Although the quadriceps femoris is the major extensor of the knee and stabilizes the patella within the trochlear groove (15,16), the role of its individual muscles during dynamic knee function is incompletely understood. This is particularly the case with the obliquely orientated, distal fibers of the vastus medialis oblique (VMO). It is unclear whether the quadriceps acts as a single unit or whether the VMO can function independently (17). This is an important issue that could aid in our understanding of patellofemoral pain, which is common in many sporting injuries and accounts for 25% of all sports-related knee injuries (8).
There are several clinical observations suggesting that the action of the VMO should be considered independently from that of the rest of the quadriceps. It is well known that degeneration of the patellofemoral region is most common laterally and that individuals with anterior knee pain often have lateral patellar tilting or lateral subluxation, which leads to increased patellofemoral contact forces (10,18). This observation suggests an imbalance exists between the action of the VMO (attempting to maintain the patella in the trochlear groove) and the forces that tend to move the patella laterally because of the Q angle. In people without knee pain, the ratio of EMG activity of the VMO relative to the vastus lateralis (VL) is approximately 1:1 throughout the range of knee movement (18), and in such individuals, the VMO and the VL are activated simultaneously (4). However, a change in the relative onset of activity of these muscles has been associated with dysfunction; people with patellofemoral pain have a reduced ratio of VMO:VL activity, and the onset of their VMO activity is delayed relative to that of the VL (4).
Although previous authors have investigated the specific role of the VMO in knee function, the majority of studies have been based on non-weight-bearing conditions (1,11,13). This most likely does not allow us to evaluate accurately the role of the VMO during dynamic, weight-bearing movements of the knee, which are common in daily life and sport. Certainly, force transmission from the quadriceps tendon-patella-patellar tendon-tibia and the contact between the femoral condyles and the patella differs greatly in weight-bearing conditions. This is apparent clinically if the contact areas between the femoral condyles and the tibial plateau are compared in x-rays of the knee joint taken in weight-bearing conditions. Recent studies have attempted to clarify the role of the lower-limb musculature, especially that of the quadriceps femoris components during weight-bearing conditions such as jumping and running (5,7). However, the primary aim of these was to investigate the possible cause of the increased incidence of ACL damage in the female. The investigators were able to show that there was differential activity in the VMO and VL, but the studies were not designed to investigate the relative contributions and temporal activity of these muscles in relation to changing angle of the knee joints and loading. The purpose of the present paper is to further clarify the relative contributions of both the VMO and the VL during dynamic weight-bearing conditions that challenge knee stability. To interpret the physiological and biomechanical findings, we also reinvestigated the gross anatomy of the distal portion of the quadriceps as a further strategy for clarifying the function of the quadriceps during weight-bearing conditions (jump landing).
MATERIAL AND METHODS
This study was conducted, with the approval of the ethics committee, on 10 male students in the University of Clermont Ferrand (France) who gave their informed consent and who had no previous history of knee pain or pathology. Their ages, heights, and body mass (means ± standard deviation) were 25 ± 3 yr, 180 ± 4 cm, and 76 ± 8 kg, respectively.
A leg press (PANATTA SPORT, DPS, Clermont Ferrand, France) was used to measure maximal isometric force (all participants were familiarized with its correct use beforehand). Further details of the procedure have been given previously (21,22).
The subjects performed three maximal isometric leg press extensions, and the best result was recorded. The knee angle was standardized at 70°-a position that was selected because pilot tests demonstrated that the force-angle relationship was maximum around this angle. Data were collected at 2000 Hz and stored in a PC computer (Intel Pentium; PSI instruments; Clermont Ferrand, France).
A force plate (9981C, KISTLER Instrument AG Winterhur, Switzerland) was used to measure force (dN; 1 dN = 10 N) during squat (SJ) and drop (DJ) jumps on both feet and one foot. The SJ consisted of a pure, maximal, concentric knee extension, and the drop jump was performed from 50 cm, according to the protocol of Komi et al. (12). Subjects performed each jump three times in random order, and all performances were recorded.
A Saga-3 3D system (Saga-3, Biogesta Denain, Clermont Ferrand, France) was used to analyze kinematic data for the greater trochanter of the hip (H), the lateral femoral epicondyle of the knee (K), and the lateral malleolus of the ankle (M) during each exercise. Subjects were filmed at 100 Hz, and the coordinated marker points were filtered using spline functions. Each jump was divided into three phases: two eccentric phases (ECC1 and ECC2) and one concentric (CON) phase (Fig. 1). The time of initiation of ECC1 was calculated using the force plate data (corresponding to the instant at which vertical force began to decrease continually). The end of the CON phase was defined as the instant at which the subject lost contact with the force platform (corresponding to the instant at which the vertical component of the ground-reaction forces was zero). The initiation of the ECC2 phase corresponded to the time when the subject first contacted the force plate on landing. The end of both ECC phases and the beginning of the CON phase was defined using the kinematic data for the knee joint.
Motor point procedures and EMG recordings.
To locate the motor points of the quadriceps femoris, each muscle was investigated by moving a monopolar probe over its belly. Electrical stimulation was delivered by a Biostim stimulator (model 6050, Mazet Electronique, France). The current was composed of symmetrical biphasic square wave pulses (duration of 200 ms) delivered percutaneously to the muscle at a frequency of 80 Hz. Electrophysiological tests showed that both parts of the vastus medialis (VM)-the proximal part of the muscle and the distal fibers that comprise the VMO-had independent motor points, in contrast to the VL, where only one motor point was observed. The EMG signal was then recorded over the motor points of the VMO and the VL. Bipolar surface electrodes (sensors, 10 mm) were placed longitudinally over the muscle belly with an interelectrode distance of 20 mm. The skin was shaved at each recording point, swabbed with alcohol, and temporarily marked for the duration of testing. A large strip of sticking plaster was used to cover and fix the electrical contact to the skin, and a reference electrode was placed on the patella. Signals were preamplified, filtered using a band-pass filter (15-1000 Hz), and sampled at 2000 Hz. RMS values were calculated and normalized to the maximal RMS values recorded during MVC, to account for differences in electrical impedance and electrode placement. Because comparing the activation of VMO and VL was of primary interest, values of the root mean square (RMS) for each muscle and the ratio of difference between VMO and VL (100 × [VMO − VL]/VMO) were obtained. Maximal RMS values and the corresponding knee-flexion angle were calculated with the knees at 80 to 60, 60 to 40, 40 to 20, and 20 to 0° of flexion, for each phase (ECC1, ECC2, and CON). The starting position (80°) was chosen because pilot tests showed that all subjects were comfortable beginning squat jumps in this position and because subjects did not go below 80° during landing.
Ten dissecting room cadavers donated to Cardiff University (UK) under the provision of the 1984 Human Anatomy Act were used to reinvestigate the gross anatomy of the quadriceps femoris. The skin and superficial fascia were removed in the region of the quadriceps muscle by making a longitudinal incision from the anterior superior iliac spine (ASIS) to the patella, followed by upper and lower transverse cuts. The upper transverse cut extended from the ASIS to the pubic symphysis, and the lower cut was made at the level of the lower pole (apex) of the patella. The femoral nerve and its branches to VM were carefully dissected beneath the deep fascia, in a craniocaudal direction, commencing in the femoral triangle.
All statistical analyses were performed with MatLab software (Version 126.96.36.1995 (R14)). Means and SD were calculated from individual measurements for each exercise. The changes between variables of each exercise were analyzed using general linear model analysis of variance with repeated measures. When significant differences were found, Fischer post hoc tests were used. In each case, the level of significance was set at P < 0.05.
Jump exercises showed that although both the VMO and VL were activated simultaneously in both the squat and drop jumps on either one or both legs (Figs. 2-5), the relative contribution of each muscle varied with the knee angle and the type of contraction (eccentric or concentric) (Tables 1 and 2).
During the concentric phase (push-off), there was no significant difference in RMS value between the VL and the VMO at 80 to 60° and 60 to 40° of flexion for squat jumps performed on either one or both legs. However at 40 to 20° and 20 to 0° of flexion, the VMO was activated to a greater extent than the VL (P = 0.003). The ratio of difference between the two muscles at 40 to 20° and 20 to 0° of flexion averaged 26.9 ± 4.5 for single-legged squat jumps and 21.3 ± 5.7 for two-legged squat jumps. The maximum values reached were 34.8 ± 3.9 (at 27 ± 8° flexion for single-legged jumps) and 22.4 ± 6.6 (at 12 ± 6° flexion for two-legged jumps). During two-legged squat jumps, both the VMO and the VL were highly activated during the eccentric phase in landing (ECC2), and there was no significant difference (P = 0.12) between the activity of the two muscles. However, in single-legged jumps, the VMO was activated more than the VL between 60 and 40° of flexion. The ratio of difference between the VMO and the VL at 60 to 40° of flexion averaged 23.6 ± 8.7 and reached a maximum of 43.5 ± 7.2 at 49 ± 8° of flexion.
In the ECC1 phase of two-legged drop jumps (landing from the box on the force plate), the VMO and VL were activated in a coordinated manner, and the ratio of difference between the two muscles approached zero. However in single-legged jumps, the VMO was activated more than the VL between 60 and 40° of flexion. At this angle, the ratio of difference between the muscles averaged 28.8 ± 5.7 and reached a maximum of 71.2 ± 12.7 at 51 ± 10°. During the concentric phase of both single- and double-legged drop jumps, there was no significant difference between the two muscles at 80 to 60° or 60 to 40° of flexion. Between 40-20° and 20-0° for both jumps, the VMO was activated more than the VL (P = 0.03). The ratio of difference between the VMO and VL at 40 to 20 and 20to 0° of flexion averaged 17.2 ± 5.2 for double-legged jumps (maximum 27.6 ± 5.6 at 22 ± 6°) and 0.28.9 ± 5.7 for single-legged jumps (maximum 40.6 ± 8.3 at 18 ± 8°). During the ECC2 phase of both jumps, the VMO was activated more than the VL between 40 to 20 and 20 to 0° flexion (P = 0.04). For double-legged jumps, the ratio of difference between the muscles averaged 19.2 ± 3.9 (maximum 38.4 ± 11.6 at 27 ± 9°), and for single-legged jumps it averaged 26.6 ± 3.5 (maximum 46.7 ± 10.4 at 33 ± 10°).
The gross anatomy of the quadriceps muscle complex and the attachment of the quadriceps tendon to the patella is shown in Figure 6. The distal fibers of the VM (i.e., the fibers of the VMO) are obliquely orientated and independent of those of the rest of the muscle. Although the VM was strongly bound to the vastus intermedius proximally, the VMO was easy to separate from the remainder of the muscle group by blunt dissection (Fig. 6E). It was also evident that the fibers of the VMO were not only attached to the medial border of the patella, but they also had a small region of direct continuity with the patellar tendon (Fig. 6A, E, and F). The VMO has an independent nerve supply from that of the rest of the muscle. The femoral nerve gave off a short branch, which rapidly ramified into four or five smaller branches that supplied the proximal part of the VM (Fig. 7). However, there was also a long thin branch (the "nerve to VM"), which traversed the subsartorial canal alongside the femoral vessels and supplied the distal part of the muscle (i.e., the VMO (Fig. 7)). This observation that the VMO is innervated by a distinct branch of the femoral nerve was made in all six cadavers that were examined for this purpose.
The combined anatomical and biomechanical/physiological approaches used herein to investigate the role of the quadriceps muscle group during dynamic knee function suggest that the structure and function of the VM is more complex than previously recognized. Although this muscle is part of the quadriceps, its distal oblique fibers (i.e., the VMO) have their own distinct nerve supply. They attach not only to the quadriceps tendon, but also to the medial border of the patella, and they can be directly continuous with the patellar tendon. Furthermore, the more proximal fibers of the VM (which comprise the bulk of the muscle) are more firmly attached to the vastus intermedius than to the rectus femoris, as others have suggested (13). Although our current anatomical dissection findings demonstrated that there is a separate branch of the femoral nerve to the VMO, as confirmed by previous authors, (14,23), it contradicts the recent findings of Peeler et al. (17), who have indicated that the nerve never enters VM at more than one site. Our observation of a distinct branch of the femoral nerve entering the VMO was made in all six cadavers examined and is thus unlikely to reflect anatomical variation between individuals. There were no special features associated with these cadavers or any specific dissection techniques employed to demonstrate the nerve. Thus, the reason for the difference between our findings and those of Peeler et al. (17) cannot be readily explained, although it was noted that the nerve was small and could easily be missed or torn in dissection. It should be noted that clinical evidence points to the likelihood of an independent nerve supply to VMO; this part of the muscle can be stimulated independently in the treatment of anterior knee pain and patellar subluxation (6). This has been confirmed by our electrophysiological tests where both parts of the VM (i.e., the proximal fibers and the VMO) were found to have independent motor points.
It is clear that activation of the VM in the weight-bearing conditions used in the present study differs from that which occurs in the isokinetic or isometric protocols used by previous workers (1,11,13,15). The fact that VMO activity was pronounced during weight-bearing conditions that prevail in landing from a jump may relate to the increase in medial and lateral knee movements and the challenges that such movements place on patellar alignment (20,24). We have shown that the relative contribution of the VMO compared with the VL varied according to the type of contraction (eccentric or concentric) and the knee angle. Our results contradict earlier studies showing that the VMO is more active at 90° of knee flexion during both concentric and eccentric contractions (3,19). Although both muscles were activated simultaneously when the knee was flexed 40 to 80°, VMO activity became more pronounced in our study between 40° of flexion and full knee extension. This may reflect the fact that the primary role of the VL is knee extension (because its fibers are predominantly longitudinal), whereas the VMO is much less effective as an extensor, in part because of its oblique fiber orientation. The finding that the EMG activity of VMO became most pronounced at full extension emphasizes the importance of this part of VM in maintaining medial patellar stability, because lateral displacement of the patella typically occurs in the initial few degrees of flexion before the patella comes to lie in the trochlear groove (9). A greater activity ratio of VMO was also found while the subject was in the air. This suggests that the VMO is required to act as a stabilizing medial force whenever the patella is in a position where it could be dislocated. It is possible that the direct attachment of the VMO fibers to the patella (rather than an indirect attachment via the quadriceps tendon) ensures that the medially directed pull of VMO maintains the position of the patella within the trochlear groove. Our dissections (Fig. 6D) contradict the earlier studies of Peeler et al. (17), who have suggested that only a small part of VM (< 10%) attaches directly to the medial aspect of the patella; the patellar attachment is frequently more extensive. We note that the VMO fibers are identified as a distinct entity at surgery, because they are dissected free from the medial border of the patella and are advanced obliquely across the surface of this bone in the Insall procedure, which is used for controlling patellar subluxation (2).
The pattern of EMG activity of the VMO differs according to whether jumps were performed on one or both legs, indicating a difference in motor unit recruitment. During single-legged jumps, there was a greater relative contribution of the VMO compared with the VL, perhaps reflecting a role of the VMO in maintaining the stability of a subject whose center of gravity is displayed laterally during a single-legged jump. Our study did not show any difference between VMO and VL activity in landing from a squat jump performed on both legs. There was only a difference in relative muscle activity when such jumps were performed on one leg, or when the subjects performed drop jumps. It is possible that one-legged squat jumps and drop jumps both destabilize the knee joint to a greater extent than two-legged squat jumps. The fact that drop jumps performed on both legs were also associated with pronounced activity of the VMO, compared with the VL, may relate to differences in jump height between drop and squat jumps. Drop jump heights (mean ± standard deviation 36 ± 7 cm) were greater than squat jump heights (mean ± standard deviation 24 ± 5 cm); this challenges knee stability during landing to a greater extent. Our results contradict those of earlier reports (5,7) where no significant differences were found between VMO and VL during landing. However, this may reflect methodological differences between the studies. It also is important to note that Fagenbaum et al. (7) did not seek specifically to compare VMO activity with that of VL, and thus they did not apply any statistical tests to the EMG data obtained from these muscles when the subjects landed at different knee angles. Moreover, the surface EMG electrodes in Fagenbaum et al.'s (7) study were attached to the skin at approximately one third of the distance from the knee joint space to the greater trochanter. Consequently, they may have been located over the proximal fibers of the VM, rather than over VMO. In our study, motor points were detected using electromyostimulation, and the surface electrodes for the VMO were located more distally.
In summary, the separate nerve supply and attachment of the VMO suggest that the distal part of the VM performs a different function compared with the proximal part. Our electromyographic results, which show an increase in VMO activity in conditions where the knee was destabilized (landing from jumping), support the idea that the VMO helps to maintain the positional stability of the patella. However, our results also show that the VMO and VL can be activated simultaneously and in a coordinated manner during certain phases of concentric and eccentric exercise. This suggests that the VMO is also an active knee extensor. Thus, the VM should not simply be considered a knee extensor or a muscle whose main role is to maintain normal patellar tracking. We suggest that it is a muscle that performs both roles according to the task performed.
This work was supported by Zimmer UK.
1. Bennell, K., M. Duncan, and S. Cowan. Effect of patellar taping on vasti onset timing, knee
kinematics, and kinetics in asymptomatic individuals with a delayed onset of vastus medialis oblique. J. Orthop. Res
. 24:1854-1860, 2006.
2. Biga, N., D. Judlin, J. M. Thomine, and H. Leger. Results of the Insall operation (plasty of the anterior cruciate ligament by transfer of the fascia lata with bone block) in the treatment of chronic anterior instability of the knee
. Rev. Chir. Orthop. Reparatrice Appar. Mot
. 72:115-120, 1986.
3. Boucher, J. P., M. A. King, R. Lefebvre, and A. Pepin. Quadriceps
femoris muscle activity in patellofemoral
pain syndrome. Am. J. Sports Med
. 20:527-532, 1992.
4. Cowan, S. M., K. L. Bennell, and P. W. Hodges. Therapeutic patellar taping changes the timing of vasti muscle activation in people with patellofemoral
pain syndrome. Clin. J. Sport Med
. 12:339-347, 2002.
5. Cowling, E. J., J. R. Steele, and P. J. McNair. Effect of verbal instructions on muscle activity and risk of injury to the anterior cruciate ligament during landing. Br. J. Sports Med
. 37:126-130, 2003.
6. Crossley, K., K. Bennell, S. Green, and J. McConnell. A systematic review of physical interventions for patellofemoral
pain syndrome. Clin. J. Sport Med
. 11:103-110, 2001.
7. Fagenbaum, R., and G. W. Darling. Jump landing strategies in male and female college athletes and the implications of such strategies for anterior cruciate ligament injury. Am. J. Sports Med
. 31:233-240, 2003.
8. Fredericson, M., and K. Yoon. Physical examination and patellofemoral
pain syndrome. Am. J. Phys. Med. Rehabil
. 85:234-243, 2006.
9. Fulkerson, J. P., and K. P. Shea. Disorders of patellofemoral
alignment. J. Bone Joint Surg. Am
. 72:1424-1429, 1990.
10. Grabiner, M. D., T. J. Koh, and L. F. Draganich. Neuromechanics of the patellofemoral
joint. Med. Sci. Sports Exerc
. 26:10-21, 1994.
11. Herrington, L., and S. Pearson. Does level of load affect relative activation levels of vastus medialis oblique and vastus laterialis? J. Electromyogr. Kinesiol
. 16:379-383, 2006.
12. Komi, P. V., and C. Bosco. Utilization of stored elastic energy in leg extensor muscles by men and women. Med. Sci. Sports
. 10:261-265, 1978.
13. Lefebvre, R., A. Leroux, G. Poumarat, et al. Vastus medialis: anatomical and functional considerations and implications based upon human and cadaveric studies. J. Manipulative Physiol. Ther
. 29:139-144, 2006.
14. Lieb, F. J., and J. Perry. Quadriceps
function. An anatomical and mechanical study using amputated limbs. J. Bone Joint Surg
. 50:1535-1548, 1968.
15. Mellor, R., and P. W. Hodges. Effect of knee
joint angle on motor unit synchronization. J. Orthop. Res
. 24:1420-1426, 2006.
16. Panagiotopoulos, E., P. Strzelczyk, M. Herrmann, and G. Scuderi. Cadaveric study on static medial patellar stabilizers: the dynamizing role of the vastus medialis obliquus on medial patellofemoral
ligament. Knee Surg. Sports Traumatol. Arthrosc
. 14:7-12, 2006.
17. Peeler, J., J. Cooper, M. M. Porter, J. A. Thliveris, and J. E. Anderson. Structural parameters of the vastus medialis muscle. Clin. Anat
. 18:281-289, 2005.
18. Powers, C. M., R. Landel, and J. Perry. Timing and intensity of vastus muscle activity during functional activities in subjects with and without patellofemoral
pain. Phys. Ther
. 76:946-955, 1996.
19. Signorile, J. F., D. Kacsik, A. Perry, et al. The effect of knee
and foot position on the electromyographical activity of the superficial quadriceps
. J. Orthop. Sports Phys. Ther
. 22:2-9, 1995.
20. Tillman, M. D., C. J. Hass, J. W. Chow, and D. Brunt. Lower extremity coupling parameters during locomotion and landings. J. Appl. Biomech
. 21:359-370, 2005.
21. Toumi, H., T. M. Best, A. Martin, and G. Poumarat. Muscle plasticity after weight and combined (weight + jump) training. Med. Sci. Sports Exerc
. 36:1580-1588, 2004.
22. Toumi, H., T. M. Best, A. Martin, S. F‘Guyer, and G. Poumarat. Effects of eccentric phase velocity of plyometric training on the vertical jump. Int. J. Sports Med
. 25:391-398, 2004.
23. Weinstabl, R., W. Scharf, and W. Firbas. The extensor apparatus of the knee
joint and its peripheral vasti: anatomic investigation and clinical relevance. Surg. Radiol. Anat
. 11:17-22, 1989.
24. Yu, B., C. F. Lin, and W. E. Garrett. Lower extremity biomechanics during the landing of a stop-jump task. Clin. Biomech
. 21:297-305, 2006.