Regional hypertrophy of muscle cannot be predicted by surface electromyography

Abstract

BACKGROUND:

In the last decades, emerging evidence has shown that muscle growth is not homogeneous along a muscle head. This phenomenon is known as regional muscle hypertrophy and has led to several questions regarding the implications it may have for health and sports performance.

OBJECTIVE:

The aim of this study was to determine whether regional hypertrophy can be predicted by surface electromyography (sEMG).

METHODS:

36 participants performed two arm exercises (preacher curls and inclined curls) in a random order to muscle failure at 70% of the 1 RM of the bicep curl exercise. As every participant performed a different number of repetitions, Peak sEMG and the integral of the sEMG of the last 3 repetitions was analyzed an compared to previously performed maximal voluntary isometric contractions (MVIC).

RESULTS:

The independent sample $t$ -tests showed no significant differences neither in the peak nor in the integral of the sEMG between exercises for any given region.

CONCLUSIONS:

sEMG cannot be used to predict regional hypertrophy.

Keywords

Arm muscle growth biceps brachii

1. Introduction

Previous research has shown that skeletal muscle growth does not happen homogeneously [1, 2, 3]. The biomechanical and physiological factors behind this phenomenon, known as regional hypertrophy, are mostly unknown. It has been suggested that regional hypertrophy may have several implications for sports performance and health issues [2]. As such, it has been shown that the rectus femoris head of the quadriceps muscle gets more involved in monoarticular movements such as leg extensions when compared to leg press [4, 5], providing evidence that hypertrophy in some regions of a muscle responds to specific movements. In fact, a recent study found that different regions inside a given muscle have different roles in different exercises [6]. Furthermore, it is suggested that some regions may be more involved in power-related actions, whilst others may have a more important role when a higher force output is demanded [7]. Considering all this evidence, it is of great clinical importance to understand how to make given regions inside a muscle grow.

Recently, Zabaleta-Korta et al. (2023) found that the distal region of the upper arm (the one next to the elbow) grew in response to the preacher curl exercise (hence, PREA) but not in response to the inclined curl exercise (hence, INC) [8]. The authors suggested that this region may be preferentially recruited to perform exercises in which the arm flexors are elongated like in the PREA, but not during exercises that are performed mostly when arm flexors are shortened, likely because the force demands on arm flexors differ between exercises. If true, this would provide strong evidence that each muscle region responds to a certain kind of external demand, and would serve as a guide for coaches and athletes that aim to develop specific strength in particular movements or exercises.

Figure 1.

RoM and resistance of both exercises in the Zabaleta-Korta et al. (2023) study.

Thus, the main aim of the present study was to understand if the PREA exercise increases the myoelectric activity of the distal region of the arm to a greater extent than the INC exercise. The secondary goal was to assess whether the same relationship exists between the proximal region and the INC exercise. Based on the results of the aforementioned study [8], our hypothesis is that the distal region of the arm will show greater myoelectric activity during the PREA exercise when compared to the INC exercise.

2. Methods

2.1 Participants

36 healthy participants were recruited for the study. Both male ( $n=$ 21) and female ( $n=$ 15) participants were recruited as the participants sex does not affect the resistance profiles between the two exercises. Participants were included in the study if they met the following conditions: 1) They were between 18 and 40 years of age and 2) They had at least 2 years of resistance training experience, which means a minimum of 2 times per week uninterruptedly. Participants were excluded if 1) They had suffered any kind of musculoskeletal injury 2 years prior to the beginning of the study; 2) They were taking anabolic steroids and 3) If they performed strenuous upper body workouts 48 h prior to the measurements. Written informed consent was obtained from each participant after a thorough explanation of the testing protocol, the possible risks involved and the right to terminate participation at will. The study was approved by the Institutional Review Board of the University of the Basque Country UPV/EHU (ref. 118/2019) and all procedures were done in accordance with declaration of Helsinki (Fortaleza, 2013).

A priori sample size calculations performed with the G*Power software, revealed that each group should have at least 36 participants to achieve a power of 0.8, an alpha of 0.05 and an increasing Muscle Thickness (Hence MT) with an effect size of 0.61. For this reason, 36 participants were recruited and performed both exercises in a random order. Half of them performed the INC exercise first, and the other half performed the PREA exercise first.

2.2 Intervention

Each participant came to the laboratory once. After the warm up, that consisted of mobility exercises and bicep curls with gradual increases in weight, 1 RM of the bicep curl exercise was calculated. For that purpose, participants were instructed on how to properly perform a biceps curl: the movement began with the forearm perpendicular to the floor and ended when the weight was lifted beyond the position in which the forearm was parallel to the floor (an elbow flexion of more than 90^∘) (Fig. 1). 1 RM was calculated as shown elsewhere [9]: after the warm up, participants were asked to lift an “easy” weight for 7 repetitions. After that, weight was increased from 2.5 to 5 kg between trials and participants only performed 1 repetition until they reported that the repetition had been hard to perform. Hence, weight increases were made in 0.5–1 kg intervals until the participants failed the repetition i.e., could no longer lift the weight enough for the elbow to reach a 90^∘ flexion. Participants were allowed enough rest between sets to perform as good as possible in the following one. After that, electrodes were placed in the proximal and distal parts of the upper-arm of the participants and maximum voluntary isometric contraction (MVIC) was measured. The MVIC served to standardize the EMG signal, and is described in the EMG section. To familiarize the participants with the tests, each of them were allowed two low-intensity trials (one per each exercise). After that, participants were trained to performed properly each exercise and each participant was randomly assigned to the INC first or the PREA first group. If the first participant was assigned to the INC first group, the second participant was assigned to the PREA first group. That way we made sure half the participants performed the PREA first and the other half performed the INC first. Sex was not considered when allocating a subject into a group.

Each participant then performed the exercises in the assigned order. They had to perform at least 5 repetitions for the measurements to be considered valid. If the participants did not reach 5 repetitions, the weight was reduced by 2.5 kilos and after a 5 minute rest, the participant performed another trial. If any participant performed a repetition with poor form (not reaching the end of the repetitions or making awkward movements) at the end of the set, he or she was asked to immediately stop the set.

The INC exercise consisted of performing bicep curls while lying on an inclined bench. The exercise began with the forearm perpendicular to the floor and ended when the weight was lifted enough for the forearm to be perpendicular to the floor (Image 2). The PREA exercise consisted on performing bicep curls on a preacher bench. The exercise began with the forearm almost perpendicular to the floor and ended when the arm returned to that position after the dumbbell was lowered to a position in which the forearm was lower than parallel to the floor. The difference between both exercises is their resistance profile. While the most difficult part of the INC exercise occurs when the elbow flexors are completely flexed, the most difficult part of the PREA group is the position in which elbow flexors are completely lengthened. A graphical representation of the difficulty of each exercise is shown on Fig. 1.

2.3 EMG measurement

Elbow kinematics were recorded using four infrared motion-capture cameras (Bonita 10, Vicon, United Kingdom) at a frequency of 100 Hz. Three reflective markers were attached to the acromion, lateral epicondyle, and radial styloid process of the participants’ dominant arm to record its motion. Kinematic data was used to identify the last three repetitions of each set before failure. The maximum and minimum elbow angles were calculated in Nexus (Nexus, Vicon, United Kingdom) and used to identify the start of the first and the end of the last repetitions, respectively.

Myoelectric activity of the distal and proximal biceps portions were recorded using a surface electromyography (sEMG) recording device (Myon 320, Myon, Switzerland) at a frequency of 1000 Hz. The recording sites were selected by measuring the distance between the acromion and the distal biceps tendon insertion and dividing it into three equal portions, each accounting for 33.3% of the total length. The proximal and distal distances limiting the middle portion (i.e. 33% and 66% of total length) were selected as the recording sites. Thereafter, two pairs of electrodes (Blusensor, Ambu, Denmark) were attached to the skin after abrasion and cleaning with alcohol, following SENIAM recommendations [10]. Three maximal voluntary isometric contractions (MVIC) trials were recorded by manually resisting elbow flexion at the distal part of the forearm with the elbow positioned at 90^∘ and full supination.

All sEMG data was processed using Matlab (Mathworks, USA). Data were band-pass filtered between 20–450 Hz using a fourth order butterworth filter, and rectified using a 50 ms window span to get the linear envelope. Lastly, sEMG data were normalized using the maximum value obtained from the MVIC trials.

Peak sEMG value and the integral of the sEMG (iEMG) from processed data. Peak sEMG value serves to compare the maximum myoelectric activity level reached on each of the recording sites, while iEMG provides a value that represents the overall myoelectric activity produced by the biceps brachii muscle in each site during the last three repetitions. The iEMG is defined as the area under the sEMG time-series curve and was calculated by integrating the curve using a trapezoidal rule:

$\displaystyle\hskip-8.535827pt\textit{iEMG}=\int^{b}_{a}\textit{rEMG dx}% \approx\frac{1}{2}h[\textit{rEMG}_{0}+$ $\displaystyle\hskip-8.535827pt2(\textit{rEMG}_{1}+\textit{rEMG}_{2}+\ldots+% \textit{rEMG}_{n-1})+\textit{rEMG}_{n}]$

Figure 2.

Position of the electrodes at 33% and 66% of humerus length.

2.4 Statistical analysis

Data was analyzed for normality and homoscedasticity. We tested all variables for normal distribution (Shapiro-Wilk test) and homogeneity of variances (Levene’s test). An independent samples $t$ -test was performed to compare the differences in the iEMG and sEMG between PREA and INC. If data from any group was not normally distributed Wilcoxon’s test was performed. Effect sizes were calculated as Cohen’s d (small $d=$ 0.2; medium $d=$ 0.5 and large $d\geqslant$ 0.8) or rank-biserial correlation, for normally distributed and non-normally distributed data respectively. An intra-class correlation coefficient test was performed to measure the reliability of the measurement protocol, in a pilot study carried out with 10 participants. All the statistical analyses were performed with JASP 0.16 for Mac.

3. Results

36 subjects (men $=$ 21; women $=$ 15; age $=$ 28.5 $\pm$ 5.3 years; height $=$ 1.68 $\pm$ 0.14; weight $=$ 74.3 $\pm$ 7.4 kg; resistance training experience = 3.1 $\pm$ 1.2 years) completed the study. In the distal region, both iEMG had a normal distribution for both groups INC ( $p=$ 0.023) and PREA ( $p=$ 0.01) but did not pass the Levene’s test ( $p=$ 0.509), the same was true for sEMG in the normality (INC, $p<$ 0.01; PREA, $p<$ 0.01) and homogeneity ( $p=$ 0.713) checks. In the distal region, neither group passed the normality and homogeneity checks, in the iEMG and sEMG parameters ( $p>$ 0.05). All data The independent samples $t$ -test’s and Wilcoxon’s Test’s revealed no statistically significant differences in iEMG and sEMG neither in the proximal (Integral: $p<$ 0.897 and Peak: $p<$ 0.857) nor in the distal (Integral: $P<$ 0.25, Peak: $P<$ 0.33) regions between groups.

4. Discussion

The aim of the present study was to test whether regional muscle growth of the arm flexors in response to two different exercises was caused by a different myoelectric activation pattern of those regions. Our hypothesis was that the distal region would be activated to a greater extent during the PREA exercise, as this region grew more in the PREA group when compared to the INC group in a previous study [8]. However, no significant differences were found in the myoelectric activation pattern of any region that could explain the differences in muscle growth seen previously.

Even if the data from this study does not support our hypotheses, there are many possible explanations for the fact that the regions that grew more do not show a greater myoelectric activation. First, it must be taken into account that for technical reasons, the areas in which the EMG was measured were not the same as those that were measured in the previous study. In that study muscle growth was analyzed according to the protocol established by Matta et al. (11, 2011). This protocol consisted on measuring arm thickness at 50%, 60% and 70% of humerus length. As the average length for a woman’s humerus is about 25 cm, arm thickness was measured in spots separated by 2.5–3 cm. But the present study involved surface EMG, and to measure it electrodes could not be placed that close (for that reason the chosen position was 33% and 66% of humerus length). This may explain the current results, but on the other hand it must be highlighted that one of the sites analyzed in this study (66% of humerus length) is very similar to the 70% of humerus length that was measured in the Zabaleta-Korta et al. (2023) study. In fact, this particular site was the one that grew in response to the PREA exercise. That 4% difference between the position of both of them means that acquisition sites were 1 cm apart in both studies, which may not be enough to explain the lack of differences in the myoelectric activation pattern between exercises. However, it is difficult to measure the effect that this may have on our findings.

It is likely that the use of other technologies to measure regional EMG may mitigate this potential errors of measurement. Previously, other researchers have used MRI [12] and an array of 16 surface electrodes that can be positioned very close to each other [13] to identify regional differences in myoelectric activity of the quadriceps femoris and pectoralis major muscles, respectively. One of the reasons why our results are not consistent with muscle growth findings may be choosing a technology that is not appropriate for our purpose. For this reason, future research should focus on the use of more advanced technology in studies with similar designs. We already know the influence that exercise selection has on regional hypertrophy, but we still do not know which mechanisms are behind this phenomenon. EMG may help us filling the gap, but while the design of our study was appropriate for this purpose, clearly surface EMG was not a wise choice.

It should be added that the relationship between surface EMG and hypertrophy is complex. Since the 1970s, many authors have attempted to predict hypertrophy with surface EMG findings [14], but the relationship between surface EMG and longitudinal outcomes is not straightforward. For example, some studies show different degrees of muscle growth with similar levels of surface EMG [15], when one group performs an exercise at long muscle lengths and the other performs the same exercise at short muscle lengths. This is probably due to the intrinsic ability of the muscle to generate tension without any myoelectric activation, caused by passive elements such as titin. It must be added that changing muscle lengths also changes the amount of myoelectric activity needed to achieve a given force level. For example, an activation of 70% at a muscle length of 0.8 will produce similar force levels to an activation of 60% at a muscle length of 1 (arbitrary units, from Vigotsky 2018).

Our study is not the first attempt to demonstrate a relationship between regional hypertrophy and myoelectric activity. Wakahara and colleagues tested whether a large acute myoelectric activity found after triceps brachii training corresponded with the increase in Anatomical Cross Sectional Area after 12 weeks of elbow extensor training [16]. Authors report that the distal region that showed lower myoelectric activity than proximal and distal ones, grew less. However, it should be noted that the method they used to measure myoelectric activity is an indirect measurement that may not accurately represent it. In addition, the “longitudinal” part of their study was completed by very few participants and because they do not report effect sizes it is difficult to know which was the real effect of their intervention. Even if the study design is correct, more studies are needed to confirm their results and it is definitely not enough to state that a greater myoelectric activity can predict muscle hypertrophy.

The same authors performed a similar study two years later in which they tested whether choosing a multijoint exercise (dumbbell press) instead of a single joint exercise (french press) would elicit a different myoelectric activity and thus a different regional hypertrophy [9]. Their hypothesis was once again confirmed, as the medial region of the triceps brachii grew more than the proximal one, and also had a greater myoelectric activity. Even if this results are promising, the authors do not discuss the fact that the activity of the medial region was also superior to that of the distal one, while the latter did not grow less than the medial. In this second study, the authors again chose the T2 weighted magnetic resonance imaging (MRI) to measure the degree of myoelectric activity. The rationale to choose this method is its correlation with EMG [17, 18], the number of repetitions with a given load [19], and exercise intensity [17, 18]. This method allows to estimate intramuscular myoelectric activity, and also allows comparisons between regions that are very close to each other. However, we should not forget that it is an indirect measurement and that factors other than myoelectric activity are involved in hypertrophy [14]. Probably for this reason, these authors found that growth differences between heads of the quadriceps muscle could not be explained with myoelectric activity in a later study [12].

Not all the trials have been that successful [7, 12]. Although Wakahara and colleagues found once again that the regions with the greatest myoelectric activity of the rectus femoris were the ones that grew more, they also reported no relationships between the growth of the different heads of the quadriceps femoris and their myoelectric activity [12]. And following a similar design to our study, Earp and colleagues (2015) tried to test if their findings on a previous study [7] could be explained by regional myoelectric activity analysis [20]. The focus of the authors was to compare the effects of training for strength gains with the effects of training to improve power.

In their first study, they found that different training types elicited different regional hypertrophy, a finding that provides an interesting rationale to understand the functional role of regional hypertrophy, that seems not to be related to acute EMG measurements. However, the authors performed comparisons between regions, which may have confounded the results. Electrode placement is usually performed following guidelines from official institutions (e.g. SENIAM), because the electrode position itself can alter surface EMG findings. For that reason, choosing a different position can make some sense if data from two different exercises is compared for that given position, which is the case in our study, but comparing data from two different positions can influence the results [21]. Therefore, we can conclude that using intramuscular measurements is more appropriate to study myoelectric activity differences between regions.

There are many possible explanations for the regional hypertrophy found on our previous study. Earp and colleagues suggested that developing the area of the quadriceps femoris that is closest to the knee can reduce the moment of inertia of the thigh during knee extension which would help to jump faster [7]. The authors also mention that muscles may be divided in “neuromuscular packs”, divisions that are more suited to develop large amount of force (or power), and thus grow in response to specific types of training. Both explanations are feasible but the point in our study is that measuring surface EMG does not serve to understand why some regions grow more than others in response to certain types of training.

Our study faced some limitations: First, the size of the electrodes did not allow us to analyze the areas we were interested in. Second, participants had to perform the exercises in the laboratory, not at their usual gyms which may have influenced their performance.

5. Conclusions

The PREA exercise does not increase surface EMG of the distal region more than the INC exercise. Thus, possibly regional hypertrophy of the arm flexors cannot be explained by the myoelectric activity patterns of the different regions of the arm. Nevertheless, technical limitations may have influenced the findings of this study, which suggests that replicating this study with more advanced technology may serve to unveil the hypothetical role of myoelectric activation on the regional hypertrophy phenomenon.

Author contributions

CONCEPTION: AZK, EFP, JSC.

PERFORMANCE OF WORK: AZK, JTU, ULE.

INTERPRETATION OR ANALYSIS OF DATA: AZK, JTU, ULE.

PREPARATION OF THE MANUSCRIPT: AZK, EFP, JSC.

REVISION FOR IMPORTANT INTELLECTUAL CONTENT: AZK, EFP, JTU.

SUPERVISION: AZK, JSC, EFP.

Ethical considerations

This study was performed according to the guidelines of the declaration of Helsinki.

Funding

This is study was funded by the research group IT1726-22.

Footnotes

Acknowledgments

Authors have no acknowledgments.

Conflict of interest

The authors have no conflicts of interest to report.

References

Franchi

Ruoss

Valdivieso

Mitchell

Smith

Atherton

, et al. Regional regulation of focal adhesion kinase after concentric and eccentric loading is related to remodelling of human skeletal muscle. Acta Physiologica. 2018; 223(3).

Zabaleta-Korta

Fernández-Peña

Santos-Concejero

. Regional hypertrophy, the inhomogeneous muscle growth. Strength & Conditioning Journal. 2020; 42(5): 94-101.

Seynnes

De Boer

Narici

. Early skeletal muscle hypertrophy and architectural changes in response to high-intensity resistance training. Journal of Applied Physiology. 2007; 102(1): 368-73.

Watanabe

Kouzaki

Moritani

. Task-dependent spatial distribution of neural activation pattern in human rectus femoris muscle. Journal of Electromyography and Kinesiology [Internet]. 2012; 22(2): 251-8. Available from: doi: 10.1016/.

Staudenmann

Kingma

Daffertshofer

Stegeman

van Dieën

. Heterogeneity of muscle activation in relation to force direction: A multi-channel surface electromyography study on the triceps surae muscle. Journal of Electromyography and Kinesiology. 2009; 19(5): 882-95.

Zabaleta-Korta

Fernández-Peña

Torres-Unda

Garbisu-Hualde

Santos-Concejero

. The role of exercise selection in regional Muscle Hypertrophy: A randomized controlled trial. Journal of Sports Sciences. 2021; 20(39): 2298-304. doi: 10.1080/02640414.2021.1929736.

Earp

Newton

Cormie

Blazevich

. Inhomogeneous quadriceps femoris hypertrophy in response to strength and power training. Medicine and Science in Sports and Exercise. 2015; 47(11): 2389-97.

Zabaleta-Korta

Fernández-Peña

Torres-unda

Francés

Zubillaga

Santos-Concejero

. Regional Hypertrophy? The Effect of Exercises at Long and Short Muscle Lengths in Recreationally Trained Women. Journal of Human Kinetics. 2023; Ahead of p.

Wakahara

Fukutani

Kawakami

Yanai

. Nonuniform Muscle Hypertrophy: Its relation to muscle activation in training session. Medicine & Science in Sports & Exercise. 2013; 45(11): 2158-65.

10.

Stegeman

Hermens

. Standards for surface electromyography: The European project Surface EMG for non-invasive assessment of muscles (SENIAM) [Internet]. Línea). Available from: http://www. med.…2007; 108-12. Available from: http//www.seniam.org/%5Cnhttp://www.med.uni-jena.de/motorik/pdf/stegeman.pdf.

11.

Matta

Simao

Freitas de Salles

Spineti

Fernandes Oliveira

. Strength training’s chronic effects on muscle architecture parameters of different arm sites. Journal of Strength & Conditioning Research. 2011; 25(6): 1711-7.

12.

Wakahara

Ema

Miyamoto

Kawakami

. Inter- and intramuscular differences in training-induced hypertrophy of the quadriceps femoris: association with muscle activation during the first training session. Clinical Physiology and Functional Imaging. 2017; 37(4): 405-12.

13.

dos Santos Albarello

Cabral

Mendonça Leitao

Halmenschlager

Lulic-Kuryllo

Torres da Matta

. Non-uniform excitation of pectoralis major induced by changes in bench press inclination leads to uneven variations in the cross-sectional area measured by panoramic ultrasonography. Journal of Electromyography and Kinesiology. 2022; Ahead of p.

14.

Vigotsky

Halperin

Lehman

Trajano

Vieira

. Interpreting signal amplitudes in surface electromyography studies in sport and rehabilitation sciences. Frontiers in Physiology. 2018; 8.

15.

Noorkõiv

Nosaka

Blazevich

. Neuromuscular adaptations associated with knee joint angle-specific force change. Medicine and Science in Sports and Exercise. 2014; 46(8): 1525-37.

16.

Wakahara

Miyamoto

Sugisaki

Murata

Kanehisa

Kawakami

, et al. Association between regional differences in muscle activation in one session of resistance exercise and in muscle hypertrophy after resistance training. European Journal of Applied Physiology. 2012; 112(4): 1569-76.

17.

Adams

Duvoisin

Dudley

. Magnetic resonance imaging and electromyography as indexes of muscle function. Journal of Applied Physiology. 1992; 73(4): 1578-83.

18.

Fisher

Meyer

Addams

Foley

Potchen

. Direct relationship betweeen proton T2 and exercise intensity in skeletal muscle MR images. Investigative Radiology. 1992; 25(5): 480-5.

19.

Yue

Alexander

Laidlaw

Gmitro

Unger

Enoka

. Sensitivity of muscle proton spin-spin relaxation time as an index of muscle activation. Journal of Applied Physiology. 1994; 77(1): 84-92.

20.

Earp

Stucchi

Demartini

Roti

. Regional surface electromyography of the vastus lateralis during strength and power exercises. Journal of Strength and Conditioning Research. 2016; 30(6): 1585-91.

21.

Mesin

Merletti

Rainoldi

. Surface EMG: The issue of electrode location. Journal of Electromyography and Kinesiology [Internet]. 2009; 19(5): 719-26. Available from: doi: 10.1016/j.jelekin.2008.07.006.