The effect of caffeine ingestion on torque and muscle activity during resistance exercise men
Michael J. Duncan, PhD, Charles, D. Thake, PhD, and Philip J. Downs, MSc
Department of Biomolecular and Sports Sciences, Coventry University, Coventry, UK Address for correspondence: Michael J. Duncan, Human Performance Laboratory, Department of Biomolecular and Sports Sciences, Coventry University, James Starley Building, Priory Street, Coventry, UK, CV 5HB. E-mail: [email protected]
Running Head: caffeine ingestion on torque
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process which may lead to differences between this version and the Version of Record. Please cite this article as an ‘Accepted Article’, doi: 10.1002/mus.24179
Muscle & Nerve
Page 2 of 19
2
Abstract Introduction: We examined the effect of caffeine ingestion on muscle torque production and muscle activity at different contraction speeds in trained men. Methods: 10 men (mean age ± SD = 22 ± 1.1 years) volunteered to participate. A double-blind, randomized cross-over design was used. Sixty minutes post ingestion of caffeine (6mg kg-1) or placebo, participants completed 6 repetitions of isokientic knee extension at 3 angular velocities (30°s-1, 150°s-1, 300°s-1) from which peak torque was determined. Electromyographic activity of the vastus medialis was also collected. Results: Repeated measures ANOVA indicated that muscle torque production was significantly higher (P = 0.02) with caffeine compared to placebo. A significant (P = 0.02) substance by velocity interaction for muscle activity indicated significantly higher vastus medialis muscle activity in the presence of caffeine vs. placebo, and this difference was amplified as angular velocity increased. Conclusions Acute caffeine ingestion improves muscle performance and increases muscle activity during short-duration maximal dynamic contractions.
Keywords: Electromyography; Isokinetics; Dynamic Contractions; Ergogenic; Nutrition; Muscle
John Wiley & Sons, Inc.
Page 3 of 19
Muscle & Nerve
3
Introduction It is well established that acute caffeine ingestion enhances endurance performance in humans.1 The effect of caffeine on muscle strength performance has been less thoroughly examined, results from research in this area have been equivocal.2
Findings have
suggested direct effects on muscle excitation-contraction coupling and motor unit recruitment, which are independent from those related to metabolic efficiency.3,4 However, data regarding the effect of acute caffeine ingestion on muscle activity during strength exercises is sparse. To date, only 2 studies have examined the impact of caffeine ingestion on strength performance alongside analysis of muscle activity. One study reported increased electromyographic (EMG) activity and enhanced muscle performance of elbow flexors during short-duration maximal dynamic contractions.
5
In contrast, the results reported by Madigan
and Willems6 found no significant difference in muscle activity of the vastus lateralis and biceps femoris between caffeine (6mg · kg-1) and placebo conditions following intermittent isometric contractions (15s contraction, 5s rest) at 50% maximal voluntary isometric torque until failure. Thus, prior research on this topic is equivocal. The potential performance enhancing effects of caffeine have been explained via multiple mechanisms. These include changes in central nervous system transmission leading to blunted perception of exertion and pain.7 Caffeine also acts as an adenosine antagonist with inhibition of A1 adenosine receptors facilitating release of neurotransmitters, including dopamine and serotonin.7,8 Moreover, caffeine ingestion has been associated with increased motor unit activation during maximal voluntary activation of knee extensors.9 Other work has also suggested that caffeine ingestion may positively enhance calcium release from the sarcoplasmic reticulum, resulting in increased muscle contraction.10,11 Prior research on this topic has also not fully considered training status of the participant group examined. Astorino and Roberson7 have noted that discrepancies in training status may explain the equivocal nature of studies on the effect of caffeine on
John Wiley & Sons, Inc.
Muscle & Nerve
Page 4 of 19
4
strength performance, as strength trained individuals potentially have different responses from other habitually physically active individuals. Thus, there is a need to examine the effects of caffeine on muscle performance in this group specifically.7 Furthermore, the majority of studies on this topic have focused on maximal isometric strength,12,13 and the effect of caffeine on neuromuscular function during dynamic contractions at different speeds has only been investigated in 1 prior study and only to a speed of 250 °s-1.5 Given that only 2 prior studies appear to have investigated this area 5,6 and have produced conflicting findings, we sought to confirm the effect of caffeine ingestion on muscle torque production and muscle activity by comparing its effect at different contraction speeds in trained men.
Materials and Methods Participants Following ethics approval and informed consent, 10 strength trained men (mean age ± SD = 22 ± 1.1 years) volunteered to participate. All participants were low habitual caffeine consumers (≤ 70 mg day-1), had extensive experience in strength training (mean ± S.D. strength training experience = 9.0 ± 5.5 years), and were free of any musculoskeletal pain or disorders. Participants were semi-professional rugby league players and were currently participating in > 10 hours week of programmed physical activity, including strength and conditioning activities. Approximately 3-4 hours per week of this programmed activity included specific strength and conditioning activity involving plyometric exercise and Olympic weightlifting. Prior to experimental trials, participants were asked to complete a brief questionnaire detailing any dietary or performance-enhancing substances they were currently consuming, or had consumed in the 6 weeks prior to testing. None reported taking any caffeine based supplements or substances containing guarana or creatine. However, all reported ingestion of whey protein supplements, and 3 of the 10 participants reported
John Wiley & Sons, Inc.
Page 5 of 19
Muscle & Nerve
5
ingestion of amino acids (leucine) within 6 weeks prior to experimental testing. Participants were asked to abstain from caffeine ingestion from 6:00 PM the night prior to testing and refrain from vigorous exercise and maintain normal dietary patterns in the 48h prior to testing.
Procedures A double-blind, randomized cross-over design was used whereby participants attended the laboratory on 3 occasions. All testing took place between 9:00 AM and noon, and each condition took place at the same time for each participant to avoid circadian variation. The first visit to the laboratory involved a briefing session, familiarization, and performance of maximal voluntary isometric contraction in order to provide baseline electromyographic data by which to normalize data gathered in subsequent testing sessions. All participants had experience performing isokinetic contractions as part of prior athletic monitoring procedures. Following this session, participants completed 2 further experimental conditions which were presented in a randomized order and were separated by 72 hours. These conditions consisted of a caffeine condition where participants consumed 6mg · kg-1 of caffeine (Bayer, UK) diluted in 250ml of artificially sweetened water and a placebo condition where they consumed 250ml of artificially sweetened water drink. Solutions were consumed 60 min before each exercise trial, as plasma caffeine concentration is maximal 1 hour after ingestion.1 Solutions were presented to participants in an opaque sports bottle to prevent the researchers who administered the solutions or the participants from actually seeing the solutions themselves. In each condition, participants completed 6 repetitions of dominant knee
extension
on
an
isokinetic
dynamometer
(Cybex
Norm,
CSMi
Solutions,
Massachusetts, USA) at 3 angular velocities (30°s-1, 150°s-1, 300°s-1) with 1 minute rest between sets. Average torque (Nm) for knee extensors was assessed at each velocity.
John Wiley & Sons, Inc.
Muscle & Nerve
Page 6 of 19
6
Activation of the vastus medialis (VM) muscle was measured using surface EMG. VM muscle activity was assessed, in preference to the other heads of the quadriceps, due to its high reliability and reproducibility in assessing EMG signals.14 Passive bipolar surface electrodes of 30mm diameter (Blue Sensor, Ltd, Denmark, Ag/AgCl) were placed on the contracted belly of the muscle in line with the muscle fiber direction, with an interelectrode distance of 1.5cm. Before electrode placement, the area was cleaned with isopropyl alcohol, shaved, and abraded to reduce skin impedance. Reference electrodes were also placed on a bony prominence on the right wrist. All electrodes remained in place until data collection was completed at the 3 intensities. Efforts were made to minimize crosstalk across the muscles of interest by selecting appropriate electrode size and interelectrode distance.15 Manual muscle testing during pilot work confirmed that crosstalk in the muscles of interest was minimised. Electromyographic activity was differentiated by pre-amplifiers and recorded via a wireless ME6000 system (MEGA Electronics, LTD, Finland), with an input impedance of less than 1015/0.2 ohm/pF, a common mode rejection ratio at 60 Hz of greater than 110dB, a noise level of 1.2 µV, a gain of 10 + 2%, and a bandwidth range from 0 Hz – 500 Hz. Muscle activity was sampled at 1000 Hz via a 16bit DAQ-516 A/D card and stored on a Sony Vaio laptop computer using MegaWin software, version 1.2 (MEGA Electronics, LTD, Finland). The raw EMG data were filtered using high and low pass Butterworth filters (5-500 Hz) and visually checked for artefacts, which were excluded from subsequent analysis. EMG data were sampled at 100 ms with a 50% frame overlap. This moving average approach in which the time windows overlap ensures that the EMG curve follows the trend of the underlying rectified EMG, but without the variable peaks that are evident in the rectified EMG. This method has been recommended for use in dynamic contractions.16 On completion of the experimental protocol the Root Mean Square (RMS) of the filtered EMG signal was full wave rectified and normalized against a maximal voluntary isometric contraction (MVC) at an angle of 15° knee extension, established during habituation sessions, as this angle elicits maximal VM activation.17 All participants had prior experience
John Wiley & Sons, Inc.
Page 7 of 19
Muscle & Nerve
7
of isokinetic dynamometry and performing MVC. Intraclass correlation coefficients for VM MVC EMG data taken 2 days apart were R = .82. This is comparable to prior research examining between-day MVC-based EMG data of the VM.17
Statistical Analysis Prior to analysis all data were checked for normality using the Shapiro-Wilk test to verify that they were normal (P >.05). Any differences in muscle torque and normalized EMG were analyzed using a series of 2 (substance ingested) X 3 (angular velocity) way, repeated measures ANOVAs. Where any differences were found, Bonferroni post-hoc comparisons were used to determine where the differences lay. Data were analyzed using Predictive Analytics Software Statistics version 18 (PASW Statistics 18). Statistical significance was set at an alpha level of P
Michael J. Duncan, PhD, Charles, D. Thake, PhD, and Philip J. Downs, MSc
Department of Biomolecular and Sports Sciences, Coventry University, Coventry, UK Address for correspondence: Michael J. Duncan, Human Performance Laboratory, Department of Biomolecular and Sports Sciences, Coventry University, James Starley Building, Priory Street, Coventry, UK, CV 5HB. E-mail: [email protected]
Running Head: caffeine ingestion on torque
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process which may lead to differences between this version and the Version of Record. Please cite this article as an ‘Accepted Article’, doi: 10.1002/mus.24179
Muscle & Nerve
Page 2 of 19
2
Abstract Introduction: We examined the effect of caffeine ingestion on muscle torque production and muscle activity at different contraction speeds in trained men. Methods: 10 men (mean age ± SD = 22 ± 1.1 years) volunteered to participate. A double-blind, randomized cross-over design was used. Sixty minutes post ingestion of caffeine (6mg kg-1) or placebo, participants completed 6 repetitions of isokientic knee extension at 3 angular velocities (30°s-1, 150°s-1, 300°s-1) from which peak torque was determined. Electromyographic activity of the vastus medialis was also collected. Results: Repeated measures ANOVA indicated that muscle torque production was significantly higher (P = 0.02) with caffeine compared to placebo. A significant (P = 0.02) substance by velocity interaction for muscle activity indicated significantly higher vastus medialis muscle activity in the presence of caffeine vs. placebo, and this difference was amplified as angular velocity increased. Conclusions Acute caffeine ingestion improves muscle performance and increases muscle activity during short-duration maximal dynamic contractions.
Keywords: Electromyography; Isokinetics; Dynamic Contractions; Ergogenic; Nutrition; Muscle
John Wiley & Sons, Inc.
Page 3 of 19
Muscle & Nerve
3
Introduction It is well established that acute caffeine ingestion enhances endurance performance in humans.1 The effect of caffeine on muscle strength performance has been less thoroughly examined, results from research in this area have been equivocal.2
Findings have
suggested direct effects on muscle excitation-contraction coupling and motor unit recruitment, which are independent from those related to metabolic efficiency.3,4 However, data regarding the effect of acute caffeine ingestion on muscle activity during strength exercises is sparse. To date, only 2 studies have examined the impact of caffeine ingestion on strength performance alongside analysis of muscle activity. One study reported increased electromyographic (EMG) activity and enhanced muscle performance of elbow flexors during short-duration maximal dynamic contractions.
5
In contrast, the results reported by Madigan
and Willems6 found no significant difference in muscle activity of the vastus lateralis and biceps femoris between caffeine (6mg · kg-1) and placebo conditions following intermittent isometric contractions (15s contraction, 5s rest) at 50% maximal voluntary isometric torque until failure. Thus, prior research on this topic is equivocal. The potential performance enhancing effects of caffeine have been explained via multiple mechanisms. These include changes in central nervous system transmission leading to blunted perception of exertion and pain.7 Caffeine also acts as an adenosine antagonist with inhibition of A1 adenosine receptors facilitating release of neurotransmitters, including dopamine and serotonin.7,8 Moreover, caffeine ingestion has been associated with increased motor unit activation during maximal voluntary activation of knee extensors.9 Other work has also suggested that caffeine ingestion may positively enhance calcium release from the sarcoplasmic reticulum, resulting in increased muscle contraction.10,11 Prior research on this topic has also not fully considered training status of the participant group examined. Astorino and Roberson7 have noted that discrepancies in training status may explain the equivocal nature of studies on the effect of caffeine on
John Wiley & Sons, Inc.
Muscle & Nerve
Page 4 of 19
4
strength performance, as strength trained individuals potentially have different responses from other habitually physically active individuals. Thus, there is a need to examine the effects of caffeine on muscle performance in this group specifically.7 Furthermore, the majority of studies on this topic have focused on maximal isometric strength,12,13 and the effect of caffeine on neuromuscular function during dynamic contractions at different speeds has only been investigated in 1 prior study and only to a speed of 250 °s-1.5 Given that only 2 prior studies appear to have investigated this area 5,6 and have produced conflicting findings, we sought to confirm the effect of caffeine ingestion on muscle torque production and muscle activity by comparing its effect at different contraction speeds in trained men.
Materials and Methods Participants Following ethics approval and informed consent, 10 strength trained men (mean age ± SD = 22 ± 1.1 years) volunteered to participate. All participants were low habitual caffeine consumers (≤ 70 mg day-1), had extensive experience in strength training (mean ± S.D. strength training experience = 9.0 ± 5.5 years), and were free of any musculoskeletal pain or disorders. Participants were semi-professional rugby league players and were currently participating in > 10 hours week of programmed physical activity, including strength and conditioning activities. Approximately 3-4 hours per week of this programmed activity included specific strength and conditioning activity involving plyometric exercise and Olympic weightlifting. Prior to experimental trials, participants were asked to complete a brief questionnaire detailing any dietary or performance-enhancing substances they were currently consuming, or had consumed in the 6 weeks prior to testing. None reported taking any caffeine based supplements or substances containing guarana or creatine. However, all reported ingestion of whey protein supplements, and 3 of the 10 participants reported
John Wiley & Sons, Inc.
Page 5 of 19
Muscle & Nerve
5
ingestion of amino acids (leucine) within 6 weeks prior to experimental testing. Participants were asked to abstain from caffeine ingestion from 6:00 PM the night prior to testing and refrain from vigorous exercise and maintain normal dietary patterns in the 48h prior to testing.
Procedures A double-blind, randomized cross-over design was used whereby participants attended the laboratory on 3 occasions. All testing took place between 9:00 AM and noon, and each condition took place at the same time for each participant to avoid circadian variation. The first visit to the laboratory involved a briefing session, familiarization, and performance of maximal voluntary isometric contraction in order to provide baseline electromyographic data by which to normalize data gathered in subsequent testing sessions. All participants had experience performing isokinetic contractions as part of prior athletic monitoring procedures. Following this session, participants completed 2 further experimental conditions which were presented in a randomized order and were separated by 72 hours. These conditions consisted of a caffeine condition where participants consumed 6mg · kg-1 of caffeine (Bayer, UK) diluted in 250ml of artificially sweetened water and a placebo condition where they consumed 250ml of artificially sweetened water drink. Solutions were consumed 60 min before each exercise trial, as plasma caffeine concentration is maximal 1 hour after ingestion.1 Solutions were presented to participants in an opaque sports bottle to prevent the researchers who administered the solutions or the participants from actually seeing the solutions themselves. In each condition, participants completed 6 repetitions of dominant knee
extension
on
an
isokinetic
dynamometer
(Cybex
Norm,
CSMi
Solutions,
Massachusetts, USA) at 3 angular velocities (30°s-1, 150°s-1, 300°s-1) with 1 minute rest between sets. Average torque (Nm) for knee extensors was assessed at each velocity.
John Wiley & Sons, Inc.
Muscle & Nerve
Page 6 of 19
6
Activation of the vastus medialis (VM) muscle was measured using surface EMG. VM muscle activity was assessed, in preference to the other heads of the quadriceps, due to its high reliability and reproducibility in assessing EMG signals.14 Passive bipolar surface electrodes of 30mm diameter (Blue Sensor, Ltd, Denmark, Ag/AgCl) were placed on the contracted belly of the muscle in line with the muscle fiber direction, with an interelectrode distance of 1.5cm. Before electrode placement, the area was cleaned with isopropyl alcohol, shaved, and abraded to reduce skin impedance. Reference electrodes were also placed on a bony prominence on the right wrist. All electrodes remained in place until data collection was completed at the 3 intensities. Efforts were made to minimize crosstalk across the muscles of interest by selecting appropriate electrode size and interelectrode distance.15 Manual muscle testing during pilot work confirmed that crosstalk in the muscles of interest was minimised. Electromyographic activity was differentiated by pre-amplifiers and recorded via a wireless ME6000 system (MEGA Electronics, LTD, Finland), with an input impedance of less than 1015/0.2 ohm/pF, a common mode rejection ratio at 60 Hz of greater than 110dB, a noise level of 1.2 µV, a gain of 10 + 2%, and a bandwidth range from 0 Hz – 500 Hz. Muscle activity was sampled at 1000 Hz via a 16bit DAQ-516 A/D card and stored on a Sony Vaio laptop computer using MegaWin software, version 1.2 (MEGA Electronics, LTD, Finland). The raw EMG data were filtered using high and low pass Butterworth filters (5-500 Hz) and visually checked for artefacts, which were excluded from subsequent analysis. EMG data were sampled at 100 ms with a 50% frame overlap. This moving average approach in which the time windows overlap ensures that the EMG curve follows the trend of the underlying rectified EMG, but without the variable peaks that are evident in the rectified EMG. This method has been recommended for use in dynamic contractions.16 On completion of the experimental protocol the Root Mean Square (RMS) of the filtered EMG signal was full wave rectified and normalized against a maximal voluntary isometric contraction (MVC) at an angle of 15° knee extension, established during habituation sessions, as this angle elicits maximal VM activation.17 All participants had prior experience
John Wiley & Sons, Inc.
Page 7 of 19
Muscle & Nerve
7
of isokinetic dynamometry and performing MVC. Intraclass correlation coefficients for VM MVC EMG data taken 2 days apart were R = .82. This is comparable to prior research examining between-day MVC-based EMG data of the VM.17
Statistical Analysis Prior to analysis all data were checked for normality using the Shapiro-Wilk test to verify that they were normal (P >.05). Any differences in muscle torque and normalized EMG were analyzed using a series of 2 (substance ingested) X 3 (angular velocity) way, repeated measures ANOVAs. Where any differences were found, Bonferroni post-hoc comparisons were used to determine where the differences lay. Data were analyzed using Predictive Analytics Software Statistics version 18 (PASW Statistics 18). Statistical significance was set at an alpha level of P
