Effect of Eccentric Training on the Plantar Flexor Muscle-Tendon Tissue Properties : Medicine & Science in Sports & Exercise

Journal Logo


Effect of Eccentric Training on the Plantar Flexor Muscle-Tendon Tissue Properties


Author Information
Medicine & Science in Sports & Exercise 40(1):p 117-123, January 2008. | DOI: 10.1249/mss.0b013e3181599254
  • Free


Traditionally, tendons were considered to be relatively nonvascular, inert, and inelastic structures. During the last decade, however, the dynamic nature of the extracellular matrix of tendon and skeletal muscle has been appreciated. It has been shown that tendons are able to respond to mechanical forces by altering their structure and mechanical characteristics (28). This process has been referred to as tissue mechanical adaptation or mechanotransduction (28). It is now clear that the amount and type of mechanical loading is very crucial in this process. Appropriate mechanical loading can result in positive changes in tendons, leading to improved performance, whereas excessive loading may induce tendon degeneration. The challenge for clinicians and trainers is to determine the most appropriate amount and type of mechanical loading. In that respect, there is a need for research that examines the underlying mechanisms associated with each type of mechanical loading.

Eccentric training is a form of mechanical loading that has become very popular in the rehabilitation of patients with Achilles tendonopathy (2). The eccentric loading exercises involve active lengthening of the muscle-tendon unit. Despite positive clinical results, it remains unknown why this program is so successful. Several hypotheses have been mentioned in the literature. Firstly, the effect of eccentric training might lead to tendon hypertrophy and increased tensile strength. Secondly, the effect of the stretching component of the eccentric exercise may have a significant influence on the elastic characteristics of the tendon (1). Thirdly, recent studies have also hypothesized that eccentric training has a sclerosing effect on neovascularization (25).

Recently, isokinetic dynamometers have been used to measure passive resistive torque associated with the range-of-motion changes following stretching programs (17,21). Furthermore dynamometer measurements, combined with ultrasonography (8,13,14,23), have allowed the appreciation of stretch within tendon structures. To date, no studies have combined these techniques in a single study to examine the effects of an eccentric training program. Yet, they may provide important information concerning the mechanisms behind the success of the eccentric training program for Achilles tendonopathies.

The objective of the present study was to investigate the effects of an eccentric heel drop exercise program on the passive resistive torque of the plantar flexors measured during isokinetic passive motion of the ankle joint and on Achilles tendon stiffness measured by ultrasound imaging.


Experimental Design

A randomized, controlled, pretest-posttest trial was set up to assess the effects of a 6-wk eccentric training program. Seventy-four volunteers were prepared to take part in the study. The subjects were randomly assigned into two groups: an eccentric training group (N = 37) and a control group (N = 37). Randomization was performed independently. Thirty-seven cards for each group were shuffled in a container. After completion of all preintervention assessments, each subject picked one card in a blinded manner. The subjects of the eccentric training group performed eccentric heel drop exercises every day. To supervise their training program, each person had to complete a personalized calendar of their exercise activity and was contacted every week by one of the investigators. The control group did not receive a training program. To supervise this group, the participants were contacted every week, and their exercise activities were discussed. They were also asked to complete a questionnaire at the end of the study. The main goal of this questionnaire was to confirm that the subjects of the control group had not undertaken additional activities during the intervention period. Unsatisfactory compliance with the prescribed regimes resulted in exclusion from the study. Before and after the 6 wk of eccentric training, all subjects were evaluated for ankle range of motion, passive resistive torque of the plantar flexors, and stiffness of the Achilles tendon.


The ethical committee of the Ghent University Hospital approved the study, and each participant gave written informed consent before participating. Subjects were informed that the study was for research purposes, and they were encouraged to give maximal effort throughout the entire testing procedure. Subjects with a history of lower-leg injuries were excluded from the study. Only recreational athletes were included in the study; competitive elite athletes were excluded. During the study, all subjects were asked to maintain normal activity. The anthropometric characteristics of the subjects are presented in Table 1.

The anthropometric characteristics of the 64 subjects who completed the testing protocol.



Before testing, all subjects completed a questionnaire to assess their medical history, their physical activity, and their experience with eccentric training. To assess possible changes in their lifestyle, and to detect the presence of injuries during the 6 wk of training, this questionnaire was completed again after the 6 wk of training. This was done to verify the compliance of each subject. The results of the questionnaires indicated that two people had not completed the eccentric program successfully, two people in the control group had done additional eccentric exercises, and six people in the control group had not participated in the posttesting session. Consequently, 64 of the 74 volunteers were included in the statistical analyses (eccentric training group N = 35; control group N = 29).

Range-of-motion measurement.

Dorsiflexion range of motion was measured with a universal goniometer by the same investigator to provide good intrarater reliability. This person did not know the group allocation of the subjects. Previous research using radiography has established the validity of goniometric measurements (11). Each measurement was repeated three times, and the mean was used for statistical analyses. The left ankle was evaluated in a weight-bearing position. The measurement was performed according to the method of Ekstrand et al. (6). The subject stood upright, with the feet parallel. The subject was asked to step back with the left foot and to bring the ankle into maximum dorsiflexion, keeping the left knee straight and the heel on the ground. The subject was aware that the front leg must be flexed, the back leg must be kept straight, and the feet must be facing forward. The weight-bearing measurement was also examined with the knee flexed (6). The subject was asked to stand on the floor, with the left foot on a bench. Then, the subject was asked to lean forward to produce a maximal dorsiflexion in the left ankle, with the heel in contact with the bench and the knee maximally flexed. The bony landmarks used for these measurements were defined using the method of Elveru etal. (7). The proximal arm of the universal goniometer was aligned with the head of the fibula. The axis of the goniometer was positioned 0.5 cm below the lateral malleolus. The distal arm was aligned parallel to an imaginary line joining the projected point of the heel and the base of the fifth metatarsal. This measurement has been found to be valid and reliable (6,7).

Passive resistive torque measurement.

To test passive resistive torque, a Biodex System 3 isokinetic dynamometer was used. The subject was placed in a supine position, with the knee maximally extended. The left foot was securely strapped to a footplate connected to the lever arm of the dynamometer. The standard Biodex ankle unit attachment, with the provided Velcro straps, was used. All subjects were asked to wear the same sport shoes with a low cut in both test sessions. The same investigator strapped the foot before and after the intervention period. The attachment of the foot was also such that the movement of the ankle joint was not impeded, to avoid overestimation of the passive resistive torque. The height and the distance of the foot attachment were registered to make the assessment reproducible in the posttest session. During the testing session, the dynamometer moved the left ankle passively through four continuous cycles of motion, from 20° plantar flexion to 10° dorsiflexion at 5°·s−1, with neutral being the line of the tibia perpendicular to the footplate. These range-of-motion limits were used in the pretesting session and the posttesting session. This range of motion is used during many functional activities. A slow joint angular velocity was used to ensure that the stretch did not elicit reflexive muscle activity. Most authors agree that 5°·s−1 achieves this purpose (10). The subjects were instructed to relax, and before data collection each person performed a test trial to become familiar with the system. During the test session, electromyographic activity from the plantar and dorsiflexor muscles was recorded (MyoSystem 1400, Noraxon USA Inc., Scottsdale, AZ). Surface electrodes with an electrical surface contact of 1 cm2 (Ag-AgCl, BlueSensor, Medicotest GmbH, Germany) were placed on the soleus, the tibialis anterior, and the medial head of the gastrocnemius muscle, according to the guidelines of Basmajian (3), with an interelectrode distance of 10 mm. The EMG tracings were monitored during the tests to ensure that calf muscle activity was less than 0.05 mV above baseline during the passive stretch cycles (10). Previous pilot work has indicated that this amount of EMG activity corresponds to approximately 2% MVC. The bandwidth of the frequency response was 20 Hz to 4 kHz. Similar to Gajdosik et al. (10), the raw EMG signals were relayed to an amplifier (×5000) and high-pass filtered at 20 Hz, and the analog signals were converted to digital data at a sampling rate of 500 Hz. The test was repeated if the subject was not relaxed sufficiently, that is, if the muscle activity was higher than 0.05 mV. The peak passive resistance torque (N·m) recorded from the dynamometer during four cycles of motion was used in the statistical analysis. A pilot study demonstrated that the reproducibility was high (ICC = 0.93-0.94, P < 0.001).

Measurement of the stiffness of the Achilles tendon.

The ratio of the calculated muscle force (Fm) and the elongation of the Achilles tendon (ELONG) provided a measure of the stiffness of the Achilles tendon. With respect to the muscle force, the measured torque (N·m) during maximal isometric plantar flexion was first converted to muscle force Fm (N), using the following equation: Fm = kTQ·MA−1 where k is the relative contribution of the physiological cross-sectional area of the medial gastrocnemius within plantar flexor muscles (18%) (9), TQ is torque in newtons, and MA is the moment arm length of triceps surae muscle at 90° of ankle joint flexion (neutral position) (50 mm) (26). Therefore, Fm = 18/100 × TQ/0.05. Secondly, the ratio of Fm and ELONG provided the stiffness of the tendon (STIFFN, N·mm−1). In this study, the calculations were based on those of Kubo et al. (12). The test-retest reliability of measuring the stiffness of the Achilles tendon using ultrasonography has been shown to be good (ICC = 0.80-0.82) (20).

Measurement of torque.

The dynamometer (Biodex System 3) was used to determine torque output during isometric plantar flexion. The subject lay prone on a bench. The left ankle was placed in a 90° position (anatomical position), with the knee joint at full extension and the foot securely strapped to a footplate connected to the lever arm of the dynamometer. The standard Biodex ankle unit attachment, with the Biodex-provided Velcro straps, was used in this study. To prevent ankle joint motion, the foot was firmly attached to the footplate of the dynamometer with a strap. The position and the height of the Biodex chair were also recorded for each subject individually and were used in the following evaluations. Before the test, the subjects performed three to five submaximal contractions, to be accustomed to the test procedure. After this warm-up, the subjects were instructed to develop an isometric maximal voluntary contraction (MVC) for 5 s. The task was repeated three times per subject, with 30 s of rest between trials. Visual examination was undertaken to ensure that the subject's ankle joint did not move during this muscle work. When motion was observed, the trial was discarded. Each subject was verbally encouraged to exert maximal voluntary effort by contracting as hard as possible. The maximal isometric strength was defined as the peak torque recorded. The force of the tendon was estimated from the plantar flexion torque, the physiological cross-sectional area ratio of the medial gastrocnemius to all the plantar flexors, and the moment arm (see formula above).

Measurement of ELONG.

To obtain a measurement of the ELONG, the method of Fukashiro et al. (8) was used. In the present study, a real-time ultrasonic apparatus (Siemens Sonoline SL-1) was used to obtain a longitudinal ultrasonic image of the medial gastrocnemius (MG) muscle at 30% of the lower leg's length (i.e., from the popliteal crease to the center of the lateral malleolus. An electronic linear-array probe of 7.5-MHz wave frequency was secured by Velcro straps on the skin. The ultrasonic images were recorded on videotape (Digital Camera, Sony). One tester who was not aware of the group allocation of the subjects visually identified the echoes from the aponeurosis and the MG fascicles. Parallel echoes running diagonally represent the collagen-rich connective tissue between the fascicles of the medial gastrocnemius. The darker areas between the bands of echoes represent the fascicles. The echo that runs longitudinally in the middle is from the aponeurosis. The point (x) at which one fascicle was attached to the aponeurosis was visualized on the ultrasonic image. This point (x) moved proximally during isometric torque output. The distance traveled by xx) is considered to indicate the lengthening of the aponeurosis and, therefore, of the tendon (23). Displacement was measured with the multimedia player Light Alloy 1.D. The mean value of the three measurements was used as a representative value for the ELONG.

Eccentric heel drop program.

The eccentric exercise program was the same as that used by the majority of the published studies on eccentric calf muscle training in patients with Achilles tendonopathy (2). All subjects were instructed on how to perform the eccentric training. They were given practical instructions and a written manual on how to progress. The eccentric exercise was a classical heel drop, executed on a step. According to Alfredson et al. (2), the subject stands in an upright body position on the left leg, with the body weight on the forefoot and the ankle in plantar flexion. The calf muscle is loaded eccentrically when the patient lowers the heel beneath the level of the forefoot in a controlled way to the point of maximum perceived stretch on the plantar flexor muscles. The subject was asked to execute the downward movement (eccentric phase) in 6 s. The other leg was used to return to the starting position (concentric phase). The exercise session included 15 repetitions performed in three sets (3 × 15 repetitions). The subjects were instructed to execute these eccentric exercises daily for 6 wk. Between the exercise sets, the subjects rested for 20 s.

Statistical Analyses

Statistical analysis was performed with Statistical Package for the Social Sciences (version 12.0; SPSS Inc., Chicago, IL). The data were assessed for normality, using the Kolmogorov-Smirnov test. One-way ANOVA were used to compare the baseline characteristics of both groups. To determine the significance of an interaction effect (time × group) or main effect for time, a two-factor (time and group) general linear model for repeated measures was performed. In these analyses, the Mauchly's test of sphericity was significant, indicating that the assumption of sphericity had been violated. Therefore, a Greenhouse-Geisser correction factor was applied to all P values. Pairwise differences were examined using Bonferroni tests, and the alpha level was set at 0.05 for all hypotheses.


Pretraining Results

Sixty-four subjects' data were included in the statistical analyses. No significant differences were observed between the groups at baseline. The baseline characteristics of both groups are presented in Table 2.

The baseline characteristics of both groups.

Posttraining Results

Range of motion.

Table 3 shows that the eccentric training group had a significantly increased dorsiflexion range of motion for both measurements, with the knee flexed and extended. The dorsiflexion range of motion of the control group did not change significantly. There were no significant interaction effects.

The results of the general linear model for repeated measures on the range-on-motion measurements.

Passive resistive torque.

The results showed a significant main effect for time. Post hoc testing revealed that the passive resistive torque decreased significantly in the eccentric heel drop group after 6 wk of training. The passive resistive torque of the control group was not changed significantly. There was also a significant interaction effect. The results of these analyses are presented in Table 4.

The results of the general linear model for repeated measures on the peak passive resistive torque measurements.

Stiffness of the Achilles tendon.

There was no significant main effect for time. The stiffness of the Achilles tendon did not change significantly after 6 wk of eccentric heel drop training. There was no significant interaction effect. Table 5 shows these results.

The results of the general linear model for repeated measures on measurements of the stiffness of the Achilles tendon.


The results of this study reveal that dorsiflexion range of motion was increased after the eccentric training program. The present study is the first study investigating the effects of eccentric training on ankle joint range of motion in a healthy population. Investigating a patient population with Achilles tendonopathy, Silbernagel et al. (27) could not find a significant increase in dorsiflexion range of motion after 12 wk of eccentric loading. A possible explanation for this finding in their study could be that the range of motion of their subjects was measured in a non-weight-bearing position by active contraction of the dorsiflexor muscles, thus limiting the amount of motion change that can be observed. Moreover, comparing our healthy population with the patient population of Silbernagel et al. (27) could be another reason for finding different results.

To assess the effects of eccentric training on the muscle-tendon tissue properties more extensively, the resistive torque of the plantar flexors during passive motion was examined together with the stiffness of the Achilles tendon.

In the present study, we observed no significant changes in tendon stiffness after 6 wk of eccentric training. No previous work has examined the effect of eccentric training on tendon stiffness. However, two studies have examined the effects of a stretching program on tendon stiffness in vivo. Firstly, Kubo et al. (14) investigated the effects of a3-wk static stretching program and found that tendon stiffness was unchanged. More recently, Mahieu et al. (19) investigated the effect of static and ballistic stretching on Achilles tendon stiffness. After 6 wk of static stretching, tendon stiffness did not change significantly. However, after 6 wk of ballistic stretching, the tendon stiffness decreased significantly.

In contrast to tendon stiffness, the results regarding the passive resistive torque show that after 6 wk of eccentric training, the passive resistive torque of the plantar flexors was significantly decreased. Like tendon stiffness, there are no previous studies to compare our results. However, in the literature concerning stretching, a number of studies have found decreases in peak passive torque after training (4,14,18,19). Because the range of motion in which the passive resistive torque was measured was the same in the pre- and posttesting session, the significant decrease in passive resistive torque after eccentric training has to be attributed to structural changes (17) as compared with stretch-tolerance changes. Although it is beyond the methods of the current study to define what structures changed, Lynn and Morgan (15) have shown that eccentric exercise in rats led to increased sarcomeres in series, thus increasing the compliance of the muscle fibers. More recently, in vivo research by Brockett et al. (5) has indicated that eccentric exercise changes the length-tension relationship of the exercised muscles, with peak torques being generated at longer muscle lengths-a finding consistent with the presence of an increased number of sarcomeres in series.

With respect to injury, whether from a prevention or rehabilitation focus, there are two perspectives from which to consider our findings. Firstly, if fiber length has increased, the increased number of sarcomeres in series will allow finer control of the muscle's length and rate of length change while under tension, particularly in the descending portion of the length-tension curve (22), hence possibly preventing stress from rising to levels that may induce injury. Secondly, it could be that the changed ratio of strain on muscle tissue to strain on the tendon tissue reduces the relative stretch on the tendon and, hence, the risk for damage to this structure. Therefore, in an early stage of Achilles tendonopathy, the eccentric training program could be used as secondary prevention.

Norregaard et al. (24) examined the long-term effect of eccentric exercises compared with stretching exercises on patients with Achilles tendonopathy. Forty-five patients were randomized either to eccentric exercises or stretching exercises for a 3-month period. Symptoms gradually improved during the 1-yr follow-up period. However, nosignificant differences could be observed between the two groups.

With respect to the findings, it should be kept in mind that the passive resistive torque and the measures of Achilles tendon stiffness cannot be compared directly, primarily because of the extremely different forces associated with these tests. In the range of motion through which passive resistive torque was measured, the forces are many times less than those associated with a maximum-effort activation of the plantar flexor muscles. There were also some limitations to the methodology of the present study. First of all, the position at which the isometric contraction was undertaken was 90° (anatomic position), and it was assumed that there was zero strain in the tendon at this point. However, Muramatsu et al. (23) have shown that this is not so, and, hence, the amount of displacement in the tendon will be underestimated, and the subsequent measurement of stiffness would be overestimated. That said, Figure 8 in their paper shows that the effect on strain between 10 and 90% MVC is relatively small. The ramifications with respect to measurements before and after training are that any decrease in stiffness might be overestimated. With respect to the calculation of Fm, we used the same moment arm for all our subjects, a technique used by others (9,12,13,14). For the measurement of individual moment arms, either direct measurements by MRI, using the Reuleaux method as previously described (16,20), or indirect measurement, involving the calculation of the ratio of change in tendon length to change in joint rotation, would be required. Similarly, individual measurements of k, which is the relative contribution of the physiological cross-sectional area of the medial gastrocnemius within plantar flexor muscles, would be more accurately assessed by MRI. Finally, it should be noted that although tendon displacement changes were measured during "isometric" muscle activation, it has been shown that small amounts of ankle joint rotation (3-7°) can take place, and these can markedly affect the displacement measurements, particularly at high levels of an MVC, leading to an overestimation of displacement and, hence, an underestimation of stiffness (17,23). In the present study, we looked for these joint-motion changes specifically by visual observation, and if they were observed, then those data were discarded, and the test was repeated.

In summary, it has been shown that the ankle dorsiflexion of healthy subjects increased after 6 wk of eccentric training. This increase was accompanied by a decrease in the passive resistive torque, the latter providing evidence that structural changes had occurred in the plantar flexor muscles. Although Achilles tendon stiffness was unchanged, the findings provide evidence for the notion that mechanical properties may play a role in the success of Alfredson's protocol for tendinosis, and future research might examine these variables in individuals with this tendon condition.


1. Alfredson H, Lorentzon R. Chronic Achilles tendinosis. Recommendations for treatment and prevention. Sport Med. 2000;29(2):117-123.
2. Alfredson H, Pietilä T, Jonsson P. Heavy-load eccentric calf muscle training for the treatment of chronic Achilles tendinosis. Am J Sports Med. 1998;26:360-6.
3. Basmajian JV. Muscle Alive: Their Functions Revealed by Electromyography. Baltimore (MD): Williams and Wilkins; 1985. p. 60-4.
4. Bressel E, McNair P. The effect of prolonged static and cyclic stretching in ankle joint stiffness, torque relaxation, and gait in people with stroke. Phys Ther. 2002;82(9):880-7.
5. Brockett C, Morgan DL, Proske U. Human hamstring muscles adapt to eccentric exercise by changing optimum length. Med Sci Sports Exerc. 2001;33(5):783-90.
6. Ekstrand J, Wiktorsson M, Öhberg B, et al. Lower extremity goniometric measurements: a study to determine their reliability. Arch Phys Med Rehabil. 1982;63:171-5.
7. Elveru RA, Rothstein JM, Lamb RL. Goniometric reliability in a clinical setting, subtalar and ankle joint measurements. Phys Ther. 1988;68:672-7.
8. Fukashiro S, Itoh M, Ichinose Y, et al. Ultrasonography gives directly but noninvasively elastic characteristics of human tendon in vivo. Eur J Appl Physiol. 1995;71:555-7.
9. Fukunaga T, Roy RR, Schellock FG, et al. Specific tension of human plantar flexors and dorsiflexors. J Appl Physiol. 1996;80:158-65.
10. Gajdosik R, Vander Linden D, Williams A. Influence of age on length and passive elastic stiffness characteristics of the calf muscle-tendon unit of women. Phys Ther. 1999;79:827-38.
11. Gogia PP, Braatz JH, Rose SJ, et al. Reliability and validity of goniometric measurements at the knee. Phys Ther. 1987;67:192-5.
12. Kubo K, Kanehisa H, Kawakami Y, et al. Influences of repetitive muscle contractions with different modes on tendon elasticity in vivo. J Appl Physiol. 2001;91:277-82.
13. Kubo K, Kanehisa H, Fukunaga T. Is passive stiffness in human related to the elasticity of tendon structures? Eur J Appl Physiol. 2001;85:226-32.
14. Kubo K, Kanehisa H, Fukunaga T. Effects of stretching training on the viscoelastic properties of human tendon structures in vivo. J Appl Physiol. 2002;92:595-601.
15. Lynn R, Morgan DL. Decline running produces more sarcomeres in rat vastus intermedius muscle fibers than does incline running. J Appl Physiol. 1994;77:1439-44.
16. Maganaris CN, Baltzopoulos V, Ball D, et al. In vivo specific tension of human skeletal muscle. J Appl Physiol. 2001;90:865-72.
17. Magnusson SP, Simonsen EB, Aagaard P, et al. A mechanism for altered flexibility in human skeletal muscle. J Physiol. 1996;497:291-8.
18. Magnusson SP, Aagard P, Simonsen E, et al. A biomechanical evaluation of cyclic and static stretch in human skeletal muscle. Int J Sports Med. 1998;18(5):310-6.
19. Mahieu N, McNair P, De Muynck M, et al. Effect of static and ballistic stretching on the muscle - tendon tissue properties. Med Sci Sports Exerc. 2007;39(3):494-501.
20. Mahieu NN, Witvrouw E, Stevens V, et al. Test - retest reliability of measuring the passive stiffness of the Achilles tendon using ultrasonography. Isokinet Exerc Sci. 2004;12:185-91.
21. McNair P, Stanley SN. Effect of passive stretching and jogging on the series elastic muscle stiffness and range of motion of the ankle joint. Br J Sports Med. 1996;30:313-7.
22. Morgan DL. New insights into behaviour of muscle during active lengthening. Biophys J. 1990;57:209-21.
23. Muramatsu T, Tetsuro M, Takeshita D, et al. Mechanical properties of tendon and aponeurosis of human gastrocnemius in vivo. J Appl Physiol. 2001;90:1671-8.
24. Norregaard J, Larsen CC, Bieler T, Langberg H. Eccentric exercise in treatment of Achilles tendinopathy. Scand J Med Sci Sports. 2007;17(2):133-8.
25. Öhberg L, Alfredson H. Effects of neovascularisation behind the good results with eccentric training in chronic mid portion Achilles tendinosis? Knee Surg Sports Traumatol Arthrosc. 2004;12(5):465-70.
26. Rugg SG, Gregor RJ, Mandelbaum BR, et al. In vivo moment arm calculations at the ankle using magnetic resonance imaging (MRI). J Biomech. 1990;23:495-501.
27. Silbernagel KG, Thomée R, Thomée P, Karlsson J. Eccentric overload training for patients with chronic Achilles tendon pain-arandomised controlled study with reliability testing of the evaluation methods. Scand J Med Sci Sports. 2001;11(4):197-206.
28. Wang J. Mechanobiology of tendon. J Biomech. 2006;39(9):1563-82.


© 2008 American College of Sports Medicine