in %CVC_max were reported during passive heating. Researchers from that study speculated that due to the mild thermal stress, the threshold for initiating reflexive cutaneous vasodilation had not been breached.

Given the consistent reports of ASA-induced alterations in SkBF [9,10,22] our group was surprised not to witness any discernable differences in SkBF in either ASA trial. Since blood platelets are known to release vasodilators (e.g. adenosine diphosphate and serotonin) [23,24], low-dose ASA therapy is thought to diminish cutaneous vasodilator activity through the following mechanism(s): 1) reduced mechanical deformation along the endothelial wall due to changes in viscoelastic properties, and/or 2) reduced platelet-derived vasodilator factors [9,10,22]. Due to the dissimilar results of the present study it is possible that the 4-day loading period was insufficient to elicit any meaningful change in blood platelet activity. However, daily ASA dosing of 20–40 mg (0.44 ± 0.06 mg/kg) has been shown to maximally inhibit up to 95% platelet activity within 6 to 12 days [25]. In the present study, we used twice the ASA (i.e. 81 mg) during the loading period and based on the mean body mass of our sample (84.7 kg, see Table 1) reached a concentration of approximately 0.95 mg/kg per day. For this reason, we speculate that the dosing strategy used in the present study inhibited the majority of platelets and our dissimilar results are a product of other factors.

Participant age is a distinct difference between the present study and previous investigations [9,10,22] reporting altered SkBF responses with oral ASA use. Minson and colleagues [26] determined that older participants (∼ 70 ± 3 years) were unable to match the blood redistribution characteristics exhibited by younger participants (∼ 23 ± 1 year) during passive heat stress. Due in part to a reduced stroke volume, less blood volume was available to perfuse the cutaneous vascular beds. Furthermore, advancing age is related to functional changes within the skin itself. Notably, there is a greater reliance on the NO-dependent component of reflexive vasodilation as individuals age [27], and as such, the relatively youthful age of our sample combined with the inherent redundancy of the thermoregulatory effectors [28] may have accounted for our dissimilar findings.

Few studies have examined the effects of ASA on body temperature regulation and sweating responses. As previously stated, we hypothesised that eccrine sweating could be influenced indirectly through a modification in SkBF responses during heat stress. However, neither ASA dosing strategy affected SR sensitivity or peak SR during heat stress (Table 3). To investigate the sensitivity of local SR, the slope was expressed as a function of . Consequently, this approach permitted a continuous examination of local sweat rate during the+ 1.0 °C heating protocol. Given the absence of thermal (i.e. , T_re, ) or SkBF differences in the present study, logic suggests local SR would be unaffected. Previous work has shown that acute ibuprofen ingestion (800 mg) did not influence local sweating responses in female participants (aged 21–32 years) at high and low hormone phases of their menstrual cycle [29]. While ibuprofen is also a COX-inhibitor, it exhibits a much greater degree of COX-2 selectivity compared to ASA [30]. While no differences in local SR were found, it is possible that eccrine sweating was not sensitive to COX-inhibition in younger individuals.

Limitations

Several limitations arose in this study. Since these data were collected during the summer months, in an area known for its hot/humid conditions, we cannot rule out the possibility of dissimilar results in cooler climates. Due in part to the heterogeneity of cutaneous vascular beds, SkBF responses are often normalised to a maximal hyperaemic response. However, in this instance no differences in baseline CVC were seen, and thus, permitted the evaluation of SkBF as a change from baseline. Irrespective of ASA treatment, inter-individual differences in body mass and aerobic fitness may have impacted individual responses to the heating protocol. ASA treatment was not titrated to body mass. However, ASA doses as low as 30 mg/day have been shown to be just as efficacious on platelet activity as 75 mg [31]. Finally, only young healthy, male participants were tested in this study, and as such, extrapolation of these data to other demographics must be done cautiously.

Conclusion

This study determined that acute and short-term (i.e. 4 days) ASA use did not influence measures of SkBF or local SR during whole-body passive heat stress. The similarities in , T_re, and between ASA and control trials support these findings. Given that previous work has shown that ASA treatment blunts SkBF responses in older participants [9,10,22], it is plausible that age may be responsible for the dissimilarity in results from previous studies [32]. Accordingly, future work should seek to understand whether different ASA dosing strategies affect eccrine sweating in other groups and special populations.

Acknowledgements

The authors thank all participants for their willingness to take part in this investigation. The entirety of this investigation was conducted at the University of Alabama.

Declaration of interest

The authors report no conflict of interest. The authors alone are responsible for the content and writing of the paper.