Plyometric training exercises consist of explosive-type movements with as short a time period as possible spent in the eccentric loading phase, the concentric propulsion phase, and the time between them (24). A variety of exercises for the lower body are used (e.g., bounding and hopping), but the most prevalent is the drop jump (6), which involves dropping from a height and jumping vertically with maximum effort immediately on landing. Plyometric training exercises, including the drop jump, are reported to improve the mechanical power output of the stretch-shortening cycle, with related enhancements in jumping, sprinting, and throwing performance (2,14,29). To date, a general rule of thumb has been to undertake drop jumps and other plyometric exercises when the neuromuscular system is not fatigued (24). The reasoning for this appears to be that if the body is fatigued, neural adaptations may be reduced (24) and the time duration of different phases of the ground contact may increase, possibly reducing the magnitude of positive adaptations (24). Nevertheless, there is a clear argument for training mechanical power output under fatigued conditions (6). Given that mechanical power is required under fatigued conditions in many sporting events, an extension of the theory of specificity of training would suggest that some plyometric training could take place under fatigued conditions to maximize task-specific adaptations. However, there may be an increase in impact loads and accelerations when the body impacts the ground in a fatigued state, as evidenced in running (4,19,27), increasing the risk of injury (13,15,19,27). When the foot contacts the ground during landings, a ground-reaction force is created that causes a transient acceleration (shock wave), which travels up the musculoskeletal system from the foot to the head (3,11,19). When excessive relative to tissue integrity, impact accelerations can cause a number of musculoskeletal overuse injuries. Such injuries include stress fractures (17,31), articular cartilage and joint degeneration (5,23,26,28), and osteoarthritis (22,28,30). Although the body's structures are able to attenuate some of the impact accelerations (e.g., calcaneal fat pad, cartilage, ligaments, tendons) (20), the action of muscles and the subsequent joint kinematic strategies they control are the main contributors to acceleration attenuation (7,21). When fatigued, the body's ability to attenuate impact accelerations may be reduced, thereby increasing the likelihood of injury (13,15). Increased impact acceleration measured on the tibia has been observed extensively in fatigued running, with associated changes in knee-joint kinematics during landing (4,19).
To address the issue of whether drop-jump training should be undertaken when fatigued, a clear understanding of the effect of fatigue on impact accelerations is required to as-certain the potential for an increase in associated injury. Given that drop jumps are undertaken from a variety of drop heights, and increased drop heights lead to increased energy and vertical ground-reaction forces at impact (16), it is important to examine the compounding effect of drop height on the relationship between impact acceleration and fatigue. To the best of the authors' knowledge, no previous studies have examined the effect of fatigue on impact accelerations in plyometric drop jumps.
The aims of the present study were to investigate the effect of fatigue and increased drop-jump height on impact accelerations measured at the proximal tibia and on the degree of knee flexion at initial ground contact and during the subsequent lowering of the body prior to the propulsion phase. It was hypothesized that impact accelerations would be greater in the fatigued versus the nonfatigued condition, with the difference between these conditions increasing with an increase in drop height. Also, it was hypothesized that knee flexion at initial ground contact and during the eccentric loading phase would be greater with fatigue.
Fifteen physically active male students (age: 21.4 ± 1.5 yr; mass: 80.1 ± 5.84 kg; height 1.78 ± 0.15 m) were recruited from a university population. Written informed consent was obtained in accordance with the university's ethical committee's guidelines. Subjects were excluded from the study if they had any history of lower-extremity injury or an abnormal resting electrocardiogram or blood pressure. Two weeks before the experimental trial, subjects were shown on three occasions, separated by at least 2 d, how to drop jump, and they practiced until they could perform drop jumps competently. They were instructed to land toe first. Competency was subjectively judged by two trainers who had taught drop-jump training for a minimum of 5 yr. Competency requirements were (i) no horizontal travel between take-off and landing, (ii) no excessive pause between the eccentric loading and the propulsion phases, (iii) a short-duration landing phase, and (iv) a toe-first landing pattern. Subjects refrained from strenuous physical activity for 24 h before the experiment. They were also asked to get adequate sleep (6-8 h) the night before the experiment and to avoid food for at least 3 h before the trial.
Subjects warmed up by running on a treadmill at 6 mph for 4 min, followed by stretching exercises of all the major muscle groups of the lower extremities. Five drop jumps from both a 30- and 50-cm-high elevated box were then carried out in a randomized order. At least 15 s of recovery was given between each drop-jump trial. Subjects were asked to keep their hands on their hips to limit movements to the lower-extremity joints. Each participant was instructed to jump vertically with maximum effort and to try to spend as little time on the ground as possible. Impact acceleration and knee-flexion data were gathered for each jump.
Whole-body fatigue was induced on a treadmill using an incremental protocol. Subjects began running at 6 mph with a 3% grade. Every minute, the grade was increased by 1.5%. Rating of perceived exertion (RPE) was taken every 2 min. An RPE of 17 ("very hard") was chosen as an indicator that the subject was in a fatigued state. RPE is a physiologically valid tool for prescribing exercise intensity (25); it has been used widely in this manner (12) and is less invasive than more direct measures (e.g., blood lactate concentration, oxygen uptake). Subjects then repeated the drop jumps in the fatigued condition in the same order as carried out in the nonfatigued condition. The time from the end of the fatiguing protocol to when the drop-jump protocol was undertaken again was always less than 2 min.
A mounted lightweight (17 g) uniaxial accelerometer (ADXL78, Analog Devices, Ireland; sensitivity 38 mV·g−1, range ± 50 g) was attached to the skin overlying the proximal anterior-medial aspect of the right tibia to measure impact acceleration at this site. The accelerometer was aligned along the longitudinal aspect of the tibia. It was attached to the skin with double-sided tape and prewrap (3M Durapore™, UK). An elastic belt wrapped around the shank pressed the accelerometer onto the skin as tightly as comfort allowed (27). This method of attachment and measurement of tibial accelerations has been used extensively previously (13,19,27,30). The attachment sites were marked on the skin so that the accelerometer could be removed for calibration and reattached at the same site. The accelerometer was calibrated before both the nonfatigued and fatigued drop jumps and was removed for the fatiguing protocol.
An electrogoniometer (XM110, Biometrics Ltd, UK) was used to measure knee-joint kinematics in the sagittal plane. It was aligned with the center of the knee joint on the lateral side of the right knee. The upper and lower arms of the electrogoniometer were attached to the leg using double-sided tape and prewrap (3M Durapore™, UK). The electrogoniometer was zeroed before both the nonfatigued and the fatigued series of drop jumps as the subject stood upright with a straight leg. The electrogoniometer remained attached to the leg throughout the experimental trial. Knee-angle data were smoothed using a dual-pass, low-pass, Butterworth filter with a cutoff frequency of 20 Hz to retain the subtleties in the velocity-time traces. Angular velocities were calculated using the first central difference method. A foot-switch apparatus (TF100, Biometrics Ltd, UK) was attached to the underside of the forefoot of each subject's right shoe and was used to register the instant of foot-ground contact. The switch was removed before the fatiguing running protocol to reduce the likelihood of damage with multiple foot contacts. Its attachment site was clearly marked so that it could be reattached in the same position after the fatiguing protocol. The foot switch was tested before and after each experiment by applying a 0.5-kg load, and it was found to work effectively on all occasions. The accelerometer, electrogoniometer, and switch were sampled at a rate of 1000 Hz (DataLINK, Biometrics Ltd, UK). Figure 1 presents a typical output of impact acceleration at the tibia, knee-joint displacement and velocity, and foot-switch contact.
The average of all five trials was used in the analysis. A 2 × 2 within-subjects ANOVA (SPSS for Windows version 10) was employed. The two factors were condition (nonfatigued and fatigued) and drop-jump height (30 and 50 cm). The dependent variables analyzed under each condition were tibial peak acceleration, knee flexion at ground contact, peak knee flexion during the eccentric loading phase, range of knee flexion between these two points, and peak knee angular velocity during the eccentric loading phase. Where a significant interaction was found, a Bonferroni adjusted post hoc analysis was performed to determine where the differences rested. The level of statistical significance in all tests was set at P = 0.05.
On average, subjects ran for 8.3 ± 2.4 min attaining a final gradient of 15.5 ± 3.6% during the fatiguing protocol.
Tibial peak acceleration.
For tibial peak acceleration, the main effects of condition (F = 4.8, P = 0.04), drop height (F = 20.4, P = 0.001), and their interaction (F = 5.3, P = 0.04) were all significant (Fig. 2). Post hoc analysis indicated that the effect of fatigue was significant at 30 cm (t = −2.6, P = 0.02) but not significant at 50 cm (t = −1.1, P = 0.30), with an increase in tibial accelerations of 24 and 5%, respectively. The effect of increased drop height, however, was significant in both nonfatigued (t = −5.6, P = 0.001) and fatigued conditions (t = −0.8, P = 0.01), with an increase in tibial acceleration of 44 and 22%, respectively. In light of these findings, an additional analysis was undertaken between the drop jumps from 30 cm when fatigued and from 50 cm when nonfatigued. Peak acceleration was less (13.4%) from 30 cm when fatigued, but the difference did not reach significance (t = −1.9, P = 0.07).
For initial knee flexion at ground contact, the main effect of condition was not significant (F = 0.1, P = 0.75) (Fig. 3). However, the main effect of drop height was significant (F = 9.8, P = 0.01) with the average initial knee flexion from 30 cm (18.4 ± 7.4°) 37% greater than from 50 cm (14.7 ± 11.8°). There was no significant interaction between condition and drop height (F = 0.0, P = 0.90).
Condition (F = 0.1, P = 0.78), drop height (F = 3.9, P = 0.07), and their interaction (F = 0.7, P = 0.43) had no significant effect on peak knee flexion (Fig. 4). However, there did appear to be a trend toward greater peak knee flexion from the higher drop height of 50 cm. This trend contributed to the significant effect of drop height on knee-joint range of motion (F = 7.2, P = 0.02) (Fig. 5). Average knee range of flexion from 50 cm (51 ± 12.39°) was 15% greater than from 30 cm (44.4 ± 11.8°). Neither the main effect of condition (F = 0.6, P = 0.45) nor the interaction between condition and drop height (F = 0.3, P = 0.59) had a significant effect on knee-joint range of motion.
For knee-joint peak angular velocity during the eccentric phase, the same pattern of results was evident as for the tibial peak-impact acceleration. The main effects of condition (F = 26.8, P = 0.00), drop height (F = 95.5, P = 0.00), and their interaction (F = 41.6, P = 0.00) were all significant (Fig. 6). Post hoc analysis indicated that the effect of fatigue was significant at 30 cm (t = 14.4, P = 0.00) but not significant at 50 cm (t = 1.6, P = 0.13), with an increase in angular velocity of 21 and 3%, respectively. The effect of increased drop height, however, was significant in both the nonfatigued (t = −4.0, P = 0.00) and fatigued conditions (t = −6.0, P = 0.00), with an increase in angular velocity of 43 and 21%, respectively.
During landing in a plyometric drop jump, a ground-reaction force is created, which results in a transient impact acceleration, or shock wave, that travels from the foot, through the lower extremities and towards the head. If the impact acceleration is excessive relative to tissue integrity, there may be an increased risk of overuse injuries, such as stress fractures (17,31), articular cartilage and joint degeneration (5,23,26,28), and osteoarthritis (22,28,30). Fatigue has been implicated as a causative factor in increasing impact accelerations, lending support to the view that plyometric drop jumps should not be undertaken when fatigued. However, given that mechanical power output of the stretch-shortening cycle is essential for performance in many sports, and that fatigue decreases this output but plyometric training may increase it (2), it can be argued that some plyometric training should be undertaken when fatigued (theory of specificity of training). The main purpose of the present study was to examine the effect of whole-body fatigue on impact accelerations measured at the proximal tibia during plyometric drop jumps. This was studied at drop heights of 30 and 50 cm to determine whether the effect of fatigue altered with increasing drop heights. These heights were selected because it is common to use drop heights ranging from 30 to 100 cm (2). The effect of fatigue on knee-joint kinematics was also investigated to determine whether changes in impact accelerations associated with fatigue could be related to changes in knee-joint motion. The knee joint was chosen because it plays a major role in landing and jumping. Fatigue was induced through a running protocol because running is a major component of many sports (e.g., basketball, volleyball, soccer, running) and therefore increases the ecological validity of the study.
In relation to tibial peak-impact acceleration, a significant interaction was found between fatigue and drop height. When dropping from 30 cm, fatigue resulted in a significant increase (24%) in peak accelerations in comparison with the nonfatigued state. This is in agreement with the hypothesis of the present study. In contrast, fatigue did not significantly increase tibial peak-impact accelerations when dropping from 50 cm. This latter finding, however, should not lead to the conclusion that fatigue does not affect acceleration attenuation per se. Rather, the nonsignificant increase (5%) from 50 cm may suggest that for this population group, impact loading associated with the higher drop height was too large for the neuromuscular system to play a major role in reducing tibial impact accelerations, whether fatigued or nonfatigued. Although the present study did not directly measure the contribution of the neuromuscular system to impact acceleration attenuation, support for this conclusion is based on three observations. First, whereas the body uses both passive (e.g., calcaneal fat pad, cartilage, ligaments, tendons) (20) and active (neuromuscular) structures to attenuate impact accelerations, the active structures make the largest contribution (7,21). Whole-body fatigue has been found to decrease neuromuscular control during the eccentric loading phase (8) because of a decrease in proprioception (18) and contractile and reflex capacities (8). In the present study, fatigue caused a significant increase in tibial acceleration from a 30-cm drop, and there is no reason to suggest that a protective neuromuscular mechanism would selectively work at a lower loading and not a higher loading unless it was incapable of doing so. Secondly, a similar pattern in statistically significant findings for peak-impact accelerations was observed for knee-joint peak angular velocities, with an increase in peak accelerations being associated with an increase in knee-joint peak angular velocities (Fig. 6). Given that neuromuscular activity via eccentric muscle actions would act to decrease knee-joint angular velocities, the results suggest that the neuromuscular activity had little effect in reducing loading during eccentric loading phase in drop jumps from 50 cm (whether fatigued or nonfatigued) but was able to affect jumps from 30 cm. Finally, in comparing results from running (19,27) with results from drop jumping gathered in the present study, the greater acceleration associated with a fatiguing protocol is reduced as the peak acceleration values (nonfatigued) increase. Mizrahi et al. and Voloshin et al. reported increases in proximal tibial peak acceleration of 62% and approximately 60%, respectively, when low-impact accelerations were employed in running (Mizrahi found nonfatigued impact accelerations of 6.9 ± 2.5 g in running) (19,27). In contrast, only an increase of 21% was evident in drop jumps (30 cm) when higher-impact accelerations were evident (nonfatigued: 15.8 ± 9.5 g). With still higher impact accelerations in drop jumps from 50 cm (nonfatigued: 24.7 ± 8.3 g), the increase in peak acceleration associated with fatigue was only 5%. This, however, requires further study using a variety of drop heights, some exceeding the 50 cm used in the present study.
Increasing drop height from 30 to 50 cm also resulted in a significant increase in tibial peak-impact acceleration (nonfatigued: 44%; fatigued: 22%) and knee-joint peak angular velocities (nonfatigued: 43%; fatigued: 21%) as hypothesized. It has been previously observed that peak whole-body vertical ground-reaction forces increase with increases in drop height (16), although there is not always a clear relationship between vertical ground-reaction force and impact accelerations measured on body segments because of changes in the body's effective mass associated with changes in technique (3). Based on the assumption that increased tibial peak accelerations may place an individual at an increased risk of tibial stress fracture and lower-extremity joint degeneration injuries (17,23), the results suggest that athletes (with little previous experience of plyometric drop jumping) should train from drop heights lower than 50 cm. The increase in impact accelerations with an increase in drop height may again be related to a decrease in reflex capabilities (short and medium latency) associated with a decrease in facilitation from muscle spindles and/or an increase in inhibitory mechanisms (8).
Of note, peak accelerations were smaller (13.4%) from 30 cm when fatigued than from 50 cm when nonfatigued. Although the difference was not statistically significant (P = 0.07), it did tend towards being so. This suggests that it may be safer to undertake drop jumps from a low drop height when fatigued than from a higher drop height when nonfatigued. Interestingly, an increase in jump height after a drop-jump training program may not be dependent on the height of drop employed (16).
Adjustments in lower-extremity angles, especially at the knee, have been obtained in previous running studies to explain changes in impact accelerations at different sites of the body (4,9,19). LaFortune et al. found that if knee flexion was increased by 40° at impact when loads were applied to the feet using a pendulum device, impact accelerations at the proximal tibia would increase significantly (9). The present study examined the effect of fatigue on the knee-joint angle at initial ground contact, the end of the eccentric phase, and the range of motion between. Fatigue was found to have no effect on these knee kinematic variables. This is in contrast to studies on running that have found that fatigue caused an increase in knee flexion at initial contact and a decrease in range of knee motion employed (4,19) and to studies on landing that found that fatigue caused an increase in knee flexion at initial contact and during landing (15). The difference in results may be attributable to the task constraint in plyometric drop jumps, requiring the performer to jump maximally and stay in contact with the ground for as short a time period as possible. The findings from the present study, therefore, indicate that changes in the knee-joint kinematics examined cannot explain the changes in tibial peak accelerations associated with fatigue and do not indicate a selective controlled attempt to attenuate them. Although not assessed in the present study, it is possible that changes in hip- or ankle-joint kinematics may have occurred.
Limitations to the present study.
Although the present study clearly shows that whole-body fatigue induced through a running protocol can result in a significant increase in peak-impact accelerations during plyometric drop jumps, it is unclear from previous studies what magnitude of impact acceleration/load rate and what number of repetitions will result in injury in humans. The general literature simply refers to "excessive"-impact accelerations/loading rates predisposing to injury (11,15,17,19,22) based on a combination of animal model studies that directly found joint cartilage and bone degeneration after high-impact loading (5,23,26,28) and epidemiological studies on subjects exposed to high loadings during their chosen activity (17). Clearly, far fewer impacts occur in drop-jump training compared with other impact events. A training session may contain between 30 and 100 ground impacts from the drop jumps, whereas a 3-km run may contain around 1339 ground impacts with each leg (based on a step length of 1.12 m) (4). However, the magnitude of impact acceleration is greater in drop-jump training. In an examination of the effect of impacts on fresh bovine and porcine femoral condyles during grafting procedures, Whiteside et al. found that the percentage of chondrocyte death was more dependent on the magnitude of loading rather than the number of impacts (28).
Accelerometers are sensitive to their orientation relative to gravity. In the present study, the shank was orientated approximately 10° relative to the vertical when tibial peak acceleration occurred, which would result in a slight underestimation (1.5%) of tibial peak-impact acceleration. The difference in shank angle between the two drop heights (approximately 2° greater from 30 cm, assuming an equal contribution to the difference in knee-joint angle from the shank and thigh) would result in only a small difference in errors in peak-impact accelerations between the two drop heights (approximately 1.1 m·s−2 greater underestimation from 30 cm). This difference, however, is relatively negligible compared with the significant differences in peak accelerations (54.5 ± 13.4 m·s−2) between the two drop heights.
The subjects used in the present study, despite being physically active and involved competitively in sport, had little or no previous experience of drop-jump training. Therefore, the results obtained may not be applicable to athletes who train regularly with plyometric drop jumps.
Measurement of tibial accelerations was made with a skin-mounted accelerometer rather than a bone-mounted accelerometer. Although a skin-mounted accelerometer may underestimate the magnitude of accelerations (10), these values are likely to be consistent across conditions (fatigued and nonfatigued) and do not require such an invasive procedure. Skin-mounted accelerometers have been used to examine impact accelerations in running (4,19,27).
Whole-body fatigue induced through running was the chosen fatiguing protocol. The type of fatiguing protocol used may significantly affect the relationship between fatigue and impact accelerations. Whereas studies employing a similar protocol have also observed an increase in impact acceleration at the tibia in running (19,27), Flynn et al. found the opposite effect of fatigue using a localized muscle-fatigue protocol (7). Therefore, the results from the present study are only applicable to situations in which fatigue of the same level and type is induced.
In light of the relationship between high-impact accelerations and various injuries (stress fractures, articular cartilage and joint degeneration, and osteoarthritis), the present study indicates that there is an increased risk of injury in performing plyometric drop jumps (30 and 50 cm) when fatigued through running. However, it may be more appropriate to undertake drop jumps from low drop heights (e.g., 30 cm) when fatigued than from higher drop heights (e.g., 50 cm) when nonfatigued.
1. Belluci, G., and B. B. Seedhom. Mechanical behaviour of articular cartilage under tensile cyclic load. Rheumatology
2. Chu, D. Jumping into Plyometrics
, 2nd ed. Champaign, IL, 1998.
3. Derrick, T. R. The effects of knee contact angle on impact forces and accelerations. Med. Sci. Sports Exerc.
4. Derrick, T. R., D. Dereu, and S. P. McLean. Impacts and kinematic adjustments during an exhaustive run. Med. Sci. Sports Exerc.
5. Ewers, B. J., D. Dvoracek-Driksna, M. W. Orth, and R. C. Haut. The extent of matrix damage and chondrocyte death in mechanically traumatized articular cartilage explants depends on the rate of loading. J. Orthop. Res.
6. Fleck, S. J., and W. J. Kraemer. Designing Resistance Training Programs
. 3rd ed. Leeds, UK: Human Kinetics, 2004, p. 392.
7. Flynn, J. M., J. D. Holmes, and D. M. Andrews. The effect of localized leg muscle fatigue on tibial impact acceleration. Clin. Biomech.
8. Gollhofer, A., P. V. Komi, N. Fujitsuka, and M. Miyashita. Fatigue during stretch-shortening cycle exercises II. Changes in neuromuscular activation patterns of human skeletal muscle. Int. J. Sports Med.
9. Lafortune, M. A., E. M. Hennig, and M. J. Lake. Dominant role of interface over knee angle for cushioning impact loading and regulating initial leg stiffness. J. Biomech.
10. Lafortune, M. A., E. M. Hennig, and G. A. Valiant. Tibial shock measured with bone and skin mounted transducers. J. Biomech.
11. Lafortune, M. A., M. J. Lake, and E. M. Hennig. Differential shock transmission response of the human body to impact severity and lower limb posture. J. Biomech.
12. Lamb, K., R. Eston, and D. Corns. Reliability of ratings of perceived exertion during progressive treadmill exercise. Br. J. Sports Med.
13. Light, L. H., G. E. Mc Lellan, and L. Klenerman. Skeletal transients on heel strike in normal walking with different footwear. J. Biomech.
14. Luebbers, P. E., J. A. Potteiger, M. W. Hulver, J. P. Thyfault, M. J. Carper, and R. H. Lockwood. Effects of plyometrics training and recovery on vertical jump performance and anaerobic power. J. Strength Cond. Res.
15. Madigan, M. L., and P. E. Pidcoe. Changes in landing biomechanics during a fatiguing landing activity. J. Electromyogr. Kinesiol.
16. Matavulj, D., M. Kukolj, D. Ugarkovic, J. Tihanyi, and S. Jaric. Effects of plyometric training on jumping performance in junior basketball players. J. Sports Med. Phys. Fitness
17. Milgrom, C. The Israeli elite infantry recruit: a model for understanding the biomechanics of stress fractures. J. R. Coll. Surg. Edinb.
18. Miura, K., Y. Ishibashi, E. Tsuda, Y. Okamura, H. Otsuka, and S. Toh. The effect of local and general fatigue on knee proprioception. Arthroscopy
19. Mizrahi, J., O. Verbitsky, E. Isakov, and D. Daily. Effect of fatigue on leg kinematics and impact acceleration in long distance running. Hum. Mov. Sci.
20. Pain, M. T. G., and J. H. Challis. The role of the heel pad and shank soft tissue during impacts: a further resolution of the paradox. J. Biomech.
21. Radin, E. L. Role of muscle in protecting athletes from injury. Acta Med. Scand.
22. Radin, E. L., D. Eyre, J. L. Kelman, and A. L. Schiller. Effect of prolonged walking on concrete on the joints of sheep. Arthritis Rheum.
23. Radin, E. L., H. G. Parker, J. W. Pugh, R. S. Steinberg, I. L. Paul, and R. M. Rose. Response of joints to impact loading - III: relationship between trabecular microfractures and cartilage degeneration. J. Biomech.
24. Siff, M. C., and Y. V. Verkhoshansky. Supertraining
. 4th ed. Denver, CO: Supertraining International, 1999, p. 500.
25. Steed, J., G. Gaesser, and A. Weltman. Rating of perceived exertion and blood lactate concentration during submaximal running. Med. Sci. Sports Exerc.
26. Torzilli, P. A., R. Grigiene, J. Borrelli Jr., and D. L. Helfet. Effect of impact load on articular cartilage: cell metabolism and viability, and matrix water content. J. Biomech. Eng.
27. Voloshin, A. S., J. Mizrahi, O. Verbitsky, and E. Isakov. Dynamic loading on the human musculoskeletal system - effect of fatigue. Clin. Biomech.
28. Whiteside, R. A., R. P. Jakob, U. P. Wyss, and P. Mainil-Varlet. Impact loading of articular cartilage during transplantation of osteochondral autograft. J. Bone Joint Surg. (Br.)
29. Wilke, K., M. L. Voight, M. Keirns, V. Gambetta, J. Andrews, and C. Dillman. Stretch-shortening drills for the upper extremities: theory and clinical application. J. Orthop. Sports Phys. Ther.
30. Wosk, J., and A. Voloshin. Wave attenuation in skeletons of young healthy persons. J. Biomech.
31. Yoshikawa, T, S Mori, AJ Santiesteban, et al. The effects of muscle fatigue on bone strain. J. Exp. Biol.