Worrell Et Al., 1994

Journal of Orthopaedic & Sports Physical Therapy Official Publication of the Orthopaedic and Sports Physical Therapy Sections of the American Physical Therapy Association

Effect of Hamstring Stretching on Hamstring Muscle Performance Teddy W. Worrell, EdD, PT, SCS, ATC, FACSM1 Troy I. Smith, MS, PT, ATCZ lason Winegardner, MS, PT3

ncreasing athletic perforniance is a common goal for athletes, coaches, and sports medicine clinicians. Most efforts to improve sprint perf o ~ m ~ a n chave e focused 011sprint training and increasing muscle perforniance. Sprinting places maximuni demands on the niusculotendinous structures of the lower extremity. Specifically, the hamstring riiuscles become riiore active than any other lower extremity niuscle during sprinting (6). In addition, Mann and Sprague (7) reported that sprinters who were faster generated the greatest hamstring niuscle moment (torque) during ground contact. Moreover, they reported that the faster sprinters had riiore hamstring injuries than the slower sprinters. In the past, much of the literature has focused on the relationship between hamstring flexibility and injury (2 1) and the most effective hamstring stretching method (1 5). Sullivan et al (1 5) reported that no significant difference existed between static and proprioceptive neuromuscular facilitation (PNF) hamstring stretching. This conflicts with ~iiost studies that report the superiority of PNF stretching conipared with static stretching (5,8- 10.12.13). Very little literature has focused on the relationship between niuscle flexibility and force production ( 1 8). During an eccentric contraction, mechanical work is absorbed bv the se-

The relationship between hamstring flexibility and hamstring muscle performance has not been reported. The purposes of this study were I) to determine the most effective stretching method for increasing hamstring flexibility and 2) to determine the effects of increasing hamstring flexibility on isokinetic peak torque. Nineteen subjects participated in this study. A two-way analysis of variance was used to compare two stretching techniques: proprioceptive neuromuscular facilitation stretch and static stretch. A one-way repeated measures analysis of variance was used to compare hamstring isokinetic values pre- and poststretching. No significant increase occurred (p > .05) in hamstring flexibility even though increases occurred with each technique: static stretch (+2 1.3%) and proprioceptive neuromuscular facilitation (+25.7%). Significant increases occurred in peak torque eccentrically at 6O0/sec (p < .05, +8.S0/o) and 12O0/sec(p < .05, 13.5%) and concentrically at 12O0/sec(p < .05, 11.2%). No significant increase occurred at 6O0/sec (p > .05, +2.5%). We concluded that increasing hamstring flexibility was an effective method for increasing hamstring muscle performance at selective isokinetic conditions. Further study is needed to determine if increasing hamstring flexibility will increase performance in closed kinetic chain activities.

+

+

Key Words: hamstring flexibility, active knee extension test, stretching

' Assistant Professor and Director of Research, Krannerf School of Physical Therapy, University of Indianapolis, 1400 E. Hanna Ave., Indianapolis, IN 46227-3697 Staff Physical Therapist, Aboite Physical Therapy, Inc., Fort Wayne, IN

'Staff Physical Thera~ist,Wishard Memorial Hosoital, Indiana~olis,IN ries elastic component of the muscle as potential energy, which is used during the immediate concentric contraction (3,4). This condition of eccentric contraction followed by concentric contraction occurs during gait and running. For example, the quadriceps femoris undergoes an eccentric contraction during heelstrike and concentric contraction at pushoff. T h e same is true for the gastrocnemius and soleus nluscles during niidstance and push-off. Factors that determine the amount of energy absorbed by muscles are the speed of the eccentric contraction and the length of the niuscle (3). Thus, if the length of the muscle can be increased, more forces will be ab-

sorbed during the eccentric contraction and more forces will be generated during the concentric contraction. Theoretically, patients with lower extremity overuse injuries would benefit by muscle stretching because greater force will be absorbed, lessening the overload on weakened and inflamed tissues. In addition, theoretically, muscle performance will be increased for activities of daily living o r sports by increasing the potential energy available for concentric contractions. Recently, Wilson et al (18) demonstrated that increases in pectoralis and deltoid flexibility resulted in significant increases ( p < 0.05) in initial concentric bench-press work if the Volume 20 Number 3 September 1994 JOSPT

RESEARCH STUDY

concentric contraction was preceded immediately by an eccentric contraction. During sprinting, concentric hamstring contractions are preceded by eccentric contractions (6.20). No reports a r e available that describe the relationship between increasing hamstring flexibility and increasing hamstring muscle performance, which has important implications for enhancing athletic performance. Therefore, the purposes of this study were 1 ) t o determine the most effective method for increasing hamstring flexibility and 2) t o determine the effects of increasing hamstring flexibility on isokinetic hamstring peak torque.

METHODS Subjects Nineteen university students without a history of knee o r hamstring muscle injury participated in this study (Table 1). In addition, subjects lacked a t least 20" (range 2252") of active knee extension with the hip in 90" of flexion during the active knee extension test. Prior t o participation, each subject read and signed a consent form approved by o u r university human subjects committee. ,

lsokinetic Testing A Biodex isokinetic dynamometer (Biodex, Shirley. New York, NY) was used to measure hamstring peak torque values. Calibration was performed prior to testing. Subjects were tested in the seated position with their arms crossed over their chest and straps for stabilization placed over their waist and distal thigh. T h e tibia1 pad was placed and secured approximately two fingerwidths proximal t o the medial malleolus. Axis of the dynamometer was aligned with the knee axis. T h e testing protocol consisted of an eccen-

Theoretically, patients with lower extremity overuse injuries would benefit by muscle stretching because greater force will be absorbed, lessening the overload on weakened and inflamed tissues.

Flexibility Assessment Hamstring muscle flexibility was assessed with the active knee extension test (15,22). Subjects were placed in a supine position with the anterior thigh touching the crossbar of a testing apparatus. T h e hip and knee angles were visually estimated at 90". In this position, a inclinometer was placed 1 inch below and parallel t o the fibular head. During the warm-up procedure, the subjects actively extended a leg four times while maintaining anterior thigh contact against the crossbar (1 6). Then, subjects actively extended the knee two additional times during which knee extension was recorded (1 5,22).

JOSPT* Volume 20 * Number 3 * September 1994

tric loading of the hamstring muscle group, followed by an immediate concentric hamstring muscle contraction. In order to accomplish this, the Biodex was set u p in the passive mode for flexion and extension. Eccentric muscle contraction occurred during passive knee extension mode, and the concentric phase occurred during the passive knee flexion mode. Subjects received standardized verbal cues of "holdn during the eccentric phase and "pulln during the concentric phase, with instruction t o not relax between the two stages but to maintain hamstring contraction throughout the arc of

movement. Both eccentric and concentric peak torque values were recorded between 0 and 90" of knee flexion a t 60°/sec (1.02 radianslsec) and 120°/sec (2.04 radianslsec). Prior to the three maximum test repetitions, the subjects underwent a series of warm-up sets. T h e initial set consisted of five repetitions of the eccentric phase only, then five repetitions of the concentric phase only at approximately 50% maximum effort. Three repetitions at 50% maximum effort of the eccentric/concentric cycle were then performed. T h e next two warm-up sets consisted of three and two repetitions at 7 5 and 100% effort, respectively. Each warm-up set was separated by 30 seconds. T h e testing trials were separated from the warm-up sets by a 1minute rest period. Subjects were tested bilaterally. Gravity effect torque was measured with the leg in full knee extension in the seated position. Testing order alternated between right and left legs. All subjects participated in a familiarization session prior t o the actual pretest session. During the familiarization session, subjects practiced the eccentric and concentric components separately and then practiced the eccentric/concentric contraction mode for several repetitions. Adequate practice was allowed until subjects were comfortable with the testing procedure.

Stretching Protocol During the familiarization session, subjects were instructed on how to perform an anterior pelvic tilt while standing (1 5). All subjects demonstrated the ability to obtain an anterior pelvic tilt in the standing position prior t o the experimental study. Subjects were told t o face a table o r chair and place the heel of the leg t o be stretched on the table o r chair seat (this was determined by subject's comfort and his/her ability to maintain an anterior pelvic tilt), keep their hands on their hips, hold their head in a neutral position look-

RESEARCH STUDY

Gender

Age

Height (c&

Males ( N = 10) Females ( N = 9)

25.7 f 2.4 26.7 f4.8

180.2 5.7 166.7 f4.1

+

Weight (kg)

80.1 f 13.7 59.8 6.1

+

TABLE 1. Description of subjects (rfSD).

ing forward, keep the stretched leg fully extended, extend their cervical and thoracic spine, and retract their scapulae while maintaining an anterior pelvic tilt. Then they were asked to move their trunks forward at the pelvis until they perceived a hamstring stretching sensation without pain (Figure 1). Each subject stretched both legs, one leg using static stretch and the other leg using contract-relax-contract (PNF) stretch. Both stretching methods were performed in an anterior pelvic tilt position. Assignment of stretching technique was randomly determined. T h e static stretch leg was stretched four repetitions of 15-20 seconds. Each repetition was separated by 15 seconds. T h e PNF stretch leg was then stretched during four bouts of 20 seconds. Each repetition consisted of 5 seconds of maximal isometric hamstring contraction, 5 seconds of rest, 5 seconds of maximal isometric quadriceps contraction, and 5 seconds of rest (contractrelax-contract). Subjects performed four repetitions of each stretching method 5 days a week (Monday-Friday) at approximately the same time each day for 3 weeks (15 stretching sessions). Subjects were monitored during each stretching session to ensure proper performance of the stretching methods.

analysis. A two-way analysis of variance (stretching method and time) was used to compare stretching techniques. A one-way analysis of variance was used to compare hamstring isokinetic values pre- and poststretching. Probability was set at p < .05. Daily attendance was kept. Subjects participated in a total of 570 individual stretching sessions (38 extremities for 15 days).

tr11 pootion. (From Sullrvan MK, Delulia /I, Worrell TW: Eifect of pelvic position and stretching method on hamstring muscle flexibility. Med Sci Sports Exerc 24:1383-1389,O The American College of Sports Medicine, 1992. with permission).

Reliability Study Prior to the experimental study, 10 subjects were tested bilaterally 7 days apart to determine intrarater

Stretching of a musculotendinous unit may affect neuromuscular transmission. intersession reliability of the active knee extension test and isokinetic dynamometry. T h e testers were blinded from the prior test results for both the active knee extension test and the isokinetic dynamometer.

Statistical Analyses For the reliability study, intraclass correlations (ICC 2.1) (14) and standard errors of measurement (SEM) (2) were calculated for the isokinetic and flexibility data. For determining active knee extension test reliability, the mean of two test trials was used for data analysis. For isokinetic reliability, the largest single peak torque value was used for data

FIGURE 1. Hamstrrng stretchrng rn an dntenor pelvrc

RESULTS For the reliability study, intratester ICC and SEM for the active knee extension test measures were .93 and 2.9 1" , respectively. For the isokinetic dynamometer, intratester ICC and SEM ranged from .95 to .97 and 8.2 to 13.2 Nm, respectively (Table 2). For the experimental study, no significant increase occurred in ham-

string flexibility for either stretching method ( p > .05). T h e hamstring flexibility of the statically stretched leg increased 8.0" and the flexibility of the PNF stretched leg increased 9.5" (Table 3). Since no significant increase occurred for either stretching method, stretching groups were collapsed to compare pre- and posttest isokinetic measures. Significant increases occurred eccentrically at 60°/sec ( p < .05) and 120°/sec ( p < .05) and concentrically at 120"/set ( p < .05). No significant increase occurred concentrically at 60°/sec ( p > .05) (Table 4). Daily attendance was 99.3% (5661570).

DISCUSSION Stretching Techniques Results indicated no significant increase in motion occurred in either stretching group. However, hamstring flexibility increases (static stretch = 8.0°, +21.3%; PNF = 9.5". +25.7%) approached significance ( p = 0.082) (Table 3). Large intersubject variation occurred as revealed by the 24.6% coefficient of variation. Consequently, statistical power was low (0.40), increasing the probability of a type I1 error, ie., the Volume 20 Number 3 September 1994 JOSPT

-

Measure

Activekneeextension test 60°/sec concentric 60a/sececcentric 12O0/secconcentric 120°/sec eccentric

ICC

SEM

0.93

2.91'

inability to detect actual differences that existed because of large subject variation and small sample size. As demonstrated in Table 5, four extremities lost flexibility and a large range of responses to stretching existed (from a loss of 2.5" to a gain of 3 1.5"). We are unable to explain such a large variation in response to the stretching protocol. Generally, subjects who were less flexible

p*

0.224

0.95 12.7 Nm 0.883 0.95 10.9 Nm 0.450 0.95 13.2 Nm 0.928 0.97 8.2 Nm 0.278

* Cornparrson oi test and retest rneature5.

TABLE 2. Reliability data for the active knee extension test and isokinetic measures for the pilot studv.

Group

Pretest

Posttest

Change

SS PNF

37.5 f 8.8' 36.7 f 8.7'

29.5 f 8.6' 27.3 f 5.9"

8.0" (21.3%) 9.5' (25.7%)

SS = Static stretch. PNF = Proprioceptive neurornuscular facilitation stretch. TABLE 3. Changes in hamstring flexibility as measured by the active knee extension test. Smaller numbers on the posttest indicate more knee extension, ie., increased hamstring flexiblity.

Velocity

Pretest

Posttest

Change (%)

P

60"sec concentric 6O0/sec eccentric 120e/secconcentric 120°/sec eccentric

115.8 f 37.0 110.1 f 37.0 112.3 f 35.3 111.7 f 39.1

118.7 f 37.7 119.5 f 43.4 124.9 f 40.3 126.7 f 41.3

2.9 Nm (2.5%) 9.4 Nm (8.5%) 12.6 Nm (1 1.2%) 15.1 Nm (13.5%)

0.322 0.016 0.002 0.000

TABLE 4. lsokinetic peak torque measures.

PNFG e t %

Static Stretch Gender

Females 1 2 3 4 5 6 7 8 9 Males 1 2 3 4 5 6 7 8 9 10

~

Pretest

Posttest

Change

Pretest

Posttest

Change

26.5 28.5 31.0 43.5 22.0 40.0 38.5 45.0 25.0

23.5 20.0 25.0 34.0 22.0 30.5 40.5 36.5 25.5

3.0 8.5 6.0 9.5 0.0 9.5 -2.5 8.5 -0.5

30.5 41.5 26.5 36.5 23.5 34.0 37.0 47.0 24.5

23.0 25.5 23.0 39.0 20.5 31.5 26.5 35.0 25.0

7.5 16.0 3.5 -2.5 3.0 2.5 10.5 12.0 -0.5

33.0 50.0 29.0 44.0 52.0 42.5 45.5 38.0 44.0 33.5

30.0 45.0 15.0 36.5 38.0 34.5 35.0 30.5 12.5 25.5

3.0 5.0 14.0 7.5 14.0 8.0 10.5 7.5 31.5 8.0

28.5 49.5 26.5 45.5 51.0 45.0 43.5 33.5 40.5 33.5

25.5 32.5 17.0 35.0 29.0 32.0 30.0 26.0 18.5 24.0

3.0 17.0 9.5 10.5 22.0 13.0 13.5 7.5 22.0 9.5

PNF = Propnoceptrve neurornuscular tac~lrtatronstretch

TABLE 5. Individual changes in hamstring stretching by gender and stretching method (active knee extension test degrees from complete knee extension). Negative numbers indicate loss of hamstring flexibility; other numbers indicate increases in hamstring flexibility.

JOSPT Volume 20 Number 3 September 1994

RESEARCH S T U D Y

(larger pretest active knee extension test) tended to gain more hamstring flexibility. However, this variability in stretching response cannot be completely explained by initial active knee extension test because the correlation between initial active knee extension test and changes in active knee extension test was low (r = 0.428, p = 0.068). Increases in flexibility in this study are in general agreement with Sullivan et al (15). who reported a 9 and 1 1" increase for static stretching and PNF stretching, respectively, after 10 stretching sessions, which were statistically significant ( P < .05). We did not find a significant difference between static stretching and PNF stretching, which is in agreement with Sullivan et al and in disagreement with others (12.1 7). Other studies (12,17) have not controlled for pelvic position when performing hamstring stretching, which we believe is a confounding factor in hamstring stretching (15). Therefore, the use of either of these stretching techniques will serve the purpose of increasing flexibility. In our experience, static stretching is much easier to teach and to perform than PNF stretching. Therefore, we recommend static stretching for increasing hamstring flexibility.

Hamstring Peak Torque Significant increases in hamstring peak torque occurred eccentrically at 60 and 120°/sec and concentrically at 120°/sec (Figure 2, Table 4). No significant increase occurred in concentric peak torque at 60°/sec. Increases in eccentric force production at 60 and 120°/sec are attributed to increases in hamstring muscle flexibility and increases in compliance of the series elastic component that results in a greater ability to store potential energy (1,3,4,18,19). Improvement in concentric peak torque production at 120°/sec resulted from the increased storage of potential energy

RESEARCH S T U D Y

these results t o closed kinetic chain (distal segment on the ground) activities. Additional studies a r e needed t o determine the effect of increasing hamstring flexibility on closed kinetic chain activities.

Peak Torquo Nm

w

100

PmT&

El Postt& p0.05

50

-

n 6O0/s Con

6O0/s Ecc

120°/s

Con

120°/s Ecc

Velocity FIGURE 2. Pretest and posttest strength values. Con = Concentric, Ecc = Eccentric

during eccentric loading, which is used in the immediate concentric contraction (1,3,4,18,19). This added potential energy must be used instantaneously following the eccentric contraction (1 1). Wilson et al (1 8) demonstrated that by increasing pectoralis and deltoid flexibility, series elastic component stiffness was significantly reduced during a 70% maximal bench press repetition. Moreover, Wilson et al (18) reported that initial concentric work of the bench press was significantly increased ( p < 0.05) after stretching. In addition, the bench load increased 5.4%. which was not significant. No significant increase in hamstring force occurred concentrically a t 6 0 "/set after eccentric loading. We are unable t o explain this phenomenon. Perhaps at a slower velocity (60 vs. 120°/sec) the instantaneous moment is lost (1 1). Further study is needed t o support o r refute this finding. T h e increases in hamstring muscle performance in this study were not d u e t o a learning effect for the following reasons: 1)subjects were familiarized with the active knee extension test and Biodex prior t o pretesting, and 2) reliability data indicated that n o learning effect occurred between testing sessions

separated by 1 week as indicated by the ICC, SEM, and probability levels (Table 2). Therefore, we conclude that the improvements in muscle performance were the result of the flexibility increases (2 1.3-25.7%), even though the increases in flexibility were not statistically significant. Stretching of a musculotendinous unit may also affect neuromuscular transmission (23). Yamashita et al (23) reported that stretching a rat soleus muscle by 10 and 20% increased posttetanic potentiation of the miniature end-plate potential, which indicates increased Ca2+conductance in the nerve terminal. This increase in intracellular free Ca2+ facilitates neurotransmitter release. Theoretically, muscle force generation should increase as a result of increased transmitter release. Therefore, the possibility exists that increases in muscle force generation seen in this study may be d u e in part t o factors other than changes in series elastic component stiffness and flexibility. However, this is speculation and further study is needed t o address the neurological effects of stretching in vivo. Hamstring muscle performance measured in the open kinetic chain (distal segment off the ground) was investigated in this study. Caution must be used when generalizing

Limitations Although three of the four isokinetic peak torque values were significantly increased, the absolute increases were small t o modest (8.513.5% o r 9.87-1 5.08 Nm). We would expect such small increases based on only 15 stretching sessions and high intersubject variation t o the stretching procedure. Even though the lCCs of the reliability study were high (.95-.97), measurement error was present, as revealed by the SEM. T h e SEM is a range of measurement error that is plus o r minus the measurement value of concern, which means that 34% of the time the SEM value will be above the value of concern, and 34% of the time the SEM will be below the value of concern. Therefore, the absolute increases in peak torque si~ouldbe compared with the SEM at each speed. Only a t 1 20°/sec eccentrically did the absolute increase in peak torque exceed the upper limit of the SEM. Again, the SEM is a range of error that can be above o r below the value of interest.

CONCLUSION Results of this study reveal that no significant increase in hamstring flexibility occurred using static o r proprioceptive neuromuscular facilitation stretching techniques. Because of the large intersubject variation, statistical power was low. However, increases in hamstring flexibility occurred, ranging from 2 1.3% for static stretch to 25.7% for proprioceptive neuromuscular facilitation. Significant increases did occur, however, in isokinetic peak torque eccentrically a t 60" and 120°/sec and concentrically at 120°/sec. No sigVolume 20 Number 3 September 1994 JOSPT

nificant increase occurred concentrically at 60°/sec. W e conclude that increasing hamstring flexibility was effective i n increasing selective hamstring isokinetic peak t o r q u e values i n t h e open kinetic chain. F u r t h e r study is needed t o determine t h e effect o f increasing hamstring flexibility o n functional activities i n t h e JOSPT closed kinetic chain.

ACKNOWLEDGMENT T h e authors thank C h r i s Ingersoll, P h D , A T C , F A C S M , o f Indiana State University f o r his critical review o f this manuscript.

REFERENCES I . Asmussen E, Bonde-Petersen F: Storage of elastic energy in skeletal muscles in man. Acta Physiol Scand 9 1:385-392, 1974 2. Baumgartner TA: Norm-referenced measurement: Reliability. In: Safrit MI, Woods TM (eds), Measurement Concepts in Physical Education and Exercise Science, pp 45-72. Champaign, IL: Human Kinetics Publishers, 1989 3. Cavagna CA: Elastic bounce of the body. I Appl Physiol 29:279-282, 1970 4. Cavagna CA, Dusman 6, Margaria R: Positive work done by a previously stretched muscle. I Appl Physiol 24:2 1-32, 1968 5. Holt LE, Travis TM, Okita T: Compar-

JOSPT *Volume 20 Number 3 September 1994

RESEARCH STUDY

6. 7. 8.

9.

10.

I I.

12.

13.

14. 15.

ative study of three stretching techniques. Percept Mot Skills 3 1 :6 1 1 616, 1970 Mann RA: A kinetic analysis of sprinting. Med Sci Sports Exerc 13:325-328, 198 1 Mann RA, Sprague P: A kinetic analysis of the ground leg during sprinting. Res Q Exerc Sport 5 1: 334-348, 1980 Moore MA, Hutton RS: Electromyographic investigation of muscle stretching techniques. Med Sci Sports Exerc l2:322-329, 1980 Osternig LR, Robertson RN, Troxel RK, Hansen P: Differential responses to proprioceptive neuromuscular facilitation (PNF) stretch techniques. Med Sci Sports Exerc 22: 106- 1 1 1, 1990 Osternig LR, Robertson RN, Troxel RK, Hansen P: Muscle activation during proprioceptive neuromuscular facilitation (PNF) stretching techniques. Am / Phys Med Rehabil66:298-306, 1987 Pousson M, Hoecken IV, Coubel F: Changes in elastic characteristics of human muscle induced by concentric exercise. Biomech 23:343-348, 1990 Prentice WE: A comparison of static stretching and PNF stretching for improving hip joint flexibility. Athl Train 18:56-59, 1983 Sady lP, Wortman M, Blanke D: Flexibility training: Ballistic, static, or PNF? Arch Phys Med Rehabil 63:26 1-263, 1982 Shrout PE, Fleiss /L: Interclass correlations: Uses in assessing rater reliability. Psycho1 Bull 86:420-428, 1979 Sullivan M, Dejulia I/, Worrell TW: Effect of pelvic position and stretching method on hamstring muscle flexibil-

16.

17.

18.

19.

20.

21.

22.

23.

ity. Med Sci Sports Exerc 24:13831389, 1992 Taylor DC, Dalton ID, Seaber AV, Carrett WE: Viscoelastic properties of rnuscle-tendon units: The biomechanical effects of stretching. Am I Sports Med l8:3OO-309, 1 990 Wallin D, Ekbom 6, Crahn R, Nordenberg T: Improvement of muscle flexibility: A comparison between two techniques. Am I Sports Med 13:263268, 1985 Wilson C/, Elliot BC, Wood CA: Stretch shorten cycle performance enhancement through flexibility training. Med Sci Sports Exerc 24: 1 16- 123, 1992 Wilson GI, Wood CA, Elliot BC: Optimal stiffness of series elastic component in a stretch-shorten cycle activity. / Appl Physiol 702325-833, 1991 Wood CR, Marshall R, Strauss C: Electro musculomechanical action of the lower limb in sprinting-Insights into hamstring injury potential. Presented at the fourth meeting of the European Society of Biomechanics, Davos, Switzerland, September 1984 Worrell TW, Perrin DH, Cansneder BM, Cieck /H: Comparison of isokinetic strength and flexibility measures between hamstring injured and noninjured athletes. 1 Orthop Sports Phys Ther 13: 1 18- 125, 199 1 Worrell TW, Sullivan MK, Dejulia I/: Reliability of an active-knee-extension test for determining hamstring muscle flexibility. I Sport Rehabil 1: 18 1- 187, 1992 Yamashita T, lshii S, Oota I: Effect of muscle stretching on the activity of neuromuscular transmission. Med Sci Sports Exerc 24230-84, 1993