Acute acetaminophen (paracetamol) ingestion improves time to exhaustion during exercise in the heat

New Findings

• •

  What is the central question of this study?

  Acetaminophen (paracetamol) is an analgesic and antipyretic, which has been shown to improve self-paced cycling performance through a reduction in pain. We sought to ascertain whether acetaminophen could improve time to exhaustion during exercise in the heat through its antipyretic action.

• •

  What is the main finding and its importance?

  An acute dose of acetaminophen allowed participants to cycle significantly longer in hot conditions by a mean of 4 min (+17%). This was accompanied by significantly lower core, skin and body temperature, and participants found the exercise less of a thermal strain. These findings suggest that acetaminophen may reduce the thermal challenge of exercise in hot conditions.

Acetaminophen (paracetamol) is a commonly used over-the-counter analgesic and antipyretic and has previously been shown to improve exercise performance through a reduction in perceived pain. This study sought to establish whether its antipyretic action may also improve exercise capacity in the heat by moderating the increase in core temperature. On separate days, 11 recreationally active participants completed two experimental time-to-exhaustion trials on a cycle ergometer in hot conditions (30°C, 50% relative humidity) after ingesting a placebo control or an oral dose of acetaminophen in a randomized, double-blind design. Following acetaminophen ingestion, participants cycled for a significantly longer period of time (acetaminophen, 23 ± 15 min versus placebo, 19 ± 13 min; P= 0.005; 95% confidence interval = 90–379 s), and this was accompanied by significantly lower core (−0.15°C), skin (−0.47°C) and body temperatures (0.19°C; P < 0.05). In the acetaminophen condition, participants also reported significantly lower ratings of thermal sensation (−0.39; P= 0.015), but no significant change in heart rate was observed (P > 0.05). This is the first study to demonstrate that an acute dose of acetaminophen can improve cycling capacity in hot conditions, and that this may be due to the observed reduction in core, skin and body temperature and the subjective perception of thermal comfort. These findings suggest that acetaminophen may reduce the thermoregulatory strain elicited from exercise, thus improving time to exhaustion.

Introduction

The conversion of metabolic to mechanical energy to produce muscle contraction results in approximately 30–70% of the total energy being liberated as heat (Edwards et al. 1975). Where exercise work rate is intense, heat production at the muscle increases and can cause a subsequent increase in core temperature (T_core). As human resting T_core is kept within reasonably narrow limits of ±1°C (Maughan et al. 2007), an effective mechanism of heat loss must be present to maintain thermoregulation. This is predominantly achieved through the transportation of metabolic heat (originating predominantly from the exercising skeletal muscle) in the blood to the skin surface, where heat can be lost to the environment through radiation, convection and evaporation (Maughan et al. 2007).

Exercise performance is progressively impaired with increases in T_core (Gonzalez-Alsonso et al. 1999), and a T_core > 40°C can result in exertional heat illness, coma and death (Nielsen, 1996). During steady-state exercise in thermoneutral conditions, T_core is generally not elevated outside of this range. However, when environmental temperatures exceed 35°C, human mechanisms of heat loss are compromised and T_core will rise as a function of exercise intensity and time (Nielsen, 1996), leading to a subsequent reduction in exercise performance. In order to moderate this effect, numerous interventions have been proposed which either reduce resting T_core before exercise, known as precooling (Booth et al. 1997), or help negate the rise in T_core during exercise, such as cold fluid ingestion (Wimer et al. 1997). These strategies can be classed as either external (application) or internal (ingestion) and have been shown to improve performance in prolonged (Wimer et al. 1997), intermittent (Castle et al. 2006) and sprint exercise in hot conditions (Castle et al. 2011) with a change in T_core of >0.3°C. Whilst these strategies have widely been shown to demonstrate an ergogenic effect, application of these techniques in a field setting is often impractical and problematic (Marino, 2002).

Acetaminophen (ACT; paracetamol) is a common over-the-counter drug, primarily used to relieve pain. Its primary mechanism of action is believed to be through the inhibition of cyclo-oxygenase (COX; Anderson, 2008), which is responsible for the production of prostaglandins (which sensitize and stimulate type II and IV afferents). Acetaminophen is also an antipyretic agent and has been shown to reduce the core temperature of febrile patients (Walson et al. 1989). Whilst the efficacy of ACT as an antipyretic to treat fever and disease is well recognized (Feldberg et al. 1972), its use in moderating T_core in healthy individuals with elevated T_core through metabolic heat is relatively unknown. However, Kasner et al. (2002) has demonstrated that ACT can reduce T_core by ∼0.22°C in afebrile stroke patients. As precooling T_core by 0.3°C has been shown to elicit an ergogenic effect during exercise (Castle et al. 2011), ACT ingestion may help to reduce the elevation in T_core during exercise and subsequently improve performance.

The aim of the present study was to investigate whether an acute dose of ACT would reduce the rise in T_core during time-to-exhaustion exercise in hot conditions. It was hypothesized that ACT would increase time to exhaustion.

Methods

Ethical approval

All eligible participants signed an informed consent form describing the experimental protocol and procedures, which were approved by the Institute for Sport and Physical Activity Research Ethics Committee (University of Bedfordshire). All subjects were provided with written instructions describing all study procedures, but were naive of its aims and hypotheses. The study conformed to the standards set by the World Medical Association Declaration of Helsinki ‘Ethical Principles for Medical Research Involving Human Subjects’ (2008).

Participants

Eleven recreationally active but untrained, healthy male adults (age, 21 ± 1 years; height, 184 ± 8 cm; weight, 72 ± 15 kg; and maximal oxygen uptake, 44 ± 6 ml min⁻¹ kg⁻¹) volunteered to participate in this study. Before selection for the study, all participants were asked to complete an acetaminophen risk assessment questionnaire. Where this questionnaire was not satisfactorily completed and consumption of acetaminophen was judged unsafe, the participant was not deemed eligible for participation. Participants who had obtained prior partial or full heat acclimation were excluded from participation.

Experimental protocol

Participants visited the laboratory on three occasions, separated by 3–5 days. All exercise tests were performed on a calibrated Monark cycle ergometer (model 824E; Monark, Varberg, Sweden). During visit 1, participants were familiarized with the laboratory, performed a graded exercise test to exhaustion (GXT) and a time-to-exhaustion (TTE) familiarization (separated by 45 min recovery) in ambient conditions (18°C, 40% relative humidity). During visits 2 and 3, participants performed a cycle TTE in hot conditions (30°C, 50% relative humidity) following the ingestion of either 20 mg (kg lean body mass)⁻¹ of acetaminophen (ACT) or a flour placebo (PLA). Visits 2 and 3 were performed in a double-blind, randomized design. Before each visit, participants were asked to keep their pre-exercise meal the same and to refrain from drinking alcohol (48 h abstinence) or caffeine (8 h abstinence) and instructed not to perform any exhaustive exercise in the preceding 48 h. This was verbally confirmed prior to each test; adherence was 100%. Participants were instructed to consume 500 ml of water 2 h preceding arrival at the laboratory, in order to ensure they were euhydrated (Sawka et al. 2007). Body mass was obtained using digital scales (BWB0800; Tanita, Arlington Heights, IL, USA) and a urine refractometer (Pocket PAL-OSMO; Alago Vitech Scientific, Partridge Green, UK) was used to measure the hydration levels of the participants. A participant was deemed to be euhydrated if urine osmolality was <600 mosmol (kg H₂O)⁻¹ (Hillman et al. 2011). Each TTE was performed at the same time of day (±1 h).

Visit 1 – graded exercise test and familiarization

Anthropometric data (height and weight) were obtained from each participant prior to completing the GXT. Following this, participants were given a 5 min warm-up period at a self-selected intensity, during which they were encouraged to determine a comfortable cadence that they could use for the GXT and TTE trials. For all participants, this cadence was between 75 and 90 r.p.m. The GXT used step increases of 40 W every 3 min, and commenced at a power output that the researcher estimated would bring about volitional exhaustion within 8–12 min. The GXT was terminated when participants could no longer maintain cadence within 5 r.p.m. of the target cadence, or when they reached volitional exhaustion. For the duration of the GXT, respiratory measures (O₂ uptake, CO₂ output and minute ventilation; Cortex Metalyser 3B, Cortex GmbH, Lepzig, Germany), rating of perceived exertion (RPE; 6–20 scale; Borg, 1998) and heart rate (HR; FS1; Polar, Kempele, Finland) were recorded. Rating of perceived exertion was recorded 15 s before the end of each stage. The researcher gave verbal encouragement throughout each GXT, and peak power output was defined as the highest stage power output sustained for >30 s. On completion of the GXT, participants were given a 45 min rest period (during which they span on the cycle ergometer, stretched and sat quietly) before commencing the familiarization TTE. The TTE was completed at the same cadence selected by the participant during the GXT, and at a power output which produced 70% of the participant's maximal oxygen uptake (determined during the GXT). The TTE was terminated when participants could no longer maintain cadence within 5 r.p.m. of the target cadence, or when they reached volitional exhaustion.

Visits 2 and 3 – time-to-exhaustion trials

Three to five days following the GXT, participants returned to the laboratory on two further occasions (separated by a further 3–5 days) to complete a TTE trial in hot conditions. On entry to the laboratory, participants ingested orally either 20 mg (kg lean body mass)⁻¹ of ACT or a placebo control (PLA; equivalent quantity of flour), which were housed in gelatin capsules. Postexperiment debriefing confirmed that participants could not determine the difference between the capsules.

Following ingestion, participants were given a period of 45 min, during which equipment to measure HR, T_core and skin temperature (T_skin) was attached/inserted. This time period was chosen because peak plasma concentrations of acetaminophen occur 30–60 min after ingestion (Medicines Compendium, 2008). A HR monitor was attached to the chest, and T_core was measured via a rectal thermistor (400H and 4491H, Henleys Medical Supplies, Brownfields, UK), which was inserted 10 cm past the anal sphincter. The skin temperature was measured using skin thermistors (EUS-U-VS5-0; Grant Instruments, Shepreth, UK), which were attached with medical tape on the belly of the pectoralis (T1), triceps brachii (T2), vastus lateralis (T3) and gastrocnemius (T4) of the right side of the body, as described previously (Castle et al. 2012). The skin temperature was calculated using the following formula: 0.3(T1 + T2) + 0.2(T3 + T4), as described by Ramanathan (1964). Body heat content (T_body) was calculated from the following formula: T_body× body mass × 3.47, as used by Tucker et al. (2006). A data logger was used to record T_skin (Squirrel 451; Grant Instruments), and T_core was recorded via a temperature monitor (YSI4600, Henleys Medical Supplies, Brownfields, UK). After 15 min stabilization, resting HR, T_core and T_skin temperatures were recorded.

Following the 45 min postingestion period, participants entered the temperature-controlled environmental chamber, which was set at 30°C and 50% relative humidity. Participants completed a 5 min warm-up at 100 W, before commencing the TTE trial in the same manner as the familiarization session. Heart rate, thermal sensation (TSS; visual scale ranging from ‘0, unbearably cold’ to ‘8, unbearably hot’; Toner et al. 1986), T_core and T_skin were recorded following the warm-up. Heart rate, TSS, RPE, T_core and T_skin were recorded every minute of the TTE trial. The TTE was terminated when participants could no longer maintain cadence within 5 r.p.m. of the target cadence, or when they reached volitional exhaustion.

Statistical analysis

All data are presented as means ± SD unless stated otherwise.

Statistical analysis was conducted using the Statistical Package for Social Science (SPSS) (version 19; SPSS Inc., Chicago, IL, USA). Assumptions of statistical tests (data distribution) were checked using conventional graphical methods (Grafen, 2002) and were deemed plausible in all instances. The Huynh–Feldt correction to the degrees of freedom was applied when violations to sphericity were present. For performance, psychological and physiological measures, a two-way (condition × time) repeated-measures ANOVA was used to determine the differences between the two conditions. Where significant differences were found, a priori one-tailed post hoc t test analysis with LSD corrections were used to determine where these were. Student's paired t test was used to analyse the difference in time to exhaustion between conditions. Effect sizes were calculated to demonstrate the size of the differences between the conditions and are reported as partial eta-squared (η_p²). To account for differences in completion time and subsequent varying frequency of data points, statistical analysis for physiological, psychological and performance data was performed using a combination of intra-individual iso-times and percentage of time to exhaustion. In this approach, the shortest time to exhaustion for each individual subject over their two respective trials was identified as 100% iso-time. This value was then compared against the equivalent time value in the subject's longer trial. Using this intra-subject iso-time as 100%, values for every 10% of both trials were calculated and analysed. Finally, the values at exhaustion for each trial were recorded. This approach provided 12 data points for statistical comparison for each subject for each trial (rest, 10–100%, exhaustion). The significance value was set at P < 0.05.

Results

Time to exhaustion

Student's paired t test revealed a significant difference in mean time to exhaustion between the ACT (1363 ± 891 s) and PLA conditions [1129 ± 787 s; t₁₀= 3.61, P= 0.005; 95% confidence interval (CI) = 90–379 s]. All but one participant achieved the same or longer time to completion following ACT ingestion compared with the placebo (see Fig. 1).

Figure 1

Differences in time to exhaustion between the acetaminophen (ACT) and placebo (PLA) conditions for all participants
The thick, black line represents the mean difference. All but one of the participants achieved the same or longer time to exhaustion in the ACT condition. *Significant difference (P < 0.05) between conditions.

Core temperature

No significant main effect for condition (F_1,10= 4.09, P= 0.071, η_p²= 0.29) was observed, but there was a significant main effect for time (F_1,10= 17.8, P < 0.001, η_p²= 0.64; ACT mean T_core= 37.62°C, 95% CI = 37.45–37.80°C; and PLA mean T_core= 37.74°C, 95% CI = 37.50–37.98°C). There was a significant condition × time interaction (F_11,110= 3.57, P < 0.001, η_p²= 0.26), with follow-up Student's paired t tests revealing that T_core was significantly lower following ingestion of ACT (than PLA) at 20% TTE (−0.15 ± 0.21°C, 95% CI = 0.007–0.28°C), 40% TTE (−0.14 ± 0.58°C, 95% CI = 0.015–0.273°C) and 50% TTE (−0.15 ± 0.20°C, 95% CI = 0.014–0.286°C; see Fig. 2A). The change in T_core over time for a representative subject is displayed in Fig. 3A.

Figure 2

Mean ± SD values for core temperature (T_core; A), skin temperature (T_skin; B), body temperature (T_body; C) and rating of thermal sensation (TSS; D) as a percentage of time to exhaustion across both ACT and PLA conditions
#Significant main effect (P < 0.05) for condition. *Significant interaction effect (P < 0.05) for condition × time.

Figure 3

Values from a representative subject for T_core (A), T_skin (B), T_body (C) and TSS (D) across exercise time (in minutes) for both ACT and PLA conditions
Note how the averaging method for time to exhaustion displayed in Fig. 2 closely represents the individual subject response shown here.

Skin temperature

There was a significant main effect for time (F_11,110= 23.97, P < 0.001, η_p²= 0.85) and condition (F_1,10= 14.38, P= 0.03, η_p²= 0.39; ACT mean T_skin= 33.27°C, 95% CI = 32.74–33.80°C; and PLA mean T_skin= 33.74°C, 95% CI = 33.13–34.34°C). There was a significant interaction effect for condition × time (F_11,110= 3.28, P= 0.001, η_p²= 0.25), with follow-up Student's paired t tests revealing that T_skin was significantly lower following ingestion of ACT (than PLA) at 30% TTE through to exhaustion (see Fig. 2B). The change in T_skin over time for a representative subject is displayed in Fig. 3B.

Body temperature

A significant main effect was observed for time (F_11,110= 3.43, P < 0.001, η_p²= 0.77) and condition (F_1,10= 6.99, P= 0.025, η_p²= 0.41; ACT mean T_body= 36.71°C, 95% CI = 36.50–36.92°C; and PLA mean T_body= 36.90°C, 95% CI = 36.63–37.18°C). There was a significant interaction effect for condition × time (F_11,110= 2.76, P= 0.003, η_p²= 0.22), with follow-up Student's paired t tests revealing that T_body was significantly lower following ingestion of ACT (than PLA) at 20% TTE through to exhaustion (see Fig. 2C). The change in T_body over time for a representative subject is displayed in Fig. 3C.

Thermal sensation

A significant main effect was observed for time (F_11,110= 24.48, P < 0.001, η_p²= 0.95) and condition (F_1,10= 8.53, P= 0.015, η_p²= 0.46; ACT mean TSS = 5.86, 95% CI = 5.40–6.31; and PLA mean TSS = 6.25, 95% CI = 5.82–6.69). A significant interaction effect was shown for condition × time (F_11,110= 4.04, P < 0.001, η_p²= 0.29). Follow-up Student's paired t tests revealed that TSS was significantly lower following ingestion of ACT (than PLA) at 10% TTE through to 60% TTE (see Fig. 2D). The change in TSS over time for a representative subject is displayed in Fig. 3D.

Heart rate

No significant main effects for time or condition were observed (P > 0.05), and no interaction effect for condition × time was observed (P > 0.05).

Discussion

The primary finding of this study was that an acute ingested dose of acetaminophen significantly increased the time to exhaustion during fixed-intensity exercise in hot conditions (as shown in Fig. 1). This increased exercise capacity occurred in the presence of a significantly reduced T_core, T_skin, T_body and thermal sensation, despite no increase in heart rate. To the authors’ knowledge, this is the first study to demonstrate that ACT improves cycling time to exhaustion in hot conditions, potentially by decreasing the thermoregulatory challenge during exercise in the heat.

Ingestion of ACT served to increase cycling time to exhaustion by an average of 17%, and all but one participant achieved the same or longer time to exhaustion in the ACT trial. This improvement in exercise capacity is less than that achieved with the ‘gold standard’ external precooling method of cold water immersion, where improvements in the region of 37% have been observed (Gonzalez-Alonso et al. 1999). However, when compared with the more logistically practical internal precooling methods, such as cold drink ingestion, ice slurry ingestion and cold air inhalation, improvement in exercise capacity is similar, i.e. 23% (cold drink - Lee et al. 2008), 23% (ice slurry - Siegel et al. 2010) and 13% (ice slurry - Siegel et al. 2012), and 15% (cold air - Geladas & Banister, 1988) improvement, respectively. The reduction in T_core (condition × time interaction) in the ACT trial in the present study, which occurred alongside the improved time to exhaustion, was small but significant (∼0.15°C). However, skin temperature was also reduced by ACT, and participants reported significantly reduced thermal discomfort throughout the ACT condition. Therefore, whilst the reduction in T_core and T_skin observed this study is of lower magnitude than other internal precooling strategies (Geladas & Banister, 1988; Lee et al. 2008), it appears as though collectively these differences were still perceived by the participants, and this may have contributed to the observed ergogenic effect.

It should be noted that the mean time to exhaustion between conditions was ∼21 min and that the mean end T_core was 38.08°C. Therefore, although participants perceived the thermoregulatory challenge as high (TSS at exhaustion was 7.38, i.e. ‘very hot’–‘unbearably hot’), the end T_core was lower than that usually associated with exhaustion due to thermoregulatory strain (Brotherhood, 2008). Therefore, whilst the small but significant reduction in T_core and T_skin may have contributed to an improved exercise capacity in the present study, other additional mechanisms altered by ACT should not be ruled out.

Exercise-induced pain may be a contributor to volitional exhaustion or changes in pacing during exercise (Mauger & Hopker, 2012; Mauger, 2013), and because ACT serves to increase pain threshold and tolerance (Pickering et al. 2006), this may have been a contributing factor to the increased time to exhaustion in the present study. Indeed, Mauger et al. (2010) have shown that ingestion of ACT prior to exercise can significantly improve performance in a 16.1 km self-paced cycle time trial. In that study (Mauger et al. 2010), ACT appeared to allow participants to maintain a higher power output during the middle section of the time trial, thereby improving overall completion time. Mauger & Hopker (2012) have also shown that an acute dose of ACT is capable of increasing corticospinal excitability in resting conditions. Whilst this effect has not yet been demonstrated in an exercising (or fatigued) state, as reductions in corticospinal excitability have been associated with fatigue, this mechanism may have a role to play. Measures of corticospinal excitability and perceived pain were not obtained in the present study, so future work should seek to employ these measures in order to help attribute performance change to a particular mechanism. The present study lends support to previous work (Mauger et al. 2010) and suggests that ACT is capable of eliciting a moderate ergogenic effect for both time-to-exhaustion and self-paced cycling exercise, and that this effect is present in nearly all participants tested. The existing literature suggests that this effect may be brought about by a variety of mechanisms, which potentially all contribute to the observed improvement in capacity/performance and may be more or less important depending on exercise modality, training status and environmental conditions.

Whilst a change in performance through the moderation of neural pathways and/or perceived pain cannot be ruled out in the present study, the significant decrease in T_core and T_skin demonstrates that ACT provided some physiological effect on participants’ thermoregulation. The mechanism of the antipyretic effect of ACT is still not completely clear, and thus requires carefully designed in vivo experiments to provide mechanistic cause and effect. However, a putative mechanism is that ACT penetrates the blood–brain barrier, where it serves to inhibit the synthesis of prostaglandin E₂ (PGE₂) in the anterior hypothalamus and thus subsequently reduces levels of COX (Botting & Ayoub, 2005). While non-steroidal anti-inflammatory drugs such as ibuprofen and aspirin act by inhibiting COX-1 and COX-2, and thus have a predominantly peripheral action, acetaminophen has been suggested to act on a splice variant of COX-1, the COX-3 enzyme (Chandrasekharan et al. 2002). As COX-3 is most abundantly expressed in the cerebral cortex, this partly explains the primarily central action of ACT and may be a potential mechanism for the antipyretic action of ACT (Chandrasekharan et al. 2002). However, whilst ACT may selectively inhibit COX-3, the COX enzyme group stimulates production of prostaglandins, and an increase in PGE₂ in the hypothalamus increases the body's thermoregulatory set-point (Aranoff & Nielson, 2001), which is the usual physiological response to fever. Therefore, the fact that ACT is able to reduce T_core in afebrile conditions suggests either that it has an additional antipyretic mechanism of action or that exercise (in the heat) increases levels of PGE₂ in the brain. Although the cytokine response to exercise differs from that elicited by severe infections (Petersen & Pedersen, 2005), it is well established that exercise leads to an acute increase in inflammatory markers such as interleukin-6 (Pedersen & Hoffman-Goetz, 2000), which is associated with PGE₂ production. Therefore, exercise in the heat speculatively may serve to increase the thermoregulatory set-point through an inflammatory response (Bradford et al. 2007) by inhibiting preoptic warm-sensitive neurons in the hypothalamus, which facilitate heat loss and suppress heat production; ACT may help to reduce T_core during exercise by moderating this inflammatory response. Future studies should seek to obtain markers of inflammation during exercise (such as interleukin-6, interleukin-10 and tumour necrosis factor-α) and ACT ingestion to test this hypothesis.

Interestingly, the skin temperature was slightly higher in the placebo group by 0.3°C from 10% of TTE, increasing to a difference of 0.5°C at exhaustion (Fig. 2B). Although convective heat loss to the environment is maintained at skin temperatures below 35°C (Brotherhood, 2008), differences in cycling time trial performance have been noted. Tatterson et al. (2000) found that mean power output was reduced by ∼6.5% during a 30 min cycling time trial, alongside a 6% greater skin temperature in hot versus temperate conditions. An unfortunate limitation of the present study is the absence of a comparison between pre- and postexercise body mass, to indicate non-urine fluid loss, which may have helped to account for this. Furthermore, it is plausible that the acetaminophen allowed a more favourable heat transfer from core to skin, and skin to environment, but was beyond the scope of this investigation. This would make an interesting line of investigation for future work to understand the influence of acetaminophen on thermoregulation.

Whilst this study has demonstrated an ergogenic effect for ACT, it should be noted that ingestion of analgesics prior to exercise has been associated with serious adverse effects (Van Wijck et al. 2012), albeit more so with non-steroidal anti-inflammatory drugs such as ibuprofen. As such, we do not condone the use of analgesics for ergogenic performance, because they can mask injury, cause gastrointestinal damage and, according to the present findings, may affect normal thermoregulatory functioning. However, the quick, easy and practical administration of ACT may provide a useful and effective means of moderating T_core during challenging occupational pursuits (e.g. the fire service, mining, military service) or in individuals suffering from exertional heat illness.

In conclusion, this study demonstrates that an acute dose of ACT is capable of significantly improving cycling time to exhaustion in hot conditions, alongside a significant reduction in T_core, T_skin, T_body and thermal sensation. The increased time to exhaustion supports previous work showing that ACT elicits a significant ergogenic effect for both time-to-exhaustion and self-paced cycling exercise. Future work needs to determine whether ACT-induced thermoregulatory changes occur in response to a reduction in the acute pro-inflammatory response to exercise, thereby serving to help moderate an increase in thermoregulatory set-point.

Appendix

Additional Information

Competing interests

None declared.

Funding

None declared.