The reverse hyper machine was originally developed to increase muscular strength of the posterior chain (low back, gluteus maximus, and hamstrings) in competitive athletes. Anecdotally, the machine is used by athletes and in the fitness industry as a tool to strengthen the posterior chain as well as to “stretch out” after compressive workouts. Zweifel recently proposed the use of the reverse hyper as a means to improve athletic performance, proposing that extension-based exercises in the horizontal plane may be useful in athletic performance, while also noting the unique qualities of the reverse hyper (11). The original manufacturer of the reverse hyper claims it will strengthen the posterior chain and is marketed as providing traction to the lower back (10). Despite these claims, and the growing popularity of the reverse hyper machine, to the best of our knowledge, there have been no scientific investigations to determine the efficacy of the reverse hyper. Therefore, the purpose of this investigation was to determine whether there are differences between the reverse hyperextension (RHE) and the back hyperextension (HE) in terms of muscle activation (erector spinae, gluteus maximus, and biceps femoris), range of motion (ROM) (pelvis to trunk and hip), and low back extension moments. We hypothesized that the RHE exercise would have greater muscle activation, ROM, and low back extension moments than the HE exercise.
Experimental Approach to the Problem
Twenty subjects performed 2 sets of 10 repetitions, each of a standard back extension and the RHE. Activity of the right erector spinae, gluteus maximus, and biceps femoris as well as kinematic data was recorded. A force transducer was used to measure the force between the pendulum of the RHE machine and the subject's lower extremities. Data were then analyzed to determine whether there were differences in muscle activity, ROM, and the moment produced at the lower back between exercises.
Twenty healthy recreationally active individuals (mean ± SD: 10 men and 10 women; age 26.8 ± 7.8 years, body mass 74.7 ± 21.5 kg, and height 1.68 ± 0.08 m) volunteered for this study. Those who had current back pain, back pain within the last 6 months, history of spinal surgery, history of any nervous system disorder, or any other back abnormalities were excluded from this study. Of the 20 participants, 8 reported having experience using the reverse hyper (5 men and 3 women). This study was approved by the Institutional Review Board at the University of New England, and all participants gave their written informed consent after the study procedures were explained to them.
Three wireless surface electromyography (EMG) sensors (2-cm interelectrode distance) (Noraxon USA Inc., Scottsdale, AZ, USA) were used to measure the muscle activation of the biceps femoris, gluteus maximus, and erector spinae sampled at 1500 Hz. All EMG sensors were placed according to SENIAM guidelines (4,6). All EMG data were normalized to values during a maximum voluntary isometric contraction, which was recorded while the subject was in the top position of a HE. Clusters of reflective markers were placed on the upper back, pelvis, thighs, lower legs, feet, and pendulum of a RH-2 RHE machine (Rogue, Columbus, OH, USA). We used the regular length strap (125.5 cm) on the RHE for this study. Individual markers were placed on bony landmarks: heels, first and fifth metatarsal head, medial and lateral malleoli, medial and lateral knee, and acromion. Eight Oqus Series-3 cameras (Qualisys AB, Gothenburg, Sweden) set at 150 Hz tracked the motion of the reflective markers during both the RHE and the HE. A force transducer was used to measure the force between the pendulum of the RHE machine and the subject's lower extremities. This force was then applied equally to both lower extremities where the strap of the RHE was attached to the person (lower calf). The orientation of the force was created to be the same orientation of the strap.
Subjects performed 2 sets of 10 repetitions for each exercise (Figure 1), and the exercise order was alternated between each subject. For the HE exercise, all subjects held a standard mass plate (20.4 kg) to their chest (arms folded around the plate). The upper body was considered to be 2/3 of the total body mass (2). Upper body mass and mass of the plate were modeled to act through the center of mass of the trunk. For the RHE, weight was added until the sum of the added weight, lower body (1/3 of the body mass) (2), and reverse hyper pendulum was equal to the load of the back extension. The subjects were asked to keep a 1:1 cadence (1 second up and 1 second down) to the best of their ability. A rest period of 2 minutes was given between each set. If the subjects were unfamiliar with either of the exercises, they were allowed for practice repetitions until comfortable with the movement. All data were analyzed using Visual 3D (C-Motion, Germantown, MD, USA). Motion of the body and pendulum was filtered with a second-order low-pass Butterworth filter with a 3-Hz cutoff. Muscle activity was smoothed by creating a linear envelope with a second-order band-pass filter (20–200 Hz), rectification, and then a low-pass Butterworth filter with a 6-Hz cutoff.
Peak, mean, and integrated muscle activity were calculated for each repetition for all subjects. Total ROM (right thigh to trunk), isolated ROM (trunk to pelvis), and peak, mean, and integrated extension moment at the low back were also calculated.
All data were analyzed in an aggregate form using SPSS Version 21 (IBM, Chicago, IL, USA). Descriptive statistics (number, means, and SDs) were used for demographic data including the subject's age, height, body mass, and sex. All data were assessed for homogeneity of variance and normality using Levene's statistic and the Shapiro-Wilk test and assessed for outliers, including visual inspection of the Q-Q plot. To assess consistency between repetitions of each exercise within the data set, intraclass correlation coefficients were used (1,2). As several data points demonstrated statistically significant variance, a Wilcoxon signed-rank test was used to assess for statistically significant differences for all pairs of data, comparing each muscle group for the 2 exercises. Cohen's d was calculated for each pair as a measure of the magnitude of difference between exercises.
Integrated muscle activity was significantly (p < 0.05) greater during the HE exercise in both the biceps femoris and the gluteus maximus (Table 1). Maximum and mean muscle activation was not different between exercises across all muscles (Table 1). The ROM between the trunk and pelvis was significantly greater (+10.7°) during the HE as compared to the RHE (Table 2). The RHE created a greater peak (+129%), integrated (+63%), and mean (+78%) low back moment as compared to the HE exercise (Table 3). Intraclass correlation coefficients were high across all variables, ranging between 0.975 and 0.998.
The main findings of this investigation are that RHE and HE with comparable loads exhibit similar muscle activation of the erector spinae; however, the HE exercise resulted in greater integrated muscle activity of the biceps femoris and gluteus maximus muscles. Interestingly, RHE produced greater low back extension moments than the HE exercise, although there was less ROM between the pelvis and trunk during RHE than HE. The results of this study agree with anecdotal evidence that the RHE is an effective low-impact exercise to target the low back musculature.
The RHE produced a greater peak (+129%), integrated (+63%), and mean (+78%) low back moment compared with the HE exercise, with large meaningful effect sizes. This large increase is likely due to the greater moment arms of the weight and body segments during the RHE. Surprisingly, activation of the erector spinae, gluteus maximus, and biceps femoris did not coincide with the increase in low back extension moment. This may be due to the nature of how the RHE exercise is performed, as the pendulum allows for momentum to be used to assist with the completion of repetitions, offsetting the need for muscular force generation. To make the exercise as natural as possible, we only attempted to constrain subjects to a set tempo and we did not control for the amount of swing at the bottom of the movement.
The RHE resulted in a significantly greater overall excursion of the thigh relative to the trunk, 76.6° compared with 64.7°. However, the RHE used a far smaller degree of lumbar flexion relative to the pelvis, 20.4° compared with 31.1° for the HE. This movement profile is preferable because it produces greater hip ROM with less angular stress applied to the lumbar spine while providing equivalent erector spinae activity. Several investigators have reported that low back pain may be related to limited ROM in the hip, resulting in excessive stresses on the lumbar spine (1,3,5,7–9). This may have practical implications in the training and rehabilitation of individuals with back pain; however, further research is required to support this suggestion. Conclusions drawn from these results should be carefully considered because there are some important limitations. Subjects were not required to have previous experience with either exercise, although they were allowed to practice both exercises until they were comfortable performing them. The amount of swing during the RHE exercise was not controlled for, as we wanted subjects to perform the exercise as naturally as possible. We did not control for body composition, and this may limit the ROM in these exercises. We used a more functional position for the normalization of the muscle activation rather than the individual test of the musculature; therefore, the values may not actually reflect true relative activation levels. As there are no published guidelines on loading parameters for the RHE, we attempted to make the load equivalent to the HE exercise; therefore, the load for the RHE is substantially lower than what is reported anecdotally (up to 50% of squat single-repetition maximum). It may be that the RHE's effectiveness and movement profiles will alter when using a load similar to those recommended anecdotally. Nevertheless, to the best of our knowledge, this is the first investigation to attempt to quantify the effects of the RHE exercise.
The results of this study support using the RHE as an effective exercise to activate the erector spinae muscles. The RHE also resulted in a more favorable movement profile, with less lumbar flexion and more hip flexion as compared to the HE exercise. Although more research is needed to determine how different tempos and loading schemes may affect ROM and muscle activation profiles, this investigation serves as the first step in determining the efficacy of the RHE.
This work was supported by a grant from the University of New England Office of Research and Scholarship.
1. Chesworth B. A comparison of hip mobility in patients with low back
pain and matched healthy subjects. Physiother Can 46: 267–274, 1994.
2. Dempster WT. Space Requirements of the Seated Operator: Geometrical, Kinematic, and Mechanical Aspects of the Body, With Special Reference to the Limbs. Springfield, OH: Carpenter Litho & Prtg., 1955. pp. 183–196.
3. Ellison JB, Rose SJ, Sahrmann SA. Patterns of hip rotation range of motion: A comparison between healthy subjects and patients with low back
pain. Phys Ther 70: 537–541, 1990.
4. Freriks B, Hermens HJ SENIAM 9: European Recommendations for Surface ElectroMyoGraphy, results of the SENIAM project, Enschede, the Netherlands: Roessingh Research and Development b.v., 1999, ISBN 90-75452-14-4 (CD-rom).
5. Friberg O. Clinical symptoms and biomechanics of lumbar spine and hip joint in leg length inequality. Spine (Phila Pa 1976) 8: 643–651, 1983.
6. Hermens H, Freriks B, Disselhorst-Klug C, Rau G. Development of recommendations for SEMG sensors and sensor placement procedures. J Electromyogr Kinesiol 10: 361–74, 2000.
7. Mellin G. Correlations of hip mobility with degree of back pain and lumbar spinal mobility in chronic low-back pain patients. Spine (Phila Pa 1976) 13: 668–670, 1988.
8. Offierski CM, MacNab I. Hip-spine syndrome. Spine (Phila Pa 1976) 8: 316–321, 1983.
9. Reiman MP, Weisbach PC, Glynn PE. The hips influence on low back
pain: A distal link to a proximal problem. J Sport Rehabil 18: 24–32, 2009.
10. Simmons L. Louie Simmons Reverse Hyper Instruction. Columbus, Ohio: YouTube, 2010.
11. Zweifel M. Importance of horizontally loaded movements to sports. SCJ 39: 21–26, 2017.