Today an article that I co-wrote with my friend Elsbeth Vaino appeared on TNation. Here’s a link to the article, it’s called, “The Best Damn Pushups Article, Period!“
The TNation editors always do a great job of sprucing up my articles and discarding what’s not necessary. I always make my articles too sciency, but for those interested in a slightly more comprehensive lit review on pushups than what was found in the article, here you go:
Pushup Literature Review
An et al. (1990) found that how you set up for the pushup affected the relative joint loading during the exercise’s movement. The three main factors that influenced the intersegmental joint loads during pushups were the location of the palms relative to the shoulder joint, the plane of the arm movement, and the relative foot positions. Furthermore, the researchers found that the speed of pushup movement affected the inertial load.
An et al. (1992) found that peak axial forces on the elbow joint during pushups averaged around 45% of bodyweight. This increases to 75% in the case of narrow base pushups. This means that narrow base pushups (with your hands closer together) increases stress on the elbow joint.
Beach et al. (2008) showed that suspended pushups activated more core musculature than regular pushups. Based on this finding you can use blast straps or a TRX to increase the efficacy of the pushup exercise.
Cogley et al. (2005) found that narrow base pushups led to higher EMG values in the triceps brachii and pectoralis major than wide base pushups. Popular belief indicates that wide base pushups activate more pec fibers, but this study showed otherwise.
Donkers et al. (1993) showed that peak elbow joint force during pushups decreased if the hands were positioned either apart (wide base) of superior (high placement) from normal position. Normal hand position produced 56% of maximum isometric torque at the elbow joint, compared to 29% with hands apart and 71% with hands together. When the hands were placed superior in a high placement, the elbow ligaments significantly opposed more torque. When ligaments oppose torque, the ligaments are stressed but muscle force requirements diminish since the ligaments are “picking up some of the slack” so to speak.
Freeman et al. (2006) studied the EMG responses in nine muscles (rectus abdominis, external oblique, internal oblique, latissimus dorsi, erector spinae, pectoralis major, triceps brachii, biceps brachii, and anterior deltoid) in ten advanced individuals during twelve pushup variations (standard, fast concentric, slow eccentric, single arm left, single arm right, uneven hands left, uneven hands right, clapping, one hand on ball left, one hand on ball right, no legs, alternating hands on ball, both hands on ball, and two hands on ball). The most challenging exercises for the various muscle groups follow: rectus abdominis – alternating hands on ball followed by two arms on two balls, external obliques – alternating followed by clapping, internal obliques – alternating followed by one arm, latissimus dorsi – one arm followed by clapping, erector spinae – one arm followed by alternating, pectoralis major – clapping followed by alternating, anterior deltoid – one arm followed by clapping, triceps brachii – clapping followed by alternating, biceps brachii – clapping followed by alternating. When performing uneven pushups, the opposite side rectus abdominis and external obliques elicited more activity than the same side of the hand placed forward.
Gouvali et al. (2005) measured the pectoralis major and tricep activity on eight advanced subjects in six pushup variants; normal position, shoulders abducted (wide base), shoulders adducted (narrow base), shoulders positioned anterior to hands (shifted forward), shoulders positioned posterior to hands (shifted backward), and from the knees. The load relative to bodyweight for a standard pushup was found to be 66.4%, while the initial load relative to bodyweight for a knee-pushup was found to be 52.9%. Wide base pushups diminished pectoralis major and tricep activity from standard form, whereas narrow base pushups increased pectoralis major and tricep activity. Shifting the torso forward relative to the hands (low placement) resulted in increased pec activity and decreased tricep activity. Shifting the torso rearward relative to the hands (high placement) resulted in slightly increased pec and tricep activity. Performing pushups from the knees decreased pec and tricep activity. The most challenging pushup variation for the pecs was the shifted-forward position (low placement) followed by the narrow base position, whereas the most challenging pushup variation for the triceps was the narrow base position followed by the shifted-rearward (high placement) position.
Hogwarth (2008) found that in order to keep the core stable during pushups, on average the abdominal muscles contributed 64%, 87%, and 4% to vertebral joint rotation stiffness (VJRS) about the flexion/extension, lateral bend, and axial twist axes, respectively (VJRS is a measure of joint stability). The rectus abdominis contributed around 43% to VJRS about the flexion/extension axis at each lumbar joint, and external oblique and internal oblique contributed around 53% and 62% to VJRS about the lateral bend and axial twist axes, respectively, at all lumbar joints with the exception of L5-S1. Due to changes in moment arm length, the external oblique and internal oblique contributed 56% and 50% to VJRS about the axial twist and lateral bend axes, respectively, at L5-S1. Transversus abdominis, multifidus, and the spine extensors contributed minimally to VJRS during the push-up exercise. This means that the rectus abdominis prevents your hips from sagging and your lumbar spine from hyperextending, whereas the obliques prevent you from shifting to one side or twisting while you perform pushups.
Janssen et al. (2000) examined muscle mass and distribution patterns in 468 men and women aged 18-88 years old. While seemingly unrelated to pushups, this study is in fact important. It showed that men have more absolute muscle mass than women (33 kg vs. 21 kg) and more relative muscle mass (38.4% vs. 30.6%). Go figure, right? But the weight distribution varied from men to women. Men have 40% more upper body muscle mass and only 33% more lower body muscle mass than women. Put another way, men have a lower body to upper body muscle mass ratio of 1.28, compared to 1.45 for women, or a 17% higher percentage of lower body muscle mass divided by upper body muscle mass. This might make the pushup exercise more challenging for the core and less challenging for the shoulders for women, and more challenging for the shoulders and less challenging for the core for men.
Juker et al. (1998) found that when standardized to a percentage of MVC, the pushup elicited 24% mean psoas, 29% mean external oblique, 10% mean internal oblique, 9% mean transverse abdominis, 29% mean rectus abdominis, 10% mean rectus femoris, and 3% mean erector spinae activity. Pushups from the knees reduced the percentages to 14, 10, 19, 7, 8, 19, 5, and 3, respectively.
Lehman et al. (2006) found that the pushups with the hands placed on a Swiss ball significantly increased tricep activation. Swiss ball pushups also increased pectoralis major, rectus abdominis, and external oblique activation compared to pushups on a bench from the same angle, whereas pushups with the feet placed on a Swiss ball did not affect muscle activity compared to pushups with the feet on a bench from the same angle. Based on these findings it appears that as long as you kept the torso angle constant, it would be more effective to perform exercises such as Swiss ball and Bosu pushups in comparison to bodyweight pushups as long as you place the hands on the unstable piece of equipment rather than the feet.
Lehman et al. (2008) found that elevating the feet above the hands had a greater influence on scapulothoracic stabilizing musculature than placing the hands on a Swiss ball. This means that it is more challenging for the shoulder girdle stabilizers to do pushups with your feet elevated onto a bench with your hands on the ground than to perform pushups with your hands on a Swiss ball and your feet on the ground.
Lou et al. (2001) showed that internal rotation of the hand position or full pronation of the forearm during pushups led to greater posterior and varus forces on the elbow joints which could produce injurious shear forces. Based on this finding, it is recommended that internally rotated hand positions should be avoided for optimal elbow health.
Sandhu et al. (2008) found that pushups with the hands placed on a Swiss ball significantly increased tricep and pectoralis major activity compared to normal pushups, but only during the eccentric phase. The researchers also found that serratus anterior and upper trapezius activity did not change from standard pushups to Swiss ball pushups.
Suprak (2011) found that subjects supported 69% of their bodyweight in the top position of a pushup and 75% in the bottom position of a pushup, compared to 54% and 62% for the knee pushup, respectively. This means that a pushup is harder at the bottom position than the top position.
Tucker et al. (2008) showed that pushups activate 27% MVC for the mid trapezius fibers, 36% MVC for the lower trapezius fibers, and 75% MVC for the serratus anterior muscles.
Youdas et al (2010) found that depending on the hand position, the push up activated between 73-109% MVC of the triceps brachii, 95-105% MVC of the pectoralis major, 67-87% MVC of the serratus anterior, and 11-21% MVC of the posterior deltoid musculature. The researchers also found that the narrow base position was the most effective position for increasing tricep contribution, and that the Perfect Push up™ device did not enhance muscular recruitment.