Journal of Orthopaedic & Sports Physical Therapy

Official Publication of the Orthopaedic and Sports Physical Therapy Sections of the American Physical Therapy Association

The Effects of 16 Months of Angle-Specific Isometric Strengthening Exercises in Midrange on Torque of the nee-~xtensor Muscles Osteoarthritis of the Knee: A Case Study
Ray Marks, MScPT, BScPT,

BSc'

ecause walking and stairclimbing deficiencies in painful osteoarthritis of the knee a r e attributable, in part, t o static weakness of the surrounding quadriceps muscles (1 l,23), open kinetic chain (tibia free) isometric quadriceps strengthening exercises are often advocated as efficacious for the treatment of this condition (22,24). However, based on studies on healthy populations (9,21), it could be argued that if performed as is often recommended at the angle of maximal weakness (3) o r in the least painful position ( 1 5). improvements in quadriceps efficiency might be limited solely t o those occurring at that angle. T h e fact that functions such as walking and stairclimbing require considerable quadriceps force generation over a large range of motion (1 7) suggests that without a transfer of training t o angles other than the training angle, functional improvements sought most by patients with knee osteoarthritis, namely walking and stairclimbing (2), might not occur. T h e objectives of this study were t o examine the transfer of training effect associated with isometric strength training of the extensor muscles surrounding an osteoarthritic knee at 60" knee flexion on knee
JOSPT Volume 20 Number 2 August 1994

Although often advocated for arthritis rehabilitation, several studies conducted on healthy subjects have demonstrated conflicting results as to whether isometric strengthening exercises performed at one joint angle can result in strength gains at other angles. The objective of this study was to determine whether midrange strength training of the quadriceps surrounding an osteoarthritic knee would increase strength in this part of the range of motion as well at other knee angles. The midrange position was chosen because the subject, a 57-year-old female with a 2-year history of moderately severe osteoarthritis of the left knee, had greatest weakness in this position. The training occurred three times per week for 16 months using a resistance training program of three sets of two maximal isometric contractions at a knee angle of 60". The dependent variable of maximal isometric knee extensor torque was recorded on an isokinetic dynamometer at knee angles of 9@, 60, and 30" at 6 weeks, I year, and 16 months. These data demonstrated a progressive torque increase at all angles over the 16-month period, with a steeper slope at the training angle. Contrary to the specificity of training concept, angle-specific strengthening in midrange may be sufficient to strengthen the extensors surrounding an osteoarthritic knee through a wide range of motion. Midrange isometric strengthening exercises might thus prove useful in the rehabilitation of patients with knee osteoarthritis who are unable to exercise their weakened quadriceps at other angles due to pain or swelling.

Key Words: osteoarthritis, knee, isometric exercise

' Director, Clinical Research, Osteoarthritis Research Centre, P.O. Box 1153, Adelaide Postal Station, Toronto, Ontario, Canada M5C 2K5; Student Mentor, Division of Physical Therapy, University of Toronto, Toronto, Ontario, Canada. This work was conducted as part ofa doctoral dissertation in the Department of Physical Therapy, University of Alberta, in cooperation with the Faculty of Health Administration. This study was supported by The Arthritis Society, Toronto, Ontario, Canada.
extensor torque a t angles of 30, 60, and 90" knee flexion; t o examine whether the training would result in improvements in walking and stairclimbing; and t o monitor the time course of the adaptation t o the present strengthening stimulus. This study was conducted as part of a larger project designed t o establish the most effective ways of providing physical therapy to improve the strength and function of the patient with knee osteoarthritis.

METHODS
Subject
T h e single subject studied was a healthy 57-year-old woman. (height = 155 cm, weight = 4 5 kg) with clinical and radiologically established

CASE STUDY

primary unilateral medial compartment osteoarthritis of the left knee who was not currently receiving any form of intervention for her condition. At study entry, she was complaining of intermittent knee pain when walking on level ground for about 15 minutes o r longer (pain score of 5.4 on a 10-point visual analog scale) and marked difficulty with stairclimbing and descending more than 8- 10 stairs d u e to pain (pain score 5.9) and subjective feelings of fatigue. She was also experiencing morning stiffness of her knee of less than 30-minutes duration. These symptoms had prevailed for approximately 2 years and were getting worse despite traditional physiotherapy, ie., straight-leg raising exercises and medication for inflammation in the initial stages of her disease. Standard clinical examination revealed crepitus on knee motion, moderate anterolateral instability, a 5" flexion contracture, and hamstring discomfort with passive knee extension. Her knee flexion range was >100, and quadriceps torque at 60" was diminished by approximately 59% on the affected side. In addition, the quadriceps force curve was abnormal on the affected side (Table 1). T h e gravity-corrected isokinetic quadriceps-to-hamstring work ratio on the affected side was a p proximately 2. T h e disease severity was assessed as moderate o r grade 2 by the investigator on a 4-point scale devised by Stauffer et al (23), where 4 = n o symptoms and 1 = very severe disease. T h e study was reviewed by a
Knee Flexion Angles 90'
44

University Hospital Ethics Committee. Prior to her participation in the study, the subject received a written explanation of the study purpose and procedures and informed consent was provided.

Design
T h e research design chosen to meet the study objectives was a single-subject, modified, time-series experiment with three baseline measures conducted biweekly before the start of the exercises and postintervention measures conducted at 6 weeks, 1 year, and 16 months. T h e independent variable for this study

training and for measuring the changes in maximal isometric quadriceps torque postexercise. T h e dynamometer was calibrated electronically before testing a subject, and the reliability and validity of the instrument had been found t o be satisfactory (4). As well, the test-retest reliability of the dynamometer for measuring the extensor torque of women with knee osteoarthritis has been found to be higher than 0.80 for knee angles of 90, 60, and 30" (26).

Testing Procedures
For each of three preexercise control assessments of knee strength conducted at 0, 3, and 5 weeks and for the postexercise strength tests, the testing protocol of Wessel et al (26) was used. That is, the subject sat on a Kin-Com chair with back s u p port, hips angled at 8 0 , and the pelvis and thigh stabilized with straps. T h e axis of the dynamometer arm was aligned with the lateral epicondyle of the test leg and secured by an ankle strap just above the medial malleolus (Figure 1). Immediately following 10 isotonic warm-up contractions of the quadriceps through the available range of motion, the subject performed two submaximal (<50% of maximal capacity) isometric warm-up contractions at angles of 90, 60, and 30, the magnitude of which was determined by restricting the height of the torque tracings visible on the computer screen. After a 1-minute rest, the subject then performed two maximal isometric quadriceps contractions of 3-seconds duration without pause at each of these angles. Compensation was made for limb segment weight, and the minimum force was set at 20 Nm. T h e primary outcome measures of maximal isometric quadriceps torque at each angle were computed from the KinCom screen using software provided by the manufacturer. In addition to the strength measurements, level walking time was reVolume 20 Number 2 August 1994 JOSPT

The training was carried out in midrange to eliminate discomfort experienced when the limb was placed in the more extended position.
was a quadriceps strength training protocol carried out isometrically at a knee angle of 60" flexion. T h e main dependent variables were measurements of maximal isometric quadriceps torque generated at knee angles of 90, 60, and 30". Also assessed were changes in level walking, stairwalking, and torque over time.

Period
Preexercise Postexercise 6 weeks 1 year 16 months

6 0 '
41

30'
44

Instruments
A Kin-Com robotic dynamometer system (Med-Ex Diagnostics, Coquitlam, British Columbia, Canada) capable of generating a visual record of torque measurements generated isometrically at different knee angles was used for the initial 6 weeks of

60 64 78

70 87 93

62
64

68

TABLE 1. Peak torque values (in Nm) recorded for the quadriceps fernoris of one subject with osteoarthritis before and after exercise training.

CASE STUDY

tractions and a 2-minute interval between sets. After the three sets were completed, the subject rested for 5 minutes and then repeated the exercises on the unaffected limb. T o attempt to control for history effects associated with the strength training regimen, the subject stopped exercising for 6 weeks after the initial 6-week exercise training session. After that time, she carried out the same protocol for a further 13 months at home using a belt for resistance. She reported to the labo-

T h e training was carried out in midrange (Figure 1) to eliminate discomfort experienced when the limb was placed in the more extended position and because the subjects' knee extensor torque was lower at this angle than it was at angles of 30 or 90". Both knees were exercised because muscle torque of an osteoarthritic asymptomatic knee may not be as strong as that of healthy subjects (23) and may attenuate more quickly with age in women with unilateral knee involvement (8), contributing to more stress on the affected knee.

FIGURE 1. Subject positioned

trarning.

for testrng and

corded twice at baseline and after training according to Marks (16) over a 13-m indoor carpeted walkway at a self-selected walking speed using a stopwatch. T o control for the effects of acceleration and deceleration, the subject walked a distance of 3 m before and after completion of the walk. After a further I-minute rest, walking time up and down four stairs without pause was recorded. T h e verbal commands, time, and order of testing remained constant throughout the study. T h e total number of stairs that could be negotiated by the subject without pain was assessed subjectively at baseline and after training.

Regarding the transfer of training effect, the present results suggest that high-intensity isometric knee extension exercises performed in midrange by an individual with knee osteoarthritis can contribute toward marked improvements in relative peak torque production.
ratory once per month during this period to ensure that the protocol was being followed rigorously. She was requested to continue her normal daily routine but not to participate in any additional activities o r forms of physical therapy for the study period. It was not possible, however, to control the duration of her regular daily activities as her condition improved, which may have increased accordingly as indicated by nonsolicited verbal feedback from the subject.

Data Analysis
T o describe the changes in peak isometric quadriceps torque recorded over the study period, the values recorded immediately before the start of the exercises (ie., at week 5 of the study) as well as those generated 6 weeks, 1 year, and 16 months after the exercises had been implemented were tabulated. A percentage change score in these values was then obtained by dividing the difference between the preexercise and postexercise scores for each period and for each angular position by the preexercise scores. T o permit visual analysis, the data were plotted graphically. T o substantiate whether quantitative as well as favorable changes in the force curve occurred during the study, given the unknown measurement properties of visually analyzed data, the preexercise and postexercise torque values at 60" were expressed as a ratio of those recorded at 30 and 90" for each respective time period. They were then compared with representative data recorded in the identical manner by a sample of older, healthy women. Differences in level walking and stairwalking time before and after 6 weeks of exercise, for which the intrarater reliability as determined by intraclass correlation coefficients was found to be greater than 0.80 for

Training Protocol
T h e training was initially carried out for 6 weeks on a Kin-Com under supervision using a similar set-up as described for the baseline testing. T h e protocol adopted was that of Marks (15), which included 10 isotonic warm-up contractions of the quadriceps through full range with no resistance, three submaximal isometric contractions of increasing intensity at 60" knee flexion followed by three sets of two maximal, 5-second isometric contractions with an interval of 30 seconds between conJOSPT Volume 20 Number 2 August 1994

CASE STUDY

both tests ( I 6), were expressed as a percentage of baseline values.

RESULTS
T h e peak torque values generated by the quadriceps of the subject's affected knee recorded over the time course of the study at knee flexion angles of 30, 60, and 90" are given in Table 1 . As indicated, these torque values increased markedly at 60" as well as at knee flexion angles of 30 and 90" at 6 weeks when assessed against their stabilized baseline values of 44. 4 I . and 44 Nm. They continued to improve steadilv at the training angle as well as at the other two angles fbr up to 16 months after the start of training. As depicted in Table 2, the improvements in maximal isometric quadriceps torque ranged from 36 t o 70% at 6 weeks, attaining a maximal value of 126% at the end of 16 months with no torque decline recorded at the training angle following 6 weeks of detraining. T h e marked increase in torque at the training angle of 60" coupled with substantial strength gains at the other angles resulted in a favorable change of the quadriceps force curve (Figure 2). As depicted in Table 3. after a year of undergoing the training, relative peak torque values generated by the quadriceps femoris over an angular range of 60" were well within normal limits recorded for the healthy knee. After 1 year of training, the peak torque values generated by the quadriceps of the symptomatic knee were almost iden-

Degrees
FIGURE 2. Graphic representation of peak isometric quadriceps femoris torque values recorded before, during, and alter training at knee angles of 30, 60, and 90' flexion.

Subject
Ratio heexercise 5 Weeks
6 Weeks

Postexercise 1 Year 16 Months

Controls (N = 5 )

TABLE 3. Ratio of isometric quadriceps femoris (QF) peak torque values (PT) (in Nm)generated at 60' and those generated at 30 and 90' by study subject and range of scores of five healthy, older control women, ages 60-72.

tical to, if not slightly higher than, those of the asymptomatic knee (Table 4). As illustrated in Table 5, the torque increments at 6 weeks were accompanied by a 7.5% improve-

ment in level walking time and a similar improvement in stairwalking time. By the end of 16 months, the number of stairs that could be negotiated by the subject without pain o r fatigue had increased by approximately 200%. indicating improved stairwalking capacity.

Postexercise Period 6 weeks 1 year 16 months

90'
36 45 77

Knee flexion Angles 60'

30'
40 45 54

DISCUSSION
Despite their widespread use in the functional rehabilitation of persons with knee osteoarthritis, knowledge is lacking as to whether maximal isometric quadriceps exercises carried out a t a single joint angle a r e efficacious for increasing muscular
Volume 20 Number 2 August 1994 JOSPT

70 117 126
X

'% improvement

change in QF P T relative to baseline QF PT at baseline

100

TABLE 2. Percentage (YO)improvement' in isometric quadriceps femoris (QF) peak torque (PT)at angles of 90, 60, and 30' knee flexion following training.

CASE STUDY

Knee Angle

QF PT

Left Knee

Riaht Knee

TABLE 4. Comparison of isometric quadriceps

femoris (QF) peak torque values (PT) (in Nm) generated at 90,60, and 30' by the quadriceps of the affected (left) and unaffected (right) knees of the patient aher I year of training.

strength at knee angles other than the training angle. More studies are also needed to determine whether isometric resistance exercises can result in functional improvements in walking and stairclimbing in this patient group. It is also necessary t o determine the optimum training duration that might be required to maximize muscle function and functional performance for osteoarthritic patients. Regarding the transfer of training effect, the present results suggest that high-intensity isometric knee extension exercises performed in midrange by an individual with knee osteoarthritis can contribute toward marked improvements in relative peak torque production of the knee extensors at the training angle as well as peak torque improvements within 30" of the training angle. AIthough exercising in the midarticular position was not equally beneficial at angles distant from the training angle, a finding of Rasch and Pierson (1 9) and Thepaut-Mathieu et al (25). the degree of carryover was inconsistent with the specificity of training principle which suggests training effects are specific to the movement pattern selected for training (6.1 2,13,21). Moreover, the

present findings were also inconsistent with the conclusions of Kitai and Sale (9),who found that when isometric training occurred in the midarticular position, a significant increase in maximal torque was confined within 10" of the training angle. While in agreement with the findings of Marks (1 5) and Fisher et al (5). who exercised the knee extensors of men with knee osteoarthritis isometrically at several angles, including midrange, and noted marked improvements in quadriceps strength and diminished walking difficulties, the present results were contrary to those of Luckhurst et al (1 4), who not only found a decrement in extensor torque at 3 weeks in patients with knee osteoarthritis but also recorded n o improvements in walking and stairclimbing following 10 maximal isometric quadriceps contractions carried out daily in full extension for 4 weeks. It is possible, however, that the latter results were attributable t o reflex inhibition of the quadriceps, which could occur if an inflamed o r painful osteoarthritic knee is exercised with the leg in an extended position (1 5-27). Moreover, the intensity and frequency of the training applied by Luckhurst et al (14) may have been excessive, given the fact that the knee extensors of persons with osteoarthritis may undergo extensive pathological changes (7), thereby undermining recovery and the training response. It is also possible that if strengthening is restricted t o a position other than the one at which the joints usually develop maximal torque, results might be less than optimal. In addition, because

Test
Level walking Stair walk in^
'% change =

Preexercise
7.9

Postexercise
7.3
4.8

% Change
7.5

5.2

7.6

preexerose-poctexercise test scores x 100 preexerclse test score

TABLE 5. Walking time (in seconds) before and after 6 weeks of training.

the shorter the muscle length at which training has been carried out, the more the gain is restricted t o the training angle (25). the degree to which a patient is then able to use the quadriceps in daily life once strength training is initiated may be undermined bv exercises conducted in extension. T h e present finding that knee extensor torque improved under all test conditions over the 16-month study period, irrespective of whether they were carried out in the laboratory o r at home, was also contrary to findings of Kreindler et al (1 0). These investigators reported inconsistent improvements in patients with osteoarthritis on a 6-week home program of either "traditional" progressive resistive exercises with o r without a progressive Kinetron program performed with the knee flexed to 30-40" with no further increase in strength during a 6-week home treatment phase. This was not the finding of Quirk et al (1 8),however, who taught patients with osteoarthritis a standardized exercise regimen in the inner range, which patients performed twice daily at home with steadv improvements in overall clinical condition u p t o 6 months after treatment. Likewise. Anderson (1 ) found progressive strength improvements of the knee extensors over a 12-month period, which were accompanied by improvements in walking capacity. Thus, it may be argued that the specificity of training concept, often attributed to neural mechanisms (9.13), may not apply in advanced training. which has the potential t o induce muscle hypertrophy. In support of this, Sale and MacDougall (2 1) suggested that any muscular adaptation, such as hypertrophy, would tend to increase the force-generating capacity at all muscle lengths (joint angles). Although no estimate was made in the present case to estimate hypertrophy. Anderson ( 1) reported increases in thigh girth for two pa-

JOSPT* \'olume 20

Number 2 * August 1994

CASE STUDY

tients with arthritic joint damage who exercised their extensors maximally a t a relatively short muscle length for about one year. In addition t o improvements over a wide range of angles that might therefore occur with prolonged training, the present findings of marked improvements in stairclimbing capacity after 16 months suggest that when an unstable osteoarthritic knee is exercised with the leg positioned comfortably with the knee extensors close t o their o p timal length (ie.. between 50-70"). the adaptive response of the muscle may be maximized. T h e author recognizes, nonetheless, that the present study used repeated isometric tests a t multiple angles as an outcome measure and that these may well have provided an additional training effect over and above the midarticular resistance training regimen. Additionally, the order of testing was somewhat limited by the capabilities of the KinCom dynamometer and may have affected the force curve in some way. Likewise, improved strength following the initiation of training may have been partially due t o improved use of the leg during daily activities, which provided a training effect. However, since the strength tests were conducted relatively infrequently and only two 3-second contractions were performed at angles other than the training angle at each, the likelihood that a learning effect (20) influenced the study results appears remote. Since the shape of the force curve in the last angle of measurement remained unchanged, it did not appear that fatigue effects due t o the order of testing had influenced the study outcome. T o the extent that knee extensor torque improved maximally as expected at the maximal point of mechanical tension generation, the present torque improvement results appeared strongly related t o the specific mode of training used. Given the present results, it is

the author's view that strengthening the quadriceps isometrically in midrange three times a week for 6 weeks under supervision may substantially enhance the physical performance of an osteoarthritic patient with weak knee extensors who cannot perform isotonic exercises o r inner range knee extension exercises due t o pain. However, to maximize the treatment effect, the exercises should probably be continued for prolonged periods at home. Since it is anticipated that as the patient's condition improves, the exercises could then be performed throughout a greater range of motion, regular follow-up of patients and their progress is recommended.

fects and t o validate the present observations is strongly recommended.

JOSPT

ACKNOWLEDGMENTS
T h e author thanks the volunteers from T h e Arthritis Society of Edmonton for their assistance in this research; Dr. J. Wessel, University of Alberta, for the use of her laboratory; and Dr. C.B. Hazlett, University of Alberta, for his contribution toward this undertaking.

REFERENCES
7. Anderson TP: Management of degenerative joint disease of the knee. Arch Phys Med Rehabil36: 754- 159, 1955 2. Bellamy N, Buchanan WW: A preliminary evaluation of the dimensionality and clinical importance of pain and disability in osteoarthritis of the hip and knee. Clin Rheumatol 5(2):23 7247, 1986 3. Bender /A, Kaplan HM: The effectiveness of isometric testing as a diagnostic aid: A hospital study. I Assoc Phys Ment Rehabil 16:737-139, 7962 4. Farrell M, Richards IC: Analysis of the reliability and the validity of the kinetic communicator exercise device. Med Sci Sports Exerc 18:44-49, 1986 5. Fisher NM, Pendergast CR, Gresham CE, Calkins E: Muscle rehabilitation: Its effect on muscular and functional performance of patients with knee osteoarthritis. Arch Phys Med Rehabil 72:367-374, 1991 6. Cardner CW: Specificity of strength changes of the exercised and non-exercised limb following isometric training. Res Q Exerc Sport 34:98-707, 7 963 7. Glasberg MR, Clasberg lR, /ones RE: Muscle pathology in total knee replacements for severe osteoarthritis: A histochemical and morphometric study. Henry Ford Hosp Med / 34(1):37-40, 1986 8. Ian M, Lai 1, Tsauo 1, Lien I: lsokinetic study of muscle strength in osteoarthritic knees of females. Taiwan I Hsueh Hui Tsa Chih 89:873-879, 1990 9. Kitai TA, Sale DC: Specificity of joint angle in isometric training. Eur 1 Appl Physiol58:744-748, 7989 10. Kreindler H, Lewis CB, Rush 5, Schaefer K: Effects of three exercise protocols on strength of persons with osteoarthritis of the knee. Top Ceriatr Rehabil 4(3):32-39, 1989
Volume 20 Number 2 August 1994 JOSPT

SUMMARY
This study supported the view that quadriceps strengthening exercises performed in midrange may be useful for the treatment of patients with knee osteoarthritis who cannot exercise in full extension due to pain o r osteoarthritic patients who have specific weakness of their quadriceps in midrange. T h e results also demonstrated that these open kinetic chain strengthening exercises may improve lower limb function, as measured by walking and stairclimbing tests. T h e results also suggest that isometric quadriceps strengthening seems efficacious on a long-term basis. It is recognized, however, that the internal and the external validity of a single subject study might be compromised d u e to the possibility of selection bias and the fact that the treatment outcome would not necessarily be typical of an unbiased sample. Therefore, the present results may not be generalizable t o all persons with knee osteoarthritis. Additionally, although withdrawing the exercises appeared t o rule out a placebo attention effect on the present study's outcome, further investigation of other individuals with knee osteoarthritis to rule out history ef-

CASE STUDY

1 1 . Lankhorst GI, Van de Stadt RI, Van der

12.

13.

14.

15.

Korst IK: The relationship of functional capacity, pain, isometric and isokinetic torque in osteoarthritis of the knee. Scand I Rehabil Med 17:167-172, 1985 Lindh M: Increase of muscle strength from isometric quadriceps exercises at different knee angles. Scand I Rehabil Med 1 1 (1):33-36, 1979 Logan CA: Differential applications of resistance and resulting strength measured at varying degrees of knee extension. Diss Abs Int 20:4027-403 1, 1960 Luckhurst B, Peppiat I, Reynolds Wl: The response of the quadriceps muscle to an isometric strength training program in rheumatoid arthritis. In: Proceedings of the World Confederation for Physical Therapy, Seventh International Congress, Montreal, Canada, lune 1974, pp 244-249. London: World Confederation for Physical Therapy, 1974 Marks R: The effect of isometric quadriceps strength training in mid-range

16.

17.

18.

19.

20.

2 1.

for osteoarthritis of the knee. Arthritis Care Res 6(1):52-56, 1993 Marks R: Reproducibility of measures of walking performance variables in persons with osteoarthritis of the knee. Physiother Can, 1994 (in press) Perry I, Antonneli D, Ford W: Analysis of knee-joint forces during flexed knee stance. I Bone loint Surg 57A(7):961967, 1975 Quirk AS, Newman Rl, Newman K/: An evaluation of interferential therapy, shortwave diathermy and exercise in the treatment of osteoarthritis of the knee. Physiother 71(2):55-57, 1985 Rasch PI, Pierson WR: One position versus multiple positions in isometric exercise. Am I Phys Med Rehabil 43:lO-12, 1964 Rutherford O M , lones DA: The role of learning and coordination in strength training. Eur I Appl Physiol 55:100105, 1 986 Sale D, MacDougall D: Specificity in strength training: A review for the coach and athlete. Can / Appl Sport

Sci 6:87-92, 198 1 22. Semble EL, Loeser RF, Wise CM: Therapeutic exercise for rheumatoid and osteoarthritis. Semin Arthritis Rheum l8(1):5 1-56, 1989 23. Stauffer RN, Chao EYS, Cyory AN: Biomechanical gait analysis of the diseased joint. Clin Orthop 126:246-255, 1977 24. Swedberg /A, Steinbauer IR: Osteoarthritis. Am Fam Physician 45:557-568, 1992 25. Thepaut-Mathieu C, Van Hoeke I, Maton B: Myoelectric and mechanical changes linked to length specificity during isometric training. 1 Appl Physi01 64(4):1500- 1505, 1988 26. Wessel I, Baron D, Baergen I: Test retest reliability of measurements of knee extensor torque of women with osteoarthritis. Physiother Can 43(Suppl): 10, 199 1 (abstract) 27. Young A, Stokes M, lles IF: Effects of joint pathology on muscle. Clin Orthop 219:21-27, 1987

JOSPT Volume 20 Number 2

August 1994