Medial opening wedge high tibial osteotomyA prospective cohort study of gait radiographic and patient-reported outcomes.код для вставкиСкачать
Arthritis & Rheumatism (Arthritis Care & Research) Vol. 61, No. 5, May 15, 2009, pp 648 – 657 DOI 10.1002/art.24466 © 2009, American College of Rheumatology ORIGINAL ARTICLE Medial Opening Wedge High Tibial Osteotomy: A Prospective Cohort Study of Gait, Radiographic, and Patient-Reported Outcomes TREVOR B. BIRMINGHAM, J. ROBERT GIFFIN, BERT M. CHESWORTH, DIANNE M. BRYANT, ROBERT B. LITCHFIELD, KEVIN WILLITS, THOMAS R. JENKYN, AND PETER J. FOWLER Objective. To evaluate the effect of medial opening wedge high tibial osteotomy on gait, radiographic, and patientreported outcomes over a 2-year postoperative period in patients with varus alignment and medial compartment knee osteoarthritis, and to identify signiﬁcant predictors of outcome. Methods. We used an observational cohort study design and prospectively administered 3-dimensional quantitative gait analysis, hip to ankle weight-bearing radiographs, and patient-reported outcomes preoperatively and 6, 12, 18, and 24 months postoperatively. Observed changes with 95% conﬁdence intervals (95% CIs) were calculated. Multivariate linear regression and cluster analysis were used to evaluate associations between patient characteristics and 2-year outcomes in dynamic knee joint load (external knee adduction moment during gait) and Knee Injury and Osteoarthritis Outcome Scores (KOOS). Results. A total of 126 patients (mean age 47.48 years) were included in the study. Mean changes suggested clinically important improvements in malalignment (change in mechanical axis angle 8.04° [95% CI 7.16°, 8.93°]), medial compartment load during gait (change in knee adduction moment ⴚ1.38 [95% CI ⴚ1.53, ⴚ1.22] percentage body weight ⴛ height), and all KOOS domain scores (change in pain 23.19 [95% CI 19.49, 26.89] KOOS points). A small (13%) increase in knee adduction moment was observed from 6 to 24 months postoperatively. Few preoperative clinical and/or gait characteristics assessed at baseline were signiﬁcantly associated with 2-year outcomes. Conclusion. A medial opening wedge high tibial osteotomy with correction to approximately neutral alignment produces substantial and clinically important changes in dynamic knee joint load and patient-reported measures of pain, function, and quality of life 2 years postoperatively. Changes in knee adduction moment observed in the ﬁrst 2 years postoperatively should be explored as potential predictors of longer-term success and subgroups of patients with poor outcomes. INTRODUCTION Knee osteoarthritis (OA) ranks among the most common disabling and costly health conditions in adults, and its prevalence is projected to increase sharply over the next 2 decades (1–3). There are currently no interventions proven to alter the course of knee OA, leaving total joint replaceSupported by the Canadian Institutes of Health Research and Arthrex, Inc. Dr. Birmingham’s work was supported by the Canada Research Chairs Program. Trevor B. Birmingham, PhD, J. Robert Gifﬁn, MD, Bert M. Chesworth, PhD, Dianne M. Bryant, PhD, Robert B. Litchﬁeld, MD, Kevin Willits, MD, Thomas R. Jenkyn, PhD, Peter J. Fowler, MD: The Wolf Orthopaedic Biomechanics Laboratory and Fowler Kennedy Sport Medicine Clinic, University of Western Ontario, London, Ontario, Canada. Address correspondence to Trevor B. Birmingham, PhD, Elborn College, University of Western Ontario, London, Ontario, Canada N6G 1H1. E-mail: email@example.com. Submitted for publication April 22, 2008; accepted in revised form February 3, 2009. 648 ment for end-stage disease the most widely accepted surgical treatment option (4). The substantial burden of knee OA provides strong support for implementing treatment strategies earlier in the disease process. High tibial osteotomy (HTO) is a biomechanically focused surgical intervention intended to improve pain and function and delay joint degeneration in patients with OA located primarily in a single knee compartment (5). Studies suggest that HTO procedures performed using a variety of techniques can result in large improvements in self-administered, kneespeciﬁc questionnaires (5–7). For example, when converted to overall scores out of 100, patient-reported status can increase from a mean of approximately 30 points preoperatively to a mean of approximately 70 points postoperatively (6,7). However, concerns regarding variability in outcomes among patients and the duration of these improvements exist, and administrative databases suggest reluctance in North America to perform these procedures (8). HTO alters knee alignment in an attempt to redistribute the weight-bearing load on the tibiofemoral joint. Classi- Outcomes of Medial Opening Wedge High Tibial Osteotomy cally, a valgus-producing osteotomy for varus gonathrosis was performed by removing a lateral wedge of bone from the proximal tibia to correct coronal malalignment (9). During the past decade, interest in a medial opening wedge valgus-producing osteotomy technique has increased, owing to claims that it offers improvements over the lateral closing technique (10). Potential detrimental effects of HTO include delayed union or nonunion and inadvertent changes in the posterior tibial slope and patellar height (11). Optimal indications for surgery, particularly age and extent of disease, and the amount of correction in the alignment necessary to adequately unload the medial compartment are controversial. Despite the proposed beneﬁts of medial opening wedge HTO, a very limited number of prospective cohort studies (12) evaluating this newer technique exist (6). Recent studies evaluating mechanical risk factors for progression of knee OA conﬁrm earlier hypotheses regarding the important role of malalignment and distribution of tibiofemoral loads (13–17). In particular, the magnitude of the external knee adduction moment during walking, measured during quantitative gait analysis, has emerged as a valid and reliable proxy for the dynamic load on the medial tibiofemoral joint and a potential risk factor for disease progression (18,19). Although decreasing the load on the medial knee compartment is the primary rationale for valgus HTO, we are aware of few studies with relatively small sample sizes evaluating changes in the knee adduction moment postoperatively. Published studies include 5 cohorts ranging in size from 7 to 32 patients, a variety of HTO techniques, and one postoperative gait analysis, often performed at variable lengths of followup (20 –25). Together, these studies provide convincing evidence that valgus HTO procedures produce statistically signiﬁcant decreases in the knee adduction moment, although the magnitude and stability of these reductions and potential changes in other gait characteristics known to affect the knee adduction moment remain unclear (20 –25). Additionally, previous studies evaluating whether preoperative quantitative gait analysis provides useful predictors of clinical outcome have produced contrary results (23,24,26,27). Our objectives were to 1) evaluate the effect of medial opening wedge HTO on gait, radiographic, and patientreported outcomes over a 2-year postoperative period in patients with varus alignment and medial compartment knee OA, and 2) identify signiﬁcant predictors of outcome. We hypothesized that medial opening wedge HTO would result in substantial, clinically important improvements in dynamic knee joint load and patient-reported pain. We also hypothesized that preoperative patient characteristics would be associated with these improvements. PATIENTS AND METHODS Study design. We used an observational cohort study design (12) to evaluate patients undergoing medial opening wedge HTO. Four surgeons (JRG, RBL, KW, PJF) at a tertiary care center specializing in adult reconstruction and orthopedic sports medicine surgery participated. We prospectively assessed gait, radiographic, and patient- 649 reported outcomes preoperatively and at 6, 12, 18, and 24 months postoperatively. Data were collected between November 2002 and December 2007. Patients provided informed consent to participate in the study. The study was approved by the Research Ethics Board for Health Sciences Research Involving Human Subjects of the University of Western Ontario. Patients. We recruited patients presenting to this center for treatment of unresolved knee pain located primarily in the medial compartment. Patients were referred from family physicians, rheumatologists, and primary care sports medicine specialists for consultation with an orthopedic surgeon regarding treatment options. Inclusion criteria consisted of mechanical varus alignment and knee OA according to the American College of Rheumatology classiﬁcation criteria (28), with the greatest severity in the medial compartment of the tibiofemoral joint. Patients with concomitant disease in the lateral compartment were considered eligible as long as pain and radiographic disease were more severe in the medial compartment. Patients with concomitant chronic anterior cruciate ligament (ACL) deﬁciency undergoing simultaneous ACL reconstruction were included. We excluded patients age ⱖ60 years with grade 4 degenerative changes in ⱖ2 knee compartments because they were considered better candidates for arthroplasty. Other exclusion criteria included inﬂammatory or infectious arthritis of the knee, end-stage disease in the patellofemoral joint, prior HTO on the contralateral extremity, multiligamentous instability, major neurologic deﬁcit that would affect gait, major medical illness with a life expectancy ⬍2 years or with an unacceptably high operative risk, pregnancy, inability to speak or read English, and psychiatric illness that limited informed consent. Intervention. Using an opening wedge osteotomy system and a nonlocking plate (Arthrex Opening Wedge Osteotomy System; Arthrex, Naples, FL), the surgeons used a similar operative technique to that previously described by Fowler et al (29). The desired angle of correction was calculated preoperatively according to the method described by Dugdale et al (30), with the goal of creating very slight valgus by moving the weight-bearing line laterally up to a maximum position of 62.5% of the medial to lateral width of the tibia, depending on the magnitude of malalignment and status of the articular cartilage in the lateral tibiofemoral compartment. The osteotomy was performed using both ﬂexible and rigid osteotomes below a ﬂuoroscopically placed guide pin, then opened slowly to the predetermined width for placement of the plate that was ﬁxed proximally and distally with cancellous and cortical screws, respectively. Cancellous allograft bone was used in osteotomies ⬎7.5 mm. Postoperative management included use of a hinged postoperative knee brace and feather-touch weight bearing with crutch use for at least 6 weeks. With clinical and radiographic evidence of healing, partial weight bearing was permitted at 6 weeks, and full weight bearing was permitted at 12 weeks postoperatively. At ⬃3 weeks postoperatively, the patients started exercises in the brace until 650 healing of the osteotomy site had occurred. Goals of these exercises during this time period of limited weight bearing were to limit swelling, joint contracture, and disuse muscle atrophy. Range of motion exercises consisted of active and passive knee and hip movements. Strengthening exercises initially consisted of isometric knee extension/ ﬂexion and hip extension. At ⬃8 weeks postoperatively, non–weight-bearing concentric exercises using weights or Thera-Band (Hygienic Corporation, Akron, OH) were added and progressed until weight bearing was permitted. At ⬃12 weeks postoperatively, weight-bearing exercises emphasizing proprioceptive control and balance were then implemented and progressed until patients demonstrated a normal walking gait pattern. Gait analysis. Gait was evaluated using an 8-camera motion capture system (Eagle EvaRT; Motion Analysis Corporation, Santa Rosa, CA) synchronized with a ﬂoormounted force platform (Advanced Mechanical Technology, Watertown, MA). We used a modiﬁed Helen Hayes 22 passive-reﬂective marker set (31). Additional markers were placed over the medial knee joint line and medial malleolus bilaterally while the patient stood on the force platform to determine body mass, marker orientation, and positions of joint centers of rotation for the knee and ankle. These 4 markers were removed prior to gait testing. Patients walked barefoot across the laboratory while 3-dimensional kinetic (sampled at 1,200 Hz) and kinematic (sampled at 60 Hz) data were recorded during the middle of several strides for at least 5 trials from each extremity. We calculated external knee moments from the kinematic and kinetic data using commercial software (Orthotrak 6.0, Motion Analysis Corporation), custom postprocessing, and previously described data reduction techniques (32,33). Based on the external knee adduction moment waveform, we identiﬁed the peak magnitudes in the ﬁrst and second halves of stance and the area under the curve (impulse), and normalized these values to body weight and height. We also calculated gait speed, toe-out angle, and lateral trunk lean because of their inﬂuence on the knee adduction moment in patients with knee OA (33–35). Walking speed was calculated as the average walking speed between successive foot contacts of the tested extremity. Toe-out angle (positive angle) was calculated as the angle between a line drawn between the center of the ankle and the head of the second metatarsal joint and the forward progression of the body. Lateral trunk lean over the stance extremity (positive angle) was calculated as the angle of a line drawn from the midpoint of the anterior superior iliac spines to the midpoint of the anterior tips of the acromion processes with respect to vertical. All gait variables were calculated by averaging across 5 trials for each patient. We have previously reported excellent test– retest reliability of the ﬁrst peak knee adduction moment (intraclass correlation coefﬁcient [ICC2,1] ⫽ 0.86) with a minimum detectable change (95% conﬁdence level) of 1% body weight (BW) ⫻ height (Ht) (36). We have also previously reported appropriate reliability of gait speed Birmingham et al (ICC2,1 ⫽ 0.92), toe-out angle (ICC2,1 ⫽ 0.69), and trunk lean angle (ICC2,1 ⫽ 0.91) measurements (34). Radiographic assessment. Long cassette bipedal standing anteroposterior and lateral digital radiographs were assessed using custom computerized software (37). For anteroposterior radiographs, the patients stood with the patellae centered over the femoral condyles and feet straight ahead to attain a true anteroposterior image and to control for effects of foot rotation on measures of lower extremity alignment (15). The x-ray beam was centered on the knee at a distance of ⬃2.5 meters and beam exposure was determined based on each patient’s leg mass. The mechanical axis angle of the lower extremity was used to quantify alignment in the frontal plane and was deﬁned as the angle formed between a line drawn from the center of the hip to the center of the knee and a line drawn from the center of the ankle to the center of the knee (38,39). Negative values indicated varus alignment. The center of the hip was identiﬁed as the geometric center of the femoral head using a circular template, the center of the knee was identiﬁed as the midpoint of the tibial spines extrapolated inferiorly to the surface of the intercondylar eminence, and the center of the ankle was deﬁned as the mid-width of the tibia and ﬁbula at the level of the tibial plafond. We have previously reported excellent reliability of mechanical axis angle measurements (ICC2,1 ⫽ 0.97) (37). Other radiographic measures of knee geometry relevant to HTO included the femoral anatomic axis angle (the angle between the anatomic axes of the femur and tibia), the proximal tibial articular angle (formed by the tibial anatomic axis and a line tangent to the tibial plateau on the medial side), the posterior tibial slope (the angle formed by a line perpendicular to the lateral longitudinal axis of the tibia and a line parallel to the medial tibial plateau joint surface), and patellar height (Blackburn-Peel and InsallSalvati ratios) (40). Measures of joint degeneration included the Kellgren/Lawrence (K/L) scale grade (41) in the medial and lateral compartments and medial and lateral joint space in millimeters. Although we have not evaluated the reliability of these additional radiographic measures, the same experienced and independent evaluator measured all radiographic variables at all followup visits. Patient-reported outcomes. We used the Knee Injury and Osteoarthritis Outcome Score (KOOS) to assess patient-reported outcomes (42,43). The KOOS is a 42-item knee-speciﬁc questionnaire with 5 separately reported domains of pain (9 items), other symptoms (7 items), function in daily living (17 items), function in sports/recreation (5 items), and knee-related quality of life (4 items). Each item has 5 response options. Domain scores represent the average of all items in the domain standardized to a score of 0 –100 (where 0 ⫽ worst and 100 ⫽ best). This instrument has face validity and has demonstrated construct validity, excellent test–retest reliability for each domain (range 0.75– 0.93), and has been shown to be responsive to change (7,42,43). Other patient-reported outcomes included the Lower Extremity Functional Scale (44), a region-speciﬁc measure of lower extremity function that Outcomes of Medial Opening Wedge High Tibial Osteotomy provides a single value for function that varies from 0 to 80 (where 0 ⫽ worst and 80 ⫽ best), and the generic physical function and mental health domains of the Short Form 36 health survey (45), standardized to scores from 0 to 100 (where 0 ⫽ worst and 100 ⫽ best). Statistical analysis. We ﬁrst calculated the mean ⫾ SD for all variables assessed preoperatively and 2 years postoperatively. We then calculated means and 95% conﬁdence intervals (95% CIs) for changes after 2 years and evaluated statistical signiﬁcance using the dependent ttest for continuous variables. We also plotted the mean and 95% CI for the mechanical axis angle, the ﬁrst peak knee adduction moment on both extremities, and each of the KOOS domain scores preoperatively and 6, 12, 18, and 24 months postoperatively. We deﬁned 2 primary outcome measures (dependent variables) observed at the ﬁnal followup: the ﬁrst peak knee adduction moment and the KOOS pain score. Linear regression was used to quantify the magnitude of the association between predictors and outcomes. Potential predictors (independent variables) were chosen based on previous reports and consisted of factors related to disease severity and knee joint loading (6,23,24,26,27), including gait characteristics demonstrated to affect the knee adduction moment (33–35). Independent variables included preoperative scores for knee adduction moment, KOOS pain domain, age, sex, body mass index, mechanical axis angle, medial compartment K/L scale grade and lateral compartment K/L scale grade (both entered as binary variables; grade 1 or 2 versus grade 3 or 4), gait speed, toe-out angle, and trunk lean angle. To speciﬁcally investigate the effect of the postoperative alignment achieved with surgery, we also included the 6-month postoperative mechanical axis angle as an independent variable. We also repeated the analyses, replacing the ﬁrst peak knee adduction moment with the area under the curve and replacing the change in the KOOS pain domain with the change in the KOOS function in daily living domain. We then repeated analyses, excluding patients who underwent simultaneous ACL reconstruction. We used exploratory cluster analysis to identify groups of patients who were similar with respect to the 2-year change in the ﬁrst peak knee adduction moment and the 2-year change in the KOOS pain score. We used a 2-step hierarchical clustering model with the unstandardized Euclidean distance algorithm as the similarity index, Akaike’s information criterion as the measure of goodness of ﬁt, and the auto-clustering feature of the software (TwoStep method, Statistical Package for the Social Sciences, version 16; SPSS, Chicago, IL). We repeated the initial analysis twice after randomizing the order of cases. Between-cluster comparisons were then made with the independent t-test, Wilcoxon’s rank sum test, and the chisquare test as required. The cluster analysis method used screened-out patients who could not be assigned to a cluster because they were not similar with respect to the variables of interest. Because of this dissimilarity, these outliers were not evaluated further. 651 Table 1. Baseline demographics and clinical characteristics (n ⴝ 128)* Characteristic Sex, no. (%) Male Female Age, years Height, meters Weight, kg BMI, kg/m2 Mechanical axis angle, degrees Femoral anatomic axis angle, degrees Medial compartment K/L scale grade, no. (%)† 1 2 3 4 Lateral compartment K/L scale grade, no. (%) 0 1 2 3 4 Value 102 (79.7) 26 (20.3) 47.48 ⫾ 9.53 1.75 ⫾ 0.09 90.54 ⫾ 16.60 29.50 ⫾ 4.82 ⫺7.50 ⫾ 4.10 ⫺2.20 ⫾ 4.50 4 (3.13) 32 (25.00) 41 (32.03) 51 (39.84) 8 (6.25) 44 (34.38) 63 (49.22) 11 (8.59) 2 (1.56) * Values are the mean ⫾ SD unless otherwise indicated. BMI ⫽ body mass index. † Kellgren/Lawrence (K/L) scale grade of osteoarthritis severity. RESULTS A total of 258 potential patients were screened for eligibility. Fifty-two patients were ineligible and 22 patients elected to not undergo HTO. Of the remaining 184 patients entered in the study, 36 were lost to followup, 3 died before completing the study, and 2 withdrew from the study. Fifteen patients elected to undergo HTO on their opposite extremity within the 2-year followup, and 2 patients required revision surgery for nonunion and loss of correction. These 17 patients were excluded from the present analysis, although we continue to follow them. A total of 126 patients with 2-year postoperative data remained. Six of these patients had some missing data and were excluded from the regression analyses (no data imputation methods were used). All 126 patients were included in the cluster analysis. Three patients had delayed union, which resulted in missing the 6-month gait test only. Four patients had infections within 2 weeks of surgery that were treated with irrigation and debridement and a short course of intravenous antibiotics. Three patients had delayed infections at their tibial incision ⬃1 year postoperatively (likely following a bacteremia) that required surgical removal of hardware and intravenous antibiotics. One patient had a hematoma 2 weeks postoperatively that required surgical washout. These complications did not require patients to miss followup tests. Table 1 shows the preoperative demographics and clinical characteristics of this sample. The patients were predominantly male and, on average, relatively young and overweight. They had substantial varus alignment and tib- 652 Birmingham et al Table 2. Gait, radiographic, and patient-reported outcome measures (n ⴝ 120)* Outcome measure Gait Knee adduction moment First peak, %BW ⫻ Ht Second peak, %BW ⫻ Ht Area under the curve, %BW ⫻ Ht ⫻ seconds Speed, meters/second Toe-out angle, degrees Lateral trunk lean, degrees Radiographic Mechanical axis angle, degrees Femoral anatomic axis angle, degrees Proximal tibial articular angle, degrees Medial joint space width, mm Lateral joint space width, mm Posterior tibial slope, degrees Blackburn-Peel ratio Insall-Salvati ratio Patient reported KOOS (range 0–100) Pain Other symptoms Function in daily living Function in sports/recreation Quality of life SF-36 physical function (range 0–100) SF-36 mental health (range 0–100) LEFS (range 0–80) Baseline, mean ⴞ SD 24 months, mean ⴞ SD Change, mean (95% CI) 2.99 ⫾ 0.92 2.37 ⫾ 1.08 1.45 ⫾ 0.49 1.10 ⫾ 0.17 12.00 ⫾ 5.77 3.45 ⫾ 2.97 1.62 ⫾ 0.69 1.30 ⫾ 0.78 0.77 ⫾ 0.36 1.16 ⫾ 0.16 13.20 ⫾ 5.40 1.96 ⫾ 2.07 ⫺1.38 (⫺1.53, ⫺1.22)† ⫺1.07 (⫺1.26, ⫺0.88)† ⫺0.68 (⫺0.76, ⫺0.60)† 0.06 (0.08, 0.04)† 1.23 (0.54, 1.92)† ⫺1.49 (⫺2.00, ⫺0.99)† ⫺8.00 ⫾ 4.09 ⫺2.15 ⫾ 4.45 81.44 ⫾ 4.51 2.86 ⫾ 1.62 6.30 ⫾ 1.82 5.15 ⫾ 4.66 0.76 ⫾ 0.14 1.05 ⫾ 0.18 ⫺0.05 ⫾ 3.05 5.11 ⫾ 3.44 88.47 ⫾ 3.76 3.30 ⫾ 1.48 5.91 ⫾ 1.81 6.37 ⫾ 4.22 0.71 ⫾ 0.14 1.05 ⫾ 0.18 8.04 (7.16, 8.93)† 7.26 (6.24, 8.28)† 7.02 (7.90, 0.44)† 0.44 (0.19, 0.68)† ⫺0.39 (⫺0.71, ⫺0.07)† 1.22 (0.24, 2.20)† ⫺0.05 (⫺0.08, ⫺0.01)† 0.00 (⫺0.03, 0.03) 51.71 ⫾ 17.25 52.62 ⫾ 17.25 60.60 ⫾ 18.64 25.88 ⫾ 20.44 25.59 ⫾ 16.86 46.99 ⫾ 24.18 75.83 ⫾ 14.94 42.62 ⫾ 13.50 74.90 ⫾ 20.25 70.76 ⫾ 19.92 82.26 ⫾ 19.32 54.04 ⫾ 29.87 53.70 ⫾ 24.21 72.20 ⫾ 24.17 82.99 ⫾ 14.69 59.64 ⫾ 16.01 23.19 (19.49, 26.89)† 18.14 (14.65, 21.63)† 21.66 (18.13, 25.20)† 28.16 (22.84, 33.49)† 28.12 (23.96, 32.27)† 25.22 (20.25, 30.18)† 7.16 (3.72, 10.61)† 17.02 (14.43, 19.61)† * 95% CI ⫽ 95% conﬁdence interval; BW ⫽ body weight; Ht ⫽ height; KOOS ⫽ Knee Injury and Osteoarthritis Outcomes Score; SF-36 ⫽ Short Form 36; LEFS ⫽ Lower Extremity Functional Scale. † P ⬍ 0.01. iofemoral degeneration primarily of the medial compartment. Although these characteristics are consistent with other studies evaluating HTO, it should be noted that ⬃10% of the present sample also had substantial degeneration in the lateral compartment (K/L scale grade ⱖ3). Table 2 shows the changes in all gait, radiographic, and patient-reported outcomes at the 2-year end point. With the exception of the Insall-Salvati ratio for patellar height, 95% CIs excluded the value 0 and were quite narrow. Changes in mechanical axis angle, ﬁrst peak knee adduction moment, and KOOS scores at all test sessions are shown in Figure 1. The mechanical axis angle plot shows that surgery produced a large reduction in varus alignment that was maintained throughout the 24 months of followup. The knee adduction moment plot also shows a large reduction in dynamic knee joint load on the operative extremity 6 months postoperatively. Unlike the mechanical axis angle, however, the knee adduction moment slightly but consistently increased throughout followup. A post hoc comparison indicated that the increase from 6 –24 months (mean increase 0.19 [95% CI 0.10, 0.28] %BW ⫻ Ht) was statistically signiﬁcant (P ⬍ 0.001). The KOOS plot also shows large changes in all domains that continued to improve throughout followup. Table 3 shows that the predictor variables explained 43% of the variation in the 2-year postoperative ﬁrst peak knee adduction moment. While adjusting for other vari- ables, higher preoperative knee adduction moment, lower lateral trunk lean, higher age, and lower 6-month postoperative mechanical axis angle (i.e., more varus) were signiﬁcantly associated with higher 2-year knee adduction moments. Analyses that excluded patients who underwent simultaneous ACL reconstruction revealed no additional insight into signiﬁcant predictors, but slightly reduced the explained variance. Similarly, there were no meaningful differences when the value for the ﬁrst peak was replaced by the area under the knee adduction moment curve. With the exception of their corresponding preoperative scores, all linear regression analyses yielded no other signiﬁcant predictors of 2-year KOOS pain or KOOS function in daily living scores. Variance inﬂation factor and tolerance levels for all regression analyses suggested acceptable levels of multicolinearity. Figure 2 shows that the cluster analysis identiﬁed 2 clusters (Akaike’s information criterion ⫽ 1,368.57) of patients who had similar values for their 2-year changes in ﬁrst peak knee adduction moment and KOOS pain score. The larger patient cluster (n ⫽ 92) shows patients with substantial decreases in both knee joint load (mean ⫾ SD change in ﬁrst peak knee adduction moment 1.33 ⫾ 0.61) and knee pain (mean ⫾ SD change in KOOS 24.47 ⫾ 14.45). The smaller cluster (n ⫽ 10) highlights a group of patients that leaned toward relatively smaller decreases in knee joint load (mean ⫾ SD change in ﬁrst peak knee Outcomes of Medial Opening Wedge High Tibial Osteotomy 653 Figure 1. Means and 95% conﬁdence intervals for A, mechanical axis angle, B, ﬁrst peak external knee adduction moment on both extremities, and C, Knee Injury and Osteoarthritis Outcomes Scores (KOOS) assessed at each test session. The ﬁgures show large, statistically signiﬁcant (P ⬍ 0.001) changes in mechanical axis angle, knee adduction moment on the operative extremity, and all KOOS domains 6 months postoperatively. Small increases in the knee adduction moment from 6 to 24 months were statistically signiﬁcant (P ⬍ 0.01). BW ⫽ body weight; Ht ⫽ height. adduction moment 0.66 ⫾ 0.73) and increases in knee pain (mean ⫾ SD change in KOOS ⫺15.27 ⫾ 14.00). Consistent results were achieved when the order of cases in the sample was randomized. Consistent with the clustering criteria, 2-year change in ﬁrst peak knee adduction moment and 2-year change in KOOS pain scores were signiﬁcantly different between these groups (P ⬍ 0.01). However, no other demographic and clinical characteristics were statistically different. Outliers (n ⫽ 24) were not evaluated. DISCUSSION We hypothesized that medial opening wedge HTO would result in substantial, clinically important improvements in dynamic knee joint load and patient-reported pain. The present ﬁndings strongly support this hypothesis. Effect sizes and standardized response means for the ﬁrst peak knee adduction moment (1.50 and 1.58, respectively) and each of the KOOS domains (0.89 –1.38) are above the threshold value for a large effect (0.8) (46). Additionally, even the lower ends of the 95% CIs for the 2-year changes in the KOOS pain, function in daily living, function in sports/recreation, and knee-related quality of life domains (Table 2) indicate improvements beyond the minimum to suggest clinically important changes have occurred in the majority of patients (43). These conﬁdence intervals are also consistent with previously reported changes in the KOOS after HTO procedures (6,7). We believe that the large improvements in both biomechanical and patientreported outcomes observed 2 years postoperatively underscore the potential beneﬁt of surgically restoring neutral lower extremity alignment. It is important to note that these changes were achieved with intentional minimal to no overcorrection of alignment. Figure 1 shows that, on average, alignment was altered to a position of only slight valgus (mechanical axis angle 1°, anatomic axis angle 5°), which was maintained throughout the 2-year followup. The absolute magnitude of change in the mechanical axis angle achieved with surgery is approximately equal to the amount of preoperative varus (Table 2). This is consistent with our aim of HTO, which generally attempts to establish neutral align- 654 Birmingham et al Table 3. Predictors of 2-year ﬁrst peak knee adduction moment (n ⴝ 120)* ␤ (95% CI)‡ Potential predictor variables† Demographic Age, years Sex (0 ⫽ female, 1 ⫽ male) BMI, kg/m2 Radiographic Mechanical axis angle, degrees Preoperative measure 6-month postoperative measure OA severity Medial compartment Lateral compartment (0 ⫽ less, 1 ⫽ more)§ Gait First peak knee adduction moment, %BW ⫻ Ht Speed, meters/second Toe-out angle, degrees Lateral trunk lean, degrees Patient-reported outcome KOOS pain domain 0.02 (0.00, 0.03) ⫺0.10 (⫺0.39, 0.20) ⫺0.01 (⫺0.04, 0.02) 0.01 (⫺0.03, 0.06) ⫺0.14 (⫺0.18, ⫺0.10) P 0.03 0.51 0.37 0.56 ⬍0.01 ⫺0.17 (⫺0.47, 0.12) 0.07 (⫺0.16, 0.29) 0.25 0.57 0.24 (0.08, 0.39) ⫺0.09 (⫺0.79, 0.60) 0.01 (⫺0.01, 0.03) ⫺0.04 (⫺0.07, 0.00) 0.01 0.79 0.45 0.05 0.00 (⫺0.01, 0.01) 0.58 * 95% CI ⫽ 95% conﬁdence interval; BMI ⫽ body mass index; OA ⫽ osteoarthritis; BW ⫽ body weight; Ht ⫽ height; KOOS ⫽ Knee Injury and Osteoarthritis Outcomes Score. † All were assessed preoperatively unless stated otherwise. ‡ Effect on 2-year peak knee adduction moment, adjusting for all other predictors in model. Adjusted r2 ⫽ 0.43. § Less severe ⫽ Kellgren/Lawrence scale grade of OA severity 0, 1, 2; more severe ⫽ 3, 4. Figure 2. Scatter plot of 2-year changes in knee adduction moment and Knee Injury and Osteoarthritis Outcomes Score (KOOS) pain domain scores shows results of the cluster analysis. The larger patient cluster (large open circles, n ⫽ 92) shows patients with substantial decreases in both knee joint load and knee pain. The smaller patient cluster (solid circles, n ⫽ 10) shows a group of patients that leaned toward relatively smaller decreases in knee joint load and increases in knee pain. Outliers did not ﬁt into either cluster (small open circles, n ⫽ 24). BW ⫽ body weight; Ht ⫽ height. ment in all patients despite the preoperative measure. Change in alignment was accompanied by a very small change in tibial slope and patellar height. However, even the upper ends of the 95% CIs for change in tibial slope (2.2°) and Blackburn-Peel ratio (⬍0.1) were extremely small and likely not clinically important (Table 2). Change in alignment was also accompanied by a very small increase in medial joint space and a decrease in lateral joint space. Overall, these radiographic ﬁndings suggest that the change in alignment produced a sustained lateral shift in weight-bearing load without producing detrimental changes in knee geometry. These ﬁndings are consistent with recent reports of favorable radiographic results with minimal complications after medial opening wedge HTO (11,47). We observed several changes in gait that are also consistent with the biomechanical rationale for this surgery (Table 2). Despite a slight increase in speed and a decrease in lateral trunk lean toward the stance extremity (both reﬂect a more normal gait pattern and would increase medial joint load), there was a large reduction in all measures of the external knee adduction moment postoperatively. Interestingly, there was also a small increase in toe-out angle that would serve to further reduce the knee adduction moment. It is unclear whether the change in toe-out angle was a structural change as a result of surgery or a gait adaptation adopted by patients. Because toe-out angle has been demonstrated to have a protective effect on future knee joint degeneration (48), the effect of medial opening wedge HTO on toe-out angle should be evaluated further. Importantly, unlike alignment, we observed an unex- Outcomes of Medial Opening Wedge High Tibial Osteotomy pected increase in the knee adduction moment from 6 to 24 months postoperatively (Figure 1). The increase was not observed on the opposite extremity, and therefore is not likely an overall response to the small increase in gait speed observed over the same postoperative time period. The magnitude of the increase in postoperative knee adduction moments from 6 to 24 months was small (13%), and its value at 2 years postoperatively was still far less than those observed preoperatively or on the opposite extremity (55% and 65%, respectively). However, we believe that these relatively early postoperative increases in the knee adduction moment are of potential concern and may be a precursor to poorer longer-term outcomes. Changes in the knee adduction moment without concomitant changes in mechanical axis angle are consistent with previous reports of the low to moderate correlation between these measures, and further emphasize the difference between dynamic measures of knee joint load and static measures of alignment (48 –50). Further research evaluating changes in knee joint load observed within 2 years postoperatively, including potential underlying mechanisms, is warranted. Most importantly, the potential for these changes to affect long-term outcomes needs to be evaluated. We also hypothesized that preoperative patient characteristics would be associated with 2-year outcomes, which was only partly supported. Linear regression analyses yielded no signiﬁcant predictors of 2-year KOOS pain or KOOS function in daily living scores other than their preoperative values. Only preoperative knee adduction moment, trunk lean, age, and mechanical axis angle at the 6-month followup (i.e., alignment achieved with surgery) explained substantial variance in 2-year knee adduction moments. It is possible that other factors not evaluated in the present study, including other clinical characteristics or more elaborate measures of neuromuscular function and motor control, contributed to changes. However, we believe that the relatively little variance in all outcomes 2 years postoperatively is the primary reason for our ﬁndings and we continue to evaluate this objective with longer-term followup. Owing to our 2 primary goals of HTO, decreasing the load on the medial tibiofemoral compartment and improving patient-reported outcomes, we performed a cluster analysis to identify patients who experienced similar changes in these outcomes. A large cluster of 92 patients experienced substantial improvements in both knee adduction moment and KOOS pain domain scores, and reﬂect the general success of this procedure. However, a cluster of 10 patients who had poor results in both outcomes was also identiﬁed. Given the small sample size, we could not identify any signiﬁcant differences in patient characteristics between these clusters. Similarly, no trends for increased prevalence of patient characteristics, including disease severity in the lateral compartment, age, sex, or correction angle, were evident. The exploratory nature of cluster analysis should be acknowledged. Given the goals of surgery, however, we believe that this small cluster represents a subgroup of patients with particularly poor results that require further investigation. Inclusion and exclusion criteria for the present study are 655 also somewhat different from previous reports and should be highlighted. We recruited patients of all ages with varus gonarthrosis and greatest radiographic severity and pain in the medial compartment, but did not exclude those with concomitant lateral compartment disease. Typically, these were young patients with substantial varus deformities who were not candidates for total knee arthroplasty. We do not advocate HTO as a widespread treatment option for patients with substantial bicompartmental tibiofemoral disease. However, 2 patients (ages 43 and 50 years) had lateral compartment OA that also met the K/L scale criteria for grade 4 severity (although medial compartment degeneration was greater) and underwent the procedure. Both patients experienced results similar to many in the sample, with reductions in the ﬁrst peak knee adduction moment of approximately 1% BW ⫻ Ht and improvements in the KOOS pain score of approximately 30 points 2 years postoperatively. We continue to closely follow patients meeting these criteria. Strengths of our study include its prospective design that adhered to published guidelines for conducting observational studies (12), its relatively large sample size, and the spectrum of validated outcome measures relevant to HTO. A limitation in this design is the inability to compare the effects of HTO with other surgical and nonsurgical treatment options. Similarly, although the meaning and expected outcome patients attribute to treatment is likely to inﬂuence outcomes (51), comparison of HTO with sham surgery is not realistic. However, the importance of comparing other less invasive surgical procedures for knee OA with medical and physical therapies (52) and even sham treatment (53) has been established. Clearly, future research comparing HTO with different treatment strategies is required. Other limitations in the present study include the relatively small number of women in our sample and the inability to generalize the results beyond 2 years postoperatively. Overall, the present ﬁndings suggest that improvements in malalignment achieved by medial opening wedge HTO produce substantial and clinically important changes in knee joint loading during gait and patient-reported measures of pain, function, and quality of life 2 years postoperatively. Changes in knee adduction moment observed within the ﬁrst 2 years postoperatively should be explored as potential predictors of long-term success and subgroups of patients with poor outcomes. AUTHOR CONTRIBUTIONS All authors were involved in contributions to study conception and design, acquisition of data, or analysis and interpretation of data, and drafting the article or revising it critically for important intellectual content, and all authors approved the ﬁnal version to be submitted for publication. Dr. Birmingham had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. REFERENCES 1. Badley EM. Arthritis in Canada: what do we know and what should we know? J Rheumatol Suppl 2005;72:39 – 41. 2. Lawrence RC, Felson DT, Helmick CG, Arnold LM, Choi H, Deyo RA, et al. Estimates of the prevalence of arthritis and 656 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. Birmingham et al other rheumatic conditions in the United States: part II. Arthritis Rheum 2008;58:26 –35. Peat G, McCarney R, Croft P. Knee pain and osteoarthritis in older adults: a review of community burden and current use of primary health care. Ann Rheum Dis 2001;60:91–7. Zhang W, Moskowitz RW, Nuki G, Abramson S, Altman RD, Arden N, et al. OARSI recommendations for the management of hip and knee osteoarthritis. II. OARSI evidence-based, expert consensus guidelines. Osteoarthritis Cartilage 2008;16: 137– 62. Brouwer RW, Raaij van TM, Bierma-Zeinstra SM, Verhagen AP, Jakma TS, Verhaar JA. Osteotomy for treating knee osteoarthritis. Cochrane Database Syst Rev 2007;3:CD004019. Spahn G, Kirschbaum S, Kahl E. Factors that inﬂuence high tibial osteotomy results in patients with medial gonarthritis: a score to predict the results. Osteoarthritis Cartilage 2006;14: 190 –5. W-Dahl A, Toksvig-Larsen S, Roos EM. A 2-year prospective study of patient-relevant outcomes in patients operated on for knee osteoarthritis with tibial osteotomy. BMC Musculoskelet Disord 2005;6:18. Wright J, Heck D, Hawker G, Dittus R, Freund D, Joyce D, et al. Rates of tibial osteotomies in Canada and the United States. Clin Orthop Relat Res 1995;319:266 –75. Coventry MB, Ilstrup DM, Wallrichs SL. Proximal tibial osteotomy: a critical long-term study of eighty-seven cases. J Bone Joint Surg Am 1993;75:196 –201. Lobenhoffer P, Agneskirchner JD. Improvements in surgical technique of valgus high tibial osteotomy. Knee Surg Sports Traumatol Arthrosc 2003;11:132– 8. Brouwer RW, Bierma-Zeinstra SM, van Koeveringe AJ, Verhaar JA. Patellar height and the inclination of the tibial plateau after high tibial osteotomy: the open versus the closedwedge technique. J Bone Joint Surg Br 2005;87:1227–32. Vandenbroucke JP, von Elm E, Altman DG, Gotzsche PC, Mulrow CD, Pocock SJ, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. Epidemiology 2007;18:805–35. Belo JN, Berger MY, Reijman M, Koes BW, Bierma-Zeinstra SM. Prognostic factors of progression of osteoarthritis of the knee: a systematic review of observational studies. Arthritis Rheum 2007;57:13–26. Brouwer GM, van Tol AW, Bergink AP, Belo JN, Bernsen RM, Reijman M, et al. Association between valgus and varus alignment and the development and progression of radiographic osteoarthritis of the knee. Arthritis Rheum 2007;56:1204 –11. Paley D. Principles of deformity correction. New York: Springer; 2002. Sharma L, Song J, Felson DT, Cahue S, Shamiyeh E, Dunlop DD. The role of knee alignment in disease progression and functional decline in knee osteoarthritis. JAMA 2001;286: 188 –95. Sharma L. The role of varus and valgus alignment in knee osteoarthritis [editorial]. Arthritis Rheum 2007;56:1044 –7. Andriacchi TP, Mundermann A. The role of ambulatory mechanics in the initiation and progression of knee osteoarthritis. Curr Opin Rheumatol 2006;18:514 – 8. Miyazaki T, Wada M, Kawahara H, Sato M, Baba H, Shimada S. Dynamic load at baseline can predict radiographic disease progression in medial compartment knee osteoarthritis. Ann Rheum Dis 2002;61:617–22. Briem K, Ramsey DK, Newcomb W, Rudolph KS, SnyderMackler L. Effects of the amount of valgus correction for medial compartment knee osteoarthritis on clinical outcome, knee kinetics and muscle co-contraction after opening wedge high tibial osteotomy. J Orthop Res 2007;25:311– 8. Noyes FR, Barber-Westin SD, Hewett TE. High tibial osteotomy and ligament reconstruction for varus angulated anterior cruciate ligament-deﬁcient knees. Am J Sports Med 2000;28: 282–96. Ramsey DK, Snyder-Mackler L, Lewek M, Newcomb W, Rudolph KS. Effect of anatomic realignment on muscle function during gait in patients with medial compartment knee osteoarthritis. Arthritis Rheum 2007;57:389 –97. 23. Wada M, Imura S, Nagatani K, Baba H, Shimada S, Sasaki S. Relationship between gait and clinical results after high tibial osteotomy. Clin Orthop Relat Res 1998;354:180 – 8. 24. Wang JW, Kuo KN, Andriacchi TP, Galante JO. The inﬂuence of walking mechanics and time on the results of proximal tibial osteotomy. J Bone Joint Surg Am 1990;72:905–9. 25. Weidenhielm L, Svensson OK, Brostrom Lk L. Surgical correction of leg alignment in unilateral knee osteoarthrosis reduces the load on the hip and knee joint bilaterally. Clin Biomech (Bristol, Avon) 1995;10:217–21. 26. Goh JC, Bose K, Khoo BC. Gait analysis study on patients with varus osteoarthrosis of the knee. Clin Orthop Relat Res 1993; 294:223–31. 27. Prodromos CC, Andriacchi TP, Galante JO. A relationship between gait and clinical changes following high tibial osteotomy. J Bone Joint Surg Am 1985;67:1188 –94. 28. Altman R, Asch E, Bloch D, Bole G, Borenstein D, Brandt K, et al. Development of criteria for the classiﬁcation and reporting of osteoarthritis: classiﬁcation of osteoarthritis of the knee. Arthritis Rheum 1986;29:1039 – 49. 29. Fowler PJ, Tan JL, Brown GA. Medial opening wedge high tibial osteotomy: how I do it. Oper Tech Sports Med 2000;8: 32– 8. 30. Dugdale TW, Noyes FR, Styer D. Preoperative planning for high tibial osteotomy: the effect of lateral tibiofemoral separation and tibiofemoral length. Clin Orthop Relat Res 1992; 274:248 – 64. 31. Kadaba MP, Ramakrishnan HK, Wootten ME. Measurement of lower extremity kinematics during level walking. J Orthop Res 1990;8:383–92. 32. Hunt MA, Birmingham TB, Gifﬁn JR, Jenkyn TR. Associations among knee adduction moment, frontal plane ground reaction force, and lever arm during walking in patients with knee osteoarthritis. J Biomech 2006;39:2213–20. 33. Jenkyn TR, Hunt MA, Jones IC, Gifﬁn JR, Birmingham TB. Toe-out gait in patients with knee osteoarthritis partially transforms external knee adduction moment into ﬂexion moment during early stance phase of gait: a tri-planar kinetic mechanism. J Biomech 2008;41:276 – 83. 34. Hunt MA, Birmingham TB, Bryant D, Jones I, Gifﬁn JR, Jenkyn TR, et al. Lateral trunk lean explains variation in dynamic knee joint load in patients with medial compartment knee osteoarthritis. Osteoarthritis Cartilage 2008;16:591–9. 35. Mundermann A, Dyrby CO, Hurwitz DE, Sharma L, Andriacchi TP. Potential strategies to reduce medial compartment loading in patients with knee osteoarthritis of varying severity: reduced walking speed. Arthritis Rheum 2004;50: 1172– 8. 36. Birmingham TB, Hunt MA, Jones IC, Jenkyn TR, Gifﬁn JR. Test–retest reliability of the peak knee adduction moment during walking in patients with medial compartment knee osteoarthritis. Arthritis Rheum 2007;57:1012–7. 37. Specogna AV, Birmingham TB, DaSilva JJ, Milner JS, Kerr J, Hunt MA, et al. Reliability of lower limb frontal plane alignment measurements using plain radiographs and digitized images. J Knee Surg 2004;17:203–10. 38. Hsu RW, Himeno S, Coventry MB, Chao EY. Normal axial alignment of the lower extremity and load-bearing distribution at the knee. Clin Orthop Relat Res 1990;255:215–27. 39. Moreland JR, Bassett LW, Hanker GJ. Radiographic analysis of the axial alignment of the lower extremity. J Bone Joint Surg Am 1987;69:745–9. 40. Rogers BA, Thornton-Bott P, Cannon SR, Briggs TW. Interobserver variation in the measurement of patellar height after total knee arthroplasty. J Bone Joint Surg Br 2006;88:484 – 8. 41. Kellgren JH, Lawrence JS. Radiological assessment of osteoarthrosis. Ann Rheum Dis 1957;16:494 –502. 42. Roos EM, Roos HP, Lohmander LS, Ekdahl C, Beynnon BD. Knee Injury and Osteoarthritis Outcome Score (KOOS): development of a self-administered outcome measure. J Orthop Sports Phys Ther 1998;28:88 –96. 43. Roos EM, Lohmander LS. The Knee injury and Osteoarthritis Outcome Score (KOOS): from joint injury to osteoarthritis. Health Qual Life Outcomes 2003;1:64. Outcomes of Medial Opening Wedge High Tibial Osteotomy 44. Binkley JM, Stratford PW, Lott SA, Riddle DL. The Lower Extremity Functional Scale (LEFS): scale development, measurement properties, and clinical application. Phys Ther 1999;79:371– 83. 45. Riddle DL, Lee KT, Stratford PW. Use of SF-36 and SF-12 health status measures: a quantitative comparison for groups versus individual patients. Med Care 2001;39:867–78. 46. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Hillside (NJ): Lawrence Erlbaum; 1988. 47. Noyes FR, Mayﬁeld W, Barber-Westin SD, Albright JC, Heckmann TP. Opening wedge high tibial osteotomy: an operative technique and rehabilitation program to decrease complications and promote early union and function. Am J Sports Med 2006;34:1262–73. 48. Chang A, Hurwitz D, Dunlop D, Song J, Cahue S, Hayes K, et al. The relationship between toe-out angle during gait and progression of medial tibiofemoral osteoarthritis. Ann Rheum Dis 2007;66:1271–5. 49. Hurwitz DE, Ryals AB, Case JP, Block JA, Andriacchi TP. The knee adduction moment during gait in subjects with knee 657 50. 51. 52. 53. osteoarthritis is more closely correlated with static alignment than radiographic disease severity, toe out angle and pain. J Orthop Res 2002;20:101–7. Specogna AV, Birmingham TB, Hunt MA, Jones IC, Jenkyn TR, Fowler PJ, et al. Radiographic measures of knee alignment in patients with varus gonarthrosis: effect of weightbearing status and associations with dynamic joint load. Am J Sports Med 2007;35:65–70. Moerman DE, Jonas WB. Deconstructing the placebo effect and ﬁnding the meaning response. Ann Intern Med 2002;136: 471– 6. Kirkley A, Birmingham TB, Litchﬁeld RB, Gifﬁn JR, Willits KR, Wong CJ, et al. A randomized controlled trial of arthroscopic surgery for osteoarthritis of the knee. N Engl J Med 2008;359:1097–107. Moseley JB, O’Malley K, Petersen NJ, Menke TJ, Brody BA, Kuykendall DH, et al. A controlled trial of arthroscopic surgery for osteoarthritis of the knee. N Engl J Med 2002;347: 81– 8.