Quadriceps femoris muscle activity patellofemoral pain syndrome
in
JEAN P. BOUCHER,* PhD, MARJORIE A. KING, MS, ATC, PT, RICHARD LEFEBVRE, AND ANDRÉ PÉPIN, MSc
From the
Department of Kinanthropology, University of Quebec at Montreal, Montreal, Quebec, Canada
pain syndrome the vastus medialis may even be less active relative to the vastus lateralis in the last degrees of extension compared to 90°. Furthermore, one may suggest that in patellofemoral pain syndrome the mechanical disturbances are exhibited first, at which time the vastus medialis atrophy, if present, would have a mechanical origin.
ABSTRACT To elucidate and attempt to dissociate the two mechanisms, neuromuscular and mechanical, underlying patellofemoral pain syndrome, 18 subjects, divided into two groups based on a diagnosis of patellofemoral pain syndrome and the knee Q angle, were studied. The control group was asymptomatic and exhibited a normal Q angle (mean, 8.25°), whereas the other group,
diagnosed as patellofemoral pain syndrome patients, reported knee pain and had an above-normal Q angle (mean, 21.05°). All subjects were tested for isometric
Patellofemoral pain syndrome (PFPS) is commonly diagnosed and spans a spectrum of athletic endeavors.5.8.13.l9,21,28 It is also referred to by a number of diagnoses such as patellofemoral chondritis, chondromalacia, anterior knee pain, quadriceps or vastus medialis obliquus (VMO) insuf-
maximum knee extension at 90°, 30°, and 15° of knee flexion while they were seated in a special restraining chair. During testing, surface electromyography at the oblique and long fibers of the vastus medialis, and at the vastus lateralis were recorded along with the knee moment of force. The integrated electromyographic signals associated with the peak torque for all of the vastus muscles, along with the vastus medialis obliquus :vastus lateralis and vastus medialis longus:vastus lateralis activity ratios showed no significant differences between groups nor between the three angles, suggesting that all vasti measured were consistently active throughout the studied range of motion. This suggests that the neural drive was not affected in the patellofemoral pain syndrome patients. However, when the five patients showing the largest Q angles were isolated, they revealed a significantly smaller vastus medialis obliquus:vastus lateralis ratio when compared to the other group. The same ratio was also significantly smaller at 15° compared to 90°. These results did demonstrate that in advanced cases of patellofemoral
ficiency, patellar subluxation, patellofemoral dysfunction, or patellar compression syndrome. Specifically, the term chondromalacia indicates a pathologic state of &dquo;softened&dquo; cartilage and is inappropriate unless visually diagnosed as such by arthroscopic examination. For an accurate diagnosis, the term chondromalacia should not be used by itself&dquo; The symptoms of PFPS
multiple.4,8.19,21 Patients com-
this may be some interaction of mechanical and neuromuscular factors. Traditionally, therapeutic exercise regimes aimed at rehabilitating the VMO focused on terminal extension exercises, since it was believed that the VMO is most active in the final degrees of extension. At present, there is some question as to the efficacy of terminal extension since some evidence suggests that the VMO fires throughout the entire knee extension range of motion.2.11.l4,l5.20,24 The goal of this study was twofold: 1) to investigate the appropriate-
*
Address correspondence and reprint requests to: Jean P. Boucher, PhD, Professor, D6partement de Kinanthropologie, Universite du Quebec a Montreal, Box 8888, Station A, Montr6al (Quebec), Canada, H3C 3P8. This work was presented in part at the 36th annual meeting of the American
College of Sports
are
plain of diffuse knee pain, with or without activity. Frequently, stair climbing or prolonged sitting produce discomfort and orthopaedic assessment usually shows a positive patella grind test and discomfort with palpation retropatella. There may also be swelling, loss of motion, and a sensation of giving way or instability. Notable VMO atrophy or dysplasia is frequently associated with PFPS.11.16.20 It is suspected that the etiology of
Medicine in 1989.
527
528 ness of a terminal extension protocol for the rehabilitation of the vastus medialis, and 2) to elucidate and try to dissociate the neuromuscular and mechanical mechanisms underlying PFPS.
MATERIALS AND METHODS
Subjects A total of 18 female subjects (mean age, 19 years) were divided into two groups based on the knee Q angle and a traditional diagnosis of PFPS rendered by a physician. Participants were not paid for their services and all signed an informed consent document.
Measurement procedures
8
The knee Q angle was measured using an automated twodimensional video-based motion analysis system (2-D SUN/ 120, Motion Analysis Corporation, Santa Rosa, CA) with retroflective markers placed on anatomic landmarks identified by palpation. These landmarks defining the Q angle are the anterosuperior iliac spine, the center of the patella, and the center of the tibial tubercule (Fig. 1). The Q angle was then quantified automatically by videotaping the subjects in the frontal plane while standing in a standardized position. The video data was then transferred to the motion analysis system through a video processor (VP110, Motion Analysis Corporation) and the Q angle was quantified using a specially designed software. The EMG signals of the VMO, the long fibers of the vastus medialis (VML), and the vastus lateralis (VL) were collected. The vastus medialis was divided into the VMO and VML based on their reported anatomic and functional independence.l5.23 As suggested by Reynolds et al., 21 the two groups of fibers of the vastus medialis can be differentiated by their respective anatomic angle relative to the long axis of the femur (VMO, 46° to 52° and VML, 15° to 18023) and through palpation during successive cycles of isometric contractions and relaxations. The EMG signal was recorded using bipolar surface electrodes (standard Beckman Ag/ AgCl, Beckman Instruments, Fullerton, CA), amplified through preamplifiers (Grass P511, Grass Instruments, Quincy, MA), and band-pass filtered at 3 Hz to 1 kHz. The electrodes were secured over the belly of the muscles in relationship with the estimated motor point, as described by Warfe126 only after the skin impedance was reduced to 10 kS2 or less. The active electrode was located over the estimated motor point, the reference electrode was secured 2.5 cm distally, and the ground electrode was overlying a bony electrically independent area (i.e., the head of the fibula). The knee extension moment of force was measured through a variable resistance strain gauge yielding an electrical signal proportional to the force applied against it. This strain gauge was then attached to a special lever to which the tested lower limb was secured. The attachment was done through straps around the ankle. Great care was taken to maintain the lower leg at 90° to the strain gauge line of
.
1. Schematic representation of the operational definition of the Q angle along with the EMG surface electrode
Figure
position. measurement. Finally, the distance between the knee axis of rotation and the attachment at the ankle was measured to quantify the moment of force from the force signal recorded through the strain gauge. The EMG and strain gauge signals were acquired on-line at a rate of 1000 Hz per channel using an analog-to-digital interface (12-bit resolution, MDAS 7000, Transera Corporation, Provo, UT) connected to a microcomputer (Amiga 2000, Commodore Business Machines Inc, West Chester,
PA).
Experimental conditions signals described above were measured during maxivoluntary contractions of the knee extensors while the subjects were seated in a special restraining chair with the hip at 90°. This chair allowed standardization of the positions and the contractions, and minimized unwanted movements. The signals were monitored at three angles: 90°, 30°, and 15° of knee flexion. There were two trials at each angle and the order of the angles tested differed randomly from one subject to another. The
mum
529
Quantification and
analysis
The EMG and strain gauge signals were quantified using a technique developed in our laboratories and presented elsewhere.’ Briefly, the EMG signal immediately preceding peak torque (i.e., 100-msec time window) was numerically fullwave rectified and integrated to obtain an integrated EMG value for each muscle and each trial. This 100-msec time window is defined by first locating the peak force in the strain gauge signal and then taking the EMG signal for a period of time starting 100 ms before, and ending at, peak force. Then, the following integrated EMG ratios were calculated : VMO:VL and VML:VL. The integrated EMG ratios were used to evaluate the muscle activity between the VMO, VML, and VL since the absolute EMG values are more variable, more subject to error due to external variables hard to control, and, hence, less reliable.’ These ratios represent quantitative measurement of the relative contribution of each muscle in the contraction. Finally, the statistical analysis needed to compare the groups in all conditions was conducted using a group x angle, two-way factorial analysis of variance with repeated measures on the last factor, the
angles.&dquo; RESULTS When comparing the groups taken as a whole, no significant differences (P > 0.05) were found between groups nor between angles (Figs. 2 and 3). Although not significantly different, an interesting trend can be observed on Figure 3 in the VMO:VL ratio for the symptomatic, or PFPS, patient group. The VMO:VL ratio reveals a tendency to decrease from the 90° angle condition to the 15° angle condition in the PFPS group, while it remains stable in the control group. To investigate this trend further, the functional difference between the groups was maximized by analyzing the five patients with the largest Q angles (>22°) and contrasting them to five control subjects (Figs. 4 and 5). In this last comparison, the difference between the knee positions (i.e., angle conditions), along with the group x angle interaction, reached statistical significance (P < 0.05). The significant group x angle interaction is revealed by a significant difference (P < 0.05) between groups only for the 15° knee angle
condition (Fig. 5). This difference shows that the VMO:VL ratio is significantly smaller at the 15° angle condition compared to the 90° angle only for the PFPS group. In the asymptomatic or control group, the same ratio was very stable. In other words, the VMO:VL ratio was found to be significantly lower in the PFPS group when compared to the control group in the terminal extension condition only. Finally, a significant difference due to the angle effect was found for the moment of force, both for the whole group and the five subjects with the largest Q angles, whereas the comparison of the PFPS group to the control group was not significant. These results are consistent with the well-documented force-angle curve.’ According to this curve, 90° is
Figure 2. Results for the integrated EMG (IEMG) of the VMO and VML, and the VL measured in all conditions for both whole groups.
the
angle
at which the muscle is the most effective and at 15° the efficiency is greatly reduced.
strong, whereas DISCUSSION
Within the limitation of this study, the results presented above failed to demonstrate that the VMO or the VML are more active in terminal extension. In fact, the VMO activity
530
Angles
Figure 3. Results for the integrated EMG (IEMG) ratios (VMO/ VL and VML/VL) and the moment of force measured in all
Angles
conditions for both whole groups.
Figure 4. Results for the integrated EMG (IEMG) of the VMO and VML and VL measured in all conditions for five control subjects and the five PFPS patients with the largest Q angles.
relative to the VL decreases in the PFPS group across the angle conditions, while the control group ratio remained stable. This decrease in the VMO:VL ratio in the PFPS group corroborates the work by Mariani and Carusol6 and Souza and Gross24 and suggests that therapeutic exercises in terminal extension may add to the neuromuscular imbalance of the VMO. Souza and Gross suggest that this reduction may interact
with biomechanical factors. Mariani and Caruso also reported a reduction in vastus medialis activity in PFPS patients, especially in the last degrees of extension, while the VL activity remained normally elevated. The same authors reported that such a reduction in the vastus medialis activity was not present in normal knee subjects. Reynolds et al. 20 also measured the VMO and VL in normal knee subjects, and corroborated the conclusions of Mariani and
531
When
comparing the five
extreme
cases
in
our
study
to
the whole group results, where the severity of the PFPS is much more variable, one may suggest that in PFPS the mechanical disturbances are exhibited first. These may include symptomatic complaints (e.g., anterior knee pain) and are associated with a moderate increase in Q angle and a stable EMG, suggesting that the vastus medialis dysplasia, if present, would have a mechanical origin. That is, a knee or ankle malalignment could put the vastus medialis in such a mechanical position that its contribution would be minimized. If the mechanical alignment is allowed to deteriorate further, the clinical picture changes. This deterioration can result in increased symptoms and is associated with a larger Q angle and a depressed VMO:VL ratio of integrated EMG, such as in the severe symptom group. Hence, a neuromuscular component is added to the syndrome, as revealed by a decreased neural drive to the VMO relative to the VL; this neuromuscular component is not present when the mechanical misalignment is less important (i.e., Q angle < 22°).
Hence, the results presented here confirm the suspicion brought forth by Mariani and Carusol6 and, more recently, by Souza and Gross,24 that the neuromuscular imbalance has a mechanical origin. This suspicion was supported by the report that when the mechanics of the quadriceps mechanisms were corrected through surgery, the vastus medialis activity regained normal values.16 The present study revealed that when the mechanical malalignment is less sethe VMO:VL ratio is similar to the normal group. When this mechanical problem is greater, such as in the five
vere,
subjects with larger Q angles, the VMO:VL ratio becomes significantly smaller, indicating a reduction of the VMO activity relative to the VL. The stability of the VML:VL ratio, also monitored in this study, may appear paradoxical at first. However, these results confirm that the VML (as with the VL), when pulling at an angle of about 15° relative to the femur, is more involved in knee extension than in tracking the patella, the latter being a primary task for the VMO, which pulls at
.--ItJ--
Figure 5. Results for the integrated EMG (IEMG) ratios (VMO/ VL and VML/VL) and the moment of force measured in all conditions for five control subjects and the five PFPS patients with the largest Q angles. Caruso, that both muscles were as active in the last degrees of extension. These results clearly demonstrate that an important neuromuscular imbalance between the VMO and the VL is associated with PFPS and that it can be investigated through VMO:VL ratio of activity.24 It has also been suggested that this imbalance could have a mechanical basis.&dquo;, 24
about 50° to the femur. This VML and VL synergy is also corroborated by the consistency of the moment of force between groups. The moment of force results revealed no group differences, which can be interpreted as PFPS having no effect on the capacity to produce knee extension force. This finding is not so surprising when interpreted in light of the stability of the VML:VL ratio and the fact that the VMO, the affected muscle, is mostly involved in tracking the patella and not extending the knee.14,20 The clinical impact of this study appears obvious. It may be inappropriate to arbitrarily choose a terminal extension angle as a standard rehabilitation protocol for patients with PFPS. If available, evaluation of torque curve tracings, both concentrically and eccentrically, may allow for an appropriate range of motion choice for short arc extension, which may not necessarily be in terminal extension. Additionally, it has been documented that the articular cartilage is adversely affected by shear forces and may actually derive its nutrition through diffusion that occurs through intermit-
532
tently applied compression.6,9,lO,l2,l7,22 This study showed an increased VMO activity relative to the VL at 90° of flexion when compared to 15° of knee flexion. Perhaps isometric knee extension at 90°, where the compression forces are maximized and the shear forces are minimized, not only assists in augmenting or facilitating the VMO compared to the VL, but also aids in the diffusion activity useful for the repair of retropatellar articular cartilage.25 Finally, when evaluating PFPS patients one may need to consider the duration of the symptomatic complaint, since it appears that the initial VMO dysplasia may be mechanical in nature. Lower extremity mechanics are critical at that stage. This may include not only bony alignment around the knee joint, but also around the ankle joint and the foot. As symptoms persist, the neuromuscular component, as demonstrated by a decreased neural drive to the VMO relative to the VL, must be addressed. CONCLUSION This study indicates that it is inappropriate to arbitrarily choose a terminal extension angle as a standard rehabilitation protocol for patients with patellofemoral pain syndrome. It does suggest that isometric contractions at 90° of knee extension may enhance not only the activity of the obliquus fibers of the vastus medialis relative to the vastus lateralis, but may also assist in retropatellar articular cartilage repair. Furthermore, when evaluating patients with patellofemoral pain syndrome, one must consider both the mechanical and neuromuscular origins of the symptomatic and orthopaedic implications. Thus, rehabilitation strategies should include a mechanical management (e.g., through orthoses) and a functional or neuromuscular management
(e.g., through functional electrical stimulation or EMG biofeedback). REFERENCES 1.
Basmajian JV, DeLuca CJ: Muscles Alive.
Fifth edition. Baltimore, Williams & Wilkins, 1985 2. Bose K, Kanagasuntherum R, Osman M: Vastus medialis oblique: An anatomical and physiological study. Orthopedics 3: 880-883, 1980 3. Boucher JP, Pépin A, Lefebvre R: Using the vastus medialis to vastus
lateralis IEMG ratio as a neuromuscular imbalance index for the diagnosis of patellofemoral syndrome. Proceedings of the 1990 International Conference on Spinal Manipulation. Arlington, VA, Foundation for Chiropractic Education and Research, 1990, pp 274-281 4. Chesworth BM, Gulham EG, Tata GE, et al: Validation of outcome measures in patients with patellofemoral syndrome. J Orthop Sports Phys Ther
11: 302-308, 1989 5. Cox JS: Patellofemoral problems in runners. Clin Sports Med 4: 699-715, 1985 6. Ekholm R: Nutrition of articular cartilage. Acta Anat 12: 177, 1955 7. Enoka RM: Neuromechanical Basis of Kinesiology. Champaign, IL, Human Kinetics Publishers, 1988 8. Garrick JG: Anterior knee pain (chondromalacia patellae). Physician
9.
Sportsmed 17(1): 75-84, 1989 Glasgow EF: Aspects of Manipulative Therapy.
New York, Churchill Livingstone, 1985, pp 21, 113-114 10. Goodfellow J, Hungerford D, Zindel M: Patello-femoral joint mechanics and pathology 1 and 2. J Bone Joint Surg 58B: 287-299, 1976 11. Hanten WP, Schulthies SS: Exercise effect on electromyographic activity of the vastus medialis oblique and vastus lateralis muscles. Phys Ther 70:
561-565,1990 12. Insall J: Chondromalacia patellae: Patellar malalignment syndrome. Orthop Clin North Am 10: 117-127, 1979 13. Ireland ML: Patellofemoral disorders in runners and bicyclists. Ann Sports Med 3: 77-84, 1987 14. Lieb FJ, Perry J: Quadriceps functions. J Bone Joint Surg 53A: 749-758, 1971 15. Lieb FJ, Perry J: Quadriceps functions. An anatomical and mechanical study using amputated limbs. J Bone Joint Surg 50A: 1535-1548, 1968 16. Mariani PP, Caruso I : An electromyographic investigation of subluxation of the patella. J Bone Joint Surg 61B: 169-171, 1979 17. McCribbon B: Nutrition, in Freeman MAR (ed): Adult Articular Cartilage. Oxford, Pitman, 1973, pp 277-286 18. Merchant AC: Classification of patellofemoral disorders. Arthroscopy 4:
235-240, 1988
Murray PB: Case study: Rehabilitation of a collegiate football placekicker with patellofemoral arthritis. J Orthop Sports Phys Ther 10: 224-227, 1988 20. Reynolds L, Levin TA, Medeiros JM, et al: EMG activity of the vastus medialis oblique and vastus lateralis and their role in patellar alignment. Am J Phys Med 62: 61-70, 1983 21. Rintala P: Patellofemoral pain syndrome and its treatment in runners. Athletic Training 25: 107-110, 1990 22. Salter RB: Prevention of arthritis through preservation of cartilage. J Can 19.
Assoc Radiol 32: 5-7, 1981 23. Scharf W, Weinstabl R, Orthner E: Anatomische Unterscheidung und klinische Bedeutung Zweier verschiedener Anteile des Musculus vastus medialis. Acta Anat 123: 108-111, 1985 24. Souza DR, Gross MT: Comparison of vastus medialis obliquus: Vastus lateralis muscle integrated electromyographic ratios between healthy subjects and patients with patellofemoral pain. Phys Ther 71: 310-320, 1991 25. Stokes M, Young A: Investigations of quadriceps inhibition: Implications for clinical practice. Physiotherapy 70: 425-428, 1984 26. Warfel JH: The Extremities: Muscles and Motor Points. Fifth edition. Philadelphia, Lea and Febiger, 1985 27. Winer BJ: Statistical Principles in Experimental Design. Second edition. New York, McGraw-Hill Book Company, 1971 28. Wittenbecker NL, DiNitto LM Jr: Successful treatment of patellofemoral dysfunction in a dancer. J Orthop Sports Phys Ther 11: 270-273, 1989
in
JEAN P. BOUCHER,* PhD, MARJORIE A. KING, MS, ATC, PT, RICHARD LEFEBVRE, AND ANDRÉ PÉPIN, MSc
From the
Department of Kinanthropology, University of Quebec at Montreal, Montreal, Quebec, Canada
pain syndrome the vastus medialis may even be less active relative to the vastus lateralis in the last degrees of extension compared to 90°. Furthermore, one may suggest that in patellofemoral pain syndrome the mechanical disturbances are exhibited first, at which time the vastus medialis atrophy, if present, would have a mechanical origin.
ABSTRACT To elucidate and attempt to dissociate the two mechanisms, neuromuscular and mechanical, underlying patellofemoral pain syndrome, 18 subjects, divided into two groups based on a diagnosis of patellofemoral pain syndrome and the knee Q angle, were studied. The control group was asymptomatic and exhibited a normal Q angle (mean, 8.25°), whereas the other group,
diagnosed as patellofemoral pain syndrome patients, reported knee pain and had an above-normal Q angle (mean, 21.05°). All subjects were tested for isometric
Patellofemoral pain syndrome (PFPS) is commonly diagnosed and spans a spectrum of athletic endeavors.5.8.13.l9,21,28 It is also referred to by a number of diagnoses such as patellofemoral chondritis, chondromalacia, anterior knee pain, quadriceps or vastus medialis obliquus (VMO) insuf-
maximum knee extension at 90°, 30°, and 15° of knee flexion while they were seated in a special restraining chair. During testing, surface electromyography at the oblique and long fibers of the vastus medialis, and at the vastus lateralis were recorded along with the knee moment of force. The integrated electromyographic signals associated with the peak torque for all of the vastus muscles, along with the vastus medialis obliquus :vastus lateralis and vastus medialis longus:vastus lateralis activity ratios showed no significant differences between groups nor between the three angles, suggesting that all vasti measured were consistently active throughout the studied range of motion. This suggests that the neural drive was not affected in the patellofemoral pain syndrome patients. However, when the five patients showing the largest Q angles were isolated, they revealed a significantly smaller vastus medialis obliquus:vastus lateralis ratio when compared to the other group. The same ratio was also significantly smaller at 15° compared to 90°. These results did demonstrate that in advanced cases of patellofemoral
ficiency, patellar subluxation, patellofemoral dysfunction, or patellar compression syndrome. Specifically, the term chondromalacia indicates a pathologic state of &dquo;softened&dquo; cartilage and is inappropriate unless visually diagnosed as such by arthroscopic examination. For an accurate diagnosis, the term chondromalacia should not be used by itself&dquo; The symptoms of PFPS
multiple.4,8.19,21 Patients com-
this may be some interaction of mechanical and neuromuscular factors. Traditionally, therapeutic exercise regimes aimed at rehabilitating the VMO focused on terminal extension exercises, since it was believed that the VMO is most active in the final degrees of extension. At present, there is some question as to the efficacy of terminal extension since some evidence suggests that the VMO fires throughout the entire knee extension range of motion.2.11.l4,l5.20,24 The goal of this study was twofold: 1) to investigate the appropriate-
*
Address correspondence and reprint requests to: Jean P. Boucher, PhD, Professor, D6partement de Kinanthropologie, Universite du Quebec a Montreal, Box 8888, Station A, Montr6al (Quebec), Canada, H3C 3P8. This work was presented in part at the 36th annual meeting of the American
College of Sports
are
plain of diffuse knee pain, with or without activity. Frequently, stair climbing or prolonged sitting produce discomfort and orthopaedic assessment usually shows a positive patella grind test and discomfort with palpation retropatella. There may also be swelling, loss of motion, and a sensation of giving way or instability. Notable VMO atrophy or dysplasia is frequently associated with PFPS.11.16.20 It is suspected that the etiology of
Medicine in 1989.
527
528 ness of a terminal extension protocol for the rehabilitation of the vastus medialis, and 2) to elucidate and try to dissociate the neuromuscular and mechanical mechanisms underlying PFPS.
MATERIALS AND METHODS
Subjects A total of 18 female subjects (mean age, 19 years) were divided into two groups based on the knee Q angle and a traditional diagnosis of PFPS rendered by a physician. Participants were not paid for their services and all signed an informed consent document.
Measurement procedures
8
The knee Q angle was measured using an automated twodimensional video-based motion analysis system (2-D SUN/ 120, Motion Analysis Corporation, Santa Rosa, CA) with retroflective markers placed on anatomic landmarks identified by palpation. These landmarks defining the Q angle are the anterosuperior iliac spine, the center of the patella, and the center of the tibial tubercule (Fig. 1). The Q angle was then quantified automatically by videotaping the subjects in the frontal plane while standing in a standardized position. The video data was then transferred to the motion analysis system through a video processor (VP110, Motion Analysis Corporation) and the Q angle was quantified using a specially designed software. The EMG signals of the VMO, the long fibers of the vastus medialis (VML), and the vastus lateralis (VL) were collected. The vastus medialis was divided into the VMO and VML based on their reported anatomic and functional independence.l5.23 As suggested by Reynolds et al., 21 the two groups of fibers of the vastus medialis can be differentiated by their respective anatomic angle relative to the long axis of the femur (VMO, 46° to 52° and VML, 15° to 18023) and through palpation during successive cycles of isometric contractions and relaxations. The EMG signal was recorded using bipolar surface electrodes (standard Beckman Ag/ AgCl, Beckman Instruments, Fullerton, CA), amplified through preamplifiers (Grass P511, Grass Instruments, Quincy, MA), and band-pass filtered at 3 Hz to 1 kHz. The electrodes were secured over the belly of the muscles in relationship with the estimated motor point, as described by Warfe126 only after the skin impedance was reduced to 10 kS2 or less. The active electrode was located over the estimated motor point, the reference electrode was secured 2.5 cm distally, and the ground electrode was overlying a bony electrically independent area (i.e., the head of the fibula). The knee extension moment of force was measured through a variable resistance strain gauge yielding an electrical signal proportional to the force applied against it. This strain gauge was then attached to a special lever to which the tested lower limb was secured. The attachment was done through straps around the ankle. Great care was taken to maintain the lower leg at 90° to the strain gauge line of
.
1. Schematic representation of the operational definition of the Q angle along with the EMG surface electrode
Figure
position. measurement. Finally, the distance between the knee axis of rotation and the attachment at the ankle was measured to quantify the moment of force from the force signal recorded through the strain gauge. The EMG and strain gauge signals were acquired on-line at a rate of 1000 Hz per channel using an analog-to-digital interface (12-bit resolution, MDAS 7000, Transera Corporation, Provo, UT) connected to a microcomputer (Amiga 2000, Commodore Business Machines Inc, West Chester,
PA).
Experimental conditions signals described above were measured during maxivoluntary contractions of the knee extensors while the subjects were seated in a special restraining chair with the hip at 90°. This chair allowed standardization of the positions and the contractions, and minimized unwanted movements. The signals were monitored at three angles: 90°, 30°, and 15° of knee flexion. There were two trials at each angle and the order of the angles tested differed randomly from one subject to another. The
mum
529
Quantification and
analysis
The EMG and strain gauge signals were quantified using a technique developed in our laboratories and presented elsewhere.’ Briefly, the EMG signal immediately preceding peak torque (i.e., 100-msec time window) was numerically fullwave rectified and integrated to obtain an integrated EMG value for each muscle and each trial. This 100-msec time window is defined by first locating the peak force in the strain gauge signal and then taking the EMG signal for a period of time starting 100 ms before, and ending at, peak force. Then, the following integrated EMG ratios were calculated : VMO:VL and VML:VL. The integrated EMG ratios were used to evaluate the muscle activity between the VMO, VML, and VL since the absolute EMG values are more variable, more subject to error due to external variables hard to control, and, hence, less reliable.’ These ratios represent quantitative measurement of the relative contribution of each muscle in the contraction. Finally, the statistical analysis needed to compare the groups in all conditions was conducted using a group x angle, two-way factorial analysis of variance with repeated measures on the last factor, the
angles.&dquo; RESULTS When comparing the groups taken as a whole, no significant differences (P > 0.05) were found between groups nor between angles (Figs. 2 and 3). Although not significantly different, an interesting trend can be observed on Figure 3 in the VMO:VL ratio for the symptomatic, or PFPS, patient group. The VMO:VL ratio reveals a tendency to decrease from the 90° angle condition to the 15° angle condition in the PFPS group, while it remains stable in the control group. To investigate this trend further, the functional difference between the groups was maximized by analyzing the five patients with the largest Q angles (>22°) and contrasting them to five control subjects (Figs. 4 and 5). In this last comparison, the difference between the knee positions (i.e., angle conditions), along with the group x angle interaction, reached statistical significance (P < 0.05). The significant group x angle interaction is revealed by a significant difference (P < 0.05) between groups only for the 15° knee angle
condition (Fig. 5). This difference shows that the VMO:VL ratio is significantly smaller at the 15° angle condition compared to the 90° angle only for the PFPS group. In the asymptomatic or control group, the same ratio was very stable. In other words, the VMO:VL ratio was found to be significantly lower in the PFPS group when compared to the control group in the terminal extension condition only. Finally, a significant difference due to the angle effect was found for the moment of force, both for the whole group and the five subjects with the largest Q angles, whereas the comparison of the PFPS group to the control group was not significant. These results are consistent with the well-documented force-angle curve.’ According to this curve, 90° is
Figure 2. Results for the integrated EMG (IEMG) of the VMO and VML, and the VL measured in all conditions for both whole groups.
the
angle
at which the muscle is the most effective and at 15° the efficiency is greatly reduced.
strong, whereas DISCUSSION
Within the limitation of this study, the results presented above failed to demonstrate that the VMO or the VML are more active in terminal extension. In fact, the VMO activity
530
Angles
Figure 3. Results for the integrated EMG (IEMG) ratios (VMO/ VL and VML/VL) and the moment of force measured in all
Angles
conditions for both whole groups.
Figure 4. Results for the integrated EMG (IEMG) of the VMO and VML and VL measured in all conditions for five control subjects and the five PFPS patients with the largest Q angles.
relative to the VL decreases in the PFPS group across the angle conditions, while the control group ratio remained stable. This decrease in the VMO:VL ratio in the PFPS group corroborates the work by Mariani and Carusol6 and Souza and Gross24 and suggests that therapeutic exercises in terminal extension may add to the neuromuscular imbalance of the VMO. Souza and Gross suggest that this reduction may interact
with biomechanical factors. Mariani and Caruso also reported a reduction in vastus medialis activity in PFPS patients, especially in the last degrees of extension, while the VL activity remained normally elevated. The same authors reported that such a reduction in the vastus medialis activity was not present in normal knee subjects. Reynolds et al. 20 also measured the VMO and VL in normal knee subjects, and corroborated the conclusions of Mariani and
531
When
comparing the five
extreme
cases
in
our
study
to
the whole group results, where the severity of the PFPS is much more variable, one may suggest that in PFPS the mechanical disturbances are exhibited first. These may include symptomatic complaints (e.g., anterior knee pain) and are associated with a moderate increase in Q angle and a stable EMG, suggesting that the vastus medialis dysplasia, if present, would have a mechanical origin. That is, a knee or ankle malalignment could put the vastus medialis in such a mechanical position that its contribution would be minimized. If the mechanical alignment is allowed to deteriorate further, the clinical picture changes. This deterioration can result in increased symptoms and is associated with a larger Q angle and a depressed VMO:VL ratio of integrated EMG, such as in the severe symptom group. Hence, a neuromuscular component is added to the syndrome, as revealed by a decreased neural drive to the VMO relative to the VL; this neuromuscular component is not present when the mechanical misalignment is less important (i.e., Q angle < 22°).
Hence, the results presented here confirm the suspicion brought forth by Mariani and Carusol6 and, more recently, by Souza and Gross,24 that the neuromuscular imbalance has a mechanical origin. This suspicion was supported by the report that when the mechanics of the quadriceps mechanisms were corrected through surgery, the vastus medialis activity regained normal values.16 The present study revealed that when the mechanical malalignment is less sethe VMO:VL ratio is similar to the normal group. When this mechanical problem is greater, such as in the five
vere,
subjects with larger Q angles, the VMO:VL ratio becomes significantly smaller, indicating a reduction of the VMO activity relative to the VL. The stability of the VML:VL ratio, also monitored in this study, may appear paradoxical at first. However, these results confirm that the VML (as with the VL), when pulling at an angle of about 15° relative to the femur, is more involved in knee extension than in tracking the patella, the latter being a primary task for the VMO, which pulls at
.--ItJ--
Figure 5. Results for the integrated EMG (IEMG) ratios (VMO/ VL and VML/VL) and the moment of force measured in all conditions for five control subjects and the five PFPS patients with the largest Q angles. Caruso, that both muscles were as active in the last degrees of extension. These results clearly demonstrate that an important neuromuscular imbalance between the VMO and the VL is associated with PFPS and that it can be investigated through VMO:VL ratio of activity.24 It has also been suggested that this imbalance could have a mechanical basis.&dquo;, 24
about 50° to the femur. This VML and VL synergy is also corroborated by the consistency of the moment of force between groups. The moment of force results revealed no group differences, which can be interpreted as PFPS having no effect on the capacity to produce knee extension force. This finding is not so surprising when interpreted in light of the stability of the VML:VL ratio and the fact that the VMO, the affected muscle, is mostly involved in tracking the patella and not extending the knee.14,20 The clinical impact of this study appears obvious. It may be inappropriate to arbitrarily choose a terminal extension angle as a standard rehabilitation protocol for patients with PFPS. If available, evaluation of torque curve tracings, both concentrically and eccentrically, may allow for an appropriate range of motion choice for short arc extension, which may not necessarily be in terminal extension. Additionally, it has been documented that the articular cartilage is adversely affected by shear forces and may actually derive its nutrition through diffusion that occurs through intermit-
532
tently applied compression.6,9,lO,l2,l7,22 This study showed an increased VMO activity relative to the VL at 90° of flexion when compared to 15° of knee flexion. Perhaps isometric knee extension at 90°, where the compression forces are maximized and the shear forces are minimized, not only assists in augmenting or facilitating the VMO compared to the VL, but also aids in the diffusion activity useful for the repair of retropatellar articular cartilage.25 Finally, when evaluating PFPS patients one may need to consider the duration of the symptomatic complaint, since it appears that the initial VMO dysplasia may be mechanical in nature. Lower extremity mechanics are critical at that stage. This may include not only bony alignment around the knee joint, but also around the ankle joint and the foot. As symptoms persist, the neuromuscular component, as demonstrated by a decreased neural drive to the VMO relative to the VL, must be addressed. CONCLUSION This study indicates that it is inappropriate to arbitrarily choose a terminal extension angle as a standard rehabilitation protocol for patients with patellofemoral pain syndrome. It does suggest that isometric contractions at 90° of knee extension may enhance not only the activity of the obliquus fibers of the vastus medialis relative to the vastus lateralis, but may also assist in retropatellar articular cartilage repair. Furthermore, when evaluating patients with patellofemoral pain syndrome, one must consider both the mechanical and neuromuscular origins of the symptomatic and orthopaedic implications. Thus, rehabilitation strategies should include a mechanical management (e.g., through orthoses) and a functional or neuromuscular management
(e.g., through functional electrical stimulation or EMG biofeedback). REFERENCES 1.
Basmajian JV, DeLuca CJ: Muscles Alive.
Fifth edition. Baltimore, Williams & Wilkins, 1985 2. Bose K, Kanagasuntherum R, Osman M: Vastus medialis oblique: An anatomical and physiological study. Orthopedics 3: 880-883, 1980 3. Boucher JP, Pépin A, Lefebvre R: Using the vastus medialis to vastus
lateralis IEMG ratio as a neuromuscular imbalance index for the diagnosis of patellofemoral syndrome. Proceedings of the 1990 International Conference on Spinal Manipulation. Arlington, VA, Foundation for Chiropractic Education and Research, 1990, pp 274-281 4. Chesworth BM, Gulham EG, Tata GE, et al: Validation of outcome measures in patients with patellofemoral syndrome. J Orthop Sports Phys Ther
11: 302-308, 1989 5. Cox JS: Patellofemoral problems in runners. Clin Sports Med 4: 699-715, 1985 6. Ekholm R: Nutrition of articular cartilage. Acta Anat 12: 177, 1955 7. Enoka RM: Neuromechanical Basis of Kinesiology. Champaign, IL, Human Kinetics Publishers, 1988 8. Garrick JG: Anterior knee pain (chondromalacia patellae). Physician
9.
Sportsmed 17(1): 75-84, 1989 Glasgow EF: Aspects of Manipulative Therapy.
New York, Churchill Livingstone, 1985, pp 21, 113-114 10. Goodfellow J, Hungerford D, Zindel M: Patello-femoral joint mechanics and pathology 1 and 2. J Bone Joint Surg 58B: 287-299, 1976 11. Hanten WP, Schulthies SS: Exercise effect on electromyographic activity of the vastus medialis oblique and vastus lateralis muscles. Phys Ther 70:
561-565,1990 12. Insall J: Chondromalacia patellae: Patellar malalignment syndrome. Orthop Clin North Am 10: 117-127, 1979 13. Ireland ML: Patellofemoral disorders in runners and bicyclists. Ann Sports Med 3: 77-84, 1987 14. Lieb FJ, Perry J: Quadriceps functions. J Bone Joint Surg 53A: 749-758, 1971 15. Lieb FJ, Perry J: Quadriceps functions. An anatomical and mechanical study using amputated limbs. J Bone Joint Surg 50A: 1535-1548, 1968 16. Mariani PP, Caruso I : An electromyographic investigation of subluxation of the patella. J Bone Joint Surg 61B: 169-171, 1979 17. McCribbon B: Nutrition, in Freeman MAR (ed): Adult Articular Cartilage. Oxford, Pitman, 1973, pp 277-286 18. Merchant AC: Classification of patellofemoral disorders. Arthroscopy 4:
235-240, 1988
Murray PB: Case study: Rehabilitation of a collegiate football placekicker with patellofemoral arthritis. J Orthop Sports Phys Ther 10: 224-227, 1988 20. Reynolds L, Levin TA, Medeiros JM, et al: EMG activity of the vastus medialis oblique and vastus lateralis and their role in patellar alignment. Am J Phys Med 62: 61-70, 1983 21. Rintala P: Patellofemoral pain syndrome and its treatment in runners. Athletic Training 25: 107-110, 1990 22. Salter RB: Prevention of arthritis through preservation of cartilage. J Can 19.
Assoc Radiol 32: 5-7, 1981 23. Scharf W, Weinstabl R, Orthner E: Anatomische Unterscheidung und klinische Bedeutung Zweier verschiedener Anteile des Musculus vastus medialis. Acta Anat 123: 108-111, 1985 24. Souza DR, Gross MT: Comparison of vastus medialis obliquus: Vastus lateralis muscle integrated electromyographic ratios between healthy subjects and patients with patellofemoral pain. Phys Ther 71: 310-320, 1991 25. Stokes M, Young A: Investigations of quadriceps inhibition: Implications for clinical practice. Physiotherapy 70: 425-428, 1984 26. Warfel JH: The Extremities: Muscles and Motor Points. Fifth edition. Philadelphia, Lea and Febiger, 1985 27. Winer BJ: Statistical Principles in Experimental Design. Second edition. New York, McGraw-Hill Book Company, 1971 28. Wittenbecker NL, DiNitto LM Jr: Successful treatment of patellofemoral dysfunction in a dancer. J Orthop Sports Phys Ther 11: 270-273, 1989
