The purpose of this study was to investigate the effects of acute massage on delayed-onset muscle soreness. A total of 20 subjects (5 men and 15 women; mean age 24 ± 3 years; height 1.7 ± 0.1 m; weight 71 ± 1.4 kg) were randomly assigned to either a massage treatment (MAS, n = 8) or control (CON, n = 12) group. Following preliminary data collection, muscle soreness was induced to both groups using identical protocols. The MAS group received a 10 min massage immediately following the muscle soreness protocol where the CON group did not. Data collected included signals from electromyography (EMG), mechanomyography (MMG), perceived muscles soreness, muscle circumference, and muscle torque. Data were collected for each subject prior to and on days 1, 2, 3, and 7 following the intervention. Repeated measures analysis of variance was used to determine significant differences in the research variables between the groups with p ≤ 0.05. A significant interaction was noted in MMG frequency during isokinetic muscle actions but all other data showed no significant interactions. Based on these data massage may not be beneficial following exercise that induces delayed onset muscle soreness.
The occurrence of delayed-onset muscle soreness (DOMS) is prevalent among individuals who participate in acute and chronic bouts of physical activity and/or training for athletic endeavors. Delayed-onset muscle soreness has been reported following strength, power, and endurance training. In most cases the influence of DOMS following an acute training session is detrimental to subsequent training bouts, specifically a loss of both strength and power. Anecdotal evidence derived from massage therapists, trainers, coaches, and athletes suggests that massage may be beneficial in reducing the effects of DOMS following acute training. Despite supporting anecdotal evidence there does not appear to be a consensus in the literature to indicate a benefit or lack thereof for the effects of massage on DOMS. Research regarding the effects of massage on DOMS has provided evidence to indicate that massage may be beneficial in reducing both edema, the intensity of perceived pain, and muscle soreness (6,7,17,19). Conversely, others have reported that massage had no effect in relieving symptoms related to DOMS (11) or on recovery of muscle strength and range of motion (19). In fact, Farr et al. (6) reported a decrease in strength and power of the massaged limb compared to the control limb at 1 and 24 hrs post intervention. Furthermore, psychological benefits of massage as opposed to physiological benefits have been postulated (8).
The recording of action potentials given off by each motor unit activated during a muscle action is known as electromyography (EMG). During recruitment of muscle fibers individual motor units are activated asynchronously which allows muscle movement to occur in a smooth pattern (4). The recording of pressure waves produced by lateral oscillations of muscle fibers is known as mechanomyography (MMG) (1,13). The synchronous use of EMG and MMG has been used to indicate the intrinsic properties of skeletal muscle. Therefore, the purpose of this study was to investigate the effects of acute massage on intrinsic properties of skeletal muscle with DOMS.
Subjects were recruited from a sample of convenience within the University setting and randomly assigned to the control group (CON) or massage treatment group (MAS). Volunteers were excluded from the study for participating in resistance training or supplementing with creatine within the previous 6 months to the study or for acknowledgement of the use of androgenic anabolic steroids. In accordance with the Declaration of Helsinki all subjects were carefully informed about the possible risks and benefits of the study, and all subjects signed a written informed consent form before participation in the study. The university’s institutional review board approved the study protocol for human subjects prior to data collection.
To induce muscle soreness, each subject began with an 11.4 kg dumbbell weight (males) or a 9.1 kg weight (females) and performed a biceps curl (elbow flexion/extension). To begin, the investigator raised the weight during the concentric (CONC) phase of elbow flexion and subjects lowered the weight using a 3-sec eccentric (ECC) muscle action of the upper arm until fatigue. Once the subject could not complete the ECC repetitions on the 3-second count, the weight was reduced by 2.3 kg increments. Each participant followed the same protocol of ECC muscle actions until volitional fatigue exhibited by failure to complete a repetition using a 2.3 kg weight within 3 sec.
Perceived muscle soreness was subjectively determined with the use of a visual analog pain rating scale (VAS), which is commonly used in the measurement of pain associated with DOMS. The VAS scale is a continuous 10-cm line with varying degrees of perceived pain intensity indicated below the line. The lower portion of the line is indicated as “no pain” whereas the upper portion of the line is indicated as “unbearable pain”. The subjects placed a pencil mark on the line at the point that they feel best describes the muscle pain they experienced. Perceived muscle soreness was obtained prior to strength testing on days 1, 2, 3, and 7.
Circumference of the biceps brachii was measured halfway between the acromion process and the antecubital space in cm using Gullick spring tension tape. This measurement was taken prior to the muscle soreness protocol on day 1 and prior to strength testing on days 2, 3, and 7.
Strength testing was performed on the dominant arm of the subjects. The test was performed on a calibrated Cybex II isokinetic device (Lumex; Ronkonkoma, New York). Participants were in a supine position with a restraining strap over the pelvis and the shoulder in 90° of abduction and the elbow in 90° offlexion. The input axis of the dynamometer was aligned with the elbow. The subject was asked to perform two submaximal warm-up trials followed by two maximal isometric voluntary contractions (MVIC) of the elbow flexors. The MVIC that resulted in the highest peak torque (PT) was then selected as the representative score. A two-minute rest was given and the subject completed randomly assigned submaximal muscle actions of 25%, 50% and 75% of MVIC. A minimum of five minutes was given prior to the fatigue protocol. During the fatigue protocol, each subject performed 25 maximal elbow flexion muscle actions at 90 deg∙sec-1.
A bipolar (7.62 cm center-to-center) surface electrode (Quinton Quick PrepAgAgCl electrode, recording diameter = 105 cm; Quinton, Bothello, Washington) arrangement was placed in a longitudinal axes of the biceps brachii muscle on the dominant arm. The recording electrodes were placed over the belly of the muscle midway between the axillary fold and the mid-point of the cubital fossa. The reference electrode was placed over the volar arch. Interelectrode impedance was kept below 5,000 ohms by abrasion of the skin. The EMG signal was preamplified (gain: 1000x) using a differential amplifier (EMG 11, BiopacSystem Inc., Santa Barbara, California). To assure consistent placement of the electrodes throughout the study, a permanent marker was used on the skin to outline the electrodes.
The surface oscillation was detected with the aid of an accelerometer (Entran EGASY- 25D, dimensions: 0.5 x 0.5 x 1.0 cm; mass: 1 g: sensitivity: 4mV/g: resonant frequency >1000 Hz). The long axis of the accelerometer was placed midway between the EMG electrodes in a transverse position to the bicepsbrachii muscle fibers (14).
A 10-minute massage was administered to the treatment group after the muscle soreness protocol was completed. A licensed massage therapist administered each massage. The massage therapy consisted of light effleurage until the muscle was relaxed from performing the perceived muscle soreness protocol.
For each isometric torque measurement, the MMG (mV) and EMG (μV) amplitude values (expressed as root mean square, RMS) were calculated for a 1.67 sec time period corresponding to the middle 1/3 of the 5 sec isometric muscle contractions to assure that the signals corresponded to the desired torque level. For the CONC isokinetic muscle contractions, the amplitude values were calculated for a time period (0.66 sec corresponding to a 60º range of motion) from approximately 60-120º of flexion at the elbow. The RMS and mean power frequency (MPF) of the highest PT was used as the representative score for both EMG and MMG. When selecting the EMG signal, the corresponding MMG signal was automatically selected so that the analyzed EMG and MMG signals were of the same time frame. The EMG and MMG signals were filtered at 10-500 and5-100 Hz, respectively, and sampled at 2,000 points per sec. All MPF analyses were performed with custom programs written with Lab view software (version 5.1 National Instruments, Austin, Texas). Each MMG and EMG data segment was processed with a Blackman window and discrete Fourier transform (DFT). This method was chosen over a fast Fourier transform because DFT is not constrained to 2n number of data points for the analysis. Therefore, the Fourier analyses in the present study were performed on the total selected data segments without having to truncate the desired data segment (down to 2n number of data points)or resort to zero padding (to increase the number of data points to 2n). The MPF was selected to represent the power density spectrum based on the recommendation of Diemont and colleagues (5) and was calculated as described by Kwatny and colleagues (10).
Separate three-way (group by day by %MVC; dependent variables = torque, EMG, and MMG;) and two-way (protocol by day; dependent variables = VAS, girth) repeated measures ANOVA (RMANOVA) were used to determine differences in torque,EMG amplitude, EMG MPF, MMG amplitude, MMG MPF, VAS and girth for the isometric and isokinetic muscle actions. Follow-up analyses included collapsing data and analyzing using two-way and one-way repeated measures ANOVA with Tukey post-hoc procedures and paired t-tests. Statistical significance was set at p < 0.05 for all analyses.
Twenty-four subjects volunteered to participate in this investigation. A total of 20 individuals completed the protocol and included 5 males and 15 females; mean age 24 ± 3 years; height 1.7 ± 0.1 m; weight 71 ± 1.4 kg. Twelve participants participated in the CON group (n=12) and eight completed the MAS protocol (n=8). No significant differences were noted between the two groups for height (p= .371), weight (p= .619) or age (p= .793).
The two-way (group by day) ANOVA for VAS indicated no interactions or significant main effect for pain (p = .708). The statistical model was decomposed by collapsing the two-way repeated measures ANOVAs and a significant main effect was found for day (p = .00) but no significant main effect for group (p = .518). Post hoc analysis for day revealed significant differences for VAS as shown in Figure 1. Tukey’s post hoc analysis revealed significance in VAS for day 1 vs. day 2, day 1 vs. day 3, day 2 vs. day 7, and day 3 vs. day 7. There was an increase in pain in days 2 and 3 in both groups despite no differences in pain responses between the groups.
Visual analog scale for control vs. treatment groups
Figure 2 displays the data for muscle circumference changes of the upper arm. The two-way (group by day) ANOVA for muscle circumference change indicated no interactions or significant main effect (p = .303). Data was subsequently collapsed into group and day. No significant main effect was noted for group (p= .381). However, a significant main effect was found for day (p =.00). Tukey’s post hoc analysis revealed significance in girth for day 1 vs. day 2, day 1 vs. day 3, and day 1 vs. day 7. This indicates an increase in muscle circumference in both groups despite no differences in muscle circumference between the two groups.
Percent change in muscle circumference size measured at bicepsbrachii
There were no significant three-way (group by day by % MVC) interactions for the isometric torque (p = .569) or isokinetic torque (p =.250). Table 1 represents mean isometric torque (Nm) at 100%, 75%, 50% and 25%of MVC. Table 2 represents mean torque values during the fatigue protocol (mean of first five repetitions and last five repetitions). Based on these data there were no differences in torque responses of the subjects between the groups.
Isometric torque (Nm) at 100%, 75%, 50% and 25% of MVC
Average isokinetic torque measurements (Nm) of first five repetitions and last five repetitions
There were no significant three-way (group by day by %EMG) interactions for EMG amplitude (p = .887) during isometric muscle actions or EMG amplitude (p = .991) during concentric muscle actions. Table 3 indicates mean EMG (µV) amplitude during isometric muscle actions and Table 4 represents mean EMG (µV) amplitude values during the fatigue protocol (mean of first five repetitions and last five repetitions). Furthermore, no significant differences were noted for EMG MPF (p = .924) during isokinetic muscle actions and these data are presented in Table 5. The subjects exhibited the same pattern of EMG responses between the groups.
Isometric EMG (µV) at 100%, 75%, 50% and 25% of MVC
Average isokinetic EMG measurements (µV) of first five repetitions and last five repetitions
Average isokinetic EMG MPF (Hz) of first five repetitions and last five repetitions
There were no significant three-way (group x day x %MMG) interactions for MMG amplitude (p = .502) during isometric muscle actions or MMG amplitude (p = .697) during concentric muscle actions. Table 6 indicates mean MMG (mV) amplitude during isometric muscle actions and Table 7 represents mean MMG (mV) amplitude values during the fatigue protocol (mean of first five repetitions and last five repetitions). The subjects exhibited the same pattern of MMG responses between the groups for MMG amplitude.
Isometric MMG (mV) at 100%, 75%, 50% and 25% of MVC
Average isokinetic MMG measurements (mV) of first five repetitions and last five repetitions
A significant three-way (group x day x MMG MPF) interaction was noted for MMG MPF (p = .017) during isokinetic muscle actions and these data are presented in Table 8. To complete further analysis, data were separated into separate two way ANOVA's but no interactions were found for the MAS group (day x MMG MPF) (p = .863) of CON group (day x MMG MPF) (p = .238). Data was further collapsed across repetitions and no significant interactions were noted for day for either the CON group (p = .142) or MAS group(p = 0.198).
Average isokinetic MMG MPF (Hz) measurements of first five repetitions and last five repetitions
No significant differences were found in torque between the treatment or control group which is consistent with other researchers (7,19). Both groups exhibited similar torque responses day 1 to day 2 with the CON decreasing by 24.7% and the MAS by 28.5%. It is interesting to note that torque had not returned to baseline for either group by day 7. Furthermore, CON increased torque from day 3 to day 7, whereas the torque for the MAS group actually decreased from day 3 to day 7. Our results indicate that massage may actually attenuate the contractile properties of the biceps brachii muscle after 7 days which is in contrast to previous research indicating that massage does not decrease the time course associated with DOMS or increase muscular activity (6). No significant differences were found in electromyography between the treatment or control group. Our data is consistent with Bobbert et al. (2) and Howell et al. (9) who reported no change in EMG following delayed onset muscle soreness. Based on our findings, torque fluctuations cannot be explained by our EMG results. For instance, the treatment group shows a decline in torque from day 1 to day 2, but there is not a change in EMG. RMS MMG was not significantly different between control and treatment groups. Both MMG and torque were not significantly different in our study. Shinohara and Sogaard (15) reported that the amplitude of MMG represents the magnitude of force fluctuations in a muscle. From day 1 to day 2, MMG decreased by 11.3% in the control group and 8% in the treatment group. Furthermore, the treatment group showed a 23.7% decrease from day 1 to day 7 for torque and a 20.3% decline in MMG from day 1 to day 7.
Shinohara and Sogaard (15) theorize that recruitment of motor units increases MMG amplitude, whereas an increase in the discharge rate of motor units contributes to an increase in mean power frequency and a decrease in the amplitude of the MMG. The MMG can reflect temporal information on fluctuations in muscle force during a steady contraction (15).
Our results did not indicate significant differences in girth with DOMS. Previous research indicates significant changes in muscle girth with DOMS (9). Although arm girth did not change significantly in our study, muscle soreness increased significantly following the DOMS protocol in both groups (CON andMAS). Smith et al. (17) concluded that DOMS does increase muscle soreness asour study also indicated through the VAS (16). In the present study, both groups responded similarly to DOMS after one week and these results are analogous to previous studies (3,18) in which no significant difference in pain ratings were found between the control group and treatment group. O’Connor and Hurley (12) concluded in a study the athletic massage 20 hours after exercise had a positive effect on muscle soreness. As mentioned in their study, athletic massage may interrupt the accumulation of neutrophils resulting in a diminished inflammatory response and a concomitant reduction in DOMS (12).
Based on the data obtained from this research cohort, massage does not significantly affect the perception of pain, muscle circumference or intrinsic properties of the muscle. However, trends in the data appear to attenuate the perception of pain obtained from a visual analog scale and the inflammation process as evaluated by the measurement of muscle circumference. A larger sample size should be considered for further research.
Based data obtained from the current project massage appears to attenuate the perception of pain following an acute bout of exercise that induces DOMS. The perception of acute pain can be a rate limiting factor in many sporting endeavors especially where repeated performance is involved. If anathlete’s perceived level of pain can be diminished with the application of massage then performance in a subsequent bout of acute exercise might be improved. Further, our data indicated that massage results in diminished inflammation associated with DOMS, which may help with the perception of pain. Evidence regarding the efficacy of massage following an acute bout of exercise remains equivocal and more research needs to be performed in order to formulate a more concrete answer to this question.
1. Barry, D. T. (1987). Acoustic signals from frog skeletal muscle. Biophysical Journal, 51(5), 769-773.
2. Bobbert, M. F., Hollander, A. P., & Huijing, P. A. (1986). Factors indelayed onset muscular soreness of man. Medicine and Science in Sports and Exercise, 18(1), 75-81.
3. Connolly, D. A., Sayers, S. P., & McHugh, M. P. (2003). Treatment and prevention of delayed onset muscle soreness. Journal of Strength and Conditioning Research, 17(1), 197-208.
4. Cram, J. R., Kasman, G. S., & Holtz, J. (1998). Introduction to Surface Electromyography. Gaithersburg, MD: Aspend Publishers, Inc.
5. Diemont, B., Figini, M. M., Orizio, C., Perini, R., & Veicsteinas, A.(1988). Spectral analysis of muscular sound at low and high contraction level. International Journal of Biomedical Computing, 23(3-4), 161-175.
6. Farr, T., Nottle, C., Nosaka, K., & Sacco, P. (2002). The effects of therapeutic massage on delayed onset muscle soreness and muscle function following downhill walking. Journal of Science and Medicine in Sport,5(4), 297-306.
7. Hilbert, J., & Mohsenin, V. (2003). Can periodic limb movement disorder be diagnosed without polysomnography? A case-control study. SleepMedicine, 4(1), 35-41. doi: S1389945702002381 [pii]
8. Hilbert, J. E., Sforzo, G. A., & Swensen, T. (2003). The effects of massage on delayed onset muscle soreness. Br J Sports Med, 37(1),72-75.
9. Howell, J. N., Chila, A. G., Ford, G., David, D., & Gates, T. (1985). An electromyographic study of elbow motion during postexercise muscle soreness. Journal of Applied Physiology, 58(5), 1713-1718.
10. Kwatny, E., Thomas, D. H., & Kwatny, H. G. (1970). An application ofsignal processing techniques to the study of myoelectric signals. IEEETransactions on Biomedical Engineering, 17(4), 303-313.
11. Lightfoot, T., Char, D., McDermott, J., & Goya, C. (1997). Immediate post exercise massage does not attenuate delayed onset muscle soreness. Journal of Strength and Conditioning Research, 11, 119-124.
12. O'Connor, R., & Hurley, D. A. (2003). The effectiveness of physiotherapeutic interventions in the management of delayed-onset musclesoreness: a systematic review. Physical Therapy Reviews, 8,177-195.
13. Orizio, C. (1993). Muscle sound: bases for the introduction of a mechanomyographic signal in muscle studies. Critical Reviews in Biomed Engineering, 21(3), 201-243.
14. Orizio, C., Diemont, B., Esposito, F., Alfonsi, E., Parrinello, G.,Moglia, A., & Veicsteinas, A. (1999). Surface mechanomyogram reflects thechanges in the mechanical properties of muscle at fatigue. European Journalof Applied Physiology and Occupational hysiology, 80(4), 276-284.
15. Shinohara, M., & Sogaard, K. (2006). Mechanomyography for studying force fluctuations and muscle fatigue. Exercise Sport Science Reviews, 34(2), 59-64.
16. Smith, L. L., Keating, M. N., Holbert, D., Spratt, D. J., McCammon, M. R., Smith, S. S., & Israel, R. G. (1994). The effects of athletic massage on delayed onset muscle soreness, creatine kinase, and neutrophil count: a preliminary report. J Orthop Sports Phys Ther, 19(2), 93-99.
17. Smith, L. L., Keating, M. N., Holbert, D., Spratt, D. J., McCammon, M. R., Smith, S. S., & Israel, R. G. (1994). The effects of atletic massage on delayed onset muscle soreness, creatine kinase, and neutrophil count: A preliminary reoprt. The Journal of Orthopaedic and Sports Physical Therapy, 19, 93-99.
18. Weber, M. D., Servedio, F. J., & Woodall, W. R. (1994). The effects of three modalities on delayed onset muscle soreness. Journal of Orthopaedic and Sports Physical Therapy, 20(5), 236-242.
19. Zainuddin, Z., Newton, M., Sacco, P., & Nosaka, K. (2005). Effects of massage on delayed-onset muscle soreness, swelling, and recovery of muscle function. Journal of athletic raining, 40(3), 174-180.