Validity of using perceived exertion to assess muscle fatigue during bench press exercise

Abstract

BACKGROUND:

Muscle fatigue is nearly unavoidable during resistance exercise, yet evaluating it in such circumstances can be challenging.

OBJECTIVE:

This study aimed to evaluate the validity of using the rating of perceived exertion (RPE) as a measure of muscle fatigue during non-explosive bench press (BP) exercise.

METHODS:

Fifteen male collegiate athletes participated in three BP tasks set at 65% of their one-repetition maximum. The RPE, spectral fatigue index (SFI), and velocity loss were measured across different experimental conditions.

RESULTS:

Significant effects were observed across different experimental conditions for the overall RPE, average velocity loss, and average SFI (all $p<$ 0.001). As the lifting tasks progressed, there were significant increases in the RPE, velocity loss, and SFI ( $p<$ 0.001). Additionally, significant differences were observed between the experimental conditions in the RPE ( $p<$ 0.001), SFI ( $p<$ 0.001), and velocity loss ( $p<$ 0.01). A significantly stronger ( $p<$ 0.05) correlation was observed between the RPE and SFI ( $r=$ 0.68, df $=$ 117, $p<$ 0.001) than between the velocity loss and SFI ( $r=$ 0.51, df $=$ 117, $p<$ 0.001).

CONCLUSIONS:

The corresponding changes observed in the RPE, velocity loss, and SFI suggest that both the RPE and velocity loss can be used as indicators of muscle fatigue during non-explosive BP exercise. However, due to the strong correlation between the RPE and SFI, RPE is more effective for reflecting muscle fatigue in non-explosive resistance exercise settings. Regarding ease of use, the RPE is more suitable than velocity loss for assessing muscle fatigue in training scenarios.

Keywords

Perceived exertion neuromuscular fatigue Borg scale surface electromyography resistance training

1. Background

Resistance exercise is effective in increasing muscle strength and power [1]. Muscle fatigue is believed to be an almost inevitable outcome during resistance exercises, which can negatively impact athletic performance, including peak power and velocity, due to a decrease in the force and/or power-generating capacity of the muscles involved [2]. In addition, muscle fatigue is associated with acute injury risks and may even lead to chronic soreness [3, 4]. Accordingly, it is very important for practitioners, such as physical therapists and strength coaches, to have a clear understanding of the muscle fatigue levels of the individuals with whom they work. Muscle fatigue is linked to numerous physiological responses and can be detected through various physiological measures, such as changes in muscle force output and blood lactate levels [2, 5]. For instance, previous studies have demonstrated that changes in blood lactate levels can serve as reliable and important indirect indicators of muscle fatigue during resistance exercises [2, 6]. However, many of these measures require invasive techniques, such as fingertip puncture, which can cause discomfort to participants. Non-invasive measures of muscle fatigue, such as those using special devices, such as isokinetic dynamometers (e.g., the Biodex dynamometer), are often required for this purpose. Therefore, many conventional measures of muscle fatigue may not be suitable for use in resistance exercise settings.

Surface electromyography (sEMG) is a widely used non-invasive technique for assessing neuromuscular function in various fields, including neuroscience, sports science, and clinical settings [7, 8]. Electromyographic signals obtained during muscle contractions can be used to detect responses in neuromuscular function induced by muscle fatigue [7, 9]. Over the past half century, numerous studies have utilized sEMG to measure muscle fatigue, and it has been demonstrated that sEMG provides reliable information on the underlying mechanisms of muscle fatigue [10]. During muscle fatigue, the power spectrum of sEMG signals tends to compress toward lower frequencies, which is thought to reflect changes in the muscle fiber conduction velocity, recruitment, and synchronization of motor units [11]. Specifically, the compression of the power spectrum causes a decrease in the frequency of the sEMG signal [7, 12]. Thus, decreases in the frequencies of the power spectrum are conventionally thought to be caused by the occurrence of muscle fatigue, and the mean and median frequencies are usually used as muscle fatigue estimators [13]. However, there are several limitations and barriers to assessing muscle fatigue using sEMG during resistance exercises, such as issues related to equipment and data reliability [7, 13]. For instance, sEMG-based muscle fatigue assessment during dynamic contractions requires considerable technical expertise and caution, and even mathematical simulations, which are almost impossible to perform in resistance exercise scenarios [7, 12, 13]. Therefore, a more accessible muscle fatigue assessment is needed in this context.

Recently, with the increased popularity of velocity-based training, velocity loss has been proposed as a reliable indicator of muscle fatigue in resistance exercises [2, 14]. Because of the widespread use of velocity measures, velocity-based muscle fatigue assessments are becoming easier to implement during resistance exercise [15, 16]. Moreover, studies have shown that velocity loss is significantly correlated with metabolic and mechanical measures of muscle fatigue during explosive resistance exercises, such as the explosive bench press (BP) [2, 14]. However, these studies only examined the relationship between velocity loss and muscle fatigue when participants were required to perform exercises at the maximal intended velocity. For muscular hypertrophy-aimed training programs, the lifting cadence is usually determined because the time-under-tension, which refers to the time the muscle spends under mechanical tension, is considered crucial for muscle protein synthesis [17]. Furthermore, for inexperienced athletes and athletes with injuries, performing exercises in an explosive manner may be inappropriate due to these individuals’ lack of strength, joint stability, and adequate lifting techniques [1]. Thus, the validity of using velocity loss as a muscle fatigue indicator remains unknown when resistance exercises are not performed at maximal intended velocity. Accordingly, the validity of using velocity loss as a muscle fatigue indicator in non-explosive settings needs further verification.

The rating of perceived exertion (RPE) scale is a subjective perception-based assessment method that combines numbers, verbal, and pictorial descriptors [18, 19, 20]. The RPE has been shown to be associated with various physiological responses, including neuromuscular fatigue levels and heart rate, during aerobic exercise [19, 21]. Over the past three decades, the use of the RPE has become more popular in resistance exercise situations. For example, it has been reported to capture important information about resistance exercises, such as the percentage of one-repetition maximum (1RM), blood lactate levels, and muscle activation levels [22, 23, 24]. Accordingly, practitioners – such as coaches and personal trainers – have started to use the RPE to quantify the intensity of resistance exercises. Despite the differences between several RPE scales (e.g., the OMNI scale and the Borg CR-10 scale), all are considered to be effective and easy-to-use tools for professionals to monitor resistance exercises. Based on the relationships between the RPE and many physiological measures, the RPE scale may also reflect muscle fatigue responses to resistance exercise. However, the relationship between the RPE and muscle fatigue has only been examined for certain exercises [12, 25, 26, 27, 28]. Therefore, further research is needed to determine this relationship in non-explosive resistance exercises.

By utilizing the new sEMG processing technique, it may be possible to establish the relationship between the RPE and muscle fatigue during dynamic contractions. This could allow professionals to monitor muscle fatigue more simply and in a timely manner. Accordingly, the aims of the present study are as follows: 1) to inspect the validity of using the RPE to assess muscle fatigue through new sEMG-based techniques and 2) to examine the validity of the RPE and velocity loss as muscle fatigue measures during non-explosive BP exercise. We hypothesized that 1) with increasing muscle fatigue, the RPE and velocity loss would change correspondingly and 2) significant relationships between the RPE and muscle fatigue and between velocity loss and muscle fatigue would be observed. However, as previous studies have shown that the RPE is a more precise muscle fatigue indicator than velocity loss in non-explosive exercises [27, 28], we hypothesized that 3) the RPE would also be more accurate than velocity loss as a muscle fatigue indicator when BP was not performed explosively.

2. Methods

2.1 Participants

The sample size was determined using a statistical power analysis employing an ANOVA model that included fixed effects, main effects, and interaction analyses (G*Power 3.1, Bonn University, Germany). The statistical input parameters were based on an effect size of 0.4 and an alpha level of 0.05, with a power of 0.95 [25, 27]. To meet the minimum required of 14 participants for this study, we recruited 15 collegiate ice hockey athletes who had no neuromuscular or skeletal muscle injuries and were not taking any medical treatment. The participants were active collegiate athletes who regularly engaged in daily training activities and participated in national- and regional-level collegiate ice hockey competitions (ranking within the top eight levels of national competition during the year of the present study). All participants were experienced with resistance exercise. The descriptive data of all participants were as follows (mean $\pm$ standard deviation): age (years) 19.4 $\pm$ 1.3; body mass (kg) 69.6 $\pm$ 7.2; height (cm) 171.8 $\pm$ 4.7; and fat (%) 12.9 $\pm$ 4.3 The 1RM of BP was measured at 86.7 $\pm$ 12.3 kg. Before the experiment, all participants were informed about the experimental protocols, assessment items, possible risks, discomforts, and benefits. Written consent was obtained from those who agreed to participate in the study. The study was carried out according to the Declaration of Helsinki’s ethical guidelines for human research, and the Human Ethics Committee of Waseda University approved the study (No. 2020-369).

2.2 Experimental design

This study utilized a counterbalanced, crossover design to examine the validity of Borg’s CR-10 scale and velocity loss as measures of muscle fatigue during non-explosive BP exercise. The choice of BP as the experimental exercise was based on its widespread use in resistance exercise programs. The protocol consisted of two separate test sessions that were at least 72 hours apart. During the initial test session, the participants were carefully instructed on the use of Borg’s CR-10 scale, and their descriptive information was obtained. An anchoring trial was then performed to establish the upper and lower limits of perceived exertion for the experimental session, which is one of the standard procedures for RPE calibration [14, 23, 29]. This trial was conducted with a single set of non-explosive BPs until physical failure [14]. The maximum number of successful repetitions was recorded to calculate the required number of repetitions for the subsequent experimental trials. During the experimental session, the participants completed three experimental conditions, consisting of 30% (low, L), 60% (medium, M), and 90% (high, H) volume, in a counterbalanced order. The volume of the experimental condition was calculated by multiplying the percentage of 1RM by the required number of repetitions. The number of repetitions needed for each condition was calculated based on the number of repetitions the participants successfully completed in the anchoring trial. The conditions were implemented in a counterbalanced order, and the RPE scores, sEMG signals, and velocity were recorded for all conditions and used for further analysis.

2.3 Initial session

The initial session began with detailed instructions on the purpose of the study, the experimental protocols, the measurement items, and the potential risks and benefits. Next, the descriptive characteristics of the participants were measured using a bioelectrical impedance device (InBody720 body composition analyzer, Biospace Co. Ltd, South Korea). Subsequently, the BP 1RM was measured according to the guidelines of the National Strength and Conditioning Association [1]. The BPs were performed using a standard Olympic barbell (Uesaka T.E Co., Ltd., Japan). The 1RM measurement included warm-up sets, followed by a progressive increase in weight until the participants successfully performed a BP with their maximal weight for one repetition.

After the initial measurements, each participant was individually instructed on the translated version of the original Borg’s CR-10 scale. The scale ranged from 0 (nothing at all) to 10 (extremely strong) and included verbal anchors for intermediate values to aid in the perception of exertion [18, 30]. The perception of exertion was defined as subjective feelings, such as effort, discomfort, and fatigue in the upper body, during the BP exercise. After that, the participants were instructed on lifting cadence and subsequently performed the specified practice with a barbell. The lifting cadence of each repetition was defined as a 2-s lowering phase (eccentric) followed by a 1-s raising phase (concentric). A 2-s pause was set between repetitions. This lifting cadence was determined according to previous work and recommendations for non-explosive settings [27, 31]. The lifting cadence was controlled using a metronome (one beep per second). After the lifting practice, an anchoring trial was conducted for all participants to establish the range of subjective exertion during the experimental trials. The participants were instructed to perform a single set of BP exercises until they were unable to complete another repetition. The relative intensity for the anchoring trial was set at 65% of the participants’ 1RM [20, 27], which amounted to 56.3 $\pm$ 8.0 kg. To minimize the influence of rebounding at the bottom position on the lifting movement, the safety bar was adjusted at a height of one level above the chest. A completed repetition was defined as lowering the barbell to approximately the safety bar level and then pushing it upward until the elbows were fully extended.

2.4 Experimental session

Before the implementation of experimental conditions, a series of warm-up lifts were performed with all participants. The warm-up lifts consisted of three sets of six repetitions of BP at 50% of 1RM [27, 32]. During the BP exercise, the safety bar was again positioned one level above the chest to prevent any impact on the sEMG signal and to eliminate barbell bouncing The participants were instructed to lower the barbell until it slightly touched the safety bar. After the warm-up lifts, they performed a practice round of RPE reporting with only the barbell. The Borg scale was placed vertically in front of the participants and was easily visible. The lifting cadence was set as 2-s for the eccentric phase and 1-s for the concentric phase, with a 2-s pause between repetitions. A metronome was used to control cadence, with one beep per second. The participants were asked to report their subjective exertion for the latest repetition using numbers from the scale during the pause between repetitions.

The experimental conditions were implemented in a counterbalanced order after the warm-up lifts and practice. The required number of repetitions for each condition was calculated based on the number of repetitions the participants successfully completed in the anchoring trial. It has been indicated that the RPE is affected by anticipatory responses to a certain exercise task [33]. To minimize the central governor effect, the participants were not informed of the required number of repetitions before the experimental trials. The participants were not informed of these numbers until the second-last repetition and were required to cease immediately after finishing the last repetition. During the experimental conditions, the participants were asked to report their subjective exertion for each repetition by selecting a number from the scale during the 2 s pause between repetitions. As the number of repetitions increased, it may have become difficult for the participants to maintain the determined cadence; consequently, they were instructed to do their best to follow the beeps of the metronome. After each experimental condition, an overall RPE score was obtained from the participants, representing their overall subjective exertion score for that condition. The participants received detailed instructions, which were consistent with those provided during the initial session before the experimental trials. A 5-min rest interval was allowed between the experimental conditions.

2.5 Surface electromyography

The sEMG signals measured from the pectoralis major (dominant side), lateral head of the triceps brachii (dominant side), and anterior deltoid (dominant side) muscles were recorded using bipolar surface electrodes (ADMEDEC Co., Ltd., Japan). Two Ag/AgCl surface electrodes were placed on each muscle, and the distance between the electrodes was set at 1 cm. Prior to electrode placement, the skin was shaved, rubbed with abrasive paper, and cleaned with alcohol swabs to reduce impedance and ensure reliable signal recording [34]. A line indicating the anatomical landmarks was used as a reference frame, and the electrodes were placed on the line according to previous recommendations for electrode placement [35]. The sEMG signals were first amplified using an active differential preamplifier configuration and then transmitted to a wireless device (MARQ MQ-8, Kissei-Com Tech, Japan). The sampling frequency was 1,000 Hz. The sEMG signals were then processed by an analog-to-digital converter and transferred to a dedicated computer. A special camera (FMVU-03MTC-CS, FLIR Systems, Inc., Canada) was connected to the computer to record a motion video synchronized with the sEMG signals during the lifting tasks. The raw sEMG signals were subsequently separated into single repetitions and exported for muscle fatigue analysis.

A fourth-order Butterworth bandpass filter (20–450 Hz) was designed to filter out background noise from the sEMG signals [27, 36]. The filtered signal was used to quantify muscle fatigue under the experimental conditions. The spectral characteristics of the sEMG signals are strongly influenced by dynamic muscle contractions, as mentioned in the Introduction section. To avoid these influences, a validated spectral fatigue index (SFI) was chosen to quantify the muscle fatigue levels in the present study [13]. The SFI has been reported to be more sensitive compared with traditional sEMG-based fatigue parameters (such as the mean and median frequencies) in assessing muscle fatigue during dynamic contractions [9]. A fast Fourier transformation was applied to calculate the power spectrum. The spectral moments were used to extract the characteristics of the power spectral density function and were calculated using Eq. (1):

$\displaystyle M_{k}=\int_{f_{\min}}^{f_{\max}}f^{k}\cdot\textit{PS}(f)\cdot df$ (1)

where $M_{k}$ is the spectral moment of order $k$ , PS( $f$ ) is the power spectrum of frequency $f$ , and $f_{\min}$ and $f_{\max}$ represent the upper and lower cut-off frequencies of the bandpass filter. The SFI value was calculated as the ratio between the spectral moments of orders $-$ 1 and 5, as in Eq. (2):

$\displaystyle\textit{SFI}=\frac{\int_{f_{\min}}^{f_{\max}}f^{-1}\cdot\textit{% PS}(f)\cdot df}{\int_{f_{\min}}^{f_{\max}}f^{5}\cdot\textit{PS}(f)\cdot df}$ (2)

The SFI was calculated for each repetition, and the relative changes in the SFI values for the different repetitions were calculated with respect to the first repetition of each experimental condition. The results obtained from the three muscles were averaged to obtain a single parameter, which was used in subsequent statistical analyses. The sEMG data processing was carried out using MATLAB R2020a (Mathworks, Natick, MA, USA).

2.6 Velocity loss calculation

To assess the validity of velocity loss as a muscle fatigue indicator in different resistance exercise programs, lifting cadence was controlled in the present study. The velocity during BP was measured using a linear position transducer (Fitro Dyne, FITRONIC s.r.o., Bratislava, Slovakia). The transducer was set directly under the barbell and connected to it with a cable. The sampling frequency was 100 Hz, and displacement changes over time were recorded using a computer. In previous studies that focused on the relationship between velocity loss and muscle fatigue assessment, participants were mostly required to perform resistance exercises explosively and/or until physical failure [2, 14]. However, in lifting programs with a non-explosive design and/or far from physical failure (e.g., at 30% of the maximum volume until failure), lifters may be able to control the weight. In these situations, muscle fatigue may occur, while velocity loss fails to increase. To assess the validity of velocity loss as a muscle fatigue indicator in non-explosive and non-failure settings, the lifting tempo and volume were modified and controlled in this study. Accordingly, the mean velocity of the concentric phase was calculated and used to determine velocity loss. The intra-set velocity loss was calculated using the velocity of each repetition against the fastest repetition of each condition, and it was calculated repetition-by-repetition [2].

2.7 Statistical analyses

Figure 1.

Overall rating of perceived exertion (RPE), average spectral fatigue index (SFI), and average velocity loss in low- (L), medium- (M), and high- (H) volume bench press tasks (averaged from 15 participants). The $p$ -value represents the differences and significance levels between the experimental conditions.

The Shapiro-Wilk test was used to examine the normality of the overall RPE, average velocity loss, and average SFI. Each variable within each condition was examined individually. The average SFI and average velocity loss were fully normally distributed. However, it was observed that the overall RPE of the M condition was not normally distributed. Therefore, a one-way ANOVA was applied to the average SFI and average velocity loss. Next, a post hoc test with Bonferroni correction was applied for multiple comparisons Friedman’s ANOVA was employed to test for differences in the overall RPE among the various conditions. Further, a post hoc test with Bonferroni correction was applied for multiple comparisons. To investigate the RPE, SFI, and velocity loss during the lifting process, the data of selected repetitions of each condition (L, M, and H conditions) were also analyzed. As the required number of repetitions was different for each experimental condition and each participant (L: 2.9 $\pm$ 0.5 repetitions; M: 5.7 $\pm$ 0.9 repetitions; H: 8.5 $\pm$ 1.4 repetitions), the first and last repetitions were extracted and used for the two-way ANOVA (3 conditions $\times$ 2 repetitions), which was used to test for the main and interaction effects of conditions and repetitions. If a significant effect was found, a post hoc test was performed with a Bonferroni correction for multiple comparisons.

The intra-set RPE, velocity loss, and SFI were used in the correlation analysis [27, 37]. As half of the data was gathered in the 0–60% volume area, only the non-overlapping portions of the three conditions were used for correlation analysis. Specifically, the data included the 0–30% volume range for the L condition, the 30–60% volume range for the M condition, and the 60–90% volume range for the H condition [37]. Next, the two correlation coefficients were compared using Fisher’s r-to-z transformation and paired two-tailed tests. The statistical analysis was performed using SPSS 28.0 (SPSS Inc., USA). The statistical significance level was considered acceptable at $p<$ 0.05.

3. Results

Figure 2.

Rating of perceived exertion (RPE), spectral fatigue index (SFI), and velocity loss during low- (L, circle with solid lines), medium- (M, square with dashed lines), and high- (H, triangle with dotted lines) volume bench press tasks (averaged from 15 participants). ${}^{*}$ Represents a significant difference between the L and M conditions, ${}^{*}p<$ 0.05, ${}^{***}p<$ 0.001; ${}^{\#}$ represents a significant difference between the M and H conditions, ${}^{\#\#\#}p<$ 0.001; ${}^{\dagger}$ represents a significant difference between the L and H conditions, ${}^{{\dagger}{\dagger}}p$ < 0.01, ${}^{{\dagger}{\dagger}{\dagger}}p<$ 0.001; ${}^{\S}$ indicates a significant difference between the last repetition and the first repetition, ${}^{\S\S\S}p<$ 0.001.

Figure 3.

Spearman’s rho between the spectral fatigue index (SFI), rating of perceived exertion (RPE) scores, and velocity loss during bench press exercises (a total of 119 repetitions from 15 participants were gathered in the correlation analysis). ${}^{*}$ Represents the significance level of the correlation coefficients, $p<$ 0.001.

Significant differences in the overall RPE ( $\chi^{2}=$ 27.763, $p<$ 0.001, Kendall’s $W$ $=$ 0.925) were observed in the L vs. M ( $p=$ 0.008) and L vs. H ( $p<$ 0.001) comparisons (Fig. 1a). Regarding the average SFI, significant differences were observed in the L vs. M ( $p=$ 0.015, Cohen’s d $=$ 1.359), M vs. H ( $p<$ 0.001, Cohen’s d $=$ 1.604), and L vs. H ( $p<$ 0.001, Cohen’s d $=$ 2.920) comparisons (Fig. 1b). Finally, significant differences in the average velocity loss were observed in the L vs. H ( $p<$ 0.001, Cohen’s d $=$ 2.038) and M vs. H ( $p=$ 0.048, Cohen’s d $=$ 0.889) comparisons (Fig. 1c).

In contrast to the average SFI, average velocity loss, and overall RPE, intra-set data were analyzed using a two-way ANOVA. As the required number of repetitions was different for each condition and participant, the first and last repetitions were extracted and used for the two-way ANOVA. A significant overall main effect of the number of repetitions on the RPE ( $p<$ 0.001, partial $\eta^{2}=$ 0.864), velocity loss ( $p<$ 0.001, partial $\eta^{2}=$ 0.688), and SFI ( $p<$ 0.001, partial $\eta^{2}=$ 0.909) was observed throughout the BP trials. Significant differences in the RPE were observed between experimental conditions at the last repetition ( $p<$ 0.001, partial $\eta^{2}=$ 0.839) (Fig. 2a). Velocity loss also significantly differed between conditions ( $p=$ 0.003, partial $\eta^{2}=$ 0.338) (Fig. 2c). However, a significant conditions $\times$ repetitions interaction was not observed ( $p=$ 0.071). Finally, significant differences in the SFI between experimental conditions were observed at the last repetition ( $p<$ 0.001, partial $\eta^{2}=$ 0.799) (Fig. 2b).

The results of the Spearman correlation analysis, in which a total of 119 BP repetitions were used, are shown in Fig. 3. The relationships between the SFI and RPE ( $r=$ 0.681, $p<$ 0.001) (Fig. 3a) and between the SFI and velocity loss ( $r=$ 0.510, $p<$ 0.001) were found to be significant (Fig. 3b). The SFI-RPE correlation was significantly stronger than the SFI-velocity loss correlation ( $z=$ 2.047, $p=$ 0.041).

4. Discussion

The purpose of the present study was to examine the validity of using the RPE to assess muscle fatigue and to examine the use of velocity loss as a muscle fatigue indicator when BP exercise was no longer performed in an explosive manner. The most important findings of the present study are: 1) the RPE and velocity loss changed significantly with increasing muscle fatigue, showing that muscle fatigue induces an increase in subjective perceived exertion and a decrease in velocity during non-explosive BP exercise, and 2) significant RPE-SFI and velocity loss-SFI correlations were observed, indicating that both the RPE and velocity loss could be used as muscle fatigue indicators during non-explosive BP exercise However, 3) the RPE was found to be a higher-precision muscle fatigue indicator than velocity loss.

Perceived exertion could correlate with various central and peripheral responses, such as an increase in motor unit recruitment and firing frequency during dynamic resistance exercises [23, 38, 39, 40]. For example, Lagally et al. reported that the RPE, blood lactate, and muscle activation levels increased correspondingly with the relative intensity of resistance exercise, demonstrating that the RPE could be used as an indirect indicator of both central and peripheral responses [23]. To eliminate the influence of relative intensity differences, a constant relative intensity was adopted for all experimental conditions. Therefore, the significant differences observed in the RPE may be related to differences in the number of repetitions performed, which may have resulted in differences in the consecutive lifting phase. Previous studies have indicated that performing more repetitions can affect intramuscular perturbation and the RPE [32, 41]. In agreement with these findings, as the number of repetitions increased in the experimental conditions, there was a greater disruption in homeostasis (e.g., in the hydrogen ion concentration). These physiological responses can affect afferent feedback from nerve fibers in the skeletal muscle, resulting in a significant increase in subjective perceived exertion [42, 43]. Moreover, severe disruptions in intramuscular homeostasis can affect muscle fiber conductivity, as indicated by the compression of the power spectrum [12, 44]. In the present study, significant changes in the SFI entailed shifts in the power spectrum, which was likely related to muscle fatigue-induced by-product concentration [45]. Consequently, the RPE and SFI increased correspondingly, indicating that subjective perceived exertion reflects muscle fatigue levels during BP exercise.

During resistance exercises, it is very important for coaches and personal trainers to be aware of the muscle fatigue levels of their clients, as mentioned in the Introduction section. Velocity loss has been reported to be a reliable indirect muscle fatigue indicator for explosive resistance exercise [2]. However, lifting explosively is inappropriate for muscular hypertrophy, rehabilitation, and beginner exercise programs. Thus, we assessed the validity of using velocity loss as a muscle fatigue indicator under a cadence-controlled lifting setting. In the present study, a significant difference between the first and last repetitions was observed. This result indicates that velocity loss increases significantly with the progression of the non-explosive BP exercise, even if the prescribed volume is far from physical failure (e.g., 30% of the volume before reaching failure). Accordingly, velocity loss is still valid as a muscle fatigue indicator during non-explosive BP exercise. However, when comparing the results with the RPE, it seems that the velocity loss was less sensitive than the RPE under the present experimental settings. Similar significant differences were observed in the RPE and SFI between all experimental conditions at the last repetition (all $p<$ 0.001), while the significant differences in velocity loss were observed only in some comparisons (L vs. M: $p<$ 0.05; L vs. H: $p<$ 0.01; conditions $\times$ repetitions interaction: $p=$ 0.071). One would expect to obtain similar results for larger sample sizes; however, velocity loss has lower precision in assessing muscle fatigue compared to the RPE. This result is not unexpected, as a previous study reported a significant decrease in velocity during fatiguing non-explosive upper-body resistance exercise [27]. The present study expanded the use of velocity loss as a muscle fatigue measure to include upper-body multi-joint exercises. While velocity-based training is becoming increasingly popular, caution is still necessary when performing velocity-based muscle fatigue assessments. In order to measure velocity with acceptable precision and accuracy, special devices, such as linear encoders, are often required. Moreover, the widespread implementation of velocity measurements is currently impractical for individual trainers and coaches. In contrast, RPE-based muscle fatigue assessment can be conducted using Borg’s CR-10 scale, which does not require any special devices and can be easily implemented. Considering the ease of use of the RPE scale compared to velocity measurement, it can be concluded that while both the RPE and velocity loss reflect muscle fatigue, the RPE is more practical for assessing muscle fatigue during non-explosive BP exercise.

In the present study, we examined the RPE–SFI and velocity loss–SFI relationships using data from each repetition. Notably, both analyses revealed a significant correlation. These results demonstrate that similar to velocity loss, the RPE can also be used as a muscle fatigue indicator with an increase in the number of repetitions. According to previous recommendations, velocity loss increases with both the number of repetitions and sets [2]. As for the RPE, most resistance exercise studies recommend assessing the RPE after a determined number of sets and/or sessions, which represents an overall subjective measure of fatigue [20, 46]. The results of the present study provide a new method for real-time muscle fatigue monitoring. For example, clients can report the RPE after a predetermined number of repetitions, such as the first, mid-point, and last repetition of a certain set, or even repetition-by-repetition. By obtaining the intra-set RPE, trainers and coaches can easily monitor and assess the level of muscle fatigue during resistance exercise, thereby minimizing the risk of injury and preventing a decrease in athletic performance induced by acute muscle fatigue [2, 47]. This new real-time muscle fatigue monitoring method can be particularly useful when velocity assessment is unavailable. These results are consistent with previous recommendations for RPE-based muscle fatigue assessment and have expanded the range of applicability from single-joint to multi-joint upper-body exercises [27]. However, it should be noted that in the present study, a 2-s pause was set between repetitions to obtain the intra-set RPE. The pause phase is similar to continuous isometric contraction, which may have affected the muscle fatigue responses in the present study [25]. For practical applications, practitioners should consider the influence of the pause phase on muscle fatigue assessment. Consequently, our results suggest that both the RPE and velocity loss could be used as muscle fatigue measures in non-explosive BP exercise.

The present study has several limitations. First, we set a 2-s pause between repetitions to obtain the intra-set RPE. It is almost impossible to avoid the influence of the pause phase on muscle fatigue responses, as discussed previously. Second, we assessed the validity of only the Borg CR-10 scale, as it is one of the most commonly used scales in research and exercise settings. However, using different types of RPE scales (e.g., the OMNI-RES scale) might yield different results. Third, the sample size was relatively small, which may have limited the significance of the statistical analysis. Finally, although we instructed the participants to consider and report only the subjective exertion induced by the BP trials based on the anchoring trial to exclude influences that were not associated with the experimental procedure, it might still have been challenging for them to separate other factors, such as breathlessness [29]. Accordingly, future studies should focus on lowering the influence of pauses on the intra-set data, assessing the differences between RPE scales, and employing a larger sample size.

5. Conclusion

The present study demonstrated that the RPE and SFI changed similarly, indicating an association between subjective exertion and muscle fatigue. Thus, we concluded that the RPE could be employed as an easy-to-use and valid muscle fatigue indicator for non-explosive resistance exercise scenarios, such as rehabilitation and beginner training. Coaches and physical therapists can use the RPE as a real-time muscle fatigue indicator. We also assessed the validity of using velocity loss as a muscle fatigue indicator during non-explosive upper-body resistance exercise. The velocity loss and SFI changed correspondingly, indicating that velocity loss is still effective as a muscle fatigue indicator when exercise is not performed explosively. Additionally, a significantly stronger correlation was observed between the RPE and SFI compared to velocity loss and the SFI. Therefore, both the RPE and velocity loss are effective for muscle fatigue assessment during non-explosive BP exercise. However, considering the differences in precision, convenience, and cost of the two methods, we can conclude that the RPE is a more practical option for muscle fatigue quantification during non-explosive upper-body resistance exercise programs.

Author contributions

CONCEPTION: Hanye Zhao and Junichi Okada.

PERFORMANCE OF WORK: Hanye Zhao and Dasom Seo.

INTERPRETATION OR ANALYSIS OF DATA: Hanye Zhao, Dasom Seo, and Junichi Okada.

PREPARATION OF THE MANUSCRIPT: Hanye Zhao, Dasom Seo, and Junichi Okada.

REVISION FOR IMPORTANT INTELLECTUAL CONTENT: Hanye Zhao and Junichi Okada.

SUPERVISION: Junichi Okada.

Ethical considerations

The experimental protocol was established, according to the ethical guidelines of the Helsinki Declaration and was approved by the Waseda University Human Ethics Committee (No. 2020-369). Informed consent was obtained from all individual participants included in the study.

Funding

No funds, grants, or other financial support was received.

Footnotes

Acknowledgments

The authors would like to thank all the participants in this study.

Conflict of interest

The authors declare that they have no conflicts of interest and competing interests, and there has been no financial support for this study that could have influenced the outcome.

References

Thomas

Roger

. Essentials of strength training and conditioning. 3rd ed. Champaign(IL): Human kinetics; 2008.

Sánchez-Medina

González-Badillo

. Velocity loss as an indicator of neuromuscular fatigue during resistance training. Med Sci Sports Exerc. 2011; 43: 1725-34. doi: 10.1249/mss.0b013e318213f880.

Roy

De Luca

Casavant

. Lumbar muscle fatigue and chronic lower back pain. Spine (Phila Pa 1976). 1989; 14: 992-1001. doi: 10.1097/00007632-198909000-00014.

Borotikar

Newcomer

Koppes

McLean

. Combined effects of fatigue and decision making on female lower limb landing postures: Central and peripheral contributions to ACL injury risk. Clin Biomech. 2008; 23: 81-92. doi: 10.1016/j.clinbiomech.2007.08.008.

Skurvydas

Brazaitis

Streckis

Rudas

. The effect of plyometric training on central and peripheral fatigue in boys. Int J Sports Med. 2010; 31: 451-7. doi: 10.1055/s-0030-1251991.

Carregaro

Cunha

Oliveira

Brown

Bottaro

. Muscle fatigue and metabolic responses following three different antagonist pre-load resistance exercises. J Electromyogr Kinesiol. 2013; 23: 1090-6. doi: 10.1016/j.jelekin.2013.04.010.

Campanini

Disselhorst-Klug

Rymer

Merletti

. Surface EMG in Clinical Assessment and Neurorehabilitation: Barriers Limiting Its Use. Front Neurol. 2020; 11: 934. doi: 10.3389/fneur.2020.00934.

Hermens

Freriks

Disselhorst-Klug

Rau

. Development of recommendations for SEMG sensors and sensor placement procedures. J Electromyogr Kinesiol. 2000; 10: 361-74. doi: 10.1016/S1050-6411(00)00027-4.

González-Izal

Malanda

Gorostiaga

Izquierdo

. Electromyographic models to assess muscle fatigue. J Electromyogr Kinesiol. 2012; 22: 501-12. doi: 10.1016/j.jelekin.2012.02.019.

10.

Yoshitake

Miyazaki

Moritani

. Assessment of lower-back muscle fatigue using electromyography, mechanomyography, and near-infrared spectroscopy. Eur J Appl Physiol. 2001; 84: 174-9. doi: 10.1007/s004210170001.

11.

Brody

Pollock

Roy

De Luca

Celli

. pH-induced effects on median frequency and conduction velocity of the myoelectric signal. J Appl Physiol. 1991; 71: 1878-85. doi: 10.1152/jappl.1991.71.5.1878.

12.

González-Izal

Malanda

Navarro-Amézqueta

Gorostiaga

Mallor

Ibañez

, et al. EMG spectral indices and muscle power fatigue during dynamic contractions. J Electromyogr Kinesiol. 2010; 20: 233-40. doi: 10.1016/j.jelekin.2009.03.011.

13.

Dimitrov

Arabadzhiev

Mileva

Bowtell

Crichton

Dimitrova

. Muscle fatigue during dynamic contractions assessed by new spectral indices. Med Sci Sports Exerc. 2006; 38: 1971-9. doi: 10.1249/01.mss.0000233794.31659.6d.

14.

Mayo

Iglesias-Soler

Kingsley

. Perceived exertion is affected by the submaximal set configuration used in resistance exercise. J Strength Cond Res. 2019; 33: 426-32. doi: 10.1519/JSC.0000000000001886.

15.

John

FFT

Kevin

Cain

CCT

Jason

Ben

Amador

G-R

, et al. Comparison of the FitroDyne and GymAware Rotary Encoders for Quantifying Peak and Mean Velocity During Traditional Multijointed Exercises. J Strength Cond Res. 2021; 35: 1760-5. doi: 10.1519/JSC.0000000000002952.

16.

Evandro

Adré

Amador

Dnica

Daniel

. Validity of the iLOAD® app for resistance training monitoring. PeerJ. 2019; 7: e7372. doi: 10.7717/PEERJ.7372.

17.

Vargas-Molina

Martín-Rivera

Bonilla

Petro

Carbone

Romance

, et al. Comparison of blood lactate and perceived exertion responses in two matched time-under-tension protocols. PLoS One. 2020; 15: 1-11. doi: 10.1371/journal.pone.0227640.

18.

Borg

GAV

. Borg’s perceived exertion and pain scales. Champaign (IL): Human kinetics; 1998.

19.

Robertson

Goss

Dubé

Rutkowski

Dupain

Brennan

, et al. Validation of the Adult OMNI Scale of Perceived Exertion for Cycle Ergometer Exercise. Med Sci Sports Exerc. 2004; 36: 102-8. doi: 10.1249/01.MSS.0000106169.35222.8B.

20.

Robertson

Goss

Rutkowski

Lenz

Dixon

Timmer

, et al. Concurrent validation of the OMNI perceived exertion scale for resistance exercise. Med Sci Sports Exerc. 2003; 35: 333-41. doi: 10.1249/01.MSS.0000048831.15016.2A.

21.

Fontes

Smirmaul

BPC

Nakamura

Pereira

Okano

Altimari

, et al. The relationship between rating of perceived exertion and muscle activity during exhaustive constant-load cycling. Int J Sports Med. 2010; 31: 683-8. doi: 10.1055/s-0030-1255108.

22.

Lagally

Robertson

. Construct validity of the OMNI Resistance Exercise Scale. J Strength Cond Res. 2006; 20: 252-6. doi: 10.1519/R-17224.1.

23.

Lagally

Robertson

Gallagher

Goss

Jakicic

Lephart

, et al. Perceived exertion, electromyography, and blood lactate during acute bouts of resistance exercise. Med Sci Sports Exerc. 2002; 34: 552-9. doi: 10.1097/00005768-200203000-00025.

24.

Eston

James

Evans

. The validity of submaximal ratings of perceived exertion to predict one repetition maximum. J Sport Sci Med. 2009; 8: 567-73.

25.

Cruz-Montecinos

Bustamante

Candia-González

González-Bravo

Gallardo-Molina

Andersen

, et al. Perceived physical exertion is a good indicator of neuromuscular fatigue for the core muscles. J Electromyogr Kinesiol. 2019; 49: 102360. doi: 10.1016/j.jelekin.2019.102360.

26.

George

Tolley

Vehrs

Reece

Fatih Akay

Cambridge

EDJ

. New approach in assessing core muscle endurance using ratings of perceived exertion. J Strength Cond Res. 2018; 32. doi: 10.1519/jsc.0000000000001915.

27.

Zhao

Nishioka

Okada

. Validity of using perceived exertion to assess muscle fatigue during resistance exercises. PeerJ. 2022; 10: e13019. doi: 10.7717/PEERJ.13019.

28.

Zhao

Seo

Okada

. Validity of using perceived exertion to assess muscle fatigue during back squat exercise. BMC Sports Sci Med Rehabil. 2023; 15: 14. doi: 10.1186/S13102-023-00620-8.

29.

Abbiss

Peiffer

Meeusen

Skorski

. Role of Ratings of Perceived Exertion during Self-Paced Exercise: What are We Actually Measuring? Sport Med. 2015; 45: 1235-43. doi: 10.1007/s40279-015-0344-5.

30.

Borg

GAV

. Psychophysical bases of perceived exertion. Med Sci Sports Exerc. 1982; 14: 377-81.

31.

Kraemer

Ratamess

. Fundamentals of resistance training: progression and exercise prescription. Med Sci Sports Exerc. 2004; 36: 674-88. doi: 10.1249/01.MSS.0000121945.36635.61.

32.

Hiscock

Dawson

Clarke

Peeling

. Can changes in resistance exercise workload influence internal load, countermovement jump performance and the endocrine response? J Sports Sci. 2018; 36: 191-7. doi: 10.1080/02640414.2017.1290270.

33.

Noakes

St Clair Gibson

Lambert

. From catastrophe to complexity: a novel model of integrative central neural regulation of effort and fatigue during exercise in humans: summary and conclusions. Br J Sports Med. 2005; 39: 120-4. doi: 10.1136/bjsm.2003.010330.

34.

Hermens

Freriks

Merletti

Stegeman

Blok

Rau

, et al. European recommendations for surface electromyography. Roessingh Res Dev. 1999; 8.

35.

Barbero

Merletti

Rainoldi

. Atlas of Muscle Innervation Zones. Lavis (TN), Italy: Springer Science & Business Media; 2012. doi: 10.1007/978-88-470-2463-2.

36.

Otto

Emery

Côté

. Differences in muscular and perceptual responses to a neck/shoulder fatiguing task between women and men. J Electromyogr Kinesiol. 2018; 43: 140-7. doi: 10.1016/j.jelekin.2018.09.010.

37.

Migliaccio

Iacono A Dello Ardigò

Samozino

Iuliano

Grgantov

, et al. Leg press vs. smith machine: Quadriceps activation and overall perceived effort profiles Front Physiol. 2018; 9: 1481. doi: 10.3389/fphys.2018.01481.

38.

de Morree

Klein

Marcora

. Perception of effort reflects central motor command during movement execution. Psychophysiology. 2012; 49: 1242-53. doi: 10.1111/j.1469-8986.2012.01399.x.

39.

Amann

Blain

Proctor

Sebranek

Pegelow

Dempsey

. Implications of group III and IV muscle afferents for high-intensity endurance exercise performance in humans. J Physiol. 2011; 589: 5299-309. doi: 10.1113/jphysiol.2011.213769.

40.

Blain

Mangum

Sidhu

Weavil

Hureau

Jessop

, et al. Group III/IV muscle afferents limit the intramuscular metabolic perturbation during whole body exercise in humans. J Physiol. 2016; 594: 5303-15. doi: 10.1113/JP272283.

41.

Gorostiaga

Navarro-Amézqueta

González-Izal

Malanda

Granados

Ibáñez

, et al. Blood lactate and sEMG at different knee angles during fatiguing leg press exercise. Eur J Appl Physiol. 2012; 112: 1349-58. doi: 10.1007/s00421-011-2090-1.

42.

Esteban

Navarro-Amézqueta

Calbet

JAL

Hellsten

Cusso

Guerrero

, et al. Energy metabolism during repeated sets of leg press exercise leading to failure or not. PLoS One. 2012; 7: e40621. doi: 10.1371/JOURNAL.PONE.0040621.

43.

Hardee

Lawrence

Utter

Triplett

Zwetsloot

McBride

. Effect of inter-repetition rest on ratings of perceived exertion during multiple sets of the power clean. Eur J Appl Physiol. 2012; 112: 3141-7. doi: 10.1007/s00421-011-2300-x.

44.

Arabadzhiev

Dimitrov

Dimitrov G

. The increase in surface EMG could be a misleading measure of neural adaptation during the early gains in strength. Eur J Appl Physiol. 2014; 114: 1645-55. doi: 10.1007/s00421-014-2893-y.

45.

Camic

Housh

Johnson

Hendrix

Zuniga

Mielke

, et al. An EMG frequency-based test for estimating the neuromuscular fatigue threshold during cycle ergometry. Eur J Appl Physiol. 2010; 108: 337-45. doi: 10.1007/s00421-009-1239-7.

46.

Gearhart

Goss

Lagally

Jakicic

Gallagher

Robertson

. Standardized Scaling Procedures for Rating Perceived Exertion during Resistance Exercise. J Strength Cond Res. 2001; 15: 320-5. doi: 10.1519/1533-4287(2001)015<0320:SSPFRP>2.0.CO;2.

47.

Sparto

Parnianpour

Reinsel

Simon

. The effect of fatigue on multijoint kinematics, coordination, and postural stability during a repetitive lifting test. J Orthop Sports Phys Ther. 1997; 25: 3-12. doi: 10.2519/jospt.1997.25.1.3.