Introduction

Muscle fatigue has many definitions, but in general it is associated with “an exercise-induced reduction in the ability of the muscle to produce force or power, whether or not the task can be sustained” ¹.^.When exercise is performed at a constant workload and high intensity until exhaustion, additional and progressive muscle fibre recruitment is required to compensate for the force loss caused by muscle fatigue. This increase in muscle fibre recruitment is manifested by an increase in electromyographic activity _(EMGslope) ^2,3. Hence, a lower _EMGslope during a constant and exhaustive power output represents a lower rate of neuromuscular fatigue and higher muscle efficiency. This simple analysis can be used to prescribe exercises and monitor training effects throughout a sports season.

The neuromuscular fatigue threshold (NFT) has been proposed to predict aerobic capacity using surface electromyography (EMG) analysis ^2,3. NFT calculates the _EMGslope of exhaustive constant-load tests and predict, by linear regression, in theory, the maximum workload intensity performed without evidence of neuromuscular fatigue ^2,3. Despite the fact that controversial results have been presented about the validity of the NFT as an aerobic index ^3,4, Mäestu et al. ⁵ showed a strong relationship of NFT with performance and suggested that NFT may be related to “local fatigue accumulation in the muscle”. On other hand, the critical power (CP) using similar protocol (i.e. constant-load tests till exhaustion) provides a valid estimation of the maximal lactate and oxygen consumption steady state ⁶, and has been shown to underestimate power outputs when compared to NFT, despite significant correlation ^2,3. The CP modelling also estimates the anaerobic work capacity (AWC) which is associated with the total amount of anaerobic energy stored and which indicates a finite amount of work that can be performed above CP ⁷. In addition, NFT and CP have been shown to be interesting methods to test the effects of ergogenic substances (e.g. creatine and beta-alanine) on exercise performance ^8,9,10.

In order to delay the fatigue process and consequently improve performance, caffeine (trimethylxanthine) has been extensively used before performing aerobic and anaerobic exercises ^11,12. Since caffeine is metabolised in several human tissues (e.g. brain, heart, kidney, muscles) ¹³, multiple mechanisms have been proposed to explain the effects on the central and peripheral systems ^14,15. Goldstein et al. ¹⁵ indicated that caffeine acts centrally as an adenosine antagonist and, peripherally, it would influence the substrate metabolism and neuromuscular function. Although caffeine ingestion has provided more consistent results of its ergogenic effects on long-term exercises ^16,17,18, controversial data has been presented showing its influence on short duration trials (anaerobic) ^{14,19,20,21,22}.^. During anaerobic exercises (<3min) evaluated by the maximal accumulated oxygen deficit - MAOD, caffeine ingestion effects have been shown to improve performance ^23,24, but not during high-intensity, short-term exercises (~30sec, i.e. Wingate test) ^22,25.^. However, most of these studies tested the caffeine effects on time-trials and all-out exercises, in which its influence on the additional muscle recruitment during high-intensity and constant workload (e.g. index of muscle fatigue) has not been shown.

Thus the present study examined the effects of acute caffeine ingestion on the _EMGslope and time to exhaustion during high-intensity and constant workload, as well as the estimation of NFT, CP and AWC. It was hypothesised that acute ingestion of caffeine would prolong time to exhaustion and attenuate the _EMGslope during the predictive trials, consequently improving CP, AWC and NFT.

Methods

Subjects

Eight healthy and physically active men (25.7 ± 3.4 years; 82.0 ± 9.1kg; 180.2 ± 5.4cm; 2x per week practitioner of recreational sports: volleyball, soccer and cycling) volunteered to participate in this study. All subjects were interviewed in order to exclude any user of anabolic steroids or any other type of supplementation. They were instructed to refrain from vigorous physical activities and alcoholic and/or caffeinated substances during the experimental period. The present investigation was approved by the local Institutional Research Ethics Committee and participants were fully informed about experimental procedures and risks before signing an informed consent form.

Study design

The study was conducted over a 6-week period, in which each volunteer reported to the laboratory on 9 occasions. All participants were completely familiar with the equipment and tests, since they had already participated in previous studies with similar design. After establishing the maximal work load by an incremental test _(WMAX), four exhaustive constant-load tests were performed in order to estimate the NFT and CP with each supplement (i.e. caffeine or placebo).

Predictive tests

Initially, a maximal incremental test was completed (20W.min^-1, ~60rpm) for peak power output assessment _(WMAX). Then, four intense constant-load bouts at 80, 90, 100 and 110% of _WMAX (Trial80%, Trial90%, Trial100% and Trial_110%, respectively) were performed using caffeine (CAF) and placebo (PLA). Testing order was randomised by condition (i.e. caffeine or placebo) and intensity, and a double-blind and placebo controlled method was adopted. All tests were completed on an electronically-braked cycle ergometer (Corival-400, Quinton Instruments, Netherlands) until exhaustion with 72h intervals between sessions. Exhaustion was defined as the incapacity to sustain the stipulated cadence for more than 5s despite strong verbal encouragement. Time to exhaustion was recorded to the nearest second. Continuous EMG data were recorded from superficial quadriceps muscle (QF) (rectus femoris - RF, vastus medialis - VM and vastus lateralis - VL) throughout all tests.

Before all predictive trials, participants warmed up for three minutes at 50W self-chosen cadence. During the tests, participants were instructed to maintain a constant pedal cadence of 60rpm. No feedback on power output or elapsed time was offered. Seat- and handlebar height were recorded and reproduced for all subsequent tests. Temperature and relative air humidity during all tests were maintained at between 21-24^oC and 40-60%, respectively. The participants were tested approximately at the same time of the day to avoid circadian effects.

Determination of the NFT and CP

The NFT was estimated according to the mathematical model proposed by deVries et al. ², in which each EMG activity for 5s integrated periods was plotted as a function of time during the four constant-load trials (Figure 1). The rates of increase of EMG _(EMGslope) from the QF muscles ([RF + VM + VL] / 3) during each predictive trial were averaged and plotted against its respective power output. The NFT was considered zero _EMGslope (Figure 2). The CP was established by fitting individual time to exhaustion from each predictive trial to the equation (non-linear model) below, where AWC is considered the anaerobic work capacity ⁷.

TE = AWC / (Power - CP)

Figure 1: Increase of EMG activity against time during the trials (80, 90, 100 and 110%). Illustrative data

Figure 2: Determination of the neuromuscular fatigue threshold (NFT) obtained via linear regression between the electromyographic signal (EMG), rate of increase, and power output. Illustrative data

EMG instrumentation and procedures

The superficial QF muscles (RF, VM and VL) from the dominant leg of the participants were recorded by EMG procedures using bipolar surface active electrodes (20mm centre-to-centre, model TSD 150TM, Biopac Systems^®, CA, USA). Prior to each test the skin was shaved and cleaned with alcohol to minimise impedance. Ink marks were made on the leg to maintain the position of electrodes between sessions following the SENIAM procedures ²⁶. The EMG activity was registered by an EMG digital amplifier (model MP150, Biopac Systems^®, CA, USA), with a 2000Hz sampling rate. The raw signals were band-passed filtered (Butterworth filter) at 20-500Hz, and integrated in 5s periods by root­-mean-square (RMS). The reference electrode was positioned over the iliac crest. The limits on signal acquisition were established at ± 5V, and the common mode rejection rate was 95dB. The AcqKnowledge 3.8.1 software was used to capture and process the signal (Biopac Systems^®, CA, USA).

Caffeine ingestion

Pure CAF (6mg.kg^-1) or PLA (pharmaceutical talc) were prepared and wrapped in gelatinous capsules and ingested 60 minutes before all tests. A randomised double-blind method was adopted. After ingestion the participants remained resting to allow absorption until the beginning of each test.

Statistical analysis

All statistical analyses were processed using the STATISTICA 6.0^TM computer package (Statsoft^®, OK, USA). Initially, descriptive statistics were applied, and data were reported as mean and standard deviation. To examine data normality the Shapiro Wilk test was used. Then the one-way analysis of variance (ANOVA), followed by the Scheffé's post hoc test, was applied to compare the workloads performed during the constant-load trials. The Student’s t-test was applied to compare the time to exhaustion, CP, AWC, _EMGslope and NFT values estimated for the CAF and PLA conditions. The significance level adopted for all analysis was P<0.05.

Results

The _WMAX reached by the subjects in the maximal incremental test was 265.7 ± 35.3W, which was used as the reference for the determination of the four different constant-load bout intensities: Trial80%, Trial90%, Trial_100% and Trial_110% (213.0 ± 26.8, 241.0 ± 31.0, 265.7 ± 35.3 and 300.3 ± 34.1W, respectively) (P<0.05).

Table 1 depicts the time to exhaustion and _EMGslope during Trial80%, Trial90%, Trial_100% and Trial110% under both supplement conditions. During the predictive trials, the CAF condition improved time to exhaustion for Trial100% and Trial110% (P<0.05); however, no significant difference for Trial80% and Trial90% was observed (P>0.05). No differences were found for the _EMGslope between all conditions for any workload (P>0.05).

Table 1: Time to exhaustion (TE) and rate of increase of RMS _(EMGslope) under caffeine (CAF) and placebo (PLA) conditions during the four predictive trials. Values are means ± SD.

Significant difference between CAF and PLA conditions (P<0.05)

The estimated values for the NFT and CP and their respective R², as well as AWC from QF during CAF and PLA conditions are shown in Table 2. The AWC for CAF was significantly higher than the PLA condition (P<0.05).

Table 2: Critical power (CP), anaerobic work capacity (AWC) and neuromuscular fatigue threshold (NFT) under caffeine (CAF) and placebo (PLA) conditions. Values are means ± SD.

* Significant difference in relation to PLA conditions (P<0.05)

Discussion

The main findings of the present study indicate that acute CAF ingestion did not alter the _EMGslope from the QF during predictive trials and consequently, the NFT estimation. Furthermore, even with the prolonged time to exhaustion on Trial_100% and Trial110% with greater anaerobic contributions, CP also did not change, but anaerobic capacity improved, as shown by the higher AWC during the CAF condition. These results contradict this paper’s original hypothesis that CAF ingestion would promote improvement in time to exhaustion for all four intensities performed, since the majority of the studies have shown CAF as an ergogenic substance to augment endurance ^16,17,18 and short-term exercises ^12,19,21, despite controversial results also being presented ^14,22,25.

Multiple mechanisms have been proposed to explain the effects of caffeine ingestion on exercise performance. Lynge and Hellsten ²⁷ showed that adenosine receptors are present at the peripheral level, especially those of the _A2A type in type I fibres. Since caffeine acts as an adenosine receptor antagonist, it could act on the peripheral receptors to increase time to exhaustion ²⁸. Centrally, caffeine also acts as an antagonist of the inhibitory effects of adenosine over the action of some excitatory neurotransmitters ^29,30. Therefore it may act to optimise the recruitment of motor units and attenuate muscle fatigue by increasing the production of maximal force, at least in isometric contractions ^31,32.^. A recent review from Astorino and Roberson ¹⁴ discussed a possible mechanism of caffeine effects in high-intensity, short-term exercise, including caffeine action in the brain (central mechanisms), which might contribute to an increase in motor unit recruitment. However, the enhancement of time to exhaustion during the more intense trials (Trial_100% and Trial_110%) during the CAF condition, with no changes in the _EMGslope, implies that the same rate of muscle activity was observed during both the CAF and PLA conditions. However, a decrease in the neural activation threshold in response to the CAF effects ³³ could have changed the absolute muscle activation, while maintaining the same rate of increase of additional muscle fibre recruitment. Unfortunately, this consideration is only speculative, since the results in this paper cannot answer this question. Further studies are required to advance this field. Another explanation suggests that CAF acts as an adenosine antagonist for antinociceptive actions ^34,35, which might decrease the perception of effort and pain during high-intensity exercise, and thus improve performance ³⁶. This mechanism of action suggests that exercise tolerance would be increased and muscle fibres would be recruited at the same rate during the prolonged period of exercise.

In addition, after the CAF ingestion, dopamine is less inhibited during exercise, stimulating the central nervous system ³⁷. This action can alter the blood flow, the modulation of the neural excitability and the synaptic transmission in the brain ^38,39.^. Meeusen et al. ⁴⁰ also suggested that the decrease in the relationship between serotonin/dopamine may promote enhancement in performance through the maintenance of motivation. Despite the possible action of the different mechanisms presented here, Jones⁴¹ suggests that the several effects of CAF on the central and peripheral structures can be expressed in an integrated way during exercise, making it difficult to identify the contributions of a single and isolated mechanism.

Moreover, similar findings from the data in this paper have been reported for cycling ^19,24 and running ²³ respectively during which significant improvement in short-term exercise performance and in maximal accumulated oxygen deficit (i.e. anaerobic capacity) was found after the ingestion of CAF. However, contradictory results have also been presented by other groups, showing no effects on supramaximal cycling performance (i.e. the Wingate test) and EMG responses ^22,25.^. Recently, a review from Astorino and Roberson ¹⁴ has suggested that large variations in individual responsiveness to CAF and/or lower doses of ingestion might explain the controversial results in the literature concerning CAF effects on exercise performance.

Contradictory physiological explanations in the literature have been presented for NFT as an index of aerobic capacity estimated by a local parameter (muscle recruitment) ^2,3,4,5.^. On the other hand, CP has been widely accepted as a valid and reliable predictor of endurance and whole body exercise tolerance ^6,7.  However, only AWC estimated by CP modelling could detect the improvements in short-term exercise with CAF ingestion, as shown by the data in this paper. Regarding practical applications, the CP method might have an advantage over the NFT test, since the equipment, time and costs needed are much less using just time to exhaustion for its calculation. However, both NFT and CP have been used in experimental models to test ergogenic resources (e.g. creatine and beta-alanine) effects on performance ^8,9,10.

Conclusion

To the best of these authors’ knowledge, the present study is the first to examine the effects of CAF ingestion on the calculation of NFT and CP. These results showed that the estimation of aerobic capacity performed by NFT nor the CP were not sufficiently sensitive to identify the performance enhancements of the more intense predictive trials. On the other hand, the significant increase in AWC indicates the positive effects of CAF on anaerobic capacity.

Address for correspondence:

Mr Eduardo Bodnariuc Fontes, Faculdade de Educação Física, Universidade Estadual de Campinas (UNICAMP), Av. Érico Veríssimo 701, Cidade Universitária Zeferino Vaz, Barão Geraldo, Campinas, SP, Brazil.

Tel.: +55 (19) 3521-6648

Email: eduardobfontes@gmail.com

References

1. Enoka RM, Stuart DG. Neurobiology of muscle fatigue. J Appl Physiol 1992; 72: 1631-1648.
2. de Vries HA, Moritani T, Nagata A, et al. The relation between critical power and neuromuscular fatigue as estimated from electromyographyc data. Ergonomics 1982; 25: 783-791.
3. Moritani T, Takaishi T, Matsumoto T. Determination of maximal power output at neuromuscular fatigue threshold. J Appl Physiol 1993; 74: 1729-1734.
4. Pavlat DJ, Housh TJ, Johnson GO, et al. Electromyographic responses at the neuromuscular fatigue threshold. J Sports Med Phys Fitness 1995; 35: 31-37.
5. Mäestu J, Cicchella A, Purge P, et al. Electromyographic and neuromuscular fatigue thresholds as concepts of fatigue. J Strength Cond Res 2006; 20: 824-836.
6. Hill DW, Smith JC. Determination of critical power by pulmonary gas exchange. Can J Appl Physiol 1999; 24: 74-86.
7. Hill DW. The critical power concept: A review. Sports Med 1993; 16: 237-254.
8. Walter AA, Smith AE, Herda TJ, et al. Effects of creatine loading on electromyographic fatigue threshold in cycle ergometry in college-age men. Int J Sport Nutr Exerc Metab 2008; 18: 142-151.
9. Kendall KL, Smith AE, Graef JL, et al. Effects of four weeks of high-intensity interval training and creatine supplementation on critical power and anaerobic working capacity in college-aged men. J Strength Cond Res 2009; 23: 1663-1669.
10. Smith AE, Moon JR, Kendall KL, et al. The effects of beta-alanine supplementation and high-intensity interval training on neuromuscular fatigue and muscle function. Eur J Appl Physiol 2009 Feb; 105(3):357-363.
11. Doherty M, Smith PM. Effects of caffeine ingestion on exercise testing: A meta-analysis. Int J Sport Nutr Exerc Metab 2004; 14: 626-646.
12. Davis JK, Green JM. Caffeine and anaerobic performance: Ergogenic value and mechanisms of action. Sports Med 2009; 39(10):813-832.
13. Sinclair CJD, Geiger JD. Caffeine use in sports: A pharmacological review. J Sports Med Phys Fitness 2000; 40: 71-79.
14. Astorino TA, Roberson DW. Efficacy of acute caffeine ingestion for short-term high-intensity exercise performance: A systematic review. J Strength Cond Res 2010, 24: 257-265.
15. Goldstein ER, Ziegenfuss T, Kalman D, et al. International Society of Sports Nutrition Position Stand: Caffeine and performance. J Int Soc Sports Nutr. 2010 Jan 27; 7(1):5.
16. Ivy JL, Kammer L, Ding Z, et al. Improved cycling time-trial performance after ingestion of a caffeine energy drink. Int J Sport Nutr Exerc Metab. 2009 Feb; 19(1):61-78.
17. Plaskett CJ, Cafarelli E. Caffeine increases endurance and attenuates force sensation during submaximal isometric contractions. J Appl Physiol 2001; 91: 1535-1544.
18. Burke, LM. Caffeine and sports performance. Appl Physiol Nutr Metab 2008; 33: 1319–1334.
19. Bell DG, Jacobs I, Ellerington K. Effect of caffeine and ephedrine ingestion on anaerobic exercise performance. Med Sci Sports Exerc 2001; 33: 1399-1403.
20. Altimari LR. Caffeine and performance in anaerobic exercise. Brazilian Journal of Pharmaceutical Sciences 2006; 42: 17-27.
21. Beck TW, Housh TJ, Schmidt RJ, et al. The acute effects of a caffeine-containing supplement on strength, muscular endurance, and anaerobic capabilities. J Strength Cond Res 2006; 20: 506-510.
22. Greer F, Morales J, Coles M. Wingate performance and surface EMG frequency variables are not affected by caffeine ingestion. Appl Physiol Nutr Metab 2006; 31: 597–603.
23. Doherty M. The effects of caffeine on the maximal accumulated oxygen deficit and short-term running performance. Int J Sports Nutr 1998; 8: 95-104.
24. Simmonds MJ, Minahan CL, Sabapathy S. Caffeine improves supramaximal cycling but not the rate of anaerobic energy release. Eur J Appl Physiol. 2010 May; 109(2):287-295.
25. Pereira LA; Curti JO; Camata TV, et al. Caffeine does not change the anaerobic performance and rate of muscle fatigue in young men and women. Medicina Sportiva 2010: 14, 67- 72.
26. Hermens HJ, Freriks B, Disselhorst-Klug C, et al. Development of recommendations for SEMG sensors and sensor placement procedures. J Electromyogr Kinesiol 2000; 10: 361-374.
27. Lynge J, Hellsten Y. Distribution of adenosine A1, A2A and A2B receptors in human skeletal muscle. Acta Physiol Scand 2000; 169: 283-290.
28. Mohr T, Van Soeren M, Graham TE, et al. Caffeine ingestion and metabolic responses of tetraplegic humans during electrical cycling. J Appl Physiol 1998; 85: 979-985.
29. Davis JM, Zhao Z, Stock HS, et al. Central nervous system effect of caffeine and adenosine on fatigue. Am J Physiol Regul Integr Comp Physiol 2003; 284: 399-404.
30. Fredholm BB, Bättig K, Holmén J, et al. Actions of caffeine in the brain with special reference to factors that contribute to its widespread use. Pharmacol Rev 1999; 51: 83-133.
31. Pinto S, Tarnopolsky M. Neuromuscular effects of caffeine in males and females. Can J Appl Physiol 1997; 22: S48.
32. Kalmar JM, Cafarelli E. Effects of caffeine on neuromuscular function. J Appl Physiol 1999; 87: 801-808.
33. Spriet LL, Howlett RA. Caffeine. In: Maughan RJ (Ed.). Nutrition in Sport. Oxford: Blackwell Scientific, 2000: 379–392.
34. Ward N, Whitney C, Dunner D. The analgesic effects of caffeine in headache. Pain 1991; 44: 151-155.
35. Migliardi JR, Armellino JJ, Friedman, et al. Caffeine as an analgesic adjuvant in tension headache. Clinical Pharmacology and Therapeutics 1994; 56: 576-586.
36. Doherty M, Smith PM. Effects of caffeine ingestion on rating of perceived exertion during and after exercise: A meta-analysis. Scand J Med Sci Sports 2005; 15: 69-78.
37. Kalmar JM, Cafarelli E. Caffeine: A valuable tool to study central fatigue in humans? Exerc Sport Sci Rev 2004; 32: 143-147.
38. Dunwiddie TV, Fredholm BB. Adenosine A1 receptors inhibit adenylate cyclise activity and neurotransmitter release and hyperpolarize pyramidal neurons in rat hippocampus. J Pharmacol Exp Ther 1989; 249: 31-37.
39. Latini S, Pedata F. Adenosine in the central nervous system: Release mechanisms and extracellular concentrations. J Neurochem 2001; 79: 463-484.
40. Meeusen R, Watson P, Hasegawa H, et al. Central fatigue: the serotonin hypothesis and beyond. Sports Med 2006; 36: 881-909.
41. Jones G. Caffeine and other sympathomimetic stimulants: Modes of action and effects on sports performance. Essays Biochem 2008; 44: 109-123.