THE EFFECT OF FUNCTIONAL KNEE BRACING ON MEDIAL COMPARTMENT OSTEOARTHRITIS by Heather L. Biagi B.P.E., University of British Columbia, 1991 A THESIS SUBMITTED IN PARTIAL FULLFILLMENT OF THE REQUIREMENTS FOR T H E D E G R E E OF MASTER OF SCIENCE in T H E FACULTY OF GRADUATE STUDIES School of Human Kinetics We accept this thesis as conforming to the required standard: T H E UNIVERSITY OF BRITISH COLUMBIA September 1999 © Heather L. Biagi, 1999 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada Date DE-6 (2/88) ABSTRACT The purpose of this study was to examine the clinical effects of the Patient-Ready Monarch Knee Brace (Smith and Nephew - Donjoy) in patients with medial osteoarthritis (OA) of the knee. Using a repeated measures cross over study design, ten subjects, eight male and two female, (mean values: age = 51.3 ± 8.63; weight = 185.2 ± 22.9 lbs., height = 69.5 ± 3.7 cm.) participated in the study. Subjects wore the knee brace for six weeks in each a neutral and valgus orientation with a three-week washout between trials. Clinical measurements included four functional tasks: timed hgure-8-run, timed 6-stair climb, incline squat, single leg hop. Each task was measured every three weeks. In addition, subjects completed the Western Ontario McMaster University (WOMAC) osteoarthritis index daily during the fifteen weeks. The WOMAC index is an OA specific quality of life questionnaire consisting of three subscales (pain, stiffness, and physical function). Statistical significance was set a priori at p< 0.05. Repeated measures Analysis of Variance (ANOVA) of the figure-8-run, incline squat, and single leg hop functional tasks, measured at testing session 1,3 and 6, indicated no significant difference between the no brace, neutral and valgus brace conditions (p>0.05). There was a significant improvement in the time to stair climb when subjects wore a brace compared to no brace (p=. 01); however, there was no difference between valgus and neutral brace conditions (p=.21). The three subscales of the WOMAC index were analyzed separately using repeated measures ANOVA. Results indicate that the perception of pain and stiffness did not significantly change between brace conditions (p=. 43 and p=. 16 respectively). There was a significant difference between conditions found in the physical function subscale (p=. 02). Post hoc analysis indicated that subjects perceived their physical function improved when wearing a brace compared to not wearing a brace. Conclusion: Patients who have medial knee O A may perceive a significant improvement in performing physical tasks while wearing a knee brace. Moreover, they are able to ascend and descend stairs quicker while wearing a brace. Results must be interpreted with caution due to the small sample size used in this study. iii TABLE OF CONTENTS ABSTRACT ii ACKNOWLEDGEMENTS VIM TABLE OF CONTENTS iv LIST OF TABLES vi LIST OF FIGURES vii CHAPTER 1: Introduction 1 1.1 Statement of the Problem 2 1.2 Hypotheses 3 1.3 Limitations 4 CHAPTER 2: Methods 6 2.1 Study Protocol 6 2.2 The Knee Brace 7 2.3 Subjects 9 2.4 Functional Tasks 10 Stair Climb : 10 Figure - 8 - Run 10 Incline Squat 11 Single Leg Hop 12 2.5 Western Ontario & McMaster (WOMAC) Pain Questionnaire 13 2.6 Statistical Analysis 14 CHAPTER 3: Results 15 3.1 Descriptive Data 15 3.2 Compliance 15 3.3 Testing Timeline 16 3.4 Functional tasks 17 Stair climb 18 Figure-8-run 19 Incline Squat. 20 Single Leg Hop , 21 3.5 WOMAC data 22 Pain 23 i v Stiffness 24 Physical Function 25 CHAPTER 4: Discussion 26 4.1 Functional Tasks 28 Stair Climbing 28 Figure-8-Run 30 Incline Squat 31 Single Leg Hop 3 2 4.2 WOMACData 34 Pain and Stiffness 34 Physical Function 34 4.3 Limitations 37 REFERENCES 42 APPENDIX A: Review Of Literature 49 The Normal Knee Joint 49 Epidemiology & Risk Factors 50 Structural changes seen with OA 52 Classification/Diagnosis 54 Pain Measurements 57 Treatment of OA of the Knee 61 Osteoarthritis and Bracing 64 Functional Knee Evaluation 68 Summary 72 APPENDIX B: Raw Data 73 Figure-8-run Data 73 Stair Climb Raw Data 73 Incline Squat Raw Data 74 Single Leg Hop Raw Data 75 WOMAC Raw Data - Pain Subscale 76 WO MAC Raw Data - Stiffness Subscale 78 WOMAC Raw Data - Physical Function Subscale 80 APPENDIX C: WOMAC Index 82 v. LIST O F T A B L E S Table 1. Study Design Timetable 6 Table 2. Subject Exclusion Criteria 9 Table 3. Physical Characteristics of Subjects 15 Table 4. Testing Timeline 16 Table 5. Carry-o ver Effect Analysis 17 Table 6. Functional test data 18 Table 7. Mean da ta for pain subscale 23 Table 8. Mean data for stiffness subscale 24 Table 9. Mean data for physical function subscale 25 Table 10. Classification Criteria for Knee Arthritis 55 Table 11. Summary of research using Single Hop Test (for distance) 70 Table 12. Summary of research using stair climbing. 71 Table 13. Time (seconds), Individual subject data 73 Table 14. Time (seconds), individual subject data 73 Table 15. Number of time in 20 s, individual subject data 74 Table 16. Distance (centimeters), individual subject data 75 Table 17. Numerical score (max=500), individual raw data 76 Table 18. Numerical score (max=200), individual raw data 78 Table 19. Numerical score (max=1700), individual raw data ....80 vi LIST O F F I G U R E S Figure 1. Comparison of the means for ascending/descending stairs 19 Figure 2. Comparison of the means for the figure-8-run 20 Figure 3. Comparison of the means for the incline squat 21 Figure 4. Comparison of the means for the single leg hop index. 21 Figure 5. Plotting of the valgus and neutral brace means for pain subscale 23 Figure 6. Plotting of the means for stiffness subscale 24 Figure 7. Plotting of the means for physical function subscale 25 vii. A C K N O W L E D G E M E N T S I w o u l d l ike to thank m y advisors: Dr . D o n M c K e n z i e for his expert direction, guidance and abi l i ty to keep me focussed th rough this thesis and Dr . D o u g Clement for his w i s d o m and encouragement th rough the ear ly stages. I w o u l d also l ike to extend m y appreciat ion to m y committee members, D r . Jack Taun ton and D r . Ross D a v i d s o n , for their cont r ibut ion and direct ion. I am grateful to Smi th & N e p h e w -Donjoy for their suppor t of this project, their wi l l ingness to supp ly the knee braces and their technical support . M y heartfelt thanks goes to Yvette Jones - I cou ldn ' t have done this thesis w i thou t her assistance and dedicat ion to this project. I w o u l d also l ike to thank K a t h r y n D u f f for her expert statistical advice and N a n c y M c L a r e n for her encouragement and adminis t ra t ive help. I w o u l d especially l ike to acknowledge m y parents w h o were there every step of the w a y to encourage, and pu t u p w i t h me d u r i n g a l l the " u p s " and " d o w n s " of m y research. viii CHAPTER 1: Introduction Osteoarthritis (OA), the most common chronic joint condition in the elderly, is a degenerative disorder of synovial joints characterized by focal loss of articular cartilage with reactive changes in subchondral and marginal bone, synovium and para-articular structures (Rottensten, 1996). It is second only to cardiovascular disease as contributing to overall disability among adults (Michet, 1993). The Arthritis Society of Canada estimated in 1994 that 2,930,000, or one in ten, Canadians were affected by osteoarthritis. Moreover, the knee is the most often involved weight-bearing joint afflicted with OA. It is not a systemic disease; it is a degenerative process that is localized to the articular cartilage, subchondral bone and marginal bone around a diarthroidal joint (Martin, 1994). The joint deterioration begins with fraying of the articular cartilage surface. This leads to pitting of the cartilage surfaces, hypertonic changes in the joint margins of the bone and reactive changes in the subchondral bone. The endstage of this disease is characterized by joint space narrowing, total destruction of articular cartilage leaving the eburnated subchondral bone exposed as the articulating surface and spur formation in the joint margins. Consequently, the joint capsule becomes scarred and fibrotic (Martin, 1994). There have been two hypotheses proposed with respect to the cause of OA. 1) normal forces acting on abnormal cartilage with an inadequate healing response 2) excessive forces acting on normal cartilage (Martin, 1994; Michet, 1993). It is 1 unknown whether it is the abnormal stress or abnormal cartilage that starts the OA process, but the biochemical changes that occur early in the condition lead to continued degradation of the cartilage and advancement of the disease. A number of treatment options are available for patients with OA, however, many focus on the symptoms of the disease, and not the possible causes (e.g. abnormal force loading). Late stage intervention of OA usually involves surgery, in the form of wedge removal or knee replacement. To postpone this invasive intervention, treatment strategies designed for the early stages may be of benefit. Specifically, treatment designed to address the abnormal loading, such as functional varus/valgus knee bracing has been suggested. The theory behind OA knee bracing is the production of opposing forces on the segments of the involved limb. For example, in medial OA, the patienf s lower limb develops a varus alignment, and the brace attempts to correct this with a valgus force to unload the medial side of the knee. 1.1 Statement of the Problem The purpose of this study was to examine the clinical effects of valgus bracing in subjects with knee osteoarthritis. This topic was selected because there is limited information in the literature with respect to OA and bracing. The literature focuses on the biomechanical analysis of valgus knee bracing (Davidson, 1994; Greenwald, 1997; Polio, 1995,1997). Recently, there have been a few studies examining specific knee braces - specifically the Generation II Unloader knee brace (Horlick etal., 1993; 2 M a r s u n o etal., 1997, K i r k l e y etal, 1999). Patients are interested i n the role bracing can have on their leve l of p a i n and improvements i n their knee funct ion d u r i n g da i ly activities. They are w i l l i n g to try non- invasive measures to p r o l o n g the possibi l i ty of surgery. S h o w i n g that valgus knee bracing, specifically the Patient Ready M o n a r c h knee brace, has a posi t ive effect on both p a i n and function, w i l l p r o v i d e g r o w i n g support that O A bracing is an effective non-surgical alternative to manag ing knee O A . 1.2 Hypotheses This s tudy w i l l test five hypotheses: (1) V a l g u s knee bracing w i l l s ignificantly decrease the percept ion of p a i n / stiffness and i m p r o v e phys ica l funct ion w h e n compared to the no-brace and neutral condit ions. (2) W e a r i n g the valgus knee brace, w i l l result i n a significant increase i n the distance hopped (and hop index) w h e n compared to the no-brace and neutral condi t ions. (3) The t ime to ascend/descend a set of stairs w i l l s ignif icant ly decrease w h i l e w e a r i n g the valgus knee brace w h e n compared to the no-brace and neutral condi t ions. (4) The number of inc l ine squats performed w i l l s ignif icantly increase w h i l e w e a r i n g the valgus knee brace w h e n compared to the no-brace and neutral condi t ion . 3 (5) The t ime to complete a figure-8-run w i l l s ignif icantly decrease w h i l e w e a r i n g the valgus knee brace compared to the no-brace and neutral condi t ion . 1.3 Limitations A major l imi t a t ion of this s tudy is patient compl iance . W e a r i n g this brace everyday and keep ing a da i ly record of p a i n and comfort relies o n the dedicat ion and suppor t to the project on the part of the subject. A s an investigator, i t is impor tant to cont inua l ly encourage and stress the importance of reliable and accurate record keep ing to the subjects. Ano the r l imi t a t ion is the accuracy of col lect ing the data. A l l funct ional tests were measured manua l l y w h i c h may have increased the r i sk for h u m a n error. For example, times to complete the stair c l imb and figure-8-run were measured us ing a m a n u a l s topwatch accurate to a 1 0 t h of a second. A n y factors changing the timer's focus or reaction t ime may have altered the accuracy of the results. A further l imi ta t ion questions whether the selected funct ional tests accurately reflect the funct ional l imitat ions experienced by subjects w i t h O A . This s tudy used innovat ive funct ional tests specific to m i d d l e aged subjects w i t h m i l d to moderate levels of O A . There is l im i t ed data on the re l iabi l i ty and v a l i d i t y of the selected functional tests on this popula t ion . F ina l ly , there is concern regarding the length of the washou t per iod . The researcher assumed three weeks was the m i n i m u m t ime requi red for the effects of 4 the previous brace condition to return to baseline levels. Extending the washout period any longer would have increased the study's duration. This would have further compromised patient recruitment and compliance. C H A P T E R 2: Me thods 2.1 Study Protocol A randomized , single b l i n d , crossover des ign was used for this study. Subjects reported to six testing sessions, and kept a da i ly record of their p a i n and abi l i ty to funct ion w h i l e wea r ing the Patient-Ready M o n a r c h Knee Brace (Smith & N e p h e w -Donjoy, Car l sbad C A , U . S . A . ) . Each subject w o r e the brace for a total of twelve weeks; six weeks i n each condi t ion interrupted by a three-week washou t pe r iod between each brace protocol . Specific instructions regard ing the appl ica t ion of the brace were g iven to each subject; however , they were not g i v e n specifics of each brace condi t ion . This experiment received pr ior app rova l f rom the Unive r s i ty of Br i t i sh C o l u m b i a Commit tee on H u m a n Exper imentat ion. Wr i t t en in formed consent was obtained f rom a l l subjects before part ic ipat ing. Table 1. S tudy D e s i g n Timetable W E E K S 0 1 2 3 4 5 6 7 8 9 10 • 11 12 13 14 15 W G R O U P 1 C o n d i t i o n A A C o n d i t i o n B S H O G R O U P 2 C o n d i t i o n B U C o n d i t i o n A T T i T 2 T 3 T 4 T 5 T 6 (T = test ing session) 6 During the first session, subjects were randomly assigned to one of the conditions (neutral vs. valgus bracing). At this time, baseline data (height, weight, age) were recorded and each subject completed a general information questionnaire on his/her condition. During this first testing session (and the 4th session), subjects performed the functional tests twice - first, in a no-brace condition and second, in the selected brace condition. In the remaining testing sessions (3, 6,12, and 15 weeks), the subjects went through the same functional knee testing once while wearing the selected brace. Prior to starting each testing session, subjects were encouraged to warm up and stretch to prevent injury. Each testing session took approximately 30 -60 minutes. Subjects were contacted by telephone during the first week; this was done to ensure that they were comfortable with the brace and were having no problems completing the daily questionnaire. 2.2 The Knee Brace Subjects wore the Patient-Ready Monarch Knee Brace (Smith and Nephew-Donjoy, Carlsbad, CA, U.S.A.). This brace is an off-the-shelf pneumatic brace that applies counter pressure to the side of the knee. It redistributes the forces acting on the joint, thereby, reducing friction and pain in the knee. It is designed with two flexible cuffs (thigh and calf) within a rigid aluminum frame to allow for a better fit to the leg. Two varus/valgus hinges located above and below the knee flexion/ extension hinge allow for the brace to be adjusted to fit the contour of each patient's leg. An inflatable pneumatic pad placed at the knee joint line enables the 7 patient to customize the pain-relieving force by adjusting the amount of air pumped in. It is designed along a 3-point principle theory of a bow and arrow - when the bow is pulled back into cocking phase it generates energy and force in the opposite direction. The Monarch acts as a bow on the leg to counteract internal pressure (Smith & Nephew-Donjoy Monarch Promotional Brochure, 1996). For this study, subjects wore the brace under two conditions. In condition A subjects wore the brace in a neutral orientation. This was achieved by adjusting the varus/valgus hinges into a neutral position and substituting a non-inflatable condyle pad in its place. In condition B, the subject wore the brace as it was designed. A valgus force was applied to the knee by way of the pneumatic airbag and minor adjustment of the varus/ valgus hinges. Upon application of the brace, subjects were required to maximally inflate the airbladder with the air pump (approx. three pumps). The brace is designed to allow the patient to adjust the amount of air depending on their symptoms. However, for the purpose of keeping continuity, the subjects in this study were asked not to vary the amount of air in the bladder. To monitor each subject/s exposure to the brace, he/she was asked to record time spent wearing the brace each day. They were instructed to wear the brace for a minimum of six hours/day; however, they were encouraged to wear the brace as long as possible during the day. 8 2.3 Subjects Eleven subjects (9 male and 2 female) were recruited from local physicians for this study. Subjects were between the ages of 35 and 62 years, were generally healthy with mild-moderate OA and were participating in some type of recreational activity (i.e. walking, jogging, golf, roller blading, etc.). All subjects were physician diagnosed with unicompartmental medial osteoarthritis of the knee. They had a history of medial joint line pain and radiographic changes (within the last three months) showing medial compartment narrowing. Subjects were excluded from the study if they displayed any of the criteria outlined in Table 2. All subjects were asked to refrain from participating in any other treatment protocols for the duration of the study and must not have participated in any treatments at least three weeks prior to the start of this study. In addition, subjects were asked to maintain their current level of activity throughout the study period. Table 2. Subject Exclusion Criteria Subjects did not display any of the following: • previous hip/knee replacement or ligament repair surgery (previous history of menisectomy or debridement okay) • any ligamentous instability of either knee • any other type of arthritis • leg length discrepancy greater that 2.0 cm • any medical condition preventing subject from performing a single leg hop, knee squats, stair climb, or figure-8 run • large varus/ valgus deformity greater than 10 degrees • any condition preventing application of brace • receiving other treatments for OA (i.e. medication, over-the-counter products (e.g. Glucosamine Sulfate, aspirin, ibuprofen etc.), viscosupplementation (e.g. Synvisc, Suplasyn), physical therapy, an exercise program, or foot orthotic devices) 9 2.4 Functional Tasks Subjects performed four functional tasks: the stair climb, the hgure-8-run, the incline squat, and the single leg hop (for distance). The order of completing each task was randomized during each session to prevent an order bias. Stair Climb The stair climb required each subject to ascend and descend a set of stairs equipped with a handrail. The stair climb consisted of six steps each with a rise of 17.8 cm and a run of 27.9 cm. There was a platform at the top allowing subjects to turn around and descend the same set of stairs. Both feet were placed on the top platform before descending. Subjects began the task by standing on a line 30.5 cm from the first step with their hand on the handrail. On the command "get set.. .go," subjects began climbing the stairs and were timed using a stopwatch accurate to a 10th of a second. Once the subjecf s toe crossed the line where they first started, the time (seconds) was recorded. To keep continuity between subjects, they were reminded to only climb one step at a time and to keep his/her hand in contact with the rail at all times during the ascent and descent. Practice trials were given in order to make sure the task was understood. Figure - 8 - Run The figure-8-run required subjects to complete one lap of the course as quick as possible. The figure-8-running course was 20 meters long with the diameter of each curve measuring 4 meters wide. Six orange cones were placed along the circumference of each curve to eliminate subjects from trying to execute 10 shorter/tighter turns. The starting point was located 2 meters from the beginning of the straightaway. Time was measured, with a stopwatch, for each subject to complete one clockwise lap of the circuit. Subjects were given a practice trial on the circuit prior to the test. Incline Squat Subjects performed an assisted squat with the use of an incline-sliding bench (Total Gym, San Deigo, California). The bench consisted of a mobile body board that displaced on a track. Varying the angle of the sliding board could modify the level of resistance. The test was performed at 37.8 degrees above the horizontal. This allowed the researcher to regulate the subject's body weight - approximately 67% of the subject's freestanding weight. Initially, goniometric measurements were taken during a single squat to 90 degrees of knee flexion. This allowed the distance limiter to be adjusted so subjects would not be able to go beyond 90 degrees during the test. To maintain position consistency, the subject's foot placement was marked on a piece of poster board secured to the foot platform. The same poster board was used for each session. Subjects were given the opportunity to practice double leg squats to become familiar with the exercise. The testing protocol consisted of each subject completing as many knee squats as he/she could on one leg during 20 seconds. The test was performed on the non-involved leg first followed by the involved leg. 11 Single Leg Hop The single-leg hop test required each subject to jump off one leg and to land on that same leg, without losing balance. Subjects were positioned with their toe behind the tape marker. The opposite leg was positioned with the knee-flexed 90 degrees and their thigh parallel to the weight-bearing leg. Their hands were placed on their hips. A single maximal hop was executed without swinging the suspended leg or removing the hands from the hips. This eliminated movement strategies involving leg and arm swing. Subjects were encouraged to achieve a maximal horizontal distance and to land on the same foot without simultaneously touching the opposite foot down. However, once the landing foot was firmly on the ground, subjects were allowed to extend the opposite leg to avoid falling. Any jump in which the subject lost balance, or landed with two feet was discarded and the subject repeated that jump. Each subject was given the opportunity to practice until comfortable. The testing protocol consisted of three recorded jumps on each leg starting with the non-involved leg. Subjects were allowed to rest between jumps and continued when ready. The distance hopped (centimeters) calculated from the toe marker to the heel placement was measured using a tape measure firmly secured to the floor. The subject was not informed of his/her distance, but was encouraged to jump as far as possible. The mean distance of the three hops and a hop index value was calculated. The hop index is a generalized measurement that expresses performance of one leg relative to that on the other leg. It is not affected by differences in gender and data analysis. A value of 100 indicates that both legs are equal. The hop index is calculated as follows: Distance hopped involved leg X 100 Distance hopped non-involved leg 2.5 Western Ontario & McMaster (WOMAC) Pain Questionnaire The WOMAC osteoarthritis index is a multidimensional self-administered instrument designed for people who have knee or hip osteoarthritis. This measure has been proven reliable (r=. 73 -.96), valid and sufficiently sensitive to detect clinically important change (Sun etal, 1997; Bellamy etal, 1988; Bellamy, 1996; Bellamy et ah, 1992). The questionnaire is separated into three sections: pain (5 questions), stiffness (2 questions) and physical function (17 questions). Subjects were asked to complete the WOMAC osteoarthritis index (Appendix C) each day for the duration of the study. The visual analog scale (VAS) version Was chosen for this study because it is slightly more responsive than the Likert scale (Bellamy, 1989). This VAS is a 100-millimeter horizontal line with the words "No Pain/Stiffness/Difficulty" on the left side and "Extreme Pain/Stiffness/Difficulty" on the right side for each respective question. Subjects were asked to place an "X" along the line that best represented the answer to each of the 24 questions. To maintain consistency, subjects were asked to complete this index each day at the 13 same time - preferably at the end of the day. Subjects were also asked to refrain from looking at the previous day's score to prevent any potential bias from one day to the next. To analyze the data, the distance the "X" was located from the left end of the line was measured in millimeters. 2.6 Statistical Analysis The functional tasks were all analyzed using repeated measures analysis of variance (ANOVA). Within group analysis for the WOMAC data was also analysed by repeated measures ANOVA. Post-hoc comparisons using Fisher7s Least Significant Difference (LSD) procedure were used to determine individual differences between brace/no brace conditions. The level of significance (a) was set at p<0.05 for all comparisons. RM ANOVAs were performed using Statview statistical package - version 5.01 (1998). 14 CHAPTER 3: Results 3.1 Descriptive Data Eleven subjects entered the study between June 1998 and January 1999. One subject dropped out because he was unable to tolerate the valgus brace. The researcher and subject both felt that there were other medical conditions present in his knee, in addition to osteoarthritis, affecting his tolerance. Therefore, ten subjects completed the study. The baseline height, weight and age are summarized in Table 3. Table 3. Physical Characteristics of Subjects A G E H E I G H T W E I G H T (years) (cm) (kg.) Male (n=8) Mean 52.25 170.44 87.16 SD 9.21 7.54 8.70 Female (n=2) Mean 47.50 165.50 72.27 SD 6.36 4.77 9.64 3.2 Compliance The subjects' compliance wearing the knee brace in both conditions was excellent. Subjects wore the neutral brace on average 11.2 + 2.8 hours per day and I wore the valgus brace on average 10.8 + 3.2 hours per day. Inspection of the data reveals that the brace condition selected first was usually worn for the longer period, 15 regardless if it was the neutral or valgus alignment. This could be attributed to the early motivation and interest in participating in a study. 3.3 Testing Timeline Subjects attended testing sessions according to the schedule outlined in Table 1 (page 6). In most cases, the timeline of 42 days in each of the brace conditions and 21 days in the washout period was adhered to (Table 4). The small variability in the numbers occurred for two reasons; first, to prevent scheduling the testing sessions on a weekend and second, to fit in with the subjects personal schedule. Subject 5 had a two-week delay from testing session 1 until the start the study because of a blistering reaction to the felt liners on the brace. Once the blisters resolved, the felt liners were changed to a non-allergenic brand and the subject was able to re-start the study. Table 4. Testing Timeline Valgus Brace Washout Neutral Brace Mean (days) 42.8 22.2 42.7 S.D. 2.6 3.7 2.2 16 3.4 Functional tasks The data was first analyzed for evidence of a carryover effect. This was performed by comparing the pre-brace data with day 19 of the washout data. A significant (p<. 05) between the two values would indicate a possible carryover effect. Comparison of the pre-brace with the post-washout data using a repeated measures A N O V A revealed no evidence of a carryover effect in the stair climbing, single leg hop or figure-8-run functional tasks (p>. 05). This suggests that it doesn't matter which brace condition the subject receives first, as it will not affect the following condition. However, analysis of the incline squat data indicated evidence of a carryover (p<. 05). Plotting the data revealed that it was the valgus brace carrying over into the neutral brace condition. This suggests that the neutral brace data may have been skewed by the valgus brace in those subjects starting with the valgus brace first. It is also possible that the significant difference could be attributed to a learning effect by the subjects from the pre-brace to washout period. The p-values are outlined in Table 5. Table 5. Carry-over Effect Analysis Stair climb Figure-8-run Incline Squat Single leg hop p-value .783 .768 .011 .901 F (1,8) 4.07 .093 10.96 .017* p=0.5 *F (1,7) = single leg hop because one subject had incomplete data 17 The functional task data were each analyzed using a repeated measure ANOVA. The final testing session data (session 3 and 6) of the two brace conditions were used for the comparison. It was assumed that both these testing sessions were a good reflection of the data. The no-brace condition was calculated as an average between the pre-brace and post washout periods. The mean data is summarized in Table 6. The results indicate that there was a significant treatment effect between brace conditions in the stair climbing (p< .05). However, there was not a significant difference between conditions in either the figure-8-run, incline squat, or single leg hop. Table 6. Functional test data. Condition Stair climb Figure-8-run Incline Squat Single leg hop (seconds) (seconds) (number) (index %) No Brace Mean 4.41 10.98 14.13 80.50 SD. 0.81 1.76 5.17 8.30 Neutral Brace Mean 3.92 10.80 19.30 80.92 SD. 0.72 1.61 3.89 11.80 Valgus Brace Mean 4.10 11.00 18.70 80.71 SD. 0.76 1.39 2.11 9.55 Stair climb There was a statistical difference in the time to complete the stair climbing protocol between brace conditions (F[2,9] = .62; p=. 006). A post hoc analysis, using 18 Fishers LSD, revealed that subjects wearing the valgus brace and the neutral brace demonstrated a significant improvement in the time to ascend/descend the set of stairs compared to the no brace (p<. 05). There was no significant difference found between the valgus and neutral conditions. Figure 1 graphically shows the means and the raw data is summarized in Table 14 (Appendix B). Stair Climb 4.5 -] • ' i ' ' ~ 4 - I J j , " 3.5 ~ 7 Mean 3 -Time (s) 2.5 " ~ 2 ~ ' ~ 1.5 " ~ 1 " " .5 ~ 7 0 1 1 — • No Brace Valgus Brace Neutral Brace Condition Figure 1. Comparison of the means for ascending/descending stairs. Figure-8-run There was no significant difference between brace conditions for the figure-8-run (i*[2,9] = .49; p=. 624). A small non-significant improvement in time was noted between no brace and the first brace condition. This can be attributed to a learning effect by the subjects. More practice time would have eliminated this result. Figure 2 graphically plots the means and Table 13 (Appendix B) summarizes the individual raw data. 19 F i g u r e - 8 - R u n 12 "| " " " 10 - " 8 _ " Mean . . Time _ (s) 6 4 " ~ 2 " " ^ N o Brace V a l g u s Brace N e u t r a l Brace C o n d i t i o n Figure 2. C o m p a r i s o n of the means for the figure-8-run. Incline Squat There was no evidence that wea r ing a brace improves the abi l i ty to perform incl ine squats. N o significant difference was found between brace condit ions (p>. 05). Results f rom this functional task must be interpreted w i t h caut ion because of the evidence of a carryover effect from the valgus into the neutral condi t ion . F igure 3 graphica l ly depicts the mean group values for each cond i t ion w h i l e Table 15 ( A p p e n d i x B) summarizes i n d i v i d u a l r aw data. 20 Incline Squat 20 17.5 Number of 15 squats mean 1 2 5 10 7.5 2.5 N o Brace Valgus Brace Condi t ion Neutral Brace Figure 3. C o m p a r i s o n of the means for the inc l ine squat. Single L e g H o p There was no statistical difference between brace condi t ions i n the distance hopped (i<[2,8] = .02; p= .98). F igure 4 shows single leg hop indexes mean data. Subject 10's data was exc luded from analysis because i t was incomplete . Therefore, the sample size for each group was unequal . I n d i v i d u a l data can be rev iewed i n Table 16 ( A p p e n d i x B). S i n g l e L e g H o p I n d e x 10 0 9 o 8 o 7 o H o p I n d e x ( m a x 1 0 0 ) 50 4 0 3 0 2 0 1 0 0 N o B r a c e V a l g u s B r a c e N e u t r a l C o n d i t i o n Figure 4. C o m p a r i s o n of the means for the single leg hop index. 21 3.5 WOMAC data The WOMAC index data was analyzed using a repeated measures ANOVA The WOMAC data was analyzed for each of the subscales: 1. Pain 2. Stiffness 3. Physical function. The no brace condition data was calculated as the average between the pre brace and day 19 of the washout period. Data averaged from day 39 and 40 of each the valgus and neutral data were compared with the no brace condition. These days were selected because they were the last two days that all subjects had data. This sampling was performed because of the variable time spent in each condition and it allowed for an equal comparison. Specifically, the pre- brace condition consisted only of one session and the washout only 21 days. Ideally, the no brace period should have been consistent with the time spent in the bracing periods, however, due to time constraints this was not possible. This variability in the days can affect the overall power of the study. Some subjects failed to complete all questions on the WOMAC questionnaire. The primary reason being the particular question did not apply to them on that day. For example, subjects usually omitted the physical function question "Getting in/out of the bath" because they used showers. Following the WOMAC user's guide (Bellamy, 1996), missing data was scored as follows: (1) If the subject failed to complete > two pain/stiffness, or > four physical function questions, the subject's response was regarded as invalid and the subscale omitted for that day. (2) If subjects missed less than the above criteria, the average value for that subscale was substituted in lieu of the missing question. 22 P a i n The subject's percept ion of p a i n d i d not s ignif icantly change between brace condi t ions (F[2, 9]=. 892; p=. 427). H o w e v e r , v i sua l inspect ion of the data reveals a smal l non-significant t rend that the subjects' p a i n percept ion decreased w h i l e w e a r i n g a brace (Table 7). T a b l e 7. M e a n data for p a i n subscale No Brace Neutral Brace Valgus Brace Mean 121.64 92.68 86.84 S.D. 56.15 125.11 44.69 There was no significant v i sua l difference between the va lgus a n d neutral brace condit ions. See F igure 5 for the g raph of the va lgus and neutral brace means. Pain Valgus Brace Neutral Brace Time (days) Figure 5. P lo t t ing of the valgus and neutral brace means for p a i n subscale 23 Stiffness Analysis of the stiffness subscale of the WOMAC index indicates no change in stiffness between brace conditions (F[2,9]=2.06, p=. 157). Again, a small non-significant trend is visually observed indicating that the subjects' perception of stiffness decreased while wearing a brace. Table 8. Mean data for stiffness subscale No Brace Neutral Brace Valgus Brace Mean 61.93 40.27 43.82 S.D. 33.40 47.06 34.26 There was no significant visual difference between the valgus and neutral brace conditions. See Figure 6 for the graph of the valgus and neutral brace means. Stiffness I ~ 60 o o W 8 40 O li < ro 20 4 0 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I CO T - CD T - CD T - CO t - T - CM CM 00 CO T i m e (days) •Valgus Brace Neutral Brace Figure 6. Plotting of the means for stiffness subscale 24 Physical Function There was a significant difference between conditions in the physical function subscale (F[2,9]=5.18, p=. 017). A post hoc analysis, using Fishers LSD, indicates a significant improvement in the subjects' perceived physical function while wearing a brace compared to not wearing a brace (valgus: p=. 016; neutral: p=. 009). The means are outlined in Table 9. Table 9. Mean data for physical function subscale N o Brace Neutral Brace Valgus Brace Mean 432.11 260.27 275.93 S.D. 203.55 255.96 133.27 There was no significant visual difference between the valgus and neutral brace conditions. See Figure 7 for the graph of the valgus and neutral brace means. P h y s i c a l F u n c t i o n CD V algus B race N eutral B race T i m e (days) Figure 7. Plotting of the means for physical function subscale o o CO o < 5- HUU o II X CO O £ 300 200 100 0 11 I 11 11 I I I I 11 11 I 11 I 11 11 I 11 I 11 11 I 11 I 11 11 1 6 1 1 1 6 21 26 31 36 25 CHAPTER 4: Discussion Once knee osteoarthritis has progressed to the point of severe articular cartilage loss with subchondral involvement, total knee replacement is usually indicated. There are two foreseeable problems with this outcome; the health care costs associated with this type of surgery, and the increasing trend that osteoarthritis is affecting a younger age group (e.g. less than 50 years of age). For this population, total knee replacement is not recommended because the prosthesis may be subjected to greater activity demands and therefore it will likely not last the patient's expected life span. The premise behind bracing an OA knee is that it mechanically unloads the arthritic compartment to slow down further progression of the disease and provide pain relief. The current study attempted to prove that knee bracing could be a favourable treatment option for this active population. A repeated measures crossover design was chosen for this investigation as this allows subjects to be their own controls and this helps to eliminate the inter-subject differences seen in parallel designs. This design is well suited to investigate treatments for chronic disease, as the underlying problem causing the disease can not be cured. Therefore, the effects of a treatment can be measured. However, a critical component of this design is the assumption that there is no carryover effect between the two conditions. Incorporating a washout period between treatment conditions usually eliminates this carryover effect. The crossover design has the added benefit of being cost and time effective because fewer subjects are required. 26 Limitations include a higher rate of dropout because of the length of the study and the possibility of a carryover effect influencing results. Horlick and Loomer (1993), in an early clinical study on valgus bracing for knee OA, used a crossover study design. In their study, subjects were randomly assigned to different sequences of no brace, valgus brace and neutral brace conditions. Two more recent studies (Matsuno etal., 1997; Kirkley etal, 1999) used a prospective case series and a parallel group randomized clinical trial design. Kirkley etal. (1999) indicated that Horlick and Loomer's (1993) study design was inappropriate because the crossover design did not allow subject blinding and they believed an unknown carryover effect may have occurred. The possibility of a carryover effect is apparent, considering Horlick and Loomer (1993) neglected to insert a washout period between trials in their study. In this study, we corrected for both of Kirkley et al.'s concerns. A 3-week washout period was inserted between brace trials, and subjects were blinded as to the brace condition. The subjects were unfamiliar with the brace design and were informed that it was unknown which condition would be useful in the treatment of OA. It would have been advantageous to have the researchers also blinded however, it was impossible because of familiarity with the brace. In addition, statistical analysis of the data revealed that other than the incline squat, there was no carryover effect between brace conditions (see Table 5). 27 4.1 Functional Tasks Stair Climbing There was a significant improvement in the time to stair climb when wearing a brace versus not wearing a brace. However, no statistical difference (p>. 05) was observed between the neutral and valgus brace conditions. Marks (1995) and Rejeski et al. (1995) both demonstrated very good reliability and validity for a timed stair climbing protocol. Marks (1995) calculated the reliability, using intraclass correlation coefficients, in knee OA patients for climbing 4 stairs to be R=. 82. While Rejeski et al (1995) further substantiated this result in knee OA subjects with a 2-week reliability R=. 93 and a 3-month reliability R=. 75 for a 5 stair protocol. In addition, they found that stair climbing as an outcome measure has both concurrent and convergent validity. In this study, a 6-stair protocol was used because it was the closest available that corresponded to the literature. The investigator assumed that 6-stair protocol used in this study would have similar validity and reliability results to the protocols used in the literature. It is very difficult to compare this study's results with the literature because there is so much variability with stair climbing protocols (Kirkley et al., 1999; Rejeski etal., 1995; Matsuno etal., 1997; Marks, 1995; Lankhorst etal., 1982). This variability is due, in part, because there is not one set protocol for using stair climbing as an outcome measure. One study used a knee scoring system to evaluate stair climbing and found significance with 12 OA subjects (Matsuno etal, 1997). Others calculated 28 the number of stairs climbed in a specified time frame and did not find significance with 119 O A subjects (Kirkley etal., 1999). As stated earlier, Marks (1995) used a 4-stair protocol. Although he used two less stairs than this study, the rise and run were similar. His results indicated the time to complete the task was almost twice as along as this study. This could be contributed to two factors: 1) the mean subject age (x=64.17) in his study being approximately 12 years older and 2) their condition had progressed further. Interestingly, Marks calculated a minimum change of 2.90 s was required to demonstrate a clinically significant change in 48 subjects. Although our study had a much smaller sample size, none of the subjects demonstrated more than a 2.34 s change between conditions. Lankhorst et al, (1982) measured the time required to climb eight stairs in knee O A subjects. He reported mean values of 9.1 s pre-treatment and 6.5 s post treatment. He also determined normative values for the 8-stair climb as 3.0 ±1 s Unfortunately, Lankhorst et al. did not indicate the mean age of their subjects, and did not specify the stair measurements to allow a thorough comparison. However, based on their limited findings and corrected for differences in stair number, this study's values are comparable. Projected values for 8-stairs would be 5.83 s pre treatment and 5.46 s with brace which is slightly faster than Lankhorst et a/s post treatment results and slower than the expected normal values. It is unclear why, but people with knee O A have an impaired ability to negotiate stairs (Marks, 1994,). There has been controversy as to possible reasons for 29 this impaired ability. Andriacchi etal. (1982) hypothesized that at the critical angle of knee flexion during stair climbing, there is a reduction in the mechanical efficiency of the knee extensors in persons with knee OA. While Lankhorst etal. (1982) found no correlation between the knee extensors and stair walking time in patients with knee joint disease. Recently, Marks (1994) found a positive correlation between pain and position sense with respect to standing and stair walking time. His results support Hurley and Newham's (1993) findings that pain and effusion associated with knee OA cause a loss or change in proprioception. Mark's results further support a study by Stauffer et al. (1977) who concluded that pain and impaired proprioception may cause a decrease in the functional demand on the OA knee during weight bearing therefore decreasing stair walking velocity. Research also indicates that slower stair walking speed might be the result of a wider based gait to compensate for the loss of proprioceptive feedback from the OA diseased joint. Moreover, slower velocity of stair climbing may be partly attributed to static weakness of the surrounding quadriceps muscles (Marks, 1994; Lankhorst etal, 1985; Stauffer et al., 1977). Figure-8-Run There was no difference found between brace conditions when compared against each other or the no brace condition in the figure -8- run outcome measure. The figure-8-run has been used extensively in the ACL deficient literature (Tegner et al., 1986; Fonseca etal, 1992). Mean values reported in the literature for uninjured 30 subjects (younger than in this study) were 8.61 ± 0.58 s by Fonseca et al. (1992) and 11.99 s (for 2 laps) by Tegner etal. (1986). The only data available for an injured population was subjects with ACL-deficient knees. Their reported mean was 8.93 ± 0.57 s (Fonseca etal, 1992). This functional task was the only test performed outside and factors such as weather and time of day may have played a factor in the results. Both are factors difficult to control for as the testing facility did not have an indoor space large enough to accommodate the circuit. Concerns have been raised in the literature whether open kinetic chain tests, such as isolated muscle testing, are appropriate to determine an injured knee/leg's functional ability. Specifically, studies have shown that open chain testing may not be sensitive enough to correlate with closed chain activities such as walking, running etc. (Munich etal, 1997; Tegner etal, 1986). For example, a subject7s performance in open chain muscle testing may be excellent, but his/her performance may be poor in functional everyday closed chain tasks. By relying on interpreting open chain test results there may be an inaccurate confidence in the subjecf s ability to function in everyday activities. Incline Squat The incline squat functional test provides an excellent estimate of knee OA function. Important to its acceptance as an outcome measure for OA is that it is a closed kinetic chain test that correlates well with many tasks patients with knee OA find difficult to perform (e.g. stairs, bending down, sitting etc.). It is also unique in 31 comparison to other closed chain tests for the knee (e.g. full knee squats and hopping) in that the subject is only partial weight bearing. This allows for safe and effective testing that might otherwise be too strenuous on this population. Although no significance was found between brace conditions in this study (p>. 05), subjects indicated that they found the task easier to complete using a brace. Compared to the normal population studied by Munich etal. (1997) this study's results, as expected, are lower. Caution must be used in interpreting these results for two reasons: 1) the incline squat test has not been proven reliable in an OA population and 2) the validity of this functional test has not been determined. Future research is needed in this area. Single Leg Hop This study did not find significance in the subjects' ability to improve their hop for distance in either brace condition. However, the results did demonstrate that subjects jumped less with their OA leg compared to their un-involved leg (mean hop index =83.3 ± 9.4). This difference may be attributed to OA subjects having unequal muscle strength between legs. Pincivero et al. (1997) concluded that concentric quadriceps and hamstring strength significantly contribute to the distance achieved in the single leg hop. They specifically found that hamstring muscles play a more important role during the propulsive phase, therefore enabling subjects to jump further. In patients with knee OA, muscle strength inequality has been demonstrated (Wessel, 1996). This inequality may be because these individuals favour their OA knee and are less likely to add extra stress to that side. Moreover, as 32 the condition progresses, patients are less active and have developed compensatory actions with the non-involved leg to accomplish certain tasks. Therefore, the weakness seen in the OA leg, compounded by the subject's decreased confidence, adds up to an expected poorer result for the OA knee in the single leg hop. There is a gap in the OA literature, using a single leg hop as an outcome measure. It is a measure more commonly used in anterior cruciate ligament (ACL) research (Kramer etal., 1992; Fonseca etal, 1992; Juris etal, 1997; Daniel etal, 1990; Tegner etal, 1986). Investigators have used this measure in these studies as an indicator of leg confidence and function. In performing the single leg hop, the integrity of the joint capsule, intra-articular structures, the surrounding musculature is tested (Juris etal, 1997). Juris etal. (1997) indicated that the single leg hop test is effective at examining the combination and interaction of motor skill and intrinsic joint stability because of its demands on motion control and force production/ absorption. These are all-important variables important to also measure in subjects with knee OA. The absence of the use of this test in the OA literature may be due to the premise that the single leg hop is too "demanding" on the OA knee. This may be true if tested in subjects where OA has progressed to a point where activities of daily living are affected, but judging by the performance in this study, it is applicable in this population. As expected, the subjects were tentative when jumping without the brace yet many indicated that with application of the brace they had more confidence performing the task. 4.2 WOMACData The WOMAC osteoarthritis index has the unique design in that it can be analyzed using two methods. Each subscale (1. pain 2. stiffness 3. physical function) can be analyzed separately or the three subscales can be aggregated together into a single score. For the purpose of this study, the former method was used. Pain and Stiffness There was no significant difference in the pain and stiffness subscales between brace conditions (p>. 05). However, there was a slight non-significant trend indicating that wearing a brace improved the subjects' perception of pain and stiffness compared to the no brace condition. A larger sample size may have found significance in this instance. Kirkley et al. (1999) estimated their sample size to be thirty-seven subjects in each group for their parallel design. They calculated this by using the standard deviation (91.06 millimeters) provided by Bellamy etal. (1992) for the change in pain score (baseline and post treatment). An alpha of 0.05 and a beta of 0.80 were used in the calculation. They indicated that with this sample size (37/ group) a clinically important difference of 60 millimeters on the pain scale would be needed for significance. Physical Function There was a significant difference between brace conditions for the physical 34 function subscale (p=. 02). Specifically, subjects' perceived that wearing a brace improved their ability to perform physical tasks better than not wearing a brace. Kirkley etal. (1999) analyzed their WOMAC data using both a single aggregate score and individual subscales. The results of this study support their findings that the valgus and neutral brace improved physical function ability compared to the no brace (control) condition. Also similar was the lack of significance between the valgus and neutral (placebo) brace conditions in the stiffness and physical function subscale. For the pain and stiffness subscales, there was no significance between brace and no-brace conditions in this study. As stated earlier, this result may have been influenced by the small sample size. There was a strong trend in both subscales indicating that pain and stiffness improved while wearing a brace. Interestingly, although Kirkley etal. (1999) found significance in both these sections comparing no brace and bracing, they only found a significant difference between the unloader and placebo brace in the pain subscale. A number of studies have looked at the relationship of knee OA and impaired proprioception (Sharma etal, 1997; Sharma etal, 1997; Hurley etal, 1997; Marks, 1994). There is evidence showing proprioception is worse in subjects with knee OA compared to aged matched controls (Barrett et al., 1991; Sharma et al, 1997; Hurley etal., 1997). However, there is a discrepancy in the literature as to what causes this impaired proprioception. Hurley et al. (1993) believe knee OA causes the impaired proprioception whereas Sharma et al. (1997) indicate that it is the impaired proprioception that may cause knee OA. Sharma et al. (1997) compared knee OA 35 subjects with elderly matched controls, and further compared the OA knee with the non-involved knee of the same subject. They hypothesized that if the impaired proprioception was the result of OA then a between knee difference would be expected. Their results indicated no between knee differences; therefore, they concluded that impaired proprioception was not exclusively the result of knee OA. Hurley et al. (1997) compared the quadriceps sensorimotor function in knee OA subjects with control subjects. They investigated whether these changes were associated with impairment of functional performance. They concluded that the reduced quadriceps motoneurone excitability and diminished proprioceptive acuity might be due to OA subject's articular damage. Thus resulting in reduced functional performances by subjects with knee OA. Research has demonstrated that application of an elastic bandage and/or a neoprene sleeve markedly improves proprioception in patients with knee OA (Kirkley etal. 1999; Marks, 1996). McNair etal. (1996) studied the effects of a "pseudo" knee sleeve on the proprioceptive ability of normal subjects in a dynamic tracking task. Their results showed an 11% increase in tracking ability when their subject wore a knee brace. A study by Kirkley et al. (1999) compared an unloader knee brace, a neoprene sleeve and standard medical treatment in patients with medial knee OA. Their findings support the hypothesis that the unloader knee brace decreases pain and improves the quality of life. However, they also demonstrated that a neoprene sleeve showed, on average, an improved quality of life and reduced pain associated with functional activities of walking and stair climbing. The authors 36 indicated that since a sleeve offers little mechanical support, the patient's feeling of stability and reduced pain may be due to the improved joint proprioception with application of the sleeve. The results from this study lend some support the idea that bracing improves knee proprioception. Although the functional task data showed no significance between bracing conditions, there was significant improvement over the no brace condition in stair climbing. As indicated earlier, this lack of significance may be due to the small sample size. However, based on Kirkley et al.'s (1999) results, it is quite possible that the significant difference from no brace to brace application but lack of significance between brace conditions occurred because of this improvement in proprioception. 4.3 Limitations The main limitation of this study was the small sample size. Subjects were recruited from local orthopedic and sport medicine physicians. Flyers, personal correspondence and word of mouth were all used to promote this study. There are three possible reasons for the limited sample size: 1) the stringent inclusion/exclusion criteria (i.e. age, no history of ligament laxity) excluded possible participants 2) the limited time available for subject recruitment 3) the competitor's brace is produced locally and is commonly prescribed by local physicians. Prior to the start of the study, a power analysis to estimate a sample size was performed. Means for the functional tasks provided in the literature were used to 37 estimate the effect size. Because much of the literature on these functional tasks was based on either normal or ACL populations, it was difficult to estimate for an OA population. Based on a power of 0.8, alpha set at 0.05 and one-tailed test, a sample size of 10 was calculated. However, upon review of the new means collected in this study, an over-estimation of the effect size occurred. Another power analysis based on means derived from this study was performed. Different effect sizes ranging from zero to two and the mean differences for each functional task found in this study were plugged into the SPSS power analysis program. It is estimated that a sample size of 74 subjects is needed to give this study a .809 power to yield a significant result. This computation assumed that the population from which the sample would be drawn has a mean difference of 2.8 with a standard deviation of 4.3. The observed value would be tested against a theoretical value of 1.0. Another limitation could be the change in weather/season over the study duration might have affected the results. Anecdotally, many OA patients' claim they can "tell the weather by how their knee is feeling." (Harris, 1987). Although limited research has been carried out on this topic, there has been support for changes in barometric pressure and temperature increasing OA symptoms (Harris, 1987/ Harris (1987) performed a quasi-experiment in which he reviewed the number of visits to doctors' offices for OA pain. The monthly figures showed an interesting distribution. There were two peaks, one in April/May and the other in September-November, in which OA patients saw their doctors office more often for pain related symptoms. He did state the results should be interpreted with caution as it was a retrospective 3 8 study and the criteria for inclusion may not have been consistent. Nonetheless, seasonal related factors might have influenced the results of this study. The first five subjects started in the summer and ended in the fall/winter. The latter five subjects started in the fall and finished in the winter. The weather in the summer months in Vancouver was warmer and had much less rain than in the fall and/or winter months. If Harris' hypotheses are true, the weather may have influenced the subjecf s perception of pain. Further study using prospective trials are needed to determine if and what influence the weather has on pain perception in OA. Another limitation may be the variance in the subjects' activity level over the study period. Apart from casually discussing with subjects what they had been doing, there was no formal record kept. Although subjects were asked not to increase or decrease their activity level from the start of the study, a daily/ weekly activity log would have helped researchers to monitor this. Conceivably, a person's level of activity may be altered by changes in the weather and or season. People tend to be more active in the summer months when the weather is better and the days are longer. As previous stated most subjects started in one season and did not finish until the next one. An increase or decrease in their activity level may have influenced the subject's function and pain therefore biasing the results. A further limitation may be the manual recording of the data. The times collected for the stair climbing and figure-8-run were manually recorded using a stopwatch accurate to a 10th of a second. Unfortunately, manual timing may generate a source of error due to the variability of the investigator's reaction time (Marks, 39 1995). Ideally, it would have been more accurate to use an electronic measuring system such as that used by other investigators (Fonseca etal., 1992; Tegner etal., 1986). Unfortunately, this type of equipment was not available for this study. However, it is important to note that most clinicians are not likely to have this electronic equipment either. Therefore, it may be more realistic to create norms using the easily accessible manual timing method to allow clinicians easy comparison. Moreover, good reliability has been demonstrated in the literature for manually recorded stair climbing in knee OA population (Marks, 1995; Rejeski etal, 1995). Finally, the inclusion and exclusion criteria for this study were quite stringent. However, even with such criteria, a major limitation to this study and many other OA studies, is group homogeneity. There is so much variability in the status and progression of OA, that it is very difficult to select a specific population all at the same stage with the same symptoms. It is common for someone with OA to have intense pain and functional impairment but have little joint destruction. Conversely, have someone with minor pain and functional impairment with severe joint destruction. Suggestions to decrease this variability are to have each subject assessed by the same physician and perform a magnetic resonance imaging (MRI) scan to clearly see the extent of the joint damage. However, costs associated with performing a MRI and the feasibility of each subject being assessed by the same physician was not realistic for this study. However, even with these suggestions group homogeneity may still be difficult to achieve. 40 4.4 Summary The results of this study support that wearing a knee brace improves the time to ascend/descend stairs compared to not wearing a brace. Moreover, they support that subjects perceive their ability to perform physical tasks easier while wearing a brace. However, the results did not show that valgus bracing is different than neutral bracing. This lack of support may due to the small sample size. The results from this study provide valuable data for further research. It is innovative in its selection of functional outcome measures and its chosen OA population. Much of the research on knee OA has focussed on subjects over age 65 who have difficulty performing activities of daily living (ADL). Specific to this older population, the outcome measures used in the literature have primarily focussed on the ability to perform these ADL. Because knee OA is affecting a younger population, outcome measures specific to the middle-aged population and their level of deterioration are needed. Future research still needs to substantiate non-invasive treatment measures that can slow the progression of this disease and help this population remain active. 41 REFERENCES Altaian, R.D. The classification of osteoarthritis. JRheumatol (suppl 43)22: 42-3, 1995. Andriacchi, T.P., Galante, J.O., Fermier, R.W. The influence of total knee-replacement design on walking and stair-climbing. J Bone Joint Surg 64-A(9): 1329-35. Badley, E.M. The effects of osteoarthritis on disability and health care use in Canada. J Rheumatol (suppl 43)22:19-22,1995. Balazs, E.A., Denlinger, J.L. Viscosupplementation: a new concept in the treatment of osteoarthritis. J Rheumatol (suppl 39)20: 3-9,1993. Balint, G., Szebenyi B. Diagnosis of osteoarthritis: guidelines and current pitfalls. Z2ra,gs52(suppl 3): 1-13, 1996. Bellamy, N. Outcome measurement in osteoarthritis clinical trials. J Rheumatol 22(suppl 43): 49-51,1995. Bellamy, N., Buchanan, W.W., Goldsmith, C.H., Campbell, J., Stitt, L.W. Validation study of WOMAC: A health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. / Rheumatol'15:1833-1840,1988. Bellamy, N., Kean, W.F., Buchanan, W.W., Gerecz-Simon, E., Campbell, J. Double blind randomization controlled trial of Sodium Meclofenamate (Meclomen) and Diclofenac Sodium (Voltaren): post validation reapplication of the WOMAC osteoarthritis index. J Rheumatol 19:153-9,1992. Bellamy, N. WOMAC osteoarthritis index: a users guide. Dr. Nicholas Bellamy. London, Ontario 1996. Bullough, P.G., Vigorita, V.H. Pathology test. Lu: The Atlas of Orthopaedic Pathology. Gower Medical Publishing, 115-120,1984. Butler, P.B., Evans, G.A., Rose, G.K., Patrick, J.H. A review of selected knee orthosis. Br JRheum 22(2): 109-120, 1983. Cawley, P.W., France, E.P., Paulos, L.E. The current state of functional knee bracing research: a review of the literature. Amer JSports Med19(3): 226-233,1991. 42 Cohen, N.P., Foster, R.J., Mow, V.C. Composition and dynamics of articular cartilage: structure, function, and maintaining healthy state. JOSPT 28(A): 203-215, 1998. Creamer, P., Flores, R., Hochberg, M.C. Management of osteoarthritis in older adults. Clin Geriatric Medl^); 435-454,1998. Daniel, D.M., Stone, M., Riehl, B., Moore, M.R. A measurement of lower limb function: the one leg hop for distance. Am JKnee Surg4: 212-4,1984. Davidson, P.L. The effects of valgus bracing on the 3D kinematics of gait in patients with osteoarthritis of the knee. Master's Thesis - University of British Columbia, 1994. Dieppe, P. Strategies for the prevention of osteoarthritis. Int J Tiss Reacl5(3): 93-97, 1993. Dieppe, P.A. Recommended methodology for assessing the progression of osteoarthritis of the hip and knee joints. Osteoarthritis and Cartilage 3: 73-77,1995. DiNubile, N.A. Osteoarthritis: how to make exercise part of your treatment plan. Phys Sportsmed25(7): 47-56,1997. Farrell, M., Richards, J.G. Analysis of the reliability and validity of the kinetic communicator exercise device. Med Sci Sports Exerc 18(1): 44-49,1986. Felson, D.T., Zhang, Y., Anthony, J.M., Naimark, A., Anderson, J.J. Weight loss reduces the risk for symptomatic knee OA in women. Ann Intern Med 116(7): 535-539,1992. Felson, D.T., Radin, E.L. What causes knee osteoarthritis: are different compartments susceptible to different risk factors? (Editorial) J Rheumatol 21(2): 181-2,1994. Felson D.T., Naimark, A., Anderson, J., Kazis, L., Castelli, W., Meenan, R.F. The prevalence of knee osteoarthritis in the elderly: the Framingham osteoarthritis study. Arthritis Rheum 30: 914-18,1987. Fisher, N.M., Pendergast, D.R., Gresham, G.E. Calkins, E. Muscle rehabilitation: its effect on muscular and functional performance of patients with knee osteoarthritis. Arch Phys Med RehabilH: 367-7'4, 1991. 43 Fisher, N.M., Gresham, G.E., Abrams, M., Hicks, J., Horrigan, D., Pendergast, D.R. Quantitative effects of physical therapy on muscular and functional performance in subjects with osteoarthritis of the knees. Arch Phys Med RehabillA: 840-7,1993. Goldberg, V.M. Osteoarthritis of the knee. Osteoarthritis: Diagnosis and Medical/Surgical Management. Edited by Roland Mosowitz. W.B. Saunders, 1992. Gracely, R.H., Kwilosz, D.M. The descriptor differential scale: applying psychophysical principles to clinical pain assessment. Pain 35: 279-288,1988. Greenwald, R.M., Self, B.P., Mecham, M.D., Pflaster, D.S. Biomechanical analysis of a medial unloading brace for osteoarthritis in the knee. Presented at: Osteoarthritis Research Society Congress Singapore June 7-9,1997. Harris, C. Osteoarthritis and the weather. Physiotherapy Practice 3: 179-184,1987. Hicks, J.E. Gerber, L.H. Rehabilitation in the management of patients with osteoarthritis. In: Osteoarthritis: Diagnosis and Medical/Surgical Management. Edited by Roland Mosowitz. W.B. Saunders, 1992. Horlick, S.G., Loomer, R.L. Valgus knee bracing for medial gonarthrosis. Clin J Sport Med'3(4): 251-5,1993. Houston, M.E., Goemans, P.H. Leg muscle performance of athletes with and without knee support braces. Arch Phys Med Rehabil 63: 431-2,1982. Hurley, M. V., Scott, D.L., Rees, J., Newham, D.J. Sensorimotor changes and functional performance in patients with a knee osteoarthritis. Ann Rheum Dis 56: 641-648, 1997 Huskisson, E.C. Visual analog scales. Pain Measurement and Assessment Edited by Melzak R. Raven Press, New York, 33-37,1983. Jawad, A.S., Goodwill, C.J. TVS brace in patients with rheumatoid arthritis or osteoarthritis of the knee. Br/Rheum 25(4): 416-7, 1986. Kazis, L.E., Anderson, J.J., Meenan, R.F. Effect size for interpreting changes in health status. Med Care 27(3 Suppl): S178-S189,1989. Kirkley, A., Webster-Bogaert, S., Litchfield, R., Amendola, A., MacDonald, S., McMalden, R., Fowler, P. The effect of bracing on varus gonarthrosis. /Bone/oint Surg 81-A(4): 539-548,1999. 44 Kramer, J.F., Dubowitz, T., Fowler, P., Schachter, C , Birmingham, T. Functional knee braces and dynamic performance: a review. Clin J Sport Med 7 (I)'- 32-9,1997. Kramer, J.F., Nusca, D., Fowler, P., Webster-Bogaert, S. Test-retest reliability of the one-leg hop test following ACL reconstruction. Clin JSport Med2(4): 240-3,1992. Lane, N.E. Exercise: a cause of osteoarthritis. /Rheumatol22(Suppl 43): 3-6,1995. Lankhorst, G.J., van de Stadt, R.J., Hinlopen-Bonrath, E., etal. Relationship of isometric knee extension torque and functional variables in osteoarthritis of the knee. ScandJRehab Med'14: 7-10,1982. Lequesne, M.G., Samson, M.M., Gerard, P. Indexes of severity for osteoarthritis of the hip and knee. Scand J Rheumatology (Suppl. 65): 85-89, 1987. Lewis, B., Bellomo, R., Lewis, D., Cumming, G. A clinical procedure for assessment of severity of knee pain. Pain 63: 361-64, 1995. Lindenfeld, T.N., Hewett, T.E., Andriacchi, T.P. Joint loading with valgus bracing in patients with varus gonarthrosis. Clin Orthop 344: 290-297,1997. Maclntyre, D.L., Reid, W.D., McKenzie, D.C. Delayed muscle soreness: the inflammatory response to muscle injury and clinical implications. Sports Med 20(1): 24-40,1995. McNair, P.J., Stanley, S.N., Strauss, G.R. Knee bracing: effects on proprioception Arch Phys Med Rehabil 77: 287-289,1996. Marks, R. The effects of 16 months of angle-specific isometric strengthening exercises in midrange on torque of the knee extensor muscles in osteoarthritis of the knee: a case study. JOSPT 20(2): 103-109,1994. Marks, R. An investigation of the influence of age, clinical status, pain and position sense on stair walking in women with osteoarthritis. Int J Rehabil Res 17: 151-158, 1994. Marks, R. Pilot study of the reproducibility of measures of walking performance variables in persons with osteoarthritis of the knee. Physiother Can 47(1): 40-44,1995. Martin, D.F. Pathomechanics of knee osteoarthritis. Med Sci Sports Exerc 26(12): 1429-1434,1994. 45 Melzack, R. The McGill pain questionnaire. Pain Measurement and Assessment Edited by Melzak R. Raven Press, New York, 41-47,1983. Michet, C.J. Osteoarthritis. Primary Care 20(4): 815-826,1993. Noyes, F.R., Barber, S.D., Mooar, L.A. A rationale for assessing sports activity levels and limitations in knee disorders. Clin Orthop 246: 238-249,1989. Pinals, R.S. Mechanisms of joint destruction, pain and disability in osteoarthritis. Drugs 52(Suppl. 3): 14-20,1996. Polio, F.E., Jackson., R.W., Gilbert, J.E. Bracing for the osteoarthritic knee. BiomechanicsJune: 21-28,1997. Polio, F.E. Biomechanical analysis of valgus bracing for the osteoarthritic knee (Abstract). Arthritis Rheum 38 Suppl.: S241,1995. Price, D.D., McGrath, P.A., Rafii, A., Buckingham, B. The validation of visual analog scales and ration scale measures for chronic and experimental pain. Pain 17: 45-46, 1983. Reading A.E. The McGill pain questionnaire: an appraisal. Pain Measurement and Assessment Edited by Melzak R. Raven Press, New York, 55-61,1983. Rejeski, W.J., Ettinger W.H. Jr., Shumaker S., Heuser, M.D., James, P., Monu, J., Burns, R. The evaluation of pain in patients with knee osteoarthritis: the knee pain scale. J Rheumatol 22:1124-9,1995. Rejeski, W.J., Ettinger, W.H., Schumaker, S., James, P., Burns, R., Elam, J.T. Assessing performance-related disability in patients with knee osteoarthritis. Osteoarthritis and Cartilage^. 157-67,1994. Rottensten, K., Monograph series on aging-related disease: IX. osteoarthritis. Chronic Dis. Canl7(3/4): 92-107,1996. Stauffer, R.N., Chao, E.Y., Gyory, A.N. Biomechanical gait analysis of the diseased knee joint. Clin Orthop 126: 246-55,1977. Schouten, J.S., Valkenburg, H.A. Classification criteria: methodological considerations and results from a 12year following study in the general population. / Rheumatol 22(Suppl 43): 44-5,1995. Scott,J., Huskisson, E.C. Accuracy of subjective measurements made with or without 46 previous scores: an important source of error in serial measurement of subjective states. Annals of Rheumatic Diseases ?>%: 558-559,1979. Scott, J., Huskisson, E.C. Vertical or horizontal visual analog scale. Annals of Rheumatic Diseases?)?,: 560,1979. Sharma, L., Yi-Chung, P. Impaired proprioception and osteoarthritis. Curr. Opinion Rheum 9: 253-258,1997. Sharma, L., Yi-Cung, P., Holtkamp, K., Rymer, W. Is knee joint proprioception worse in the arthritic knee versus the unaffected knee in unilateral knee osteoarthritis? Arthritis Rheum 40(8): 1518-1525, 1997. Spector, T.D., Harris, P.A., Hart, D.J., Cicuttini, F.M., Nandra, D., Etherington, J., Wolman, R.L., Doyle, D.V. Risk of osteoarthritis associated with long -term weight -bearing sports. Arthritis Rheum 369(6): 988-995,1996. Steiner, L.A., Harris, B.A., Krebs, D.E. Reliability of eccentric isokinetic knee flexion and extension measurements. Arch Phys Med Rehabil 74:1327-35,1993. Stucki, G., Sangha, O., Stucki, S., Michel, B.A., Tyndall, A., Dick, W., Theiler, R. Comparison of the WOMAC osteoarthritis index and a self report format of the self-administered Lequesne-algofunctional index in patients with knee and hip osteoarthritis. Osteoarthritis Cartilage 6(2): 79-86, 1998. Sun, Y., Sturmer, T., Gunther, H , Brenner, H. Reliability and validity of clinical outcome measurements of osteoarthritis of the hip and knee - a review of the literature. Clin Rheumatol16(2): 185-198,1997. Tan, J., Balci, N., Sepici, V., Gener, F. A. Isokinetic and isometric strength in osteoarthritis of the knee: a comparative study with healthy women. Am J Phys Med Rehabil 74: 364-369,1995. Tegner, Y., Lysholm, J., Lysholm, M., Gillquist, J. A performance test to monitor rehabilitation and evaluate anterior cruciate ligament injuries. Am J Sports Med 14(2): 156-159,1986. Wegener, L., Kisner, C , Nichols, N. Static and dynamic balance responses in persons with bilateral knee osteoarthritis. JOSPT 25(1): 1 3-18,1997. Wessel, J. Isometric strength measurements of the knee extensors in women with osteoarthritis of the knee. /Rheumatol 23: 328-31,1996. 47 Wessel, J. The reliability and validity of pain threshold measurements osteoarthritis of the knee. ScandJ Rheumatol'24: 238-42,1995. APPENDIX A: Review Of Literature The Normal Knee Joint The articulating surfaces of the knee joint that are affected by OA are the condyles of the femur, the tibial plateau, and the posterior aspect of the patella. The surrounding support structures that play key roles from a mechanical and nutritional standpoint are the menisci, ligaments, joint capsule, muscles, tendons, retinacula and the synovium. The menisci aid in stability and contribute to stress distribution during weight bearing. The ligaments also aid in stability while aligning the opposing articular surfaces throughout the knee movements. The joint capsule, muscles, tendons and retinacula all add secondary restraints to the knee joint. Moreover, the muscle and tendons further contribute to absorbing stress in the knee by being more flexible than the ligaments. Therefore, all structures contribute to an important role in approximating and stabilizing the articular surfaces throughout the range of motion of the knee (Martin, 1994). The synovial membrane lines the knee joint and has three functions. The first is to regulate the content of the synovial fluid, the second is to remove waste products by phagocytosis, and the third, is to secrete synovial fluid and other macromolecules. It is an extremely vascular tissue that is made up of predominantly two cell types; Type A which are mainly phagocytic and Type B which are secretory and produce synovial fluid. It is this synovial fluid that covers the articular surfaces 49 of the joint and is responsible for lubrication and nutrition of the articular cartilage (BuUough et al., 1984). This articular cartilage is supported by underlying subchondral bone and further supported by the trabeculae which is responsible to absorb and offer support for the majority of stresses of the knee. The cartilage is composed mainly of water and helps absorb the shock through the joint. The water content is highest at birth and decreases with age. Other components of the articular cartilage are Type II collagen and proteoglycans. The collagen, secreted by the chondrocytes, provides the articular cartilage with tensile strength and stiffness by way of crosslinking its polypeptide chains in a triple helix configuration. Proteoglycans are complex hydrophilic glycoproteins that aid in creating an expansive osmotic effect and contributes to the weight bearing ability of cartilage (Martin, 1994). Specifically, water is released gradually at the articular surface when a compressive load is applied and the process reverses when the load is removed (Pinals, 1996). It is all these components, that allow the articular cartilage to have a viscoelastic nature to withstand forces up to seven times body weight (Martin, 1994). Epidemiology & Risk Factors Several factors have been found to either initiate or perpetuate the development of OA (Michet, 1993). Increasing age is one of the strongest determinants of this condition. This is supported by evidence of radiographic changes seen in the 40+ age group and the increasing occurrence in the elderly 50 (Felson, 1987). More importantly, radiographic evidence of OA was observed more frequently in the older person, irrespective of symptoms. The mechanism underlying the influence of aging is unclear. However, it appears that during aging, the joint has a decreased ability to withstand compression and this puts more stress on the subchondral bone and surrounding support structures. The articular cartilage loses water and proteoglycan concentrations, and therefore, some of its weight bearing properties (Pinals, 1996). However, OA is not merely caused by the natural aging process of cartilage. Other possible factors include cumulative biomechanical trauma to the cartilage, changes in the chondrocyte functioning or possible decline in the neuromuscular protective mechanisms about the joint. Trauma has a variable impact on the risk of OA. A joint injury such as ligament tears, or joint fractures may lead to increased laxity or incongruity to the joint, predisposing the cartilage to extra stress. This leads to the question - does exercise increase the risk of OA? There is no evidence that recreational exercise itself increases the risk (Lane, 1995). However, there is concern that elite level athletes are at more risk due to excessive loading and greater frequency of joint injuries (Spector, 1996). Lane (1995) concluded that exercise or activities that repeatedly expose joints to high levels of impact or torsional loading appear, over time, to be associated with an increase risk. It has also been suggested that stresses placed on a joint due to repetitive occupational activities can influence the OA risk. For example, a job requiring continual knee bending may increase the prevalence. For these reasons, it 51 has been more prevalent for men to experience OA at a younger age than women (Michet, 1993, Rottensten, 1996). Recent literature has suggested that the prevalence of OA increases with age because of an increase in joint load caused by a decline in joint position sense and/or proprioception (Sharma etal., 1997; Barrett etal., 1991). Barrett etal, 1991 demonstrated that knee position sense is worse in subjects with knee OA versus aged matched controls. However, the "chicken and egg" phenomenon is applied here, because researchers are not sure whether the impaired proprioception is caused by the OA or vica versa. Hurley et al. (1997) believe knee OA causes the impaired proprioception, whereas Sharma et al. (1997) have indicated that the impaired joint position sense causes the OA. Obesity may also play a possible factor in knee OA. This is especially evident in older women due to the greater hip angle in females. It had been suggested that joint impairment, induced by the obesity, may contribute to the greater incidence; however, Michet (1993) argued that it is likely the result of increasing biomechanical stress on the joint. Studies have indicated that weight loss in the obese patient will reduce the risk of knee OA (Felson, 1992; Felson, 1994). Structural changes seen with OA In an osteoarthritic knee there are some key structural changes observed compared to the "normal" knee. Fissuring, erosion, softening and ulceration occur to the articular cartilage. The lesions appear first in the load bearing areas, but 52 during progression of the condition, there is a thinning, loss of cartilage and, ultimately, subchondral bone sclerosis (Pinals, 1996). In addition, the joint space undergoes narrowing while the joint capsule thickens and the synovial tissue becomes inflamed. As weight bearing continues in this abnormal situation, the final stage is characterized by a complete breakdown of articular cartilage resulting in a bone-on-bone weightbearing situation (Martin, 1994). It has been postulated that the progression of OA is related to the degree of imbalance between the degradation and repair of cartilage. The cartilage damage is the result of increase enzyme release from the chondrocytes into the matrix leading to increase cartilage matrix degradation (Pelletier, 1993; Pinals, 1995). This is evident in that an increase synthesis rate of enzymes such as collogenase and stromelysin is proportional to the severity of cartilage matrix lesions. Stromelysin breaks down proteoglycans and collagenase is responsible for degrading the type II collagen fibers of cartilage. The breakdown of these structures releases fragments into the synovial fluid which may act as irritants and may be partially responsible for the inflammatory changes that occur in the synovial membrane (Pelletier, 1993). During the course of OA the synovial fluid loses its viscoelastic properties. Specifically, a component of the fluid, called hyaluronic acid, loses its viscoelasticity, which contributes to poor lubricating ability of the synovial fluid on the cartilage surface. This alters the mechanical force transmission to the cartilage, increasing the wear and tear. 53 Pinals and colleagues (1996) indicated that early in this destructive process there might be accompanying efforts to repair cartilage. The chondrocytes attempt to generate new cells in a very limited capacity. The proteoglycan component may also increase which leads to an increase in fibrocartilage> however, this type of cartilage has inferior mechanical properties compared to the "real" cartilage. This attempted restorative process may cause overgrowth and remodeling of the bone; one of the earliest signs may be the formation of osteophytes (Pinals, 1996). These bony changes alter the joint configuration and decrease the range of motion in response to mechanical forces between the calcified cartilage adjacent to the bone and overlying hyaline cartilage (Pinals, 1996). Classification/Diagnosis Diagnosis of OA is not straightforward. It is important to rule out other diseases and conditions such as neurological disorders (e.g. Parkinson's disease), referred pain as observed in neuropathies, soft tissue disorders independent of OA, arthritis of some other origin, pathological changes of adjacent bone (e.g. tumor) or mechanical injuries such as fractures. The American College of Rheumatology has set out criteria to aid in the proper classification of knee OA. 54 Table 10. Classification Criteria for Knee Arthritis (Altaian, 1995) American College of Rheumatology Classification Criteria 1. Knee pain (on most days of the month) 2. Crepitus 3. Morning Stiffness < 30 minutes 4. Age greater than 38 years 5. Boney enlargement of the joint Classification is made if the patients elicit the following combinations of the above criteria: 1,2,3,4, or 1,2,5 or 1,4,5. The classification criteria are different from diagnostic criteria, as the latter is a symptom directed process. The clinical diagnostic process consists of analysis of symptoms, laboratory findings and imaging results. A common diagnostic tool for detection of osteoarthritis is x-ray imaging. In early stages of the condition the presence of osteophytes exists. As the condition progresses, there is evidence of the joint space narrowing and/or hardening of the subchondral bone (Balint, 1996). However, as stated earlier in this paper, patients may exhibit some of these initial radiographic changes but still be asymptomatic. Currently, laboratory findings do not play a very important role in the diagnosis of OA. Sedimentation rate, joint fluid properties and synovial fluid volume are all at, or close to, normal. However, biochemical markers such as cartilage and bone derived components, and indicators of synovial activity are under investigation, and may prove helpful in the early diagnosis of this disease (Balint, 1996). 55 Clinical symptoms of OA usually include pain, crepitus, swelling, and possible decrease in range of motion. Upon a knee examination, there may be some abnormal passive movements due to joint laxity as a result of ligamentous or capsular injury. A varus or valgus deformity and/or bony growths may be present depending on the affected side. These deformities originate from the gradual destruction of the articular structures such as the cartilage, synovium and subchondral bone. The pain associated with OA is usually mechanical in nature, increasing with movement and joint loading while subsiding with rest. It is usually worse when initiating a movement. For example, getting out of a chair after sitting, or getting up out of bed after sleeping. As the condition progresses the pain is evident during extended periods of walking and movement because of the bone-on-bone grating. The cause of the pain in an individual with osteoarthritis may be complex and is not easy to explain. For example, it is not uncommon to have individuals in the advanced stages with deteriorating radiographs exhibiting minimal pain and individuals with mild OA experiencing intense pain. The pain is thought to have an articular or periarticular origin rather than the articular cartilage itself. For example, soft tissue damage such as synovial inflammation, joint capsule distention, ligament damage, muscle spasm or bursitis could be the cause of the pain. Conversely, the pain may be completely independent of this, and be related to factors unrelated to the articular events. For example, factors such as lifestyle, depression, anxiety, social reasons, job related physical stress or obesity may all influence pain perception. 56 Pinels and colleagues (1996) also indicated that the possible breakdown of the sensory reflex mechanisms might contribute to OA pain by allowing motion beyond the normal limits. Pain Measurements Pain is a response that consists of a sensory and psychological component. In the laboratory setting pain measurement usually involves the sensory components whereas in the clinical setting pain assessment predominantly involves the affective components. The literature is extensive on general pain scales available for research (Gracely, 1988; Huskisson, 1983;Melzack/1983). Three of the more commonly used scales are the Descriptor Differential Scale, The McGill Pain Questionnaire, and the Visual Analog Scale. The Descriptor Differential Scale (DDS) method of pain assessment, collects multiple responses minimizing response variability and scaling error (Maclntyre, 1995). The DDS measures both the sensory and affective component of pain, and was designed to randomly present twelve descriptors so as to reduce the effect of memory. It requires a 21-point scale response in relation to each descriptor as the subject is asked to rate each descriptor through a range of -10 to +10. Criticism made with respect to the DDS implies that it is sometimes difficult to rate some of the descriptors on the DDS, and subjects may find themselves attempting to make comparisons between descriptor items. For example, deciding where to put the check mark in relation to the descriptor "mild" and "very mild", or "weak" and 57 "very weak" can be a very difficult task, especially if there is confusion interpreting the meaning of the words. This may also lead to inter-subject variability because each subject may interpret the meaning differently. The McGill Pain Questionnaire (MPQ) provides a reliable and valid indices of pain sensitive enough to detect differences between different pain relief methods (Reading, 1983). The MPQ has 20 groups of various descriptive words that are categorized under 4 major sections: sensory, affective, evaluative and miscellaneous. Subjects are asked to indicate which word(s) in the group(s) best describes their pain and a rank value is given for each word. Criticism towards this questionnaire includes its prognostic precision, and questions have been asked with respect to its incremental validity to alternative measures. In addition, it has not been proven whether tailored abbreviated formats prove as reliable or valid (Reading, 1983). The Visual Analog Scale (VAS) is a simple, robust, sensitive and reproducible method of measuring pain. It allows the patient to express the severity of pain in such a way that it can be given a numerical value. It is commonly used to compare pain severity in the same patient at different times and/or receiving different treatments. The VAS is a line, usually measuring 10 cm in length, in which the line represents the continuum of some experience (i.e. pain). Therefore, one end is often marked "no pain" and the opposite end is marked "severe pain". The subject is asked to put a single mark on the line that corresponds to the severity of pain. The distance measured from the end of the line represents the pain severity. Scott and Huskisson (1979) reported that there was no difference whether a vertical or 58 horizontal line is used, however, the horizontal line is more commonly used in the literature. This scale is easy to understand and the subject is able to quickly evaluate their pain (Huskisson, 1983). It is an ideal scale to be used in crossover experiments where subjects are comparing different treatments. The main criticism is that the VAS perceives pain as a simple unitary dimension varying only in intensity (Gracely, 1988). In addition, it possibly allows the subject to remember a past response and may therefore suffer from scaling error (Maclntyre, 1995). However, although it can not be denied that previous scores may influence the subject, knowledge of previous scores may lead to greater precision and less variation in the results. Particularly during, or after, prolong treatment periods when the patienf s memory of the initial state has faded (Scott & Huskisson, 1978). There is a new trend for researchers to develop outcome questionnaires specific to conditions. For example, for knee OA, Rejeski and colleagues (1995) developed a Knee Pain Scale (KPS). They attempted to capture both the frequency and intensity of pain experienced during the patient's daily living activities. This scale is better suited for patients who display disability in their daily living activities. Lequesne et al (1987) developed indexes of severity for both knee and hip OA. In the knee index, there are three sections: pain or discomfort, maximum distance walked, activities of daily living. In the pain subscale patients answer 5 questions regarding night pain, stiffness duration, pain during walking, standing and sitting. In the second subscale subjects are asked to select the distance able to walk. The final section has the patients rate their ability in four activities of daily living. In each 59 section the patients are awarded points based on their answer. The higher the point score the greater the disability (e.g. >14 equals extremely severe handicap and 1-4 equals mild handicap). This measure has been reported to have good inter-rater reliability and has demonstrated responsiveness in NSAID trials. Similarly, Bellamy and colleagues (1988) developed an activity based pain subscale called the Western Ontario and McMaster University Osteoarthritis Index (WOMAC). This index separates evaluation of the knee into three categories: pain, stiffness, and physical function. Subjects answer twenty-four questions using either the Likert or Visual Analogue scaled formats. WOMAC is a valid and reliable measure that is sensitive enough to detect clinically important health status changes following an intervention (Bellamy, 1996). Reliability values of the three subscales ranged from .73 to .96 for the Visual Analogue and .75 to .97 for the Likert scaled versions (Bellamy, 1996). In a validation study done by Bellamy etal. (1988), each subscale of the WOMAC fulfilled conventional criteria for face, content and construct validity, reliability, responsiveness and relative efficiency. Potential limitations include its restriction to pain intensity assessment and it does not consider pain during transfer activities (i.e. getting in/out of a car). Comparisons between the Lequesne and WOMAC indices have been addressed in the literature (Bellamy etal, 1991; Sun etal, 1997; Stucki etal, 1998). In a double blind randomized trial comparing two NSAID medications by Bellamy et al. (1992), the study demonstrated the relative efficiency of the WOMAC was similar to that of the Lequesne index. Stucki et al. (1998) compared the German version of 60 WOMAC and the Lequesne in knee and hip OA patients. They concluded that the all subscales of the WOMAC were internally consistent but found only the function subscale (not the symptoms) internally consistent in the Lequesne index. Sun et al. (1997) in a comparative literature review of knee and hip OA outcome measures indicated that the WOMAC subscale showed relatively high levels of correlation with the Lequesne items, but indicated that the Lequesne probed different dimensions of health - specifically, the type of pain and duration of stiffness. Conversely, the WOMAC measures the severity of pain and stiffness and is more sensitive to change. The World Health Organization and the American Association for Orthopedic Surgery has recommended both the WOMAC and Lequesne indices as the primary efficacy measures in OA treatment studies (Sun etal, 1997). Trea tmentof OA of the Knee The general goals in treating individuals with OA are (1) to increase function (2) to maintain current function (3) to prevent dysfunction or preserve normal function (Hicks, 1992). Treatment programs are usually tailored to the individual depending on the amount of pain and disability he/she experiences. Early intervention may consist of a physical therapy rehabilitation program. In this situation, the goal is to use modalities to aid in pain relief. Active rehabilitation, in the form of exercise and strengthening, can maintain joint range of motion and to improve mechanical stabilization of the joint. Muscles perform an important protective function in the knee by maintaining normal alignment and help in 61 absorbing shock. If the muscles are too weak and unable to perform this function, the loading stress that is not absorbed will impact directly on the opposing articular cartilage (Tan, 1995). Research has demonstrated that effectively strengthening the muscles around the knee joint will decrease pain and increase function (Fisher, 1991; Fisher 1993). Another early treatment intervention has been the use of non-steroidal anti-inflammatory drugs (NSAID) to help in pain relief by decreasing inflammation. However, OA is not an inflammatory disease and NSAID's are only effective if the patient has an associated inflammatory condition (Michet, 1993). A more appropriate chemical intervention is an analgesic (i.e. acetaminophen) to help in pain control. Nonetheless, the use of a medication only serves to treat the symptoms and will not delay or heal the progressively deteriorating condition. New research has been focussing on other areas of treatments such as viscosupplementation, glucosamine sulfate and even transplanting cartilage cells developed in the laboratory setting. Viscosupplementation attempts to restore the homeostasis of the viscloelastic properties of the synovial joint. A synthetic hyaluronan derivative is injected into the joint in an attempt to act as a lubricant to provide protection of the joint surfaces. There is some evidence indicating it may stimulate the body's own production of hyaluronan, but this has not been substantiated (Balazs etal., 1993). The effect of viscosupplementation does not last indefinitely. This is primarily due to the underlying problem not being resolved and the supplement breaking down and being absorbed by the body. 62 Glucosamine sulfate is a nutritional supplement taken orally by OA patients in an attempt to supplement and/or possibly stimulate the body's production of glycosaminoglycans in the articular cartilage (Creamer etal, 1998). Research is ongoing, in that much of the evidence is anecdotal or is not methodologically sound. Other, more invasive, treatment options include surgery; (1) debridement, (2) tibial/femoral osteotomies, (3) knee replacement. These options are not usually considered until the disease has progressed to a disabling state and all other treatment options have been exhausted. Debridement, as the name implies, consists of debriding the knee joint of any loose bodies, rough edges, and/or bony growths in an attempt to re-align the congruity of the two joint surfaces (Goldberg, 1992). It is usually a temporary treatment option, as the joint will eventually return to pre-debridement state due to the same forces still present on the articular surfaces. This procedure is commonly performed on OA patients less than 65 to prolong the time until they require joint arthroplasty. The goal of the osteotomy surgery is to cut a wedge out of the tibia or femur (depending which bone is creating the varus/ valgus deformity) to re-align the extremity and to decrease the stress on the single degenerated compartment. Knee arthroplasty, replacement with a synthetically manufactured joint, is usually done as the last resort. It can be either partial (replace only the affected compartment) or complete (replace the whole joint). The prognosis is good for arthroplasty, however, long-term effects are just becoming known. Generally, this 63 type of surgery is only considered in patients over 65 as the replacement usually wears out after 10 years. Relating to this study, a treatment option becoming more widespread in its use is knee bracing. Specifically, the use of a valgus/varus brace for medial/lateral OA, respectively. There is little in the literature with respect to bracing and OA. Much of the research focuses on knee bracing for the anterior cruciate deficient knee (Cawley etal., 1991; Houston etal, 1982; Kramer etal, 1997). However, there are two types of knee braces, explained in the literature, for knee OA. Osteoarthritis and Bracing The first arthritic knee brace to be designed was the CARS-UBC brace or the slightly modified version of it called the TVS brace (Butler, 1983; Jawad etal, 1986). It was designed to hold a medially or laterally unstable knee from moving into a painful position. However, this brace was only effective in patients who exhibited a small degree of varus/valgus deformity (Jawad etal, 1986; Butler, 1983). The brace consisted of two plastic cuffs, with velcro closures, joined together by a flexible nylon rod, and a telescopic metal tube. A leather knee sling is joined to the cuffs by two front straps and connected behind the knee to the telescopic tube by clips. Jawad and Goodwill (1986) reported that pain reduction in patients wearing the brace either occurred in the first three days or not at all, and that 14 of 18 patients with OA who found the brace relieved their pain continued to wear the brace. There is no 64 current literature on this knee brace, and it is assumed it is no longer in production due to its cumbersome appearance, difficult application, and patient discomfort. The second type of knee brace used to treat OA is the varus/valgus knee brace. This brace design, first introduced in 1989, is the one referred to in the current literature (Horlick etal., 1993; Polio etal, 1997; PoUo, 1995; Davidson, 1994; Greenwald, 1997; Matsuno etal, 1997; Kirkley etal, 1999). The premise behind this type of bracing is that it utilizes a three-point system to reduce the load across the painful compartment by shifting the load to the unaffected compartment. It basically involves applying a medial or lateral force to the center of the knee in conjunction with two opposing forces, transmitted through cuffs above and below the knee joint, to provide this unloading design. There are many features differentiating the array of OA braces on the market Some are custom made, while others are over-the-counter. The cuffs can be made of plastic, aluminum or composite materials, while the application of the center force can be achieved through condylar pads, inflatable air bladders or a spiral strap. In addition, brace designs differ in single vs. double upright supports, single vs. polyaxial hinge design and the ability to adjust the brace for patient comfort. A clinical comparison of the various designs has not been addressed in the literature. Polio and colleagues (1997) did state that although the dual upright design is a more rigid brace, the materials used to manufacture the single upright design is more than rigid enough to withstand the loads it requires. Furthermore, the polyaxial hinge design will allow greater motion, similar to that of the normal knee joint. 65 Horlick and Loomer (1993) tested the efficacy of a valgus knee brace (Generation II, Vancouver, B.C.) in patients with medial gonarthrosis. They looked at two valgus brace designs - one with a lateral hinge and the other with a medial hinge. The medial hinged design was better accepted by the patients and resulted in more patients continuing to wear the brace after the study. Results indicated significant pain relief (from both designs) when compared to baseline. No significant changes were observed between the brace vs. unbraced radiographs. The authors concluded that valgus knee bracing is a useful treatment modality to reduce the pain associated with OA. Matsuno and colleagues (1997) concurred with Horlick and Loomer (1993) that Generation II bracing caused significant reduction in pain relief in medial OA patients. Muscle strength and performance in the walking test and stair climb also significantly improved. Other variables, the femorotibial angle and mean angle, measured at the pre-brace and final observation period both decreased. Kirkley et al. (1999) in the most recent study compared a custom valgus OA knee brace (Generation II, Richmond, BC), a neoprene sleeve and no brace (control) condition. The authors used a parallel - group, randomized clinical trial design with the following outcome measures: WOMAC osteoarthritis index, McMaster-Toronto Arthritis Patient Preference Disability Questionnaire (MACTAR), six-minute walk and thirty-second stair climbing task. One hundred and nineteen subjects were randomized between groups and scores at baseline and six-months were analyzed. Results indicated that subjects in both brace conditions significantly improved in all 66 outcome measures compared to the control group. There was also a significant difference between the valgus brace and neoprene sleeve in the functional tests; however, no difference was found between braces for the disease specific questionnaires. Greenwald and colleagues (1997) studied the forces applied to the lateral side of the knee and compared the varus moment at the knee during level gait with and without the valgus knee brace. They used the Monarch Knee Brace (Smith & Nephew - Donjoy, Carlsbad, CA). They concluded that this brace significantly reduced the varus moment at 20 and 25% of the stance phase compared to the unbraced condition. Furthermore, the application of the brace resulted in a relatively constant level of valgus force being applied during the stance phase. Polio and colleagues (1997) measured the effect of valgus knee bracing (Generation II, Bothell, WA) on gait mechanics at the knee initially and three months after bracing treatment. They reported that the brace condition resulted in a significant reduction in the external varus moment experienced by the knee from 5% to 43% of the stance phase. Subjects showed no treatment effect when walking unbraced after three months, but showed improvement in pain and function parameters. Therefore, they felt application of the brace resulted in less varus loading to the knee. Davidson (1994) studied how knee bracing (Generation II, Vancouver, BC) altered the alignment of the shank segment with respect to the thigh segment. The results showed that the brace did not have an effect on the thigh coronal angle, but 67 was significant in reducing the shank coronal angle into varus at toe-off. He also reported that bracing prevented full extension during mid-stance and created significant external rotation of both the thigh axial angle throughout the stance phase and of the shank axial angle during knee flexion. However, he attributed this to the brace's helical strap design. Functional Knee Evaluation Functional evaluation of the knee is an important factor in the analysis of evaluating the success of a treatment option. Many OA studies observe patients performing active daily functions such as getting in and out of a chair, or walking a specified distance. Unfortunately, these types of measures may be difficult to quantify or have not been proven reliable or valid. However, there are some valid and reliable functional tests assessing knee function in OA patients. Further a number of tests that have been used more extensively in the A C L literature. An appropriate functional assessment measurement applicable for mild OA patients is the one-legged hop test. Originally described by Daniel et al (1990), this test requires the subject to hop from and land on the same leg without losing balance. The one leg hop test has been reported as a good indicator of leg confidence and function, and has been given support by the International Knee Documentation Committee to be included in their knee test protocol (Kramer etal, 1992). Most of the literature assessing the performance of the one-leg hop test is reported using patients who have experienced anterior cruciate ligament (ACL) deficiencies. 68 Kramer and colleagues (1992) studied the test-retest reliability of the one leg hop test using two measures - the distance hopped, and the hop index score in patients following ACL reconstruction. The results indicated an acceptable test-retest reliability of .75. Fonseca and colleagues (1992) examined a selection of knee functional tests to assess A C L repair. The four performance tasks he reviewed were the slalom circuit, straight running, figure- 8-run and single leg hop test. He reported that the best test for distinguishing normal and ACL deficient knee was the figure of 8 test (expressed as a ratio of time obtained in the straight run test). He also concluded that this test and the hop index appear to complement each other and should be incorporated into studies evaluating functional outcomes of A C L injuries and their treatment. 69 Table 1 1 . Summary of research using Single Hop Test (for distance) Reference Subjects Results Bolgla etal. (1997) - 20 males/females - no lower extremity dysfunction - Reported intraclass correlation coefficients (ICC) of .95 to .96 suggesting high level of reliability Fonseca etal. (1992) - 10 ACL deficient subjects -10 normal controls - both the figure of 8 run and hop index complement each other and should be used to evaluate functional outcomes of ACL injuries Kramer etal. (1992) - 22 men/16 women who had undergone ACL repair - acceptable reliability; - ICC>0.75 for any one occasion. ICC greater if averaged over two occasions Daniel et al. (1990) -100 normal subjects - one leg hop for distance - serves as a objective measure of knee function A common complaint by the osteoarthitic patient is the difficulty climbing and descending stairs. For this reason, a functional test measuring the time for subjects to ascend/descend a set of stairs is applicable. Stair climbing has been often used in the literature (Fisher, 1993; Matsuno etal, 1997; Rejeski etal, 1995; Houston etal, 1982; Kirkley et al., 1999), however, there is little information regarding reliability and validity of this functional test. Rejeski et al(1995) revealed a 3-month test/ re-test reliability of .75 and .87 using a 5-stair protocol at two individual testing 70 sites, but gave no information on the validity. In addition, there is concern over the lack of general agreement with respect to a standardized test protocol for the stair climb. This includes measurements for the stair rise/run and the number of stairs involved in the test. Table 12. Summary of research using stair climbing. Reference Method Results Rejeski etal. (1995) - 148 knee OA subjects > 60 years - assessing physical activity restrictions - 5 stair protocol - good test-retest reliability for stair climb (R=.75 and .87 for two different groups) Fisher etal. (1993) - 40 knee OA subjects - studied the effects of rehabilitation program - significant increase inability to climb stairs after treatment Matsuno etal. (1997) - 20 knee OA subjects - Studied effect of valgus bracing - used knee scoring system to evaluate stair climbing function - number of stairs not given - no reliability/validity data given for knee scoring system - significant improvement seen in the stair ascent/ descent Houston etal. (1982) - 7 males with past ACL - each wearing a ACL knee knee brace to compare braced vs unbraced performance in tests - stair protocol consisted of measuring vertical velocity betw. the 8 t h and 12th step Kirkley et al. (1999) - 119 knee OA subjects - compared Gil, neoprene sleeve and no brace -stair protocol: number of stairs in 20 s - bracing better than no bracing - valgus bracing better than neoprene in functional task grps. - no difference between brace conditions in both questionnaires 71 In a recent study, Munich and colleagues (1997) reviewed the test - retest reliability of an incline squat protocol. This functional test relates well to OA patients because it closely replicates the stair climbing activity as well as many other daily functions, (i.e. such as getting in/out of a chair or car or bending down td get something out of the cupboard). This test uses an incline slide board apparatus to help subjects perform a knee squat to 90 degrees. The incline feature allows a partial weight-bearing environment for the subjects (approximately 65% depending on the degree of incline) so as to minimize the stress to a painful knee or weak muscle group. The authors tested two protocols: (1) how long it would take the subject to do 50 squats and (2) how many squats the subject could perform in 20 seconds. The intraclass correlation coefficient for a one-week repeated test was .80 and .89 respectively. These results indicate acceptable test-retest reliability and the authors advocate the use of this protocol to evaluate functional ability during rehabilitation of lower extremity conditions Summary There is a large volume of research available with respect to knee osteoarthritis. Much of the literature focuses on an older population at the end stage of the disease process and the treatment options available at this severe stage. The information is scarce for those patients with mild to moderate OA and the treatment strategies best initiated at these stages. It is the younger OA population that we must concentrate our future research. Questions such as: How can we prevent their disease from getting worse? How can we prevent this disease? should be an area of focus. 72 APPENDIX B: Raw Data Figure-8-run Data Table 13. Time (seconds), Individual subject data Subject Pre- Test 1 Test 2 Test 3 Post Test 4 Test 5 Test 6 Brace Wash 01 B/A 13.00 13.81 12.22 12.72 14.34 14.32 13.00 13.38 02 B/A 9.25 9.38 8.87 9.16 9.00 8.97 9.41 10.25 03A/B 13.81 12.44 11.41 11.54 11.90 11.22 11.75 12.47 04A/B 9.97 10.03 10.60 10.09 9.43 9.62 10.06 10.53 05 B/A 9.00 9.31 8.50 9.37 8.84 8.81 9.53 8.88 06 B/A 12.13 12.03 12.19 11.72 12.03 11.69 11.97 13.09 07A/B 9.63 9.69 10.33 9.57 9.25 9.44 9.97 9.47 08A/B 13.66 12.13 11.94 11.75 11.03 11.13 did not finish study 09A/B 9.25 9.59 9.62 9.32 10.00 9.81 9.34 9.36 10B/A 12.19 12.91 12.25 11.81 12.50 11.59 11.19 11.41 11A/B 12.25 11.75 12.90 12.47 12.38 11.97 11.62 11.41 • A / B ; B/ A indicates order of crossover (A= neutral brace, B= valgus brace) • Subject 08 withdrew from study - data not used in analysis Stair Climb Raw Data Table 14. Time (seconds), individual subject data Subject Pre-Brace Test 1 Test 2 Test 3 Post Wash Test 4 Test5 Test 6 01 B/A 5.91 6.34 5.59 5.85 5.85 5.47 4.97 5.22 02 B/A 3.93 4.06 3.63 3.84 3.59 3.56 3.71 3.68 03A/B 6.78 4.78 3.91 4.25 4.38 3.44 3.81 3.97 04A/B 5.00 4.65 4.56 4.34 3.34 4.08 4.06 4.16 05B/A 4.32 3.84 3.38 3.31 3.25 3.31 3.41 2.97 06B/A 4.37 4.34 4.03 4.13 3.69 3.68 3.87 3.88 07A/B 3.56 3.91 3.78 3.06 3.18 2.94 3.16 3.00 08A/B 4.66 4.60 4.19 4.54 4.09 4.50 did not finish study 09A/B 3.75 4.16 3.65 3.22 4.16 3.81 3.50 3.90 10B/A 4.98 5.03 4.63 4.31 4.25 4.37 4.03 . 3.94 11 A/B 4.29 4.69 5.22 4.65 4.97 4.78 4.38 4.50 • A / B ; B / A indicates order of crossover (A= neutral brace, B= valgus brace) • Subject 08 withdrew from study - data not used in analysis o o o o o o o o o — o CD CO CD Oi CO rO — ^ B/A B/A B/A B/A B/A 00 B/A ro ro ro B/A B/A ro ro B/A B/A O O l O O O O l 0 3 0 0 0 ) o i O ) U - ^ ^ . - ' • r O - ^ - ^ h J r O c o ->•->• ->-->-roro-^->-ro->-->-->--1-C O C T J O O C O C O C O C D C n C O - t ^ - i r O r O r O M - ^ M - i o - ^ r o Q o u o ^ _ i . _ > . l s j - > . - > . r O r O r O - » - - » - - » -4 ^ 0 ) r 0 C D S - i U O 0 H 0 0 ) _ » . 0 _ j . _ > . r O r O r O - > - - ^ r O - > . CO * r O O O - ^ C O - ^ - v l r O - > - C O J ^ . C 0 C n O C D - » - O C D C 0->-a) Ol - » - - » - K ) - » - r O r O r O - » - - » - r O - » -r o r o r o r o r o ^ - v - i - r O r O r O r O r O - ^ r o - ^ C n c O O D r O C O r O r O - » - C O C O C O o _ > . _ i . r o ^ r o - » - r o r o - ^ r o - > -U l 0 0 O 5 ' - i l D W - i N 4 ^ 0 0 co' zr ->• f n f n r o r o r o ->•->• r o ->• • v j c n u i r o r o -r^ c o a> _ > . _ > . _ > . r o r o t o ->• r o ^ CD CD —^ CO CO CD CO -&>• CO co c CD' o > CD O 03 <: CD CD CO r o CD to I—t-CO > c/> I—*-I o of CO > —I CD CO > —I CD _ co O c n > —I CD _, to O CD » cr M z i - i O i-h 3 ro 5" r o o I .cr1 D-£0 (0 3 o i-t-cn ro _k o o O o o o o o o o to oo en cn 4k CO ro k >ro;>;>>roro;>>roro 03>D0O000>>00D0>> o> OS cn _k OS CO cn cn o cn •vl o cn o o p ro cn © O l ro to o O l CO b CO 1A b b 'cn b o » o » "vl CO -v l o "vl o CO . A - A -k o _k o to CO CO cn v j o CO OS y4 4> OS cn ;v l CO cn p ro CJ CO CO cn CO o CO 08 b 1A b CO CO CO o co o CO CO •vl •vl o 00 O l o to u> OS CO CO v j •vl OS 00 cn CO 4k 4k ro p O l p ba ro co bo CO OS cn co to ro 1A ro * cn IO OS 4k "vl cn OS to Q . ro -vl CO cn cn ro cn 4k CO — ^ a cn 4k cn p 4k o 4k Is i s "o Ik 1A CO lo. b 'cn 1A CO o o o C J •vl o o -vl 5' . A . A _k 55' _k o o _k 4k 08 cn vj . OS O l 00 o p 4^ 4k en CO 4k O l ro 4k Is OS O cn 1A CO O l b 1A CO - j CO o - J . -vl CO o •vl -v l OS o> -vl oo 08 OS 08 •vl -vl v4 4k OS to ; v i cn 08 O 4k p p b 4k ro b cn CO CO b to io en 4^ cn ro -vi cn cn O l CO 4k _k OS 00 ro 08 cn O l 08 ro cn ro p p cn ro CO CO ro b cn o LA CO O l cn b CO CO o * o -vl CO - i -v l CO CO CO _k _k o k o CO CO cn ~ J . k O l •vl 4k — ^ ; v l 00 ~«i ro cn p o co b CO Lk o O 08 o b b In o CO -vi o o CO o o ~ - l o -vl cn OS to oo to •vl -vl 08 -vl OS CO to CO CO o _k 08 OS CD b *4k b •v l cn b i b 4k cn » cn CO o o cn O l cn CD O 3 0 O 3 & 1 ^ fl) fl) § CO » O cu > 3 3 t fl) 4^ fit 3 CD O fl) > 3 o 3 14 cn O > 3 CD fi) 3 3 o a> B> 3 . A _k o o o O o o o o o - A o co oo ~g at O l 4k CO ro k > ro > ;> ro ro DO DO DO > ro ro ro > > 00 DO . A CO 08 -vl k k OS k en -vl o 4k p ;v | o 4k ro O l p O l — * ro p CO b b ^vl 1A 4k CO b i CO b b CO o C J o CO o CO CO C J •vl o ^ _k CO k OS -v l ro o ro •v l to CO 03 OS ro 4k CO p »k cn v j ^vl p ro b i b b to b i b 4k 1A b ro b o v j CO CO o v j CO o v j CO -vl 08 CO OS CO OS •vl -vl cn — * vJ. CO ID o 4^ p 4k p p O l "vl N> In b -vl b i b i CO b i b 4^ — * ro o OS CO o -vl en 4k cn o »k _ ^ CO OS OS C J o 08 CO cn O l o OS p ^ i 4k p p p ro 00 00 ro p CO CO b i b 1A •v l 00 b b b CO CO CO o -vl o CO o o o -vl CO o ro OS •vl ro CO CO en CD CO CO — * CO O l p p p ro to — ^ p CO 'cn b CO CO CO b CO b 4k CO •vl o o CO CO CO •vl o o O CO CO _k 08 •v l CO O l 00 CO o CO cn •v l CO 00 a to — ^ p ro p •vl 4k cn ro CO b b 1A 1A 4k CO 4k 1A b i 1A CO o C J CO CO o OS ro »k O l o _k OS CO OS 4k ro to CO o> _k •vl p p — * o 4k o 4k v g ro ro to b b b i b b 00 b 'co b io io -vl o o CO -vl -v l o CO •vl •vj o _k . A _k o ro CO oo CO CO 4k OS 08 4k CO 4k p ro p to en ro - A v j — ^ cn Ik b b i b b 4k b b b CO cn -v l -v l o CO o o CO o •vl CO -v l ^ i OS 4^ 00 CO CO CO •vl 00 C J p p to 4k p p ro CO ro 4k 4k 1A •v l b ro b ro io 4k b 4k ro — * o CO en ro — * o 4k -k -vl o cn CO 00 CO en O l CO •vl ~J ro — ^ 4k 4A p p p p — ^ CO b b i b i b CO to b b b CO * -vl o o CO CO CO o v j _k o k _k 00 4k CO 4k OS 08 ro CO C J 4k o p ;v i ro ro CO C J o CO b 1A b b b b b 4k b b i CO o o o CO o o o o CO -vl cn ro CO •vl CO OS CO 00 en 00 -vl 4k p ro — ^ p O l _k v~l o to b ro 08 1A CO 4k "vl CO b 1A ro 1A 4^ » CO O l CD ro — ^ CO O l ro •vl CO c CD" o CD O fi) > 3 ro 00 CD O fi) > 3 O -* 3 3 fl> B) 3 o > 3 3 CD fi) 3 O -% 3 3 ro u H fi) M sr > CO CO > 3 § 5 S o e Pre Ne Wa: Va Pre CO c sr 00 03 EU ral o c ai n ral c Ul n fD 00 O Cn I O w M N " S Ul Ol M U l to OJ r o OJ O OJ OJ GN o sO O OJ OJ t o so OJ o \ ] i—i OJ OJ ro O U l O l 4 ^ OJ O l O N O OJ s o ro O l t—» b i b i ro b i b i 0 0 sO 0 0 b i s i b i b i sO b i ro b i 0 0 b i b i b i b t ON b i i—i ,_, i—i i—i i—i i—i U l NJ ro o i—i 0 0 i—i s ] U l 4 ^ o o OJ i—i ro sO S J OJ 0 0 o so o O l Oo s o b i b l U l b i b i b i O N b i b i OJ b i b i U l U l b i U l b i b i b i >~i t-i ,_k h-* ro OJ o OJ . C o i—1 Cn ON 00 O l U l w 1-1 a s i ro M p Co CO Ln Ln Ln cn to cn CO Ul SI CO o i ce ^ O l N ] OS Cn Ov \ 1 00 W \ 0 O Ln Ln Ln vo Cn Ln g - -. O i—' Ul OS CO 9 s o ui cn u i Jv CO 00 CO 0 1 t ° cn ui cn Ul fO VI Ul vO Ul ^ 2 M » * • Ji cn cn co co u i Ji vo o Ln Ln co o cn M os u i ? vo v] cn to V O Ul sj to Ln _ to 00 vo i—i CO Ul O S Ln Ul CO 1—1 , . CO to o ' 1 Ji Ji Ln 1 CO o 00 to 00 00 to 1-1 sj CO o Ji Ji Ln CO ,_, CO 1—1 Ji Ul Ov ON Ln vo Ln ro ro ro o O l OJ OJ cn n O x Se cr. Si, o rrj o 3 P -< P -i-i P -&> sr ^ >° vo Cn Cn os s i vo Ln Ln h-i °} oo to to o sj Ln co oo bi s i oo o ui Co co Ji o Ji OS •-. i-i M >e M OS to Cn CO Ov Ul vo OS SI s i to OS Ul Ln o Ln Ln Ln Ul O S Ji Ln CO cn P ^ to cn cn vo to to CO CO ^ b i o i vo to co O S ] 00 1—1 Ji vO OS Ul OS ji OS vO CO 00 VO o Cn Ul Ov Ji o to HM to ji VO CO OS 00 VO Ln o Ln Ln ON Ln Ln OS bi Ln O Ln b i to u i >—i Ul Ln to b i Ln CO CO Jsl j i Ji . to OS Ul >_i I— 1 to to P° P 1 O Ol Ul SJ H-1 Ji s j i—i i b i b i Ln to i-i i-i Ji OS to j i Ji sj "—1 j i sj to Ln bi Ln Ln vO bi oi i—1 3 % S P 0 0 U l O l oi oi ro to s j Ul o CO CO s i NJ Ji o V0 Ol Ln Ul s i Ul Ln Ln CO s j •—i OS OS O Ov CO OS s i 00 cn Ln O to to bi Ln s0 Ol sO b i b i b i Ol so oi bi S P O l CD O l OJ OJ ^ sj O P° oi ro oi h-l i-i —1 i—i to I—i o CO s i CO Ji Ji Ul o 1—1 00 vo to Ol OS Ji ro to CO I Ji Ul SJ CO oo VO Ji 00 b i to Ln b i 00 bi vO Ln OS Ul bi b i CO bi b i b i vo ro i—i i-i i—i o i-i CO vO Ol s i Ji Ul Ov ji Ji Ul Ul •—1 CO VO 00 CO S J Ul to CO 00 o O ji s i Ul ro p OS u i 00 Ul Ln b i Ln b i b i Ln ro Ji bi Ul u Ul —1 ON Ln Ln VO Ul OS CO 00 o t-1 i-i to j i VO s i Cn CO CO OS s j Ji s i CO b i b i b i Ln Ln Ln CO s i OS CO Ji Ji b i b i b i j i OS vo >° O SI Ol OS vO N L O oo Ol Ol 00 co N M Ji co w CO Ul Ol 00 Ji Ul b i b i b i CO Ji CO Cv to 00 VO SI Ul Ul Ov vo Ol Ol 0 0 Ji ro o i vo o i ro o i o i SI N O o vO U l ON Lo u i u i vO o vo OJ IO fo N vO U l £ s Ln Ln vi _ Ln M ON 00 |3 1- £ 1-1 v*> 1-1 00 ro •-i U l UJ i t ro 1-1 U l i—i U l Ln ON ON Ln Ln Ln vO >-J { , ro ro VO i—i ife OJ ro Ln Ln UJ o VI i-i 1-1 VD vO i-i , ro 00 00 OJ o O J Ln Ln Ln Ln o 1-1 i-i ro o } , o ON U l VO ro 1—1 it* Ln Ln Ln *. VI U l i-J Ul 31.5 34.5 OJ ro 110.5 ro o OJ v j OJ >-> ON vO JA v i O l O l Ln Ln Ln Ln OJ 00 i-i i—i OJ J> 00 00 v i OJ O l VO ON o O l i C OJ ll> o 101.5 00 vo OJ 00 45.5 96.5 ON ro ro O l 0J 0J 00 65.5 ro v i OJ ro ON l-l OJ 00 ro Ln Ln Ln Ln ON OJ Ln i-i Ol vO Ul ro Ul J> J> VO i—i 00 ON Ul Ln Ln Ln Ln Ln H-l OJ i—i V) OJ 1-1 o v l Ln Ln i_i O l t , 00 OJ o it. Ln it^ Ln OJ o 00 00 Ln U l U l OJ vo NO JA Ln o OJ vo VO o po v l Ln Ln Ln t-i ro o o vO OJ 1-1 ro 00 U l v l 1-1 ON it» o ro VI vo Ln 1JA U l 1—1 v l o ro o v l O l Ln U l v l o ro o 00 vo Ln Ln 1-1 v l i—i o OJ • J IO ro o u i vO vO u N ? O Ol Ul i—i ON OJ U l ON Ul Ul u i ?° ro u i u i ° N i-i VO iti U l i-i JA Ul o 129.5 OJ Ul ON ro i-i o Ul OJ ro ON Ln 58.5 vO Ul OJ Ul OJ Ul V ] ON O J ON ON VD ON 1-1 OJ ON ro Ul IO VI Ul o Ul ON • V I v i O J K 53.5 ON 1—1 OJ Ul 00 Ul 69.5 1-1 o Ul OJ v l 1-1 Ul VO 1-1 o NO OJ ON 00 Ln ON vO 1—1 ON OJ v i Ul V ] vo OJ vO OJ Ln Ln Ln ON ON vo lf^ Ln OJ 00 00 ro o I >> & i v > • i-i —i > 0 3 S < » EL jo 00 o 5 N P N bi o bi rfv ui (Jl O VO U l 00 po U l O U l 00 H J v o U l U l U l p ° H J _ > U l H J o ca > 7 z a £L U l O U l V I Ov CO v o N J U J U l U l U l N N U vO 00 > CO w > ? Z J < 0> t±L-£. oo o S U l 00 Ov O Ov b i v i u i u i H J 00 VO v i U l U l U l vi VO V I U l VJJ 00 O U l v o U l 00 U l U l > cd Bi E L a> oo HT <= o ft ED E L i» oo ? c N J H J U J V I H U l ~j y > IN , j ; , ^ C N v O U J U l N J i lA U l U l U J U l U l 00 v O H J U J N J H-" 0 0 P U J W W ? U l U l U l H J 00 U l V I _ U l 00 U l U l U l U J U J N J <*> W U l U l U l N J U l N J N J N J o * M U l U l U J N I H J U l N J Ov H J ^ H J O U l J A 0 J H J ^ H J O J U l 00 U l U l 00 V I V I 00 Ov V I N C > ? U l U l U l 00 Ov Ov OJ V I i t * u i cn b i U l H J i U U l U l U l H J ><>• ^ H J >C V I U l U l OJ OJ >IA b i NJ b i OJ H J b i p b i u i OJ =L S •z s < £9 i ? 2L 5- !£. oo & -a (T) 0 3 00 s ^ e 3 HJ 0) O J H-i-o &) n O >-t fD N J O U l it itv J> hJA o O l I sr I fr Si. < e Co X I N J O o Pu so sr P H J U l VO O V I I - 1 00 0 J O l O J O l v l iJA U l 00 U l v o O J v l * - * • H J U l U l M U M U l U l U l 00 oo — ,_ , 00 VO H J U l U l U l U l N J HJ O U l U l 00 OJ J A U l U l N) O JA OJ _ 0 J O J O l H J O J 00 p u i b i vo N J >4A O v I O ^ b i u i H I U l U l U l N J V I b i b i v i H J H J ? » o vo cn U l VO V I U l U l >JA 00 VO VO b i u i u i N J 00 U l U l U l P N N J U l U l |JA V I v o U l U l y i OJ b i u i ov 0 J 0 J Ov 00 b i b i o 00 00 U l V I U l b i b i o v v O v O P O Ov U l 0 J U l U l U l V JA U l N J VO V I ON U l U l U l VO M O J U l U l U l O J * . U l U i l l 0 J _j ? O vo O l i t * N J H J rfA 00 OV O l U l V I H J N Ov O U l V I J A VO v j N J U l O J U l V ] VO U l VO U l O U l O l J A U l N J U l U l 00 N J 00 U l U l U l U l U l Ov OJ NJ VO JA U l U l H J N J U l U l Ov H J y i Ov U l U l U l N J )° JA V I U l 0 v VO _ O b i o b i U l 00 O l U l Ov b i v i u t H J OJ CO M T b i b i OJ O l J J O l b i b i b i N J O l b i b i JA O O l N J ,_ i 00 O ( J l M 00 V I V I b i b i b i N J N J ^ N JA O l U l U l W H J N J U l U l U l NJ U l O VO vO bi s] j i i—i i - i ro 0 0 s i 23.5 OJ ro 12.5 ro ov 55.5 ov Ov o bi s i 53.5 3.5 s i o J i 1-1 j i Ov OJ J i 03.5 VD 0 0 sj. ro *-h-l bi o i—i ro a- vD ro JA s i ro Ov bi O l O l . J i Ov p bi OJ O l J i OJ si OJ Ov J i OJ ro vo vO O l 1-1 vO b i s i OJ O l O l i—i o b i O l ro ro bi O i 0 0 bi OJ OJ Ov O l s i Ov s) ro OJ Ov VD OJ b i 0 0 0 0 b i OJ VO J i 0 0 OV ro J i s i vO vO bi s i vO b i O l O l b i 0 0 b i j i OJ o ro J i Ov OJ s i bi s i ro bi s i bi i - i OJ 0 0 O J i O l s j OJ vO b i Ov Ov »—i Ov O l O l J i 0 0 b i 0 0 b i 0 0 o b i O l Ov b i i—i p b i O l vo O l O l ro OJ bi OJ s i O l vD vD o OJ O J i Ov OJ b i O l SI si b i 0 0 vo 11.5 i—i o Ov vo bi O! s j s j b i SJ ro J i b i OJ O l b i ro O l OJ o bi Ov vO CO s i bi t—i OJ bi s i o ro Ov O J i O l s i o i - i i - i j i i—i o OJ b i vO bi vO s j 1-1 j i Ov Ov b i Ov ro p b i J i ro bi ro ro O l Ov J i bi Ov s j o j> bi s i o o O l po b i J i 0 0 j i i - i ro o 0 0 ro VO 0 0 sj OJ bi J i j i VO s j OJ OJ OJ O l b i ro j i ro OJ bi s j O b i s j ro bi b i OJ bi ov O l OJ * O l o b i O l OJ i—i OJ 1—1 O l s j vO s j b i s j O l j i j i b i s j b i SI ro O l ro oo s j ro VO 0 0 . 0 0 bi J i 0 0 o bi OJ O l ro ro O l o O l VD b i 1—1 OJ 1—1 VO OJ 0 0 SI b i Ov s i b i 55 b i s j OJ b i ro •-1 s i ro b i O l b i IO Ov vo OV O l ro JA' bi Ov _ i j i ov Ln * O l I—1 Ov J i 1-1 b i i—i i—i SI 10.5 vO s i i - i b i it* vo Ov ro s j b i OJ j i bi VD bi OJ s j s i o o bi j i o J i s i s i VD i—i ro Ov ro OJ o 0 0 b i s i ro b i O l i - i b i SI Ov ro ro bi si i - i Ml 1—1 J i s j o 1—1 1-1 vO O l io bi O l o O l vO i - i ro J i b i o VD vO bi h-l VO bi j i s i bi Ov i—i b i . ro J> bi 0 J o bi 1—1 vO vo s j i - i bi Co J i o bi OJ bi o b i i—i o VO J i OJ Ov OJ i - i ro o vO O l 0 0 bi vo SI ro O l Ov I - i o i - i p b i ro Ov bi ro 0 0 bi i - i J i Ov bi s i OJ O l OV bi o j i bi i - i bi s j OV b i O l J i Ov SI i—i j i Ov Ov i - i bi s i i—i i - i s i i - i bi O l S] b i Ov b i 0 0 OJ i - i b i j i j i O l bi •—j j i b i 0 0 ro O l ro bi OJ o b i O l OJ O l ro 0 0 vO b i O l vO Ov s j VO Ov vO OV O 0 0 SI b i ro VO bi j i ro bi O l OJ Ov VO Ov O l bi o bi Ov o bi j i O l 52 s j 1-1 O l SI O l O VO b i l - i o s j bi Ov i - i b i VO b i s j b i ro ro j i Ov ro ro i—i sj O l VO O l o bi j i b i b i 0 0 Ov O l ro 1—1 1—1 o 1-1 J i SI Ov O l 0 0 1-1 o b i b i 52 bi i - i o b i O l b i ro vO OJ ro bi i—i s i bi VO Ov i—i O l o bi OJ Ov o 0 0 O l O l o b i 0 0 ro 1-1 Ov O b i 0 0 ro 0 0 i—i s j j i 0 0 b i vO O l b i ro Ov OJ vO 0 0 b i s i Ov J i bi O l OJ O l ro b i Ov SI j i s i b i Ov O l vo bi vO oo ro bi O l O l b i SI b i Ov ro o OJ o 0 0 ro bi O l Ov b i Ov 1-1 i—i i—i ro ro OV OJ i - i o 0 0 b i i - i o b i s i ro b i s j O l Ov O l b i Ov O l 0 0 s l bi O l J i bi Ov bi i - i j i i - i b i I ! vfl 2 i No Ov Ov b i si —1 vO z rt) C Ln > ft O c v i 5° o> OJ M J O OO Is) ">1 o \ N J >0 OJ N M bo O U l C O s D N J 23 v o ">> VO O l NO O Ln vj Lo £ vo S i—i O l Ov Ln ON fc vi r-> oo ° » 8 i O O 00 ^ ^ s \ j N J vo Ln Co Ln \D ON 5 N J ui oo O l W s j N J v i vo bo v i a fc v v i '— N J V J N J fc O J O J Ln C O OJ fc v O _ N M n o 03 O O VO 00 > 00 0 3 > Z 5 < P o7 SL E Si. oo 0) o w £. oo o ft c "T3 0 3 OJ NJ Ov O J VO * • N J ^ O J O J O J O J 00 O J O J N J S a 2 N J O l O J O J O J co N J U l P* V I v o O l O J C-i £ O J r a o O l to * v OJ is O l O J O J O J o oi O J „ . Ln NJ Ln O J O J Ov O J N J VO O l O J O l O l 0 0 v j 8 w r v cn fc U l Ln O J fc P-i O J • v i U l v o 0 0 o v U l Ln U l Jv, P >D U l U l U l U J O J O l O l Ov V ] O l VO N J JV. v^j v5 r1 Ln U l O J O l 0 J N J i—1 V O O I J J K I H O I - I H N S ^ N J ^ P ^ N O O ^ N ^ u i w ^ o i u i ^ u i o i N J £ 00 oo oo Ln O J Ln NJ N J U l O J Ov - H J O U l O NJ N J jo •• v O VO fc ' O l & co ^ o oi NJ N J N J Ov V l o v 00 N J N J NJ N J N J fc V I i - i 00 h-> V I N J Ln v i v o N J N J ->*" N » N J ON fc 0 J O & 0 0 N J > - i U l 0 0 ON fc N J Ln 0 0 Ln N J K-l O J U l O l N J NJI _ O J v j fc O l O l * • o v i > o Ov 03 Z c : < . era" Z S M y c 0) O ui — C — i - ui 0 3 ^ ~ 3 Qi C n —• c O J 00 p J i b o i J i o v O u i H i o d i — i H A C O co V O V I O J O J l — i O l O v v l N J V I O l fc N J ! *~ V ] H-> O l Pi 1 0 „ N io r N J U l O l O J O J O l O J U l >—i fc P 0 0 >*• 00 fc fc N J N J *• v , v O ON OJ N J N ) C O O l V O i - i O J O J o O J oi N J O J O l Lo 141* v o N J ON Lo >i> O J o N J fc VJ Ov N O N J Ln v j v o N J N J U l , , r v fc 0J P • 0J fc U l N J N-> £ O J 00 oi oi N J OJ H- l 00 O N V J i - i W £ 2 -oi N m CO M J i O J fc o P 0 0 00 * M W N vo ° ON C O N J O l N J V I VO -<] O J A O ON H J NJ ON N J O N ^ V I v o v o H J N J Ov N J 1—1 P O 0 0 V 0 VO H J V I O J oo NO N J H J N J H J 00 v l N J H J fc v j OV ON Ln C O v O N J 1^ H J N J v j O l V I U l V I N J H J 00 VO " V J O J O J VO V I N J N J O v H J O J O v o O v O v i VO v O VJ Ln _ N J H J ON O J 0J J A 00 N J ON VO fc O N O J O J 1 N J ON O J O l O J 0 O J 0 1 V ] fc ON fc o „ _ _ Ln O J Ln fc _ N J N J N J H J N J O J fc VO v o VO O J O J H J ON O l V I U l CO ON ON i-J Ln Ln co Ln O l * -ON v o > < I? Z 0) — UL 00 0 v. -7 Z S O J ON H J H J 0 J " S K & N J N O K O I ^ O J O J S O l O J O J V I oo » O l N J O l J J _ J O l O J O J — v j U l O l V I o i O J 0J H J v -Ov H J OJ Ln ON Ln OJ ro O J a vO fc >C J A VJ O l N J O J C oo ui N VO U l O l O J O J O l O l U l J J O J U l ON O J O J V O V] O - A O J ON v l O N CO N J • Ov ON U l 00 NS O N ~ N J fc v C Ln U l Ln s -oi ui OJ fc fc v i oo J J O0 0 0 Ln OJ ^ HJ OJ U l OJ U l N J ON —i P P H J O l U l N J o 0J H J v O Ln O Ln — O l — ON fc p° 00 U l 00 CO ^ j Ln O o i vi P Ln fc v o O J ON ^ Ln U l O i VI H J Ln U l OJ XT a, 0 — c o O J > •x 0 3 J ON O J C OJ ON — bi ON O Ln bo Ln •r oo O l O N J A N J ~vl >-± O J v O O J Ln O l O l o b o O J fc o fc Z 5 < £ i" co o N J _co > 7 z 0 3 £ H( OJ 01 n J " rt) N J 00 O J V I H J H J O J O O 00 OJ b o b i b o fc * H J N J ON 00 O J 00 fc H-l O J N J O J » H J O J O J O O J KI p o ? o cn oi 00 v l N J H J O i ! i v m J O PO ON fc 00 V I 00 Ln fc V I fc N J >-t 0> > < HP a 00 ifc Lo ON v i O l or 2L K. oo o ft v j o v j N J ON O J v j v i ON O J O J Lo O v U I U l ON H J fc V I U l N J VO ui 9? V O O J ° N N J NJ fc O J O N O J ON r* > CO ID Ov fc 00 Q N V J U l o N J v l O J fc fc H - l O l fc Ln N I g; oi v o u i o ON N J O J v i C O V I 0 0 85 NJ o S C O 00 O l O J ON C O VO co j i NJ H-? O J t i O J O J V J v i ON P O 00 to ^ -I CD Cu » li 3 OJ X o o 03 § a. § ST I p-CD sr N J VO O l H J O l p o o O J fc O J U l O J CO N J CO O J N J p i H J ON H J H J O J O l H J N J oi1 H J fc H J N J N J O N N J O N J U l H J O l N J v O 0N co v e 3 fc fc O J N J V J v l ON N J N J O l U l o H J ON o fc 00 i—i fc CO o N J i fc O J fc v O O N VO N J b i Ln i o v O i - J Lo v ] U l 00 O l Ln U l O l Ln U l i - J b o U l ON VO * 00 456.9 v j v j O v vO O 00 CO o ON O J co V O CO N J o N J fc H J N J V O O N J Ln N J H J U l N J O J v j O N J ON v o v j H J ON N J N J H J V I H-1 N J CO U l fc v o CO fc O l o fc o j - fc * — i p U l O l O l N J O l CO fc O J v l O N H J 0O O J b o fc v j oo N J v i N J VO ON O O l co oo N J CO U l 00 ON SB" £ fc 0 0 v j C O N J H CO u i N J O l v j v O N O O J O l O l O l CO VJ N J v fc v O O l O l H J O ) M H J O 00 VO ON V I 0 0 o fc H ^ 1^ I 1 H J N J N J KJ V ] fc co CO O l H J » K po O i N J fc O N N J O N O l V O ° ° b i O J u i u i O J ro U l O N VO i - i O J ^ v O N J ON U l ON vO H J N J N J N J fc O l N J * ; ON o o CO H J VO V I N I CO O l c CO i—i V I U i ON b i OJ ov OJ Ol O NJ — ) ° HJ b i b i vi ON ON V j fc OJ v j v i O J P° O l O l oo \Q 00 U l 4 ^ $ CO O N J O J o v O 4ii \ D b i Lo * * v i N J t—' ON fc i — i N J H J v l v l fc ON ON ON fc bo fc b o vO i-J N J N J fc O l v i V ] 00 o o ON v l 00 * ON N J i-1 fc N J \ ] CO ON v O fc N J fc ON fc * bo N J Lo b o _ 00 OJ S J N O N O 1-1 S J l b O N 00 O N b i 00 00 O N s j b> j i 00 00 to 1-1 NO OJ 00 N O OJ 1—1 j i 00 1—1 to O l o to o b i bo ON 00 N O O N O N O N i—i s j J i j i N O N O 00 1—1 00 j i to o N O bo s j i—i po O l 1—1 O N OJ OJ OJ OJ o N O £ to OJ OJ o to OJ o OJ OJ 1—1 N O I O 5*1 I N ) IN) >J *• —• oo b i U l b i to to g O J OJ O N P O l O l O l i-j r° to OJ 00 b i to o O l OJ O l SJ b i OI O l O l 296.5 to Ji OJ 1-1 00 s j OJ i—i to to 00 )-i 0 0 o b i b i b i 329.5 to to OJ O l o b i OJ i—i to to ji OJ o 00 to o SI 1—» S J o 311.5 1—1 S J s j OJ O l O l 1-i a 1-1 s j N O OJ ji O l s j O l OJ S I 00 b i b i b i 373.5 to OJ N O ro 00 O N b i OJ O l 00 00 O l ro i—i i-i OJ o 1—1 ro to i-i s j O l to O l N O i—i N O s j i-i OJ 1—1 281.5 ro to o 1-1 OJ O l b i N O O N 325.7 135.4 135.43 ro ro i—i OJ OJ O N o CO j i O N b i OJ N O OJ ro o o OJ ro i—i i—i l—i O N ro O l bo N O N O OJ ro O l i-i J i s j O i-i OJ o ro ON 00 OJ b> ro O N ro OJ s j s j OJ OJ CO I—I 00 o O N j i N O ro OJ to o J i J i s j s j O 00 NO _ 244.9 391.6 OJ ro OJ 199.8 319.5 578.5 Zl 16.5 269.3 339.5 185.4 134.3 328.5 Ol j i 00 j i b i i-i Ol 224.7 OJ ro OJ 192.7 i-i Ol 365.5 J i SI N O 00 bi 18.5 227.4 365.5 i—i i-i N O 172.1 j i Ol 00 331.5 14.5 19.5 ro s j ro 351.3 158.8 209.7 J i o 1—1 OJ O N Ol SI bi i—i 00 225.8 481.1 189.7 227.4 OJ s i J i 477.5 OJ ro O N 243.8 370.3 163.9 109.9 348.5 O N O N bi 29.5 215.7 268.3 191.9 160.3 385.5 511.5 Ol 12.5 227.8 358.6 127.5 166.4 362.5 OJ OJ OJ Ol 23.5 227.4 330.4 127.5 134.8 j> h-l NO 473.5 10.5 ro o ro i-i ro 406.9 i—i OJ io 143.9 OJ ro OJ O N o to Ol b i 29.5 220.4 375.1 304.4 146.2 315.5 352.5 10.5 ro ro 214.1 353.3 152.5 109.9 OJ O N 355.5 j i b i 25.5 179.6 307.1 170.5 108.2 to s j ON 364.5 12.5 29.5 213.6 301.8 196.1 130.9 to NO o 329.5 CO bi — _ i ui p) s i v O — o-> — ' ON 2 £ ro O J 1 — 1 o ' Co oo o-U l O N O J O N u i ! ° s j J i >0 O J to P 1 NO * . OJ o 1—1 ON O N ji ro ro ro OJ ON r oo *^ O l s j v j O l OJ ON J > 0= Is) J> OJ ~> oo O J j i vi i-i oo ro O l O N OJ O l » OJ j i 00 ro O N i—i io bo J i j i OJ i—i p N O » Kl J i S I J i N O 1—1 OJ o ro 1—1 OJ N O bo ro O l ro ro N O O N OJ j i ro ro i-i 338.4 ro J i O l O J ro o j i N O IO ON O OJ OJ 1-1 * bo b i ro O l OJ OJ OJ 00 N O N O j> j i S I ro O l to OJ ON S I b i OJ O N ro s j po ro O l i—i j i O O l bo N O ro OJ j i ro s j N O j i to N O i-i O l 00 SI O N o ro o N O 1-1 s j O b i OJ N O OJ ro OJ N O J i S I ro s j i-i o ON N O bo * OJ 00 o j i ro s j I & b 5 & a N>5 rs i t s > S J O l ON i-i OJ Ji o i—i 00 A P P E N D I X C : W O M A C Index Section A INSTRUCTIONS TO PATIENTS The following questions concern the amount of pain you have experienced due to arthritis in your study joint(s). For each situation please enter the amount of pain experienced in the last 48 hours. (Please mark your answers with an "X".) QUESTION: How much pain do you have? 1. Walking on a flat surface. No Pain 2. Going up or down stairs. 3. At night while in bed. No Pain 4. Sitting or lying. No I 5. Standing upright. J Extreme lpain J Extreme I Pain J Extreme lpain Section B INSTRUCTIONS TO PATIENTS The following questions concern the amount of joint stiffness (not pain) you have experienced in the last 48 hours in your study joint(s). Stiffness is a sensation of restriction or slowness in the ease with which you move your joints. (Please mark your answers with an "X".) 6. How severe is your stiffness after first awakening in the morning?. 7. How severe is your stiffness after sitting, lying or resting later in the day? Nol Stiftness | O Dr. Nicholas Bellamy. All rights reserved 1996. STIFF6 STIFF7 Section C INSTRUCTIONS TO PATIENTS The following questions concern your physical function. By this we mean your ability to move around and to look after yourself. For each of the following activities, please indicate the degree of difficulty you have experienced in the last 48 hours due to arthritis in your study joint(s). (Please mark your answers with an "X".) QUESTION: What degree of difficulty do you have? 8. Descending stairs. Difficulty | 9. Ascending stairs. Nol Ditficutty | 10. Rising from sitting. No I Difficulty [ 11. Standing. No I Difficulty I 12. Bending to floor. Difficulty | 13. Walking on flat. Difltculty J 14. Getting in/out of car. Nol Difficulty J 15. Going shopping. Ditficutty | Difficulty Extreme Difficulty Extreme Difficulty Extreme Difficulty Extreme Difficulty Extreme Difficulty Extreme Difficulty Extreme Difficulty PFTN8 PFTN9 PFTN10 PFTN11 PFTN12 PFTN13 PFTN14 PFTN15 © Dr. Nicholas Bellamy. All rights reserved 1996. 16. Putting on socks/stockings. Nol Difficulty I 17. Rising from bed. Difficulty | 18. Taking off socks/stockings. Difficulty | 19. Lying in bed. DifTicutty I 20. Getting in/out of bath. Difficulty | 21. Sitting. Ditficutty [ 22. Getting on/off toilet. Difficulty 23. Heavy domestic duties. DifTicutty I 24. Light domestic duties. No Difficulty | © Or. Nicholas Bellamy. Ail rights reserved 1996. [Difficulty [Difficulty Extreme Ditficutty Difficulty Ditficutty | Difficurty [Difficulty PFTN16 PFTN17 PFTN18 PFTN19 PFTN20 PFTN21 PFTN22 PFTN23 PFTN24 THANK YOU FOR COMPLETING THE QUESTIONNAIRE