Comparison of Muscle-Activation Patterns During the Conventional Push-Up and Perfect· Pushup™ Exercises

Abstract

Youdas, JW, Budach, BD, Ellerbusch, JV, Stucky, CM, Wait, KR, and Hollman, JH. Comparison of muscle-activation patterns during the conventional push-up and perfect·pushup™ exercises. J Strength Cond Res 24(12): 3352-3362, 2010-Manufacturers of Perfect·Pushup™ handgrips claim enhanced muscular recruitment when compared with the conventional hand-on-floor push-up exercise. Electromyographic (EMG) data were recorded using surface electrodes from the right-sided triceps brachii, pectoralis major, serratus anterior (SA), and posterior deltoid muscles during push-ups performed from 3 different hand positions: (a) shoulder width, (b) wide base, and (c) narrow base (NB). Push-ups were performed under 2 conditions: (a) standard push-up and (b) Perfect·Pushup™ handgrips. We recruited 20 healthy subjects, 11 men (24.9 ± 2.6 years) and 9 women (23.8 ± 1 years). Subjects completed 3 consecutive push-ups for each hand position under both conditions. Push-up speed was controlled using a metronome, and testing order was randomized. We recorded peak EMG activity for each muscle during each of the push-ups and normalized EMG values by maximum muscle contractions (% maximum voluntary isometric contraction [MVIC]). Electromyographic data were analyzed with 3 (hand position) × 2 (condition) repeated-measures analysis of variance with a post hoc Bonferroni-adjusted simple effects test to detect significant position effects for position by condition interactions (α = 0.05). A push-up required considerable muscle activation of the triceps brachii (73-109% MVIC), pectoralis major (95-105% MVIC), SA (67-87% MVIC), and posterior deltoid (11-21% MVIC) whether performed using the conventional hand-on-floor position or the Perfect·Pushup™ handgrips. The NB hand position was most effective for preferentially activating the triceps brachii and posterior deltoid muscles. Based upon EMG activation from 4 muscles, the Perfect·Pushup™ handgrips do not appear to preferentially enhance muscular recruitment when compared with the conventional push-up method.

Introduction

The conventional push-up is a popular method for assessing a person's muscular endurance or as a way to improve muscle performance of the shoulder, arm, and trunk. The American College of Sports Medicine (¹) advocates using the push-up as a test procedure for measurement of muscular endurance. Similarly, the Army Physical Fitness Test assesses an individual's fitness level using a variety of tests including the conventional push-up (²¹). Using electromyographic (EMG) procedures, investigators have documented the push-up as an effective method for activating muscles of the shoulder, arm (^{7,15,16,22-24,31}), and trunk (^{4,14,18,20,21,25}). The push-up's popularity (¹⁵) for assessing muscle endurance or upper extremity strength presumably arises because it is simple to learn, does not normally involve any equipment, and can be modified to fit various levels of physical fitness. Manufacturers of novel strength training devices often advertise that their product will produce improved results when compared with conventional strength training procedures (⁵). Unlike conventional push-up bars whose handles remain stationary, the new Perfect·Pushup™ handgrips are designed to rotate via ball bearings on a stationary base. Recently, the originator of Perfect·Pushup™ handgrips claimed that the device's distinctive rotation handles move with the arm's natural motion to improve strength, balance, flexibility, and endurance, while reducing risk of muscle strain and injury (⁵). Presently, no one has investigated these claims.

At first glance, the conventional push-up with hand-on-floor appears to be a straightforward exercise to perform, although, the mechanical demands placed upon the muscles of the shoulder, arm, and trunk may be modified by simply changing the standard shoulder width (SW) position of the weight-bearing hands to either a wide base (WB) or narrow base (NB) position. Donkers et al. (¹²) studied the mechanical demands created at the elbow joint during hand-on-floor push-ups from SW, WB, and NB hand positions. They discovered that elbow flexion torque is greatest when the push-up is performed from the NB hands together position. Although Donkers et al. (¹²) did not examine muscle-activation patterns at the shoulder or elbow, they did confirm that biomechanical differences exist at the elbow when hand position is varied. More recently, 2 independent studies used surface EMG to examine the influence of hand position on muscle activation of the triceps brachii and pectoralis major muscles during a conventional push-up. Cogley et al. (⁷) studied 3 hand positions (SW, WB, and NB) during the push-up exercise in 40 (11 men and 29 women) healthy young adults, whereas Gouvali and Boudolos (¹⁵) examined 6 different hand postures that they termed exercise variants on 8 healthy men. Both investigators reported that if an exercise specialist desires to create greater muscle activation of the triceps brachii and pectoralis major during the push-up, then he or she should advise a client to perform the push-up in an NB position-shoulder adduction compared with WB position-shoulder abduction. The studies by Cogley et al. (⁷) and Gouvali and Boudolos (¹⁵) were the first to substantiate statements in the popular fitness literature that advocated that hand position can influence myoelectric activity of arm and shoulder muscles during a push-up. Nevertheless, the investigators only examined 2 muscle groups, so there is a scarcity of information describing muscle activation patterns for scapular stabilizers and humeral internal and external rotators during varied hand positions of the conventional push-up. Furthermore, no one has reported how the Perfect·Pushup™ handgrips would influence muscle-activation patterns of the shoulder and arm muscles during varied hand placement positions during the push-up exercise. Because the Perfect·Pushup™ handgrips elevate the user's hands from the surface of the floor when compared with the conventional hand-on-floor push-up position, the shoulder and elbow joints may move through a greater range of motion during the lowering and raising of the chest. Therefore, the Perfect·Pushup™ handgrips could create a greater external demand on the muscles of the shoulders and arms resulting in greater EMG activation than a conventional hand-on-floor push-up. In summary, muscle activation data obtained from this study would enhance the exercise professional's clinical decision making when prescribing variants of the push-up for strengthening selected muscles of the arm and shoulder girdle.

The purpose of this study was to examine muscle activation of the triceps brachii, pectoralis major, serratus anterior (SA), and the posterior deltoid muscles during push-ups performed from 3 selected hand positions: SW, WB, and NB using both Perfect·Pushup™ handgrips and the conventional hand-on-floor push-up method. We hypothesized that push-ups, both conventional hand-on-floor and those using the Perfect·Pushup™ handgrips, performed from the NB hand position would produce significantly greater EMG activation of the triceps brachii, pectoralis major, SA, and posterior deltoid muscles than either SW or WB hand positions. We also hypothesized that the use of the Perfect·Pushup™ handgrips in all 3 hand positions would significantly increase the EMG activation of the triceps brachii, pectoralis major, SA, and posterior deltoid muscles when compared with a conventional hand-on-floor push-up.

Methods

Experimental Approach to the Problem

Electromyography is a valuable device because it allows a fitness professional to record motor unit activity within a muscle or group of muscles and observe differences in muscle activity between exercise conditions (^11,13). Investigators assume that exercises that produce higher EMG signal amplitudes also generate larger strengthening effects (³). Traditionally, the hand-on-floor push-up is performed with the hands in an SW position. Two variants of the standard push-up include raising and lowering the chest with the hands in a WB or NB position. In this study, we examined which hand position generated the greatest EMG activity from triceps brachii, pectoralis major, SA, and posterior deltoid muscles during performance of a typical push-up vs. using the Perfect·Pushup™ handgrips. Electromyographic data were collected with surface electrodes affixed to the skin according to Cram and Kasman (⁸), processed with the root mean square algorithm, and normalized to a maximum voluntary isometric contraction (MVIC). We used a 2-factor repeated-measures design to test our null hypothesis that the EMG recruitment pattern of the triceps brachii, pectoralis major, SA, and posterior deltoid muscles is equivalent between the conventional push-up and the Perfect·Pushup™ handgrips. The independent variables in this study included 2 separate push-up conditions, the Perfect·Pushup™ handgrips and the conventional push-up, and 3 different hand conditions: SW, WB, and NB. The dependent variable was the normalized EMG value of the triceps brachii, pectoralis major, SA, and posterior deltoid muscles. Testing order was randomized to reduce potential order threats to the study's internal validity. Specifically, the study design attempted to answer the following research question: “Do the Perfect·Pushup™ handgrips show increased EMG activation patterns across the 3 hand positions when compared with the conventional hand-on-floor push-up?”

Subjects

Twenty healthy subjects, 11 men (mean age ± SD = 24.9 ± 2.6 years; mean body mass ± SD = 79.1 ± 13.4 kg; mean height ± SD = 182.4 ± 7.2 cm; mean body mass index [BMI] ± SD = 23.7 ± 3.2 kg·m⁻²) and 9 women (mean age ± SD = 23.9 ± 1.1 years; mean body mass ± SD = 58.4 ± 7.1 kg: mean height ± SD = 164.4 ± 6.3 cm; mean BMI ± SD = 21.5 ± 1.7 kg·m⁻²) volunteered to participate in this study. A sample size of 20 subjects permitted us to detect an absolute mean difference in EMG recruitment of 15%- normalized to the MVIC-between conditions while protecting against type II error with a statistical power (1 − β) of 0.80 at α = 0.05. Subjects comprised a sample of convenience and were recruited from the graduate program in physical therapy at the Mayo School of Health Sciences in Rochester, MN. Body mass index was calculated using data obtained from self-reported height and body mass. Subjects included in the study demonstrated normal active range of motion (ROM) of the shoulder, elbow, wrist, and hand. During the screening process, each subject demonstrated the ability to correctly perform 4 consecutive conventional push-ups with a rigid spinal posture for each of the 3 hand positions. At the time the study was completed, 50% (n = 10) of the subjects self-reported that they were currently engaged in a weight training program that included some form of chest press exercise (bench, incline, or decline) using either a barbell or dumbbells. Furthermore, 60% (n = 12) of the subjects also reported that they were performing standard push-ups at least 2-3 times per week. Volunteers reporting a history of the following upper extremity conditions were excluded from the study: (a) previous subluxation, dislocation, or fracture; (b) a history of joint instability, tendinitis, bursitis, impingement, adhesive capsulitis, neurovascular complications, or any condition that limited physical activity for greater than 2 days over the last 6 months; and (c) current complaints of neuromuscular pain, numbness, or tingling in the upper extremity, neck, or back during the push-up screening process. The study procedures were approved by the Mayo Institutional Review Board, Mayo Clinic, Rochester, MN. All subjects read and signed an approved inform consent form before their participation in the study.

Instrumentation

Raw EMG signals were collected with D-100 bipolar surface electrodes (Therapeutics Unlimited, Inc., Iowa City, IA). The active Ag-AgCl electrodes have an interelectrode distance of 22 mm and are encased within preamplifier assemblies measuring 35 × 17 × 10 mm. The preamplifiers had a gain of 35. The combined preamplifier and main amplifier permitted a gain of 100 to 10,000 with a bandwidth of 40 Hz to 6 kHz. The common mode rejection ratio was 87 dB at 60 Hz, and input impedance was greater than 15 MΩ at 100 Hz. Data were collected at a sampling frequency of 1,000 Hz. Raw EMG signals were processed with WinDaq data acquisition software (DATAQ Instruments, Inc., Akron, OH). A metronome set at 60 Hz was used to regulate the cadence of the push-up.

Procedure

Each subject's skin was prepared by vigorously rubbing the electrode attachment site area with an alcohol wipe. After preparing the subject's skin, the electrode preamplifier assemblies were attached with double-sided, padded adhesive tape. The tape had wells that were aligned with the electrodes, in which conductive gel (Signa Creme® Electrode Cream, Parker Laboratories, Inc.) was used to conduct the myoelectric signal to the electrode. The electrodes were placed parallel to the line of action of the triceps brachii, pectoralis major, SA, and posterior deltoid muscles on the right side of each subject. The triceps brachii electrode was positioned at the midpoint between the posterior aspect of the acromion and the olecranon processes. The pectoralis major electrode was placed at the midpoint of the distance between the sternal notch and the axillary fold, whereas the SA electrode was positioned just anterior to the border of the latissimus dorsi muscle at the level of the inferior tip of the scapula. The electrode for the posterior deltoid was placed 2 cm below the lateral border of the spine of the scapula and angled obliquely toward the deltoid tuberosity. The ground electrode was positioned over the ulna immediately proximal to the styloid process. These methods are consistent with the procedures described by Cram and Kasman (⁸).

Once the electrodes were placed, gross MVIC was assessed for all muscles in a seated position to calibrate the amplifier and avoid saturation of EMG signals. After this, subjects performed 2 standard push-ups to again check for signal saturation. Once calibration was completed, a 5-second MVIC for each of the 4 muscles was determined using manual muscle-testing techniques described by Hislop and Montgomery (¹⁷).

Testing order was randomized to reduce threats to the study's internal validity. The start position for both push-up exercise conditions began with the chest wall elevated from the floor, spine straight, and shoulders flexed 90° relative to the trunk's longitudinal axis. The exercise was initiated with controlled lowering of the trunk (eccentric phase) until each subject's chest wall at the level of his or her hands reached a distance of 8-10 cm from the floor. This distance was visually estimated by the examiner. At this point, the eccentric phase was completed and the up or concentric phase of the push-up began. A subject was required to repeat the push-up if the examiner felt he or she did not descend to the correct depth. Conventional push-ups are presented in Figures 1-3 and were performed with the subject's forearms pronated, wrists and fingers extended, and palms on the floor and fingers facing forward. The Perfect·Pushup™ device consisted of a soft cell-foam handle, 11.5 cm in height mounted to a circular 18.8-cm diameter, nonslip rotating base. The SW (Figure 4) and NB (Figure 5) exercise using the Perfect·Pushup™ handgrips began in the up position with forearms pronated, wrists neutral, and fingers flexed. In the completed or down position, the subject's forearms and wrists were in the neutral position and fingers flexed. The WB exercise (Figure 6) in the up position with Perfect·Pushup™ handgrips was begun with the subjects' forearms and wrists in neutral position and fingers flexed, whereas in the completed or down position, the subject's forearms were pronated, wrists in neutral position, and fingers flexed. Hand placement was normalized to each individual. The SW hand position for the conventional push-up (Figure 1) and Perfect·Pushup™ (Figure 4) was determined by instructing the subject to assume a quadruped position by placing his or her hands on the floor with the third digit in line with the acromion process. An examiner then recorded the distance in centimeters between the tips of right and left third digits, which represented the value for the SW hand position. The WB hand position for the conventional (Figure 2) and Perfect·Pushup™ (Figure 6) was designated by marks placed 50% wider and spaced equally, on either side of the SW hand position. The NB hand position for the conventional (Figure 3) and Perfect·Pushup™ (Figure 6) was measured to be 50% narrower, spaced equally from the SW hand position. Marks were then made on the floor for the subjects to align their third digits with when performing the NB push-up. Subjects completed 3 consecutive push-ups for each hand position in both the conventional and Perfect·Pushup™ handgrip conditions. In an attempt to standardize the cadence at which the volunteers performed the push-up movements, a 2-second rate for descent and ascent of an individual push-up cycle was maintained by a 60-Hz metronome.

Figure 1: Conventional push-up performed at the shoulder width (SW) hand position where the third digit of each hand is positioned directly inferior to each shoulder's acromion process. (A) Start or up position for the push-up and (B) down position. An examiner measured and marked the distance in centimeters between the tips of right and left third digits for the SW hand position on a piece of tape.

Figure 2: Conventional push-up performed at the wide base (WB) hand position. (A) Start or up position for the push-up and (B) down position. An examiner measured and recorded the distance in centimeters between the tips of right and left third digits for the shoulder width (SW) hand position. Using a piece of tape, the WB hand position for the conventional push-up is designated by marks placed 50% wider and spaced equally on either side of the SW hand position.

Figure 3: Conventional push-up performed at the narrow base (NB) hand position. (A) Start or up position for the push-up and (B) down position. An examiner measured and recorded the distance in centimeters between the tips of right and left third digits for the shoulder width (SW) hand position. Using a piece of tape, the NB hand position for the conventional push-up is designated by marks placed 50% narrower and spaced equally, on either side of the SW hand position.

Figure 4: Perfect·Pushup™ performed at the shoulder width (SW) hand position where the third digit of each hand is positioned directly inferior to each shoulder's acromion process. (A) Start or up position for the push-up and (B) down position. Each handgrip is aligned with a mark placed on a piece of tape that corresponds with the distance in centimeters between the tips of right and left third digits. The SW exercise using the Perfect·Pushup™ handgrips begins in the up position with forearms pronated, wrists neutral, and fingers flexed. In the completed or down position, the subject's forearms and wrists are in the neutral position and fingers flexed.

Figure 5: Perfect·Pushup™ performed at the narrow base (NB) hand position. (A) Start or up position for the push-up and (B) down position. Each handgrip is aligned with a mark placed on a piece of tape that corresponds to the distance in centimeters that is 50% narrower and spaced equally, on either side of the shoulder width hand position. The NB exercise using the Perfect·Pushup™ handgrips begins in the up position with forearms pronated, wrists neutral, and fingers flexed. In the completed or down position, the subject's forearms and wrists are in the neutral position and fingers flexed.

Figure 6: Perfect·Pushup™ performed at the wide base (WB) hand position. (A) Start or up position for the push-up and (B) down position. Each handgrip is aligned with a mark placed on a piece of tape that corresponds with the distance in centimeters that is 50% wider and spaced equally, on either side of the shoulder width hand position. The WB exercise using the Perfect·Pushup™ handgrips begins in the up position with subjects' forearms and wrists neutral and fingers flexed, whereas in the completed or down position subject's forearms are pronated, wrists neutral, and fingers flexed.

Data Processing

Electromyographic signals were processed with the root mean square algorithm at a time constant of 55 milliseconds. Peak EMG activity recorded during the up or concentric phase of each test condition was normalized to the muscle's respective peak activity level in the MVIC trial and therefore expressed as a percentage of MVIC.

Reliability of Electromyographic Measures

We did not determine the test-retest reliability of the normalized surface EMG data for the triceps brachii, pectoralis major, SA, and posterior deltoid muscles during the 2 push-up conditions and 3 hand positions. Nevertheless, previous investigators reported clinically acceptable test-retest EMG values. Cappato de Araújo et al (⁶) assessed the normalized EMG activation values of the triceps brachii, pectoralis major, SA, and posterior deltoid muscles during a conventional push-up using the intraclass correlation coefficient (ICC_2,1). ICC values demonstrated excellent intraday consistency for normalized (ICC 0.78-0.99) values of the upper limb muscles. On the basis of this information, we believe our EMG activation data would be reliable because we used similar electrode placement and skin preparation procedures as did Cappato de Araújo et al (⁶).

Statistical Analyses

Descriptive statistics-including means and SDs-for EMG recruitment in the triceps brachii, pectoralis major, SA, and posterior deltoid during the push-ups were calculated. The magnitudes of EMG recruitment were analyzed with 3 × 2 (3 hand positions by 2 push-up conditions) repeated-measures analyses of variance (ANOVAs) for each of the muscles included in the study (α = 0.05). Post hoc simple effects tests with the Bonferroni-adjusted α were used to control α for multiple comparisons and to analyze pairwise comparisons in EMG recruitment between hand positions and between push-up conditions when either the main effects or the position by condition interactions were significant. Greenhouse-Geiser adjustments for degrees of freedom were applied if assumptions of sphericity were violated. All data were analyzed with SPSS 15.0 for Windows software (SPSS Inc, Chicago, IL).

Results

Triceps Brachii

There was no difference in EMG activation between push-up conditions; however, the effect of hand position was statistically significant (F (2,36) = 13, p < 0.001 [Figure 7]). Post hoc Bonferroni-adjusted simple effects tests revealed the following significant pairwise differences in triceps brachii muscle activation because of hand position: (a) NB hand position greater than the SW hand position (mean difference = 18% MVIC, p = 0.001, effect size = 0.35); (b) NB hand position greater than the WB hand position (mean difference = 33% MVIC, p = 0.003, effect size = 0.69); and (c) SW hand position greater than WB hand position (mean difference = 15% MVIC, p = 0.04, effect size = 0.33).

Figure 7: Mean normalized electromyographic (EMG) activity in the triceps brachii muscle obtained during push-ups performed from the shoulder width (SW), wide base, and narrow base (NB) hand positions for both the conventional push-up and a push-up performed with the Perfect·Pushup™ handgrips. There is no difference between push-up conditions. The EMG activity is significantly greater in the NB hand position compared with the SW (p = 0.001) and wide base hand position (p = 0.003). The SW hand position (p = 0.04) is significantly greater than the wide base hand position.

Pectoralis Major

Because the assumption of sphericity was violated, we used the Greenhouse-Geisser adjustment for the degrees of freedom. A significant position by condition interaction was detected (F (2,19) = 3.9, p = 0.049) although the Bonferroni post hoc comparison showed no significant difference in the pairwise comparison of hand position and push-up condition (Figure 8). The most relevant finding was greater activation of the pectoralis major in the Perfect·Pushup™ condition than the conventional push-up in the WB hand position (mean difference = 9.9% MVIC, p = 0.065, effect size = 0.18).

Figure 8: Mean normalized electromyographic (EMG) activity in the pectoralis major muscle obtained during push-ups performed from the shoulder width, wide base, and narrow base hand positions for both the conventional push-up and a push-up performed with the Perfect·Pushup™ handgrips. There is no difference in EMG activity between hand position and push-up condition.

Serratus Anterior

A statistically significant (F (2,36) = 7.3, p = 0.002) hand position effect was detected. For the conventional push-up condition, Bonferroni post hoc comparisons revealed differences in EMG activation (Figure 9) between the SW and NB hand positions (mean difference = 16.4% MVIC, p = 0.003, effect size = 0.40) and the WB and NB hand position (mean difference = 14.9% MVIC, p = 0.04, effect size = 0.42). During the Perfect·Pushup™ condition, there was a statistically significant difference between the SW and WB hand positions (mean difference = 19.9% MVIC, p = 0.012, effect size = 0.53). An interaction effect was detected between hand position and push-up condition (F (2,36) = 5.1, p = 0.012 [Figure 9]). The conventional push-up showed greater EMG activation than the Perfect·Pushup™ in the SW hand position (mean difference = 19.7% MVIC, p = 0.004, effect size = 0.52).

Figure 9: Mean normalized electromyographic (EMG) activity in the serratus anterior muscle obtained during push-ups performed from the shoulder width (SW), wide base, and narrow base (NB) hand positions for both the conventional push-up and a push-up performed with the Perfect·Pushup™ handgrips. For the conventional push-up condition, there is greater EMG activation in the SW than in the NB hand position (p = 0.003) and the wide base (WB) than the NB hand position (p = 0.04). During the Perfect·Pushup™ condition, there is a statistically significant difference between the SW and WB hand positions (p = 0.012). The conventional push-up has greater EMG activation than the Perfect·Pushup™ in the SW hand position (p = 0.004).

Posterior Deltoid

The effect of hand position was statistically significant (F (2,36) = 6.7, p = 0.003). Bonferroni post hoc comparisons indicated greater EMG activation of the posterior deltoid in the NB vs. the SW hand position (mean difference = 5.2% MVIC, p = 0.039, effect size = 0.40 [Figure 10]) and the NB vs. the WB hand position (mean difference = 5.3% MVIC, p = 0.012, effect size = 0.44 [Figure 10]). The effect of push-up condition was also significant (F (1,18) = 13, p = 0.002). Greater EMG activation of the posterior deltoid was detected in the conventional push-up when compared with the Perfect·Pushup™ condition (mean difference = 4.4% MVIC, p = 0.005, effect size = 0.39 [Figure 11]).

Figure 10: Mean normalized electromyographic (EMG) activity in the posterior deltoid muscle obtained during push-ups performed from the shoulder width (SW), wide base, and narrow base (NB) hand positions for both the conventional push-up and a push-up performed with the Perfect·Pushup™ handgrips. There is significantly greater EMG activation of the posterior deltoid in the NB vs. the SW hand position (p = 0.04) and the NB vs. the wide base hand position (p = 0.012).

Figure 11: Mean normalized electromyographic (EMG) activity in the posterior deltoid muscle obtained during push-ups. The effect of push-up condition is statistically significant. Greater EMG activation of the posterior deltoid occurs in the conventional push-up (p = 0.005) when compared with the Perfect·Pushup™ condition.

Discussion

In this study, we examined the EMG activity in 4 muscles required to perform a conventional push-up and a push-up using the Perfect·Pushup™ handgrips. It has been demonstrated by numerous authors that the push-up is an effective method to activate shoulder, arm, and trunk musculature (^{7,15,16,22-24,31}). Strengthening effects are believed to be enhanced by the push-up exercise that yields the greatest amount of EMG activity. Hand position has been shown to influence myoelectric activity of the shoulder and arm muscles during a push-up (^7,15). Recently, according to the manufacturer of the Perfect·Pushup™ handgrips (⁵), a push-up performed with this new device would enhance glenohumeral joint stability and muscular recruitment when compared with the conventional hands-on-floor push-up. Using EMG analysis, we sought to examine the claim of increased muscular recruitment made by the manufacturer. Additionally, we attempted to substantiate the results reported by Cogley et al. (⁷) and Gouvali and Boudolos (¹⁵) whereby hand position impacted EMG activity of arm and shoulder muscles during the performance of a push-up. Electromyography analysis of muscle activation is not intended to provide a direct of measure muscle strength; instead, it is a snapshot of motor-unit recruitment beneath the active electrode during muscle contraction. Muscle recruitment can be used to demonstrate which hand position and push-up condition placed the highest external demand upon a muscle. On the basis of EMG activation data, DiGiovine et al. (¹⁰) initially classified relative muscular demand into 4 categories: low (<20% MVIC), moderate (20-40% MVIC), high (41-60% MVIC), and very high (>60% MVIC). Moreover, the scheme proposed by DiGiovine et al. (¹⁰) agreed with that of other investigators who contended the threshold value for promoting muscle-strength gains during therapeutic exercise required muscle activation greater than 50-60% MVIC (^2,3,26). Therefore, we chose to use the EMG signal amplitude (% MVIC) as a method to discuss the relative exercise intensity of the shoulder and arm muscles when comparing push-ups performed from 3 selected hand positions using both Perfect·Pushup™ handgrips and the conventional hand-on-floor push-up method.

The triceps brachii muscle is the primary extensor of the forearm at the elbow joint, and because the long head of the muscle has its proximal attachment on the scapula, the triceps brachii also extends the arm across the glenohumeral joint (^19,27). Our first research hypothesis was confirmed because the triceps brachii had significantly greater EMG activation in the NB hand position when compared with both SW and WB hand positions. Electromyography results from our study are in accord with findings of Donkers et al. (¹²) who studied the conventional push-up in 9 healthy young men and reported external elbow flexion torques across the elbow joint were 71% of maximum isometric torque in the NB hand position, 56% of maximum isometric torque in the SW position, and 29% of maximum isometric torque in the WB position. Peak muscle activation of the triceps brachii in the NB hand position during the push-up occurred in response to the greatest external demand created across the elbow joint. Similarly, Cogley et al. (⁷) and Gouvali and Boudolos (¹⁵) also reported peak values of triceps brachii muscle activation when the conventional push-up was performed from an NB position. On the other hand, in contrast to Cogley et al. (⁷) and Gouvali and Boudolos (¹⁵), we discovered that triceps brachii muscle activation was significantly greater in the SW position when compared with the WB hand position. Such a finding is not surprising because Donkers at al. (¹²) reported a 27% greater elbow flexion torque demand on the triceps brachii during a push-up performed in the SW hand position when compared with the WB hand position. Lastly, we found no significant interaction effects between hand position and push-up condition. This information opposed our second research hypothesis that predicted Perfect·Pushup™ handgrips in all 3 hand positions would significantly increase the EMG activation of the triceps brachii muscle when compared with a conventional hand-on-floor push-up. Nonetheless, despite the lack of a statistical difference in EMG activation between push-up conditions, both the conventional hand-on-floor push-up and the push-up performed with Perfect·Pushup™ handgrips generated triceps brachii EMG activation ratios that exceeded 60% MVIC. Muscle strength gains of the triceps brachii would be expected when performing upper extremity weight-bearing exercises using a conventional push-up or Perfect·Pushup™ handgrips (^2,3,10,11,26).

The pectoralis major muscle is recognized as a horizontal adductor and internal rotator of the arm (^19,27). Our electrode placement attempted to maximize recording of both clavicular and sternocostal segments involved in the motion of horizontal arm adduction. In contrast to Cogley et al. (⁷), who reported a significant increase in EMG activation of the pectoralis major in the NB hand position during a conventional push-up, our results failed to show muscle activation of the pectoralis major was differentially influenced by a change in hand position during both the hand-on-floor push-up and the push-up performed with Perfect·Pushup™ handgrips. Results from the present study contradict our first research hypothesis that pectoralis major EMG activity would have its greatest amplitude when push-ups are performed from the NB hand position. Similarly, data from the present study also challenged our second research hypothesis that predicted muscle activation of the pectoralis major would be greater during push-ups performed with the Perfect·Pushup™ handgrips when compared with the standard hand-on-floor position. Both the conventional hand-on-floor push-up and the push-up performed with Perfect·Pushup™ handgrips generated pectoralis major EMG activation ratios that exceeded 80% MVIC. Muscle strength gains of the pectoralis major would be expected when performing upper extremity weight-bearing exercises using a conventional push-up or Perfect·Pushup™ handgrips (^2,3,10,11,26).

The SA muscle is an effective upward rotator of the scapula because of its large internal moment arm (²⁷). Furthermore, the SA also protracts the scapula and stabilizes or fixes the scapula's medial border against the chest wall (^19,27) during arm movements. For the conventional push-up condition, our results revealed increased SA muscle activation in both the SW and WB hand positions when compared with the NB hand position, whereas in the Perfect·Pushup™ condition, we found significantly greater activation of the SA in the WB position when compared with the SW position. Hand-position data for both push-up conditions contradicts our first hypothesis because muscle activation of the SA was the least in the NB hand position. Although we did not perform a kinematic analysis of scapular motion during the 3 hand positions over the 2 conditions, we believe greater scapular protraction range of motion and SA muscle activation would occur in the SW and WB hand positions compared with the NB position. Several investigators have previously reported muscle activation of the SA in healthy subjects during a conventional push-up performed with the hands at SW, and their findings were similar to our results (SW = 86% MVIC). The least SA activity (69% MVIC) was reported by both Moseley et al. (²⁵) and Tucker et al. (³⁰), whereas SA activation values of 80 and 100% MVIC were reported by Ludewig et al. (²³) and Decker et al. (⁹), respectively. Clearly, the conventional push-up performed at the SW hand position elicited a substantial amount of SA muscle activity. With respect to the second research hypothesis, our results failed to show that the use of Perfect·Pushup™ handgrips in all 3 hand positions would significantly increase EMG activation of the SA muscle when compared with a conventional hand-on-floor push-up. EMG activation of the SA was comparable between the 2 push-up conditions for the WB and NB hand positions, although the conventional push-up was more active (86% MVIV) than the Perfect·Pushup™ handgrips (67% MVIC) during the SW hand position. Both the conventional hand-on-floor push-up and the push-up performed with Perfect·Pushup™ handgrips generated SA EMG activation ratios that exceeded 60% MVIC. Hence muscle-strength gains of the SA would be expected when performing upper extremity weight-bearing exercises using a conventional push-up or Perfect·Pushup™ handgrips (^2,3,10,11,26).

The posterior deltoid is an extensor and adductor of the arm at the glenohumeral joint (^19,27). Of the 4 muscles examined in the present study, the posterior deltoid demonstrated the least amount of muscle activation (% MVIC). Greater activation within the posterior deltoid occurred during the NB hand position (18% MVIC) compared with both SW and WB (13% MVIC) hand positions. These data confirm our first research hypothesis where activation of the posterior deltoid (% MVIC) within each push-up condition was significantly greater in the NB hand position when compared with both WB and SW hand positions. In the NB hand position during the concentric phase of a push-up, the humeral head is exposed to a posterior translation force because of the external shoulder extension moment created by the reaction force of the floor-on-the weight-bearing hand. In response to this external moment of force, the posterior deltoid would presumably be activated to help stabilize the glenohumeral joint (²⁸). Our level of posterior deltoid muscle activation during the conventional push-up position at the SW hand position (13% MVIC) was similar to the value (18% MVIC) reported by Uhl et al. (³¹). However, in contrast to the present study, subjects in the Uhl et al. (³¹) investigation did not perform a dynamic push-up; instead, they assumed the push-up position with elbows in full extension and shoulders flexed to 90°. Lastly, greater posterior deltoid muscle activation (17% MVIC) was observed in the conventional push-up vs. the Perfect·Pushup™ (12.7% MVIC) condition. This information contradicted our second research hypothesis. Perhaps the rotational component of the arm and forearm associated with the Perfect·Pushup™ handgrips placed the shoulder joint in a more efficient posture requiring less activation of stabilizing muscles, including the posterior deltoid. Based upon muscle activation data from the present study, if a fitness professional desired to prescribe an exercise to selectively strengthen the posterior deltoid muscle, then he or she should refrain from using a push-up and select a non-weight-bearing exercise (²⁹) that required horizontal arm abduction (93% MVIC).

Results of this investigation should be interpreted with caution because of 4 study limitations. The primary limitation of this study was the absence of 3D kinematic measurements of the arm and forearm motion during each push-up condition at the 3 specific hand positions. Such information would be particularly helpful to fitness professionals when prescribing upper extremity weight-bearing exercise to promote shoulder and elbow muscle activation and range of motion. Second, because our EMG amplifier had a limited number of channels, we elected to only examine shoulder muscle activation. However, future research is warranted to investigate the influence of hand position on activation of the intrinsic back muscles and abdominal muscles during the performance of a conventional push-up and one with the Perfect·Pushup™ handgrips. Third, data from the present study may have limited external validity because our volunteers were lean, healthy, physically fit young men and women capable of performing 3 successive repetitions of a conventional hand-on-floor push-up at 3 different hand positions. Lastly, despite the utility of EMG, there are limitations to the use of EMG as the principle indicator of muscle function. Crosstalk may exist particularly when using surface electrodes (¹¹). For example, we presume the EMG signal obtained with electrodes over the posterior deltoid came solely from the posterior deltoid. However it may be possible that muscle activity from either the long head or lateral head of the triceps brachii or both contributed to the EMG signal because these extensor muscles of the arm are in close proximity to the posterior deltoid. Nevertheless, we minimized the potential for crosstalk and movement artifact by using standardized techniques for surface electrode placement.

Practical Applications

The push-up is a well-known exercise designed to generate muscular activity in the shoulders and arms. Therefore, the result of hand position on muscle activity would be valuable information to any practitioner who wanted to recommend the push-up as a form of upper extremity strength training. Results of our study indicated the NB hand position should be used if triceps brachii or posterior deltoid strengthening is desired. In contrast, hand position lacked a differential effect on muscle activation of the pectoralis major, whereas the SW position was optimal for activating the SA muscle. Based upon normalized EMG activity alone, the Perfect·Pushup™ handgrips are not preferable to a conventional hand-on-floor push-up when the goal is to produce muscle strengthening of shoulder and arm muscles using weight-bearing exercise. Regardless of hand position, both push-up conditions generated EMG activation ratios that exceeded the 60% MVIC threshold value for promoting muscle strength gains in the triceps brachii, pectoralis major, and SA muscles (^{2,3,10,11,13,26}).

Acknowledgments

Funding for this project came from the Mayo Program in Physical Therapy, Mayo Clinic, Rochester, MN. The authors disclose they have no professional relationship with the company that manufactures the Perfect·Pushup™ device. The results of the present study do not constitute endorsement of the Perfect·Pushup™ device by the authors or the NSCA.