ARTICLE IN PRESS

Journal of Bodywork and Movement Therapies (2008) 12, 312–317

Bodywork and

Journal of

Movement Therapies

1360-8592/$ - sdoi:10.1016/j.j

�Correspond63501, USA. Te

E-mail addr

www.elsevier.com/jbmt

CLINICAL METHODS STUDY

A comparison of two muscle energy techniques forincreasing flexibility of the hamstring muscle group

Madeleine Smith, B.Clin.Sc., M.H.Sc.(Osteo)a, Gary Fryer, Ph.D.,B.App.Sc.(Osteo.), N.Da,b,c,�

aSchool of Biomedical and Clinical Sciences, Victoria University, Melbourne, AustraliabCentre for Ageing, Rehabilitation & Exercise Science, Victoria University, Melbourne, AustraliacA. T. Still Research Institute, A. T. Still University of Health Sciences, 800 W. Jefferson St.,Kirksville, MO 63501, USA

Received 17 April 2008; received in revised form 15 June 2008; accepted 17 June 2008

KEYWORDSMuscle;Hamstring;Stretching;Isometric;Osteopathicmedicine

ee front matter & 2008bmt.2008.06.011

ing author at: A. T. Stilll. +1 660 626 2530 Fax:ess: gfryer@atsu.edu (G

Summary Variations in the application of muscle energy technique (MET) forincreasing the extensibility of muscles have been advocated, but little evidenceexists to support the relative merit of a particular approach. This study investigatedtwo types of muscle energy techniques that have been advocated in the osteopathicliterature that differ primarily in the duration of the post-contraction stretch phase.Forty asymptomatic participants (mean age ¼ 22.173.5, male female ¼ 1:4) wererandomly allocated to one of two groups (Group 1: MET with 30-s post-isometricstretch phase; Group 2: METwith 3-s post-isometric stretch phase). Hamstring lengthwas measured using active knee extension (AKE). Participants received an initialapplication of the allocated intervention, and then a second application 1 weeklater. Analysis with a split-plot ANOVA revealed a significant effect of time(F3,36 ¼ 42.30;po0.01), but no significant time*group interaction (F3,36 ¼ 0.12;p ¼ 0.95). Post-hoc analysis revealed that the significant differences over timeoccurred between pre- and post-measurements at both weeks, and between post-Week 1 and pre-Week 2 measurements.

Both techniques appeared to be equally effective in increasing hamstringextensibility, and there appeared to be sustained improvement 1 week followingthe initial treatment. The findings suggest that altering the duration of the passivestretch component does not have a significant impact on the efficacy of MET forshort-term increases in muscle extensibility.& 2008 Elsevier Ltd. All rights reserved.

Elsevier Ltd. All rights reserved.

Research Institute, A. T. Still University of Health Sciences, 800 W. Jefferson St., Kirksville, MO+1 660 626 2099.. Fryer).

ARTICLE IN PRESS

A comparison of two muscle energy techniques for increasing flexibility 313

Introduction

Muscle energy technique (MET) is a manual proce-dure that uses controlled, voluntary isometriccontractions of a targeted muscle group and iswidely advocated by authors in the field ofosteopathy. MET is claimed to be useful forlengthening a shortened muscle, improving rangeof motion at a joint and increasing drainage of fluidfrom peripheral regions (Greenman, 2003). Muscleenergy procedures, and related post-isometricprocedures such as proprioceptive neuromuscularfacilitation (PNF), have been demonstrated to bemore effective than static stretching for improvingthe extensibility of shortened muscles (Handelet al., 1997; Magnusson et al., 1996a; Sady et al.,1982).

Passive stretching of various muscle groups,particularly the hamstrings, has been reported toimprove the length and extensibility of muscles inboth short and long-term periods of stretching(Bandy et al., 1997; Bandy et al., 1994; Felandet al., 2001; Roberts and Wilson, 1999). Addition-ally, many researchers have reported that post-isometric stretching techniques, such as MET andPNF, produce greater changes in range of motionand muscle extensibility than static or ballisticstretching, immediately following treatment(Cornelius et al., 1992; Moore and Hutton, 1980;Tanigawa, 1992; Wallin et al., 1985) and in thelonger term (Handel et al., 1997; Magnusson et al.,1996a; Sady et al., 1982; Wallin et al., 1985). Theexact mechanism by which increased muscleextensibility occurs is still unclear, and probablyinvolves both neurophysiological (including changesto stretch tolerance) and mechanical factors (suchas viscoelastic and plastic changes in the connec-tive tissue elements of the muscle) (Fryer, 2006).

Although there are many variations of theapplication of MET, with most authors in the fieldof osteopathy advocating a post-isometric stretchfor increasing muscle length, the recommendedduration for the passive stretch component varies.A typical application of MET for the purpose oflengthening a shortened muscles involves thefollowing steps: (1) stretch the muscle to apalpated ‘barrier’ or to the patient’s tolerance ofstretch, (2) the patient produces a voluntaryisometric contraction of the muscle under stretchagainst the clinicians’ controlled and equal coun-terforce, (3) the muscle is allowed to relax, whilethe clinician maintains a stretch for a definedperiod, (4) the clinician ‘takes up the slack’following relaxation so that the muscle has beenlengthened to a new barrier, (5) this process isrepeated several times. It is possible to alter the

application of MET by with variations to thecomponents of the technique: the force andduration of the isometric contraction phase, theduration of the post-contraction stretch phase, andthe number of repetitions. The literature currentlyoffers little guidance as to the most efficaciousapplication (Fryer, 2006).

In the osteopathic literature, two markedlydifferent applications of MET for increasing muscleextensibility have been advocated by Greenman(2003) and Chaitow (2006), with differences in thenumber of repetitions (3–5 and 3, respectively),and the period of passive stretching between theisometric contractions. Chaitow suggests a stretchduration following isometric contraction to be heldfor at least 30 and up to 60 s for chronicallyshortened muscles, whereas Greenman (2003) andMitchell et al. (1979) recommend only enough time(several seconds) for patient relaxation and tensionto be taken up in the affected tissue. The relativemerit and efficacy of these different approacheshave not been investigated.

Ballantyne et al. (2003) and Lenehan et al.(2003) used techniques similar to the Greenmanprotocol, both following a 5–7 isometric contrac-tion with a passive stretch lasting only severalseconds until the new barrier was engaged. Otherresearchers have used PNF techniques similar tothe Chaitow method, such as Wallin et al. (1985)and Handel et al. (1997), who used a maximalisometric contraction with a 15 s rest period. Whilethese techniques were similar to the methodadvocated by Chaitow, the duration of the stretch(15 s) was shorter than the recommended minimumof 30 s (Chaitow, 2006). The longer passive stretchof the Chaitow approach may make the techniquemore effective, given that passive stretching for a30-s (Bandy and Irion 1994) or 60-s (Feland et al.,2001) period have been reported to be moreefficacious for increasing muscle extensibility thanshorter durations.

The relative efficacy of the Greenman andChaitow approaches for increasing myofascial ex-tensibility should be investigated. Most researchinvolving MET has focused on a single application oftreatment (Ballantyne et al., 2003; Mehta andHatton, 2002; Magnusson et al., 1996a), butpractitioners typically deliver more than onetreatment for a patient complaint, and anticipatethat there will be carry-over changes still presentfrom the previously delivered treatment. This studyaimed to determine the relative efficacy of the twoapproaches for increasing the extensibility of thehamstring muscles, and determine if there wereany carry-over changes in hamstring length, orchanges in responsiveness to treatment, when the

ARTICLE IN PRESS

M. Smith, G. Fryer314

hamstrings were treated on a second occasion1 week later.

Figure 1 Active knee extension measurement.

Materials and methods

Participants

Forty participants (mean age ¼ 22.173.5 years,male: female ¼ 1.4) were recruited from studentsenrolled at Victoria University, Melbourne, Austra-lia. Twenty participants were required for eachgroup to achieve 80% power, based on a large effectsize, and analysis using a split-plot ANOVA (SPANO-VA) (Aron and Aron, 1999). Volunteers under theage of 18 and over the age of 65 were excluded, aswere those with lower extremity or lower back painat the time of the study. Participants were includedif they presented with shortened hamstrings,determined as o751 of active knee extension(AKE). Volunteers with 4751 of AKE were excludedfrom the study, resulting in 15 volunteers (meanage ¼ 22 years, SD ¼ 1.4, male: female ratio of0.15) being excluded. Written, informed consentwas obtained from all participants and the studywas approved by the Victoria University HumanResearch Ethics committee.

Participants were randomly assigned to Group 1(MET with 30-s stretch) (N ¼ 20; mean age ¼ 21years, SD ¼ 2.0; male: female ratio ¼ 1.2) or Group2 (MET with 3-s stretch) (n ¼ 20; mean age ¼ 23.2years, SD ¼ 4.7; male: female ratio ¼ 1.5).

Measures

AKE was used to measure hamstring extensibility.AKE has been commonly used by researchers for themeasurement of hamstring length (Handel et al.,1997) and has been demonstrated to be a reliablemeasure (Sullivan et al., 1992).

The participants were requested to lie supineand the experimental hip flexed to 901 andstrapped to a stabilizing bar. The thigh of theopposite leg was firmly secured to minimizerotation of the pelvis (Sullivan et al., 1992). Jointmarkers were placed on the greater trochanter,lateral femoral condyle, head of the fibula andlateral malleolus to provide a point of reference tomeasure degree of knee extension (Figure 1). Theparticipants were asked to extend the knee as faras possible, and a photograph was taken of thisposition using a digital camera by Researcher 1.This procedure was performed three times. Thedigital images were analyzed using ‘‘SiliconCOACHPro’’ software to determine the angle of AKE, andthe mean of the three measures used for analysis.

Pilot reliability studyAn AKE pilot reliability study was conducted on 15participants using 2 measurements of AKE, 10minapart, where the participant remained positionedand attached to stabilsing bar in order to establishan accurate measure of the reliability of themeasurement procedure. Repeatability was deter-mined using an intra-class correlation (ICC) andfound to be highly correlated (ICC ¼ 0.99). Thestandard error of mean (SEM) for the differencescores between the two trials, which reflects thevariability of measurements due to repetition andrandom error and provides an indication of theabsolute reliability of the measurement procedure,was calculated to be 0.79.

Procedure

Researcher 1 measured the AKE of the investigatedleg and then left the room. Researcher 2 (GF; aregistered osteopath with 15 years clinical experi-ence) entered the room and assigned participantsto treatment groups via lottery draw. Researcher 2(MS; a final-year osteopathic student) treated theexperimental leg of the participants according tothe group allocation (Figure 2), and then left theroom. Researcher 1, who was blinded to thetreatment allocations, re-entered the room andperformed the post-treatment AKE measurement

All participants returned 1 week later, receivingthe same measurement and treatment procedureas previously described. There were no restrictionsto participant activity between treatments.

InterventionSubjects allocated to Group 1 (n ¼ 20; meanage ¼ 21, SD ¼ 2; male: female ¼ 1:2) received

ARTICLE IN PRESS

Table 1 Mean AKE values (SD) for MET groups

Week Treatment Pre-treatment

Post-treatment

Pre–post

1 Group 1 145.52(8.64)

154.00 8.484

Group 2 142.23 150.23 7.89

2 Group 1 147.62 153.7 6.05Group 2 145.12 151.48 6.37

155.00

152.50

150.00

147.50

Group 2Group 1

Treatment Allocation

Ave

rage

AK

E

Figure 2 Application of muscle energy technique.

A comparison of two muscle energy techniques for increasing flexibility 315

an application of MET as advocated by Chaitow(2006): the supine subject’s hip was passivelyflexed and the leg extended until tension wassensed by the researcher and the subject reporteda moderate stretching sensation. The participantprovided a moderate (approximately 40% of max-imal contraction) knee flexion isometric contrac-tion, against the researcher’s shoulder for 7–10 s.This was followed by 2–3 s of relaxation, and thenthe leg was passively stretched to the palpatedbarrier and/or tolerance to stretch and held for30 s. The leg was then lowered to the table for ashort resting period (approximately 10 s). Thisprocedure was repeated two more times.

Those subjects allocated to Group 2 (n ¼ 20;mean age ¼ 23.2, SD ¼ 4.7; male: female ¼ 1:5)received an application of MET as advocated byGreenman (2003): the supine subject’s hip waspassively flexed and the leg extended until tensionwas sensed by the researcher and the subjectreported a moderate stretching sensation. Theparticipant then provided a moderate (approxi-mately 40% of maximal contraction) knee flexionisometric contraction against the researcher’sshoulder for 7–10 s. The subject was allowed torelax for 2–3 s with the stretch maintained, andthen the leg was further extended to the palpatedbarrier and/or tolerance to stretch. This procedurewas repeated three more times.

Week 2Post PPT

Week 2Pre PPT

Week 1Post PPT

Week 1Pre PPT

145.00

142.50

Treatment Time

Figure 3 Comparative means of AKE (degrees) over time.

Analysis

All data were collated and analyzed using SPSSprofessional version 16. Pre- and post-interventionROM measurements were analyzed for both groupsusing a multi-variate (Hotelling’s T) SPANOVA.

Statistical significance was set at the po0.05 level.Post-hoc analysis for significant differences be-tween the time periods were conducted withpaired t-tests.

Results

The means and standard deviations for eachmeasurement period (pre-treatment, post-treatment and pre–post) are shown in Table 1 andFigure 3. Pre–post gains were similar for bothGroups 1 and 2 at Week 1 (8.48 and 7.89) andWeek 2 (6.05 and 6.37) (Table 2).

Analysis with the SPANOVA revealed a significanteffect of time (F3,36 ¼ 42.30; po0.01), but nosignificant time*group interaction (F3,36 ¼ 0.12;p ¼ 0.95). Further post-hoc analysis of the entiregroup over each time interval using paired t testsrevealed significant differences between Week 1pre-measurement and all other measurement

ARTICLE IN PRESS

Table 2 Post-hoc analysis of changes over time

Treatment phase Mean (SD) p-value

Week 1 pre–post �8.24 (4.89) 0.00Week 1 pre—Week 2 pre �2.49 (7.19) 0.04Week 1 pre—Week 2 post �8.70 (9.15) 0.00Week 1 post—Week 2 pre 5.75 (7.85) 0.00Week 1 post—Week 2 post �.46 (8.68) 0.74Week 2 pre—Week 2 post �6.21 (4.99) 0.00

M. Smith, G. Fryer316

periods, and between Week 1 post- and Week 2 pre-treatment measurements, and between Week 2pre- and post-measurements, suggesting bothshort- and long-term significant changes in AKE(Table 2).

Discussion

Many variations of post-isometric stretching havebeen advocated for the lengthening of shortenedmuscles, but little research is available to helpclinicians choose the most efficacious method.Although a few researchers have examined theeffect of varying the application of certain compo-nents in these techniques (Feland and Marin, 2004;Mehta and Hatton, 2002; Rowlands et al., 2003),little can yet be concluded concerning the mostefficacious application. A direct comparison be-tween the methods advocated by Chaitow andGreenman, two of the most commonly advocatedapproaches to muscle stretching in the osteopathicliterature, had not previously been investigated.

The present study found that the immediatepre–post comparison of AKE at both treatmentperiods were significantly increased, demonstratingan increase in the extensibility of the hamstringmuscle group. Although there was a significantwithin-group change over time, no significant differ-ences were evident between the two treatmentgroups. Neither method in the present study wasfound to produce significantly greater gains in rangeof motion on either treatment day or betweentreatment days, suggesting that both treatmenttechniques were equally effective for increasingrange of motion. This in itself is an important finding,and implies that while MET may be more effectivethan static stretching (Handel et al., 1997; Magnussonet al., 1996a; Sady et al., 1982), variations in theduration of post-isometric stretch do not significantlyalter the efficacy for increasing muscle length.However, it is worth noting that the Greenmanapproach may be more time efficient because theperiod between contractions is substantially shorter.

In the present study, the two treatments wereseparated by a period of 1 week, and there was asignificant difference between the initial pre-intervention values and both the pre- and post-intervention values of the second week. While onlytwo treatments in 8 days is not representative oftypical clinical practice, the results suggest a carry-over effect from the first treatment. These resultsalso suggest that a longer period of treatment mayproduce further gains in range of motion, and this isworthy of investigation in the future.

Bandy and Irion (1994) explored the effect ofvariations in the duration of each individual stretchover a series of treatments, and found that a 30-sduration of passive stretching produced increases inrange of motion greater than shorter durations, butno different from longer ones. This suggests thatthe longer passive stretch of the Chaitow approach(30–60 s) may make the technique more effica-cious. This was not demonstrated, however, in thepresent study. It may be that application over alonger period of treatment, such as the 8 weeksused by Bandy and Irion (1994), may be needed todemonstrate differences in the efficacy of the twotechniques for increasing muscle extensibility, andthis is an area that should be further investigated.

The mechanism of action of isometric contrac-tion or static stretching is still unclear, andprobably involves both neurophysiological andmechanical factors (Fryer, 2006). AKE provides areliable measure of range of motion, but does notmeasure the torque generated to produce thestretch and does not give any evidence of abiomechanical change to the muscle property.The few studies that have measured torque appliedin pre- and post-passive knee extension (PKE)following isometric stretching strongly suggest thata change in tolerance to stretch, rather thanviscoelastic change, is the main mechanism forincreased extensibility (Magnusson et al., 1996b;Ballantyne et al., 2003). It would be interesting toexamine the effect of a longer term of treatmentusing torque-controlled PKE in order to determineif changes to the viscoelastic property of themuscle was affected over a longer time course.

It should be noted that while the subjects in thepresent study displayed limited hamstring flexibil-ity, all were asymptomatic and likely to be youngerthan a typical patient population. This is animportant limitation of the study, because it ispossible that differences in the efficacy of the twoMET approaches may be found when treating amore representative patient population, such as anolder group, those with hamstring pain or healinghamstring tears. It is feasible that these techniquesmay help to prevent adhesion and cross-link

ARTICLE IN PRESS

A comparison of two muscle energy techniques for increasing flexibility 317

formation in the repairing muscle (Lederman,2005), and future study of these populations withthese stretching approaches may yield informationconcerning relative treatment effectiveness thatproves valuable for clinicians.

Although the present study investigated METwithout comparison to either a control or staticstretching group, other researchers have reportedsignificant differences between contract–relaxtreatments and passive stretching (Tanigawa,1992; Wallin et al., 1985; Moore and Hutton,1980). While there were significant differencesbetween pre-initial and post final AKE measure-ments, a comparison to both a control and staticstretching would have added greater validity to theresults, and clearly demonstrated that both thesepost-isometric techniques were superior to passivestretching in this group of subjects.

The present study demonstrated that both theGreenman and Chaitow approaches to MET resultedin increased AKE, both immediately following thetreatments and 1 week after treatment. Furtherstudy relating to the relative efficacy of the specificcomponents of MET techniques will be required todetermine most appropriate clinical application.

Conclusion

This study found that both Greenman and Chaitowmuscle energy approaches produced increased AKEimmediately after intervention, and demonstrateda carryover effect 1 week later. There was asignificant increase in range of motion of the kneeimmediately following both treatments at bothweeks, and a significant increase at the pre-treatment measurement at Week 2. There was,however, no significant difference between the twoapplications. This suggests that variations in theelements of the techniques, such as the duration ofpassive stretch, may not have a significant influ-ence on the efficacy of the technique for increasinghamstring extensibility.

References

Aron, A., Aron, E.N., 1999. Statistics for Psychology, second ed.New Jersey, Prentice-Hall.

Bandy, W.D., Irion, J.M., 1994. The effect of time on staticstretch on the flexibility of the hamstring muscles. PhysicalTherapy 74 (9), 845–850.

Bandy, W.D., Irion, J.M., Briggler, M., 1997. The effect of timeand frequency of static stretching on flexibility of thehamstring muscles. Physical Therapy 77, 1090–1096.

Ballantyne, F., Fryer, G., McLaughlin, P., 2003. Journal ofOsteopathic Medicine, April 6 (1), 37.

Chaitow, L., 2006. Muscle Energy Techniques, third ed. ChurchillLivingstone, Edinburg.

Cornelius, W.L., Ebrahim, K., Watson, J., et al., 1992. Theeffects of cold application and modified PNF stretchingtechniques on hip joint flexibility in college males. ResearchQuarterly in Exercise and Sports 63 (3), 311–314.

Feland, J.B., Marin, H.N., 2004. Effect of sub maximalcontraction intensity in contract-relax proprioceptive neuro-muscular facilitation stretching. British Journal of SportsMedicine 38 (4), E18.

Feland, J.B., Myrer, J.W., Schulthies, S.S., et al., 2001. Theeffect of duration of stretching of the hamstring musclegroup for increasing range of motion in people aged 65 yearsor older. Physical Therapy 81, 1100–1117.

Fryer, G., 2006. Muscle energy technique: research and efficacy.In: Chaitow, L. (Ed.), Muscle Energy Techniques, third ed.Churchill Livingston, Edinburgh, pp. 109–132.

Greenman, P.E., 2003. Principles of Manual Medicine, third ed.Lippincott Williams and Wilkins, Baltimore.

Handel, M., Horstmann, D., Dickhuth, H., et al., 1997. Effects ofcontract-relax stretching training on muscle performance inathletes. European Journal of Applied Physiology andOccupational Physiology 76 (5), 400–408.

Lederman, E., 2005. The Science and Practice of ManualTherapy, second ed. Churchill Livingstone, London.

Lenehan, K.L., Fryer, G., McLaughlin, P., 2003. The effect ofmuscle energy technique on gross trunk range of motion.Journal of Osteopathic Medicine 6 (1), 13–18.

Magnusson, P., Simonsen, E., Aagaard, P., Dyhre-Poulsen, P.,Malachy, P., McHugh, M., Kjaer, M., 1996a. Mechanical andphysiological responses to stretching with and withoutpreisometric contraction in human skeletal muscle. PhysicalMedicine and Rehabilitation 77 (4), 373–378.

Magnusson, S.P., Simonsen, E.B., Aagaard, P., Sorensen, H.,Kajer, M., 1996b. A mechanism for altered flexibility inhuman skeletal muscle. Journal of Physiology 487, 291–298.

Mehta, M., Hatton, P., 2002. The relationship between theduration for sub maximal isometric contraction and improve-ment in the range of passive knee extension. Journal ofOsteopathic Medicine 5 (1), 40.

Mitchell Jr., F.L., Moran, P.S., Pruzzo, N.A., 1979. An Evaluationand Treatment Manual of Osteopathic Muscle Energy Proce-dures. Institute for continuing Education in OsteopathicPrinciples, Missouri.

Moore, M., Hutton, R., 1980. Electromyographic investigation ofmuscle stretching techniques. Medicine and Science in Sportsand Exercise 12 (5), 322–329.

Roberts, J.M., Wilson, K., 1999. Effect of stretching duration onactive and passive range of motion in the lower extremity.British Journal of Sports Medicine 33 (4), 259–263.

Rowlands, A.V., Marginson, V.F., Lee, J., 2003. Chronic flexibilitygains: effect of isometric contraction duration duringproprioceptive neuromuscular facilitation stretching techni-ques. Research Quarterly Exercise and Sport 74 (1), 47–51.

Sady, S.P., Wortman, M., Blanke, D., 1982. Flexibility training:ballistic, static or proprioceptive neuromuscular facilitation?Archives of Physical Medicine and Rehabilitation 63 (6), 261–263.

Sullivan, M., Dejulia, J., Worrell, T., 1992. Effect of pelvic positionand stretching method on hamstring muscle flexibility. Medicineand Science in Sports and Exercise 24 (12), 1383–1389.

Tanigawa, M.C., 1992. Comparison of the hold-relax procedureand passive mobilization on increasing muscle length.Physical Therapy 52, 725–735.

Wallin, D., Ekblom, B., Grahn, R., Nordenborg, T., 1985. Improve-ment in muscle flexibility. A comparison between two techni-ques. American journal of Sports Medicine 13 (3), 263–268.