Exercises to improve the strength and endurance of trunk muscles are often included in fitness and rehabilitation programs because properly conditioned trunk muscles will reduce the risk of low back injuries during physical exertions (16) and allow for arm and leg movements to be performed more forcefully and accurately (6,9). Indeed, it is no surprise that a plethora of studies have been conducted to identify trunk exercises that optimize trunk muscle activity levels and patterns (21). Nearly all of these studies have been conducted on land, which is fitting given the terrestrial nature of humans. However, some populations such as those with low back pain (LBP) are more frequently being prescribed aquatic exercises (14), although there is limited knowledge regarding how the exercises influence trunk muscle activity levels.
The basic premise for prescribing aquatic exercises for LBP patients is that buoyancy will reduce spinal loads and hydrostatic pressure and temperature of water will assist with balance and pain control, respectively (29). Accordingly, patients with LBP who find it difficult to perform land-based exercises may successfully perform aquatic exercises first and then progress to more functional land-based exercises after a period. Although it would seem intuitive that aquatic-based programs would lead to improved clinical outcomes in LBP patients when compared with land-based programs, the research generally finds that both programs lead to similar results in terms of pain reduction and strength improvement (29).
It is possible that aquatic trunk exercise programs for LBP patients are not any more effective than land-based programs because the aquatic environment may suppress trunk muscle activity levels (19). A recent review article by Masumoto and Mercer (19) summarized the effect of human locomotion in water and its effect on muscle activity levels as measured by EMG. These authors concluded that lower extremity muscles were less active in water than on land when gait speeds were self-selected and when maximal voluntary contractions were performed. In all of their studies reviewed, none focused solely on trunk muscle responses between environments. Only recently have trunk muscles been examined during deep-water running and shallow-water walking (4,17,20). For example, Kaneda et al. (17) recorded EMG activity of muscles rectus abdominis (RA), external oblique, and erector spinae (ES) during deep-water running, shallow-water walking, and land walking at three self-selected speeds (slow, moderate, and fast paces). They observed greater muscle activity in the ES trunk muscle during both deep- and shallow-water walking when compared with land walking at moderate and fast paces. However, when the walking pace was slow, activity levels of muscles RA, external oblique, and ES were not different between environments suggesting that activation of the trunk muscles may not follow the same declining trend as those observed for lower extremity muscles in an aquatic environment. Whether these same observations across environments also apply to trunk exercises commonly used for patients with LBP is yet to be determined.
It would be of value to know whether trunk muscle activity levels are different during exercises performed on land and in water to better understand the potential advantages and disadvantages of aquatic programs for LBP patients. In addition, a better understanding of how trunk muscles respond to specific trunk exercises between environments will assist the physical rehabilitation specialist in deciding which environment would be most effective for their client by applying the principle of progressive overload. Accordingly, the purpose of this study was to determine whether trunk muscle activity levels are different during trunk exercises performed in water than on land in healthy adults. A secondary purpose was to assess if the findings were consistent for a single-case patient with chronic LBP.
Eleven physically active males were asked to volunteer for the experimental aspect of the study. The number of subjects chosen was calculated using SamplePower software (SPSS, Inc., Chicago, IL) and was based on an effect size of 0.25 SD (10) with an α level of 0.05 and power at 0.80. Participants were recruited from a university population via word of mouth. Participants were included in the experimental study if they were free of musculoskeletal injury and pain for the previous 12 months. We asked only males to volunteer for this initial study to minimize the effect of subcutaneous fat on EMG signal fidelity. The participants displayed the following physical characteristics (mean ± SD): age = 25.7 ± 5.53 yr, mass = 77.8 ± 16.2 kg, and height = 1.82 ± 0.06 m.
The participant for the case study aspect of the study was referred by a university athletics team doctor. The patient was a male 23 yr old (mass = 72.6 kg) who was active in recreational sports. In terms of medical history, the patient reported that he had back pain in high school but never received medical attention; his pain was intermittent. A recent episode of pain was 2 wk before testing in the laboratory where his pain was 4/10 at worst and he could not participate in normal recreational activities. Pain decreased after one visit to a physical therapist and was 2/10 at the time of testing. No gross instability of the lumbar spine was noted, but some coordination problems with the deep lumbar stabilizer muscles were present. Before taking part in the study, all participants read and signed an informed consent form approved by the university institutional review board.
Procedures and research design.
Procedures and instrumentation for the experimental and case studies were identical, although the case study LBP patient did not perform maximal voluntary contraction (MVC). Participants attended a familiarization session and a testing session separated by <48 h. The experimental aspect of the study was a crossover design where the same healthy participants were tested both in water and on land. The dependent measure was normalized root mean square (RMS) values for select trunk muscles during four trunk exercises that were randomly assigned for each environment with the aquatic exercises being performed after the land exercises. By not randomizing environments, shivering due to temperature changes was minimized, and the integrity of the waterproofing adhesives was better maintained. The independent variable in the study was the environment (water and land).
During the familiarization session, participants were introduced to four trunk exercises that were performed in water and on land. The four exercises were 1) abdominal hollowing: participants assumed a standing semisquat position with their back against a wall, feet shoulder width apart, and knees bent to approximately 30°. Participants were instructed to maintain a neutral spine position and to maximally activate the abdominals while drawing the navel up and in toward the spine, holding for 5 s. 2) Abdominal bracing: in a standing neutral spine position, participants were instructed to maximally activate the abdominals without hollowing the lower abdomen and holding for 5 s. 3) Anteroposterior (AP) pelvic tilts: participants were seated in a neutral spine position on either a Swiss ball (land) or kickboard (water) that was stabilized by a tester. Participants' feet were flat on the surface, shoulder width apart, and knees were bent to 90°. Participants were instructed to tighten their abdominals and to roll their tailbone (coccyx) backward (anterior pelvic tilt) and forward (posterior pelvic tilt) for five continuous repetitions. 4) Mediolateral (ML) pelvic tilts: performed identical with AP pelvic tilts except participants were instructed to tilt their hips side to side by simultaneously dropping and lifting opposite hips. These exercises were chosen because they could be performed in both environments (water and land) using similar techniques and because they are frequently recommended for patients with LBP or spinal instability (1,22,28).
Participants performed a minimum of three sets of each exercise in water and on land during the familiarization session. More time was provided if the clinician instructing the participant felt the exercises were not being performed properly. For the AP and ML exercises, participants were instructed to move at a slow self-selected cadence, which is typical for LBP patients. Aquatic exercises were performed in a HydroWorx 2000 pool (HydroWorx 2000TM; Middletown, PA) at a water depth equal to the xiphoid process with the water temperature set to 30°C. The land exercises were performed at poolside with the air temperature set to 24°C.
During the test session, EMG activities of muscles RA, external oblique (EO), lower abdominals (LA), multifidus (MT), and ES were recorded during reference EMG tests and for each of the four exercises in water and on land. EMG signals were recorded using a pair of passive BIOPAC EL503 surface electrodes connected to BIOPAC LEAD110 leads (BIOPAC Systems, Inc., Goleta, CA). The electrode configuration included two Ag-AgCl circular electrodes with 1-cm conductive areas each. The interelectrode distance was 2 cm with each lead connected to a BIOPAC TEL110C amplifier. The system's typical common-mode rejection ratio was 110 dB, and the input impedance was 2 MΩ. A common reference electrode was placed on the skin over the acromion process. The guidelines described by Cram and Kasman (11) were followed for positioning the electrodes over muscles RA, EO, MT, and ES, whereas muscle LA location was determined using procedures described by Marshall and Murphy (18). Note the LA location of 2 cm inferior and medial to the anterior superior iliac spine is where the internal oblique muscle blends with the transversus abdominis muscle; therefore, the activity of these later two muscles cannot be separated, hence the term LA.
EMG activity from the selected muscles was recorded on the right side only. To ensure high fidelity of the EMG signals on land and particularly in water, the skin at each site was shaved, lightly abraded with sandpaper, and cleaned with rubbing alcohol before electrode placement. Because water infiltration can contaminate the EMG signal (5), careful attention was paid to waterproofing electrodes. The same waterproofing technique described by Silvers and Dolny (27) was followed, which required systematic applications of waterproof adhesive (OPSITE™ Smith and Nephew, Largo, FL) and silicone sealant (DAP, Inc., Baltimore, MD). The Silvers and Dolny (27) waterproofing technique is reportedly reliable between environments (intraclass correlation coefficients ≈ 0.97) for lower extremity muscle activation at a depth similar to the current study. The waterproofing adhesives were applied after completion of the land exercises to prevent damage to the adhesives due to sweat or excessive movements. The electrodes were not repositioned between environments, and the addition of waterproofing adhesives has shown not to influence EMG amplitudes during land exercises (8).
Once participants were prepared for EMG data recording, reference EMG data were obtained before collecting EMG data for each exercise. Participants were asked to perform dry land MVC for each muscle. For muscles RA and LA, subjects lay supine on a stable plinth with hips and knees flexed, feet flat, and secured by an assistant. Participants crossed their arms over their chest and on the command "go," attempted to maximally curl up against manual resistance. For muscle EO, the same aforementioned procedures were followed but the curl up included a twist to the right. For muscles MT and ES, subjects lay prone with feet secured by an assistant; the hands were positioned behind the head, and on the command "go," they attempted to extend at the trunk maximally against manual resistance. Each test was performed twice, and each effort was held for 5 s with a 30-s rest between repetitions.
After collecting reference MVC data, the exercises described in the familiarization session were performed. Participants performed one set of three repetitions for the abdominal hollowing and bracing exercises and five repetitions for the pelvic tilt exercises. No encouragement was provided during the exercises, and proper execution of each exercise was visually determined by the instructing clinician. If the exercise was performed incorrectly, it was repeated. Participants began each repetition on the verbal command "go." The EMG system was manually triggered before the command to record 10 s of data for each repetition between a bandwidth of 20 and 500 Hz. For all reference tests and exercises, an event marker was manually triggered before and after the completion of each repetition. EMG signals were amplified by a factor of 10,000, and the amplified signals were sampled at 1024 Hz.
To determine whether the exercises were performed similarly between environments, the instructing clinician visually reviewed sagittal plane videos of the exercises for all subjects. This inspection led to no exercises being rejected for post-EMG processing. EMG signals were then visually inspected for false components (e.g., high-frequency bursts). This inspection led to the dismissal of <1% of the original data. With the remaining data, RMS values during the middle 3 s of the EMG data collected for each repetition of the reference tests and abdominal hollowing and bracing exercises were computed. Similarly, RMS values for each repetition of the pelvic tilt exercises were computed between event markers. Next, an average RMS value over three repetitions of each exercise was computed. The average RMS value of each muscle was normalized to the average RMS value computed for each muscle's reference (MVC) test. EMG data for the case study LBP patient were simply expressed as the average RMS value (mV) and were not normalized.
The effect of environment (water and land) on mean normalized EMG values for each muscle was assessed with paired-samples t-tests. α levels of 0.05 were Bonferroni adjusted and used for all calculations to determine significance. Effect sizes were also quantified to appreciate the meaningfulness of any statistical differences. The effect sizes were calculated with the following formula: effect size = (high value − low value) / SD of high value, and the Cohen (10) convention for effect size interpretation was used (<0.41 = small, 0.41-0.7 = medium, and >0.7 = large). EMG data from the case study was descriptive.
Normalized EMG values for muscles RA, EO, LA, and MT were significantly greater for all exercises performed on land than in water (P = 0.029-0.001, effect size = 0.55-1.61; Figs. 1 and 2). Similar to the other muscles, ES values were significantly greater for all exercises performed on land than in water (P = 0.026-0.008, effect size = 0.52-0.85; Fig. 2) with the exception of ML pelvic tilts; muscle ES values were not different between environments (P = 0.098). It was observed that EMG values for the LBP patient followed the same trend as those for the healthy group. Examples of the EMG trends for the LBP patient are reported in the Table 1.
The aim of this study was to determine whether trunk muscle activity levels were different during trunk exercises performed in water than on land for healthy male adults. A case study was also included to assess if the findings were consistent for a patient with chronic LBP. It was observed that trunk exercises, such as abdominal hollowing, abdominal bracing, and pelvic tilts elicited greater EMG values for most muscles on land than in water despite replicating the exercises similarly between environments. This trend was consistent with a single-case patient with LBP.
The trend in the results of this study are in agreement with studies that observed greater lower extremity muscle activity levels during various nonspecific trunk exercises performed on land than in water (19). In comparison with studies that examined trunk muscle activity levels (4,17,20), our results are not in agreement suggesting the type of exercise performed (walking or running vs abdominal hollowing/bracing or pelvic tilts) may be an even larger factor than the environment (water vs land) for influencing trunk muscle activity levels.
Various mechanisms have been proposed to address why muscle activity levels are often lower in water than on land. For instance, Silvers and Dolny (27) have argued that inadequate methods for waterproofing electrodes may in part explain decreased EMG values in water. This contention was supported by Rainoldi et al. (26) and Carvalho et al. (8) who observed decreased EMG activity levels in water when electrodes were not covered with adhesive tape. This limitation does not likely apply to the present study because careful attention was paid to waterproofing electrodes with adhesive tape using reliable methods that reportedly prevented a decrease in EMG values in water (27).
It is also possible that with hydrostatic pressure and buoyancy, trunk muscles play less of a stabilizing role in the aquatic environment, which minimizes their EMG activity levels. This idea was proposed by Deban and Schilling (13) who observed that trunk muscle activity levels in salamanders were substantially lower during trotting in water than trotting on land. However, the influence of an aquatic environment on stabilization of the trunk may depend on the dynamics of the exercise (e.g., abdominal hollowing vs walking) because fluid resistance increases as a function of a body's relative velocity squared (15). In another work, Pöyhönen and Avela (24) have argued that EMG reductions in water may be reflex related via impulses from pressure mechanoreceptors that are disturbed over the entire body. As a body becomes immersed in water, hydrostatic pressure may stimulate mechanoreceptors whose impulses produce presynaptic inhibition with interneuron pathways. This mechanism was proposed by Pöyhönen and Avela (24) after they observed a substantial reduction in the Hoffman reflex during water immersion. Muscle temperature also influences frequency and sometimes magnitude of an EMG signal (23); however, water and land temperatures in the present study were considered thermoneutral (27), and large fluctuations in trunk muscle temperatures were unlikely because of the large mass that insulates this region of the body (23).
From a clinical perspective, the magnitudes of reduction we observed in muscle activity during aquatic exercises were meaningful on the basis of the effect sizes that ranged from moderate to high. This observation does not imply that aquatic trunk exercises are not useful for maintaining or improving neuromuscular conditioning of the trunk muscles. In fact, researchers have found that exercises that produce normalized EMG values of 25% or less are sufficient to improve motor control and endurance aspects of some trunk muscles (28) and are a level of intensity that may maximally stiffen segmental joints of the spine (12). Some have argued that when normalized EMG values exceed approximately 40%, a greater risk of joint pain or injury to the spine may occur (2,3,7). In this view, aquatic exercises may have produced more appropriate activation levels for patients with LBP because normalized EMG values in water were generally <25% (Figs. 1 and 2). In contrast, normalized EMG values on land were often >40% for the EO and LA muscles (Fig. 1). More practically, the results of this study suggest the aquatic environment may be the safest medium to begin a rehabilitation program in patients with LBP who are unable to adequately exercise the trunk muscles on land because of pain or balance issues. However, because of the substantial reduction of EMG activity in water, a patient must progress toward land-based exercises to more fully activate the trunk muscles as evidenced in Figures 1 and 2.
There are several limitations of this study that should be noted. The results are specific to the abdominal hollowing, abdominal bracing, and pelvic tilt exercises as performed in this study. Whether reduced activity is observed in other trunk muscles for other trunk exercises performed in water cannot be determined in this study. For example, the findings of Kaneda et al. (17) and others would suggest that walking or running in water will increase the activity of muscle ES when compared with similar land exercise. In addition, the results apply only to healthy young physically active males. Although our case study participant with LBP displayed a similar EMG trend to our healthy sample, the results cannot be generalized to an LBP population because various between-individual factors, such as the severity of back pain, were not represented in our sample size of one.
In regards to future research, it would be of value to explore how various other aquatic exercises influence trunk muscle activity levels. As noted in a systematic review by Waller et al. (29), there are a large number of therapeutic aquatic exercises used in the treatment of LBP, yet there is no consistency in use of the exercises between studies. The lack of consistency is likely because it is unclear which exercises are most effective at activating trunk muscles in an aquatic environment. It would also be of value to determine whether the results of this study are present after a training period. Pöyhönen et al. (25) observed that a 10-wk aquatic training program designed to improve strength in the knee extensors and flexors resulted in improved neural activation of the knee extensors, suggesting neuromuscular pathways are equally adaptable in an aquatic environment. Although a single familiarization session was included in the present study, it may be that chronic exposure to the aquatic environment will desensitize any inhibitory reflexes that are stimulated by the lack of gravity and hydrostatic pressure. This assertion obviously needs to be researched in a systematic manner; the challenge will be the methodological limitations of reliably assessing EMG between days.
In conclusion, when healthy adults perform abdominal hollowing, abdominal bracing, and pelvic tilt exercises in water, most trunk muscles display substantially lower EMG activity levels when compared with performing the same exercises on land. This trend seems to be consistent when these exercises are performed by a patient with LBP.
This study was supported by grants from the National Swimming Pool Foundation.
The authors would like to thank the technical support of Katy Martin, Coby VandenBerg and Ryan Porter.
The authors declare that they have no conflict of interest.
The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Allison GT, Godfrey P, Robinson G. EMG signal amplitude assessment during abdominal bracing and hollowing. J Electromyogr Kinesiol
2. Arokoski JP, Kankaanpää M, Valta T, et al. Back and hip extensor muscle function during therapeutic exercises. Arch Phys Med Rehabil
3. Arokoski JP, Valta T, Kankaanpää M, Airaksinen O. Activation of paraspinal and abdominal muscles during manually assisted and nonassisted therapeutic exercise. Am J Phys Med Rehabil
4. Barela AM, Stolf SF, Duarte M. Biomechanical characteristics of adults walking in shallow water and on land. J Electromyogr Kinesiol
5. Benfield RD, Newton ER, Hortobágyi T. Waterproofing EMG instrumentation. Biol Res Nurs
6. Butcher SJ, Craven BR, Chilibeck PD, Spink KS, Grona SL, Sprigings EJ. The effect of trunk stability training on vertical takeoff velocity. J Orthop Sports Phys Ther
7. Callaghan JP, Gunning JL, McGill SM. The relationship between lumbar spine load and muscle activity during extensor exercises. Phys Ther
8. Carvalho RG, Amorim CF, Perácio LH, et al. Analysis of various conditions in order to measure electromyography of isometric contractions in water and on air. J Electromyogr Kinesiol
9. Cholewicki J, Ivancic PC, Radebold A. Can increased intra-abdominal pressure in humans be decoupled from trunk muscle co-contraction during steady state isometric exertions? Eur J Appl Physiol
10. Cohen J. Statistical Power Analysis for the Behavioral Sciences
. 2nd ed. Hillsdale (NJ): Lawrence Erlbaum; 1988. p. 1-567.
11. Cram JR, Kasman GS. Introduction to Surface Electromyography
. Gaithersburg (MD): Aspen Publication; 1998. p. 237-387.
12. Cresswell AG, Oddsson L, Thorstensson A. The influence of sudden perturbations on trunk muscle activity and intra-abdominal pressure while standing. Exp Brain Res
13. Deban SM, Schilling N. Activity of trunk muscles during aquatic and terrestrial locomotion in Ambystoma maculatum. J Exp Biol
14. Dundar U, Solak O, Yigit I, Evcik D, Kavuncu V. Clinical effectiveness of aquatic exercise to treat chronic low back pain: a randomized controlled trial. Spine (Phila Pa 1976)
15. Hamill J, Knutzen KM. Biomechanical Basis of Human Movement
. Baltimore (MD): Lippincott Williams & Wilkins; 2003. p. 346-8.
16. Hodges PW, Richardson CA. Altered trunk muscle recruitment in people with low back pain with upper limb movement at different speeds. Arch Phys Med Rehabil
17. Kaneda K, Sato D, Wakabayashi H, Nomura T. EMG activity of hip and trunk muscles during deep-water running. J Electromyogr Kinesiol
18. Marshall P, Murphy B. The validity of and reliability of surface EMG to assess the neuromuscular response of the abdominal muscles to rapid limb movement. J Electromyogr Kinesiol
19. Masumoto K, Mercer JA. Biomechanics of human locomotion in water: an electomyographic analysis. Exerc Sport Sci Rev
20. Masumoto K, Takasugi S, Hotta N, Fujishima K, Iwamoto Y. Muscle activity and heart rate response during backward walking in water and on dry land. Eur J Appl Physiol
21. Monfort-Pañego M, Vera-García FJ, Sánchez-Zuriaga D, Sarti-Martinez MA. Electromyographic studies in abdominal exercises: a literature synthesis. J Manipulative Physiol Ther
22. O'Sullivan PB, Phyty GD, Twomey LT, Allison GT. Evaluation of specific stabilizing exercise in the treatment of chronic low back pain with radiologic diagnosis of spondylolysis or spondylolisthesis. Spine (Phila Pa 1976)
23. Petrofsky J, Laymon M. Muscle temperature and EMG amplitude and frequency during isometric exercise. Aviat Space Environ Med
24. Pöyhönen T, Avela J. Effect of head-out water immersion on neuromuscular function of the plantarflexor muscles. Aviat Space Environ Med
25. Pöyhönen T, Sipilä S, Keskinen KL, Hautala A, Savolainen J, Mälkiä E. Effects of aquatic resistance training on neuromuscular performance in healthy women. Med Sci Sports Exerc
26. Rainoldi A, Cescon C, Bottin A, Casale R, Caruso I. Surface EMG alterations induced by underwater recording. J Electromyogr Kinesiol
27. Silvers WM, Dolny DG. Comparison and reproducibility of sEMG during manual muscle testing on land and in water. J Electromyogr Kinesiol
28. Vezina MJ, Hubley-Kozey CL. Muscle activation in therapeutic exercises to improve trunk stability. Arch Phys Med Rehabil
29. Waller B, Lambeck J, Daly D. Therapeutic aquatic exercise in the treatment of low back pain: a systematic review. Clin Rehabil