The Effect of Open and Closed Chain Exercise and Knee Joint Position on Patellar Tracking in. Lateral Patellar Compression syndrome1

The Effect of Open and Closed Chain Exercise and Knee Joint Position on Patellar Tracking in w Lateral Patellar Compression syndrome1 Susan A. Doucette, PT, MS Douglas D. Child, MD I t is generally accepted that conservative rehabilitation brings relief of symptoms to the majority of patients with patellofemoral pain. It has also been demonstrated radiographically that physical therapy can improve patellar tracking (6). The basis of this rehabilitation is to strengthen the vastus medialis oblique whose fibers act as a medial dynamic stabilizer of the patella. Conservative treatment for patellofemoral pain generally consists of strengthening, stretching, motor control, external appliances, therapeutic modalities, and anti-inflammatories. There are numerous rehabilitation protocols for patellofemoral problems, but there is no consensus regarding the proper way to conservatively treat this disorder. Open chain leg extension exercises have been the traditional way to strengthen the quadriceps because this exercise allows for quadriceps isolation. Examples of open chain quadriceps strengthening includes straight leg raises, short arc quadri- ceps, and sitting knee extensions. In open chain exercise, the end segment is free, the axis of motion is distal to the joint, and the muscle action is primarily concentric. Closed chain exercise is considered more functional because there is a normal physiological load through the skeletal system, muscle contractions are synergistic, and normal proprioceptive feedback mecha- There are numerous rehabilitation protocols for patellofemoral problems, but there is little objective data to determine the most effective exercise program to conservatively treat this disorder. The purpose of this study was to determine the effect of open and closed chain exercise and knee joint position on patellar tracking in lateral patellar compression syndrome. Computed tomography scans of the patellofemoral joint were performed with the leg in three muscle conditions and at five knee angles in 16 subjects with lateral patellar compression syndrome. Patellar tracking was evaluated by measuring congruence angle. Relaxed and closed chain conditions demonstrated improved congruence as compared with the open chain condition at 0, 10, and 20' of knee flexion (p<.0001). Open chain strengthening techniques appear to be most appropriate after 30' of knee flexion. The three muscle conditions demonstrated progressively improved patellar congruence from 0 to 40' of knee flexion. Key Wonls: patella, computed tomography, movement ' 1995 Breg Excellence in Research Award Paper. Presented at the 16th Annual Physician-Thera,oist Team Concept Conference, Dallas, TX, November 3-5, 1995. ' Physical Therapist, Mountain West Physical Therapy, 850 East 1200 North, Logan, UT 84341. ' Radiologist, Western Medical Center, logan, UT nisms are utilized. Examples of closed chain quadriceps strengthening exercises include the squat, leg press, and step exercises. In these exercises, the end segment is fixed, the axis of motion is proximal and distal to the joint, and concentric, eccentric, and isometric muscle action is utilized. Biomechanics and muscle function associated with closed and open chain conditions are considerably different. Open chain subtalar joint pronation produces eversion, abduction, and dorsiflexion of the calcaneous and the foot with the leg and talar unit held stable (8). In closed chain subtalar joint pronation, the leg will internally rotate, the knee will flex, the talus will adduct and plantar flex, and the calcaneous will evert (8). In open chain supination, the calcaneous and foot will adduct, invert, and plantar flex with the leg and talar unit held stable (8). With closed chain supination, the leg externally rotates, the knee extends, the talus will abduct and dorsiflex, and the calcaneous will invert (8). The quadriceps muscle group acts as a knee extensor in an open chain condition and as a decelerator of knee flexion and anterior stabilizer of the knee in the closed chain condition. Significant differences exist between open chain and closed chain exercises and clinicians are not in agreement as to when it is most a p propriate to use them with the patellofemoral pain patient. Short arc quadriceps sets are often advocated for vastus medialis oblique strength- Volume 23 Number 2 February 1996 JOSPT

ening (1,4,5,11,13). However, the use of short arc quadriceps sets to selectively recruit the vastus medialis oblique is unsubstantiated (7). In normal subjects, the patella assumes a more lateral orientation when short arc quadriceps sets are used (10). Some clinicians feel that only open chain exercises should be used initially because they cause less joint compression (15,23,25). Others have reported that open chain exercise often increases patellofemoral pain and may actually damage the articular cartilage by placing excessive stresses on the joint (8,21). The importance of knee angle or position should be considered when rehabilitating the patellofemoral joint. It has been reported that with patellofemoral pain, the quadriceps should be isometrically strengthened initially with the knee in an extended position because the damaging compressive force on the patellar cartilage and the mechanical stress on the often painful extensor mechanism are much lower in extension (15). Since there is no articulation from 10" of knee flexion to full extension, some believe it is generally safe to - -- The patella assumes a more lateral orientation when short arc quadriceps sets are used. - - - -- - - - -. -..--.-- --. - strengthen in this range with any type of exercise (9,25). Others advocate open chain exercise from 0 to 30" of flexion of the knee because of the lower patella femoral contact forces and pressures that occur in those ranges of movement (15,23, 25). However, patellofemoral joint stresses were shown to be significantly greater in open chain exercise at 0 and 30" than with closed chain exer- cise and that patellofemoral joint stresses were significantly greater in closed chain exercise at 60 and 90" than in open chain exercise (21). It has been demonstrated that vastus medialis oblique electromyographic activity was the greatest between 60 to 90" arc of movement and vastus medialis oblique electromyographic activity was the lowest in the 0-30" arc of motion (3). Little objective data exist to sup port or refute the use of open or closed chain exercise to enhance rehabilitation in patients with patellar dysfunction. The purpose of this study was to determine the effect of closed chain and open chain exercise and knee joint position on patellar tracking in lateral patellar compression syndrome as evidenced by computed tomography. Computed t c~ mography scanning has not previously been used for evaluating the knee in a functional or closed chain position. MATERIALS AND METHODS Population and Sample Sixteen knees were studied in 14 women and two men. Lateral patellar compression syndrome was diagnosed in these subjects by orthopaedic surgeons after clinical and radiological evaluation. Inclusion in the study required the presence of patellofemoral pain for a minimum of 6 months and evidence of patellar tilt on axial view radiographs. There were no restrictions on age or gender. The average age of the group was 23 years, ranging from 14 to 37 years. Most of the following symptoms were demonstrated during the physical examination: vastus medialis oblique atrophy, reduced transverse play, pain with patellofemoral joint compression in varying degrees of flexion, patellar crepitus, medial or lateral patellar facet tenderness, increased Q angle, positive apprehension test, tightness in the lateral retinaculum and iliotibial band, and excessive lateral patellar tracking with quadriceps contraction and throughout the range of knee motion. Design The dependent variable used to quantify patellar tracking was congruence angle. This was evaluated in a 3 X 5 factorial with three muscle conditions and five knee angles. Each subject was tested in three muscle conditions: with lower extremity relaxed, holding an open chain, and closed chain exercise position. Knee angles were evaluated in the fully extended position and in 10, 20, 30, and 40" of knee flexion because this is the range in which most activities are performed. Subjects were in a supine position. Open chain exercise consisting of a quadriceps set was performed at full knee extension and at 10, 20, 30, and 40" of knee flexion with 3 kg of resistance placed on the ankle. The closed chain condition was obtained by using the positioning device, illustrated in Figure 1, in which resistance cords attach to the main frame at one end and to a footplate at the other end. Subjects were instructed to push the foot-plate into the stabilization bar. Sized rolls were placed under the knee to hold the 10-40" knee angle positions. Goniometric knee range of motion readings were taken for proper knee placement. A pilot study was performed to determine the amount of resistance used in the closed chain condition. Ten subject's knees (five women and five men) with patellar tracking dysfunction were placed in the positioning device at 0 and 20" of knee flexion with 18 and 32 kg of resistance. Congruence angle, measured using computed tomography, was used to evaluate patellar tracking. There were no significant differences in patellar tracking with 18 and 32 kg resistance at 0 and at 20" of knee flexion. Eighteen kg of resistance were utilized since it was easier for subjects to hold this position. JOSPT Volume 23 Number 2 February 1996

*- ------..---- ------a-.-- - - a paired t test analysis, there was no significant difference between the two sets of measurements (mean difference = 0.2"). The reproducibility of the scan itself was also evaluated by rescanning two patients 1 month following the initial scans at two knee angles and comparing sulcus angles. No significant difference in sulcus angle measurements was observed based on a paired t test (mean difference = lo). RESULTS FIGURE 1. Positioning device used to produce closed chain muscle contractions. Radiology Plain axial radiographs cannot image the patellofemoral joint at less than 30" of knee flexion (17). Since the first 30" of knee flexion is where most instability occurs, imaging over this range should best display relevant tracking abnormalities (1 9). Computed tomography is a sensitive and accurate method for evaluating patellar maltracking (18). Imaging Images were created using a Picker 2000 (Picker International Inc., Cleveland, OH) computed tomography scanner. A single 5-mm thick axial cut was obtained through the middle of the patella for each of the five knee angles in each of the three muscle conditions. Using the positioning device, it was found that there was no need to vary the angle of the scanner gantry from vertical in order to create cuts perpendicular to the midpatellofemoral joint at each position. Since highly detailed images were not required for evaluating patellar maltracking, a relatively low dose technique (100 rna, 1.5 second scan time, 80 kv) was used in order to minimize the radiation dose. Irradiation to the gonadal area was mini- mized by placing a lead apron around the torso between the X-ray source. Consent of the patient to undergo examination by computed t e mography was obtained after the nature of the study was fully explained. Female patients were excluded if there was any possibility of pregnancy. Measurements Merchant's congruence angle was used to measure patellar tracking. All measurements were made by a radiologist using the electronic calipers of the monitor of the scanner. Merchant's congruence angle is obtained by bisecting the sulcus angle to establish a zero reference line. A line is then drawn from the apex of the sulcus angle to the lowest point on the articular ridge of the patella. These two lines create the congruence angle. Values medial to the zero reference line are designated as negative and those lateral are positive as shown in Figure 2 (14). Merchant et al (14) found no significant difference when analyzing for sex, age, and side. To examine the reproducibility of these measurements. 13 congruence angles, randomly picked from the data, were remeasured. Based on A split block, randomized block design consisting of 16 subjects (blocks) and each observed under three muscle conditions (whole plot treatment) was examined using an analysis of variance procedure. Under each condition, observations were made at five knee angles (subplot treatments) over the range of 0-40" at 10" increments. For all tests, alpha was set at 0.05. Analysis of variance demonstrated significant differences among the muscle conditions with a p value of <.0001 and significant differences among the knee angles with a p value of <.0001. Multiple mean comparisons tests were used to determine differences between muscle conditions and knee angles. The Fisher's Protected Least Significant Difference and Scheffe Congruence angle FIGURE 2. Merchant's congruence angle. Volume 25 Number 2 Febn~ay 1996 JOSPT

RESEARCH STUDY Lateral Z Medial -e- relaxed --a - closed chain - + - open chain 0 5 10 15 20 25 30 35 40 Knee angle (degrees) FIGURE 3. The relationship between muscle condition and knee joint angle as measured by congruence angle. tests demonstrated significant differences at 0, 10, and 20" of knee flexion between closed chain and open chain conditions and also between the relaxed and open chain conditions. With the knee fully extended, the congruence angle of the relaxed (mean = 34") and closed chain (mean = 37") conditions were significantly less than the open chain (mean = 52") condition. This represents tracking for the open chain condition. At 10" of knee flexion, the relaxed (mean = 29") and closed chain (mean = 31") conditions were again significantly less than the open chain (mean = 45") condition, representing tracking for the open chain condition. At 20" of knee flexion, the relaxed (mean = 23") and closed chain (mean = 21") were significantly less than the open chain (mean = 36") condition, again representing tracking for the open chain condition. At 30 and 40" of knee flexion, no significant differences were seen in congruence angle among the three muscle conditions (Figures 3 and 4). The Fisher's Protected Least Significant Difference and Scheffe tests also determined differences between JOSFT Volume 23 Number 2 February 1996 knee angles and the muscle condition. Closed chain, open chain, and relaxed conditions demonstrated progressively improved patellar congruence from 0 to 40" of knee flexion. For closed chain, open chain, and the relaxed conditions, the 0" knee angle demonstrated significantly more lateral patellar tracking than 20, 30, and 40". In the closed chain, open chain, and the relaxed conditions, the 10" knee angle demonstrated tracking than 30 and 40". For open chain and the relaxed conditions, the 20" knee angle demonstrated tracking than 30 and 40". In the closed chain condition, the knee angle of 20" demonstrated significantly more lateral patellar tracking than the 40" angle. DISCUSSION Our results are in agreement with those of Koh et a1 (12) and van Kampen and Huiskes (24) and indicate that quadriceps contraction has less influence on patellar tracking at 30" knee flexion than at 0" flexion. This likely represents the increased stability of the patella as it moves into the intercondylar groove with increased knee flexion. Flexion beyond 30" increases tension by means of the extensor mechanism, drawing the patella deeper into the trochlear groove and causing the subluxing patella to return to a normal position (19). Closed chain, open chain, and relaxed conditions demonstrated progressively improved patellar congruence from 0 to 40" of knee flexion. Our measurements of congruence angle indicate significantly more lateral patellar tilt and glide for open chain vs. the closed chain and relaxed conditions from 0 to 20" of knee range of motion. Koh et al (12) demonstrated that at 0" of knee flexion, stimulation of the vastus medialis oblique produced medial patellar rotation, tilt, and glide while isometric quadriceps contraction produced lateral patellar rotation, tilt, and glide. They felt knee extensors pull the patella laterally (primarily due to the vastus lateralis) and act to counter and dominate the medial rotation, tilt, and glide produced by the vastus medialis oblique. In open chain exercises, one can isolate the contraction of the quadriceps, allow- FIGURE 4. Axial view radiographic comparison of the patellofemoral joint in one subject with relaxed (leh), closed chain (center), and open chain (right) muscle conditions with the knee in full extension.

ing the strongest component of the muscle to be the major determinant of patellar tracking direction. The closed chain condition (utilizing synergistic muscle activity) and the relaxed muscle condition demonstrated significantly improved patellar congruence than the open chain condition (quadriceps muscle isolation) at 0, 10, and 20" of knee flexion. Open chain strengthening may be most beneficial after SO0 of me tion, the range when no tracking differences in open chain and closed chain conditions were seen. Steinkamp et al (21) mathematically demonstrated that patellofemoral joint reaction force and patellofemoral joint stress were significantly greater in leg press (closed chain) exercise than leg extension exercise (open chain) at 60 and 90" of knee flexion. Brownstein et al (3) demonstrated that vastus medialis oblique electromyographic activity was greatest between 60 to 90" arc of motion and lowest in the 0-30" arc of motion. Patellar tracking does not ap pear to be affected by exercise type at knee flexion greater than 30". although there may be less patellofemoral stress and more vastus medialis oblique electromyographic activity with open chain activities when the knee is flexed from a 60 to 90" angle. The knee range of motion from 0 to 40" is where most functional activities are performed and is also where the patella is least stable, so exercise type is important during this range. The patellar "play" that is evident during the first part of flexion may be explained by the fact that the patella is not yet confined to its femoral groove and is affected by tibial rotations through tensioning of the patellar ligament and the lateral and medial retinacululae (24). The type of muscle condition may also be a factor in patellar position. Patellofemoral joint reaction force and patellofemoral joint stress were mathematically demonstrated to be signifi- cantly greater in leg extension exercise (open chain) than leg press (closed chain) exercise at 0 and 30" of knee flexion (21). Steinkamp et al concluded that patients with patellofemoral joint arthritis may tolerate rehabilitation with leg press better The knee range of motion from 0 to 40" is where most functional activities are performed and is also where the patella is least stable. than with leg extension in functional ranges of motion because of lower patellofemoral joint stresses. From 0 to 20" of knee flexion in closed chain exercise, there is less lateral patellar tracking and less patellofemoral stress. Our results demonstrated significant differences in patellar tracking in open chain vs. closed chain muscle conditions from 0 to 20, possibly due in part to differences in internal rotation of the tibia during open and closed chain conditions. van Kampen and Huiskes (24) found that the patella displays complex but consistent threedimensional motion patterns during flexion, including flexion, rotation, tilt, and a lateral glide relative to the femur, and that the motion patterns are very much affected by tibial rotations accompanying knee flexion. Tibial rotations highly influence patellar tilt and glide in the first part of knee flexion, and patellar re tation is more common in the second part of knee flexion. They found that patellar glide is coupled to the tilt and that internal and external tibial rotation of up to 25" of knee flexion strongly influenced patellar tracking. In closed chain activity, the synchronous actions of the knee and subtalar joint are interdependent m e tions and the rotation of the lower leg is an obligatory action that is necessary for normal kinematics of both joints (22). When the knee flexes in a closed chain position, internal rotation of the tibia and femur occur as well as subtalar joint pronation. As the knee flexes in open chain exercise, less tibial rotation occurs and no compensations are seen at the foot or thigh. This study found that with closed chain activities there is less lateral patellar tracking during the first 0-20" of knee flexion than with open chain exercise, possibly due to the effects of internal tibial and femoral rotation and synergistic muscular activity of the lower extremity (Fig ure 5). We feel that in functional ranges of motion, closed chain exercise allows for better patellar positioning and less joint imtation when performing activities to strengthen the vastus medialis oblique in patients with lateral patellar compression syndrome. Pain and knee joint effusion are contraindications to exercise since they cause reflex inhibition of the quadriceps (20). All three muscle conditions demonstrated progressively improved patellar congruence from 0 to 40" of knee flexion. Initial strengthening activities with the knee in 40" of flexion will allow the patella to be more centrally positioned if strengthening throughout the knee range of motion is not tolerated. Nonweighted progression to partially weighted closed chain activities may be indicated in the initial stages of rehabilitation to allow pain-free strengthening of the lower extremity with the patella most appropriately positioned through a functional range of motion. It is also important to consider secondary joints for muscular imbalances (ie., subtalar joint supinator weakness, iliotibial band tightness, etc.) and all planes of m e tion (sagittal, frontal, and transverse). Open chain strengthening techniques 108 Volume 23 Number 2 Febn~aq 1996 JOSPT

should be avoided during the first 30" of knee range of motion unless used in conjunction with electromye graphic biofeedback techniques to ensure a proper one-to-one relationship of vastus medialis oblique to vastus lateralis. Open chain strengthening techniques appear to be most appropriate after the patella centralizes in the trochlea after 30" of knee flexion. Functional activity is composed of both open and closed chain components and proper usage of each is important in the rehabilitation of the patellofemoral pain patient. Brown and Bradley (2) and Shellock et al (16) indicate that it is important to perform kinematic magnetic resonance imaging knee studies with a quadriceps loading device because this increases the sensitivity of detection of patellar tracking abnormalities. The loading devices that these researchers used produced open chain muscle conditions. We found that during 0-20" of knee flexion there was significantly more lateralization with the open chain condition. Since there was no significant differences between closed chain and relaxed conditions, the relaxed condition may be more indicative of function or a true condition than an open chain condition. We feel that there could possibly be false positive subluxators radiologically with open chain contractions due to the lack of synergistic muscle contraction and lower extremity biomechanical adap tations. Further research is warranted in this area before radiologists recommend surgical procedures to the orthopaedist based on open chain muscle condition findings. CONCLUSIONS FIGURE 5. Axial view radiographic comparison ofthe patellofemoral joint in one subject with open chain muscle condition (on lehj and closed chain muscle condition (on right) from top to bottom: 0, 10,20, 30, and 40" of knee flexion. The effect of open and closed chain exercise and knee joint position on patellar tracking in lateral patellar compression syndrome patients was investigated using com- JOSPT e Volume 23 e Number 2 e Februaq 1996

-. puted tomography. It was found that closed chain exercise allows for better patellar tracking than open chain exercise during full knee extension and through 20" of knee flexion. This may be due to the effects of internal tibia1 and femoral rotation and synergistic muscular activity of the lower extremity. Open chain strengthening techniques appear to be most appropriate after 30" of knee flexion. Patellar congruence improved progres sively from full knee extension to 40" of knee flexion in all three muscle conditions (closed chain, open chain, and relaxed). These results provide a ratie nale for designing an exercise protocol for patellofemoral pain patients. JOSm ACKNOWLEDGMENTS The authors express their appreciation to Ron Rhees and Stacie Parker for computed tomography scanning; Brian Balls, Jeff Ohwiler, Roger Olson, Tom Rees, and William Doucette, PhD for technical assistance, Don Sisson, PhD for statistical advice, and Mountain West Physical Therapy Inc. for financial assistance. REFERENCES 1. Antich TJ, Brewster CE: Modification of quadriceps femoris muscle exercises during knee rehabilitation. Phys Ther 66~1246-1251, 1986 2. Brown SM, Bradley WG Jr: Kinematic magnetic resonance imaging of the knee. MRI Clin North Am 2(3):441-449, 1994 3. Brownstein BA, Lamb RL, Mangine RE: Quadriceps torque and integrated electromyography. 1 Orthop Sports Phys Ther 6:309-3 14, 1985 4. Brunet ME, Stewart GH: Patellofemoral rehabilitation. Clin Sports Med 8:3 19-329, 1989 5. Dehaven KE, Dolan WA, Mayer PJ: Chondromalacia patellae in athletes: Clinical presentation and conservation management. Am J Sports Med 75-1 1, 1979 6. Doucette SA, Goble EM: The effect of exercise on patellar tracking in lateral patellar compression syndrome. Am J Sports Med 20:434-440, 1992 7. Grabiner M, Koh T, Miller G: Fatigue rates of vastus medialis oblique and vastus lateralis during static and dynamic knee extension. J Orthop Res 9:391-397, 1991 8. Gray G W: Successful strategies for closed chain and open chain testing and rehabilitation. Presented at the Chain Reaction Plus Course, San Diego, CA, Sept 23-25, 1994 9. Hungerford DS, Barry M: Biomechanics of the patellofemoral joint. Clin Orthop 144:9-15, 1979 10. lngersol CD, Knight KL: Patellar location changes following EMG biofeedback or progressive resistive exercises. Med Sci Sports Exerc 23:1122-1127, 1991 1 1. Knight KL: Rehabilitating chondromalacia patellae. Ph ys Sportsmed 7: 147-148, 1979 12. Koh TJ, Grabiner MD, DeSwart RJ: In vivo tracking of the human patella. J Biomech 25(6):637-643, 199 1 13. Kramer PC: Patella malalignment syndrome: Rationale to reduce excessive lateral pressure. J Orthop Sports Phys Ther 8:30 1-309, 1 986 14. Merchant AC, Mercer RL, Jacobsen RH, Cool CR: Roentgenographic analysis of patellofemoral congruence angles. / Bone Joint Surg 56A:l39 1-1 396, 1974 15. Pevsner DN, Johnson JR, Blazina ME: The patellofemoral joint and its implications in the rehabilitation of the knee. Phys Ther 59(7):869-874, 1979 16. Shellock FG, Foo TK, Deutsch AL, Mink JH: Patellofemoral joint: Evaluation during active flexion with ultrafast spoiled GRASS MR imaging. Radiology l80:581-585, 1991 17. Shellock FG, Mink JH, Deutsch AL: Patellofemoral joint: Identification of abnormalities with active-movement, "unloaded" versus "loaded" kinematic MR imaging. Radiology 188575-578, 1993 18. Shellock FG, Mink JH, Deutsch AL, Fox JM: Patellar tracking abnormalities: Clinical experience with kinematic MR imaging in 130 patients. Radiology 172:799-804, 1989 19. Shellock FG, Mink JH, Fox JM: Patellofemoral joint: Kinematic MR imaging to assess tracking abnormalities. Radiolog y 168551-553, 1988 20. Spencer JD, Hayes KC, Alexander I/: Knee joint effusion and quadriceps reflex inhibition in man. Arch Phys Med Rehabil65: 171-1 77, 1984 2 1. Steinkamp LA, Dillingham MF, Markel MD, Hill ]A, Kaufman KR: Biomechanical considerations in patellofemoral joint rehabilitation. Am 1 Sports Med 2 1 (3):438-444, 1993 22. Tiberio D: The effect of excessive subtalar joint pronation on patellofemoral mechanics: A theoretical model. Orthop Sports Ph ys Ther 9(4): 160-165, 1987 23. Tria A], Palumbo RC, Alicea ]A: Conservative care for patellofemoral care. Orthop Clin North Am 23(4):545-554, 1992 24. van Kampen A, Huiskes R: The threedimensional tracking pattern of the human patella. 1 Orthop Res 8:372-382, 1990 25. Woodall W, Welch J: A biomechanical basis for rehabilitation programs involving the patellofemoral joint. / Orthop Sports Phys Ther 1 1(11):535-542, 1990 Volume 23 Number 2 February 1996 JOSF'T