Validity and Reliability of Radiographic Knee Osteoarthritis Measures by Arthroplasty Surgeons

Transcription

1 Validity and Reliability of Radiographic Knee Osteoarthritis Measures by Arthroplasty Surgeons Daniel L. Riddle, PhD; William A. Jiranek, MD; Jason R. Hull, MD abstract Full article available online at Healio.com/Orthopedics. Search: Most orthopedic surgeons do not routinely use radiographic classification systems to grade the extent of joint space narrowing in patients considered for total knee arthroplasty. The authors compared the validity and reliability of radiographic measures of tibiofemoral osteoarthritis by 2 experienced and 2 inexperienced orthopedic surgeons on individuals who subsequently underwent total knee arthroplasty. The Kellgren-Lawrence and the Osteoarthritis Research Society International classification systems were used by all surgeons to score the radiographs in 116 individuals in the Osteoarthritis Initiative, a federally funded cohort study of individuals with or at risk of knee osteoarthritis. Validity was judged based on comparison with the criterion centrally adjudicated consensus measures obtained by Osteoarthritis Initiative investigators. Weighted kappa, a chance corrected agreement index, was used to describe validity and reliability. Validity and intrarater reliability were substantial to almost perfect for 1 experienced and 1 inexperienced surgeon, with weighted kappas ranging from 0.76 to 0.96 for the surgical knees. The other experienced and inexperienced surgeons demonstrated moderate to substantial validity, with weighted kappas ranging from 0.43 to 0.70 and lower intrarater reliability. Interrater reliability was generally less than intrarater reliability. With minimal training, some surgeons can obtain valid and reliable measurements of knee osteoarthritis status in individuals who eventually undergo total knee arthroplasty. Measurement quality does not appear to be dependent on extent of surgeon experience. Some surgeons require additional training to become proficient in the radiographic classification systems, and future research should examine this issue. The authors are from the Department of Physical Therapy (DLR) and the Department of Orthopaedic Surgery (WAJ, JRH), Virginia Commonwealth University, Richmond, Virginia. The authors have no relevant financial relationships to disclose. The OAI is a public-private partnership comprising 5 contracts (N01-AR ; N01-AR ; N01-AR ; N01-AR ; N01-AR ) funded by the National Institutes of Health, a branch of the Department of Health and Human Services, and conducted by the OAI Study Investigators. Private funding partners include Merck Research Laboratories, Novartis Pharmaceuticals Corporation, GlaxoSmithKline, and Pfizer, Inc. Private sector funding for the OAI is managed by the Foundation for the National Institutes of Health. This manuscript was prepared using an OAI public use data set and does not necessarily reflect the opinions or views of the OAI investigators, the National Institutes of Health, or the private funding partners. The funding sources played no role in the current study. The authors thank Drs Robert Neff and Derek Whitaker for their assistance during the study. Correspondence should be addressed to: Daniel L. Riddle, PhD, Department of Orthopaedic Surgery, Virginia Commonwealth University, PO Box , Richmond, VA (dlriddle@vcu.edu). doi: / JANUARY 2013 Volume 36 Number 1 e25

2 Orthopedic surgeons routinely interpret knee radiographs of patients considering total knee arthroplasty (TKA). Knee osteoarthritis severity is considered to be one of the more important criteria for judging whether a patient may be a candidate for surgery 1,2 and has been shown to influence outcomes following TKA. 3,4 To take full advantage of this prognostic evidence, radiographic interpretations made by surgeons should demonstrate a clinically acceptable level of validity and reliability. Although substantial evidence was used to examine the reliability of preoperative planning procedures 5,6 and implant alignment, 7 the current authors found only 1 study examining surgeons agreement on the extent of knee arthritis in surgical patients. Mehta et al 8 examined the reliability of radiographic findings using the International Knee Documentation Committee classification system from 5-year follow-up examinations conducted on 205 knees with reconstructed anterior cruciate ligaments. The authors reported that interrater reliability ranged from an intraclass correlation coefficient (ICC) of 0.20 to A recently published systematic review of 24 radiographic reliability studies of the knee was reviewed, and no studies examined the reliability of orthopedic surgeon judgments of osteoarthritis. 9 To the current authors knowledge, no evidence exists indicating the extent of validity or reliability of surgeons judgments of the extent of knee osteoarthritis in individuals considering TKA. Common clinical approaches to radiographic assessment provide a semiquantitative judgment of the extent of osteoarthritis, the most common being the Kellgren-Lawrence (KL) 10 or the Osteoarthritis Research Society International (OARSI) classification systems. 11,12 In the authors experience, many orthopedic surgeons do not routinely use either of these systems, but rather use a variety of qualitative and quantitative approaches to describe the extent of radiographic joint space narrowing when judging whether a patient may be a candidate for TKA. The KL and OARSI classification systems include joint space width in their scales and provide a system for standardizing descriptions and potentially enhancing communications among surgeons, other physicians, and patients. Valid and consistent judgments of the extent of knee osteoarthritis would likely benefit patients and surgeons. Patients would benefit because they sometimes see multiple surgeons, and, ideally, different surgeons should agree on the extent of knee osteoarthritis. Communication among surgeons would likely also benefit if it were known that radiographic judgments could be reliably obtained. The authors were interested in determining the extent of validity and reliability of surgeons judgments of the severity of knee osteoarthritis in individuals who had standardized fixed-flexion knee radiographs taken prior to TKA. The purpose of the current study was to examine the validity and intra- and interrater reliability of radiographic assessments made by experienced and inexperienced joint arthroplasty surgeons. The authors hypothesized that the more experienced surgeons would demonstrate a higher degree of validity and reliability than surgeons with minimal clinical experience. They previously reported on the extent of tibiofemoral knee osteoarthritis in medial and lateral compartments in individuals prior to TKA using data from the Osteoarthritis Initiative (OAI) 13 but did not explore validity and reliability for experienced vs inexperienced surgeons. The current article addresses this evidence gap. The study was approved by the institutional review board at the University of California at San Francisco, San Francisco, California. Materials and Methods The OAI is a National Institutes of Health funded and privately funded prospective 4-year longitudinal cohort study of individuals aged 45 to 79 years with radiographically confirmed knee osteoarthritis or a risk for knee osteoarthritis. 14 Participants received no treatment as part of the study, nor were they solicited for treatment by the investigators. The study was approved by the institutional review board at the OAI Coordinating Center, the University of California at San Francisco. Included patients were those who subsequently underwent TKA during the 3-year follow-up. Participants were recruited from communities near the following institutions: (1) the University of Maryland in Baltimore, Maryland, (2) The Ohio State University in Columbus, Ohio, (3) the University of Pittsburgh in Pittsburgh, Pennsylvania, and (4) Memorial Hospital of Rhode Island in Pawtucket, Rhode Island. Participants in the OAI were excluded if they met 1 or more of the following criteria: (1) diagnosis of rheumatoid arthritis; (2) bilateral TKA or preexisting plans to undergo bilateral TKA during the same surgical visit in the next 3 years; (3) bilateral end-stage knee osteoarthritis; (4) use of ambulatory aids other than a single straight cane more than 50% of the time; (5) comorbid conditions that might interfere with 4-year participation; and (6) unlikely to reside in clinic area for at least 3 years. In addition, because the OAI study required repeated magnetic resonance imaging, men weighing more than 130 kg and women weighing more than 114 kg were excluded because they were unable to undergo 3.0-T magnetic resonance imaging. Over the course of 3 years of followup, a total of 116 individuals underwent TKA, and this sample was included in the study (Table 1). A total of 27 individuals underwent TKA during year 1, thirtyeight during year 2, and 51 during year 3. A total of 19 of the 116 patients underwent more than 1 TKA, with 13 undergoing bilateral TKA in the same year and 6 undergoing either repeated unilateral TKAs in different years or contralateral TKAs in e26 ORTHOPEDICS Healio.com/Orthopedics

3 Radiographic Knee Osteoarthritis Measures Riddle et al different years. Only 1 involved knee was selected per patient to ensure that each patient provided independent data. Surgeons examined the radiographs of both knees of each patient and were blinded to the surgical knee. To identify the involved knee for study, a number generator was used to randomly select either the right or left knee for patients who had surgery on both knees during the 3-year period. For the unilateral TKAs, the radiograph prior to the first surgery was selected. Power Analysis Power analysis was conducted to estimate the adequacy of the sample size, particularly for assessing validity because validity is critical when judging the usefulness of measurements. 15 A commonly described approach was used for estimating power using kappa. 16 Assuming that the proportion of individuals with end-stage knee osteoarthritis of the surgical knee is 0.40, a large-sample, level 2-sided test of the null hypothesis that intraclass kappa was 0.50 will have 80% power to detect an alternative kappa of 0.80 when the sample size is 68. This level of statistical power will differentiate moderate from substantial agreement in the validity and interrater reliability analyses. 17 Radiographic Data Collection and Interpretation Radiographs were obtained using a SynaFlexer frame (Synarc, San Francisco, California) to reproducibly position the standing patient s knees in a 20 flexed position and to equally distribute body weight while the radiograph beam, set at a standardized distance from the knee, was angled caudally at This approach was chosen to optimally capture the tibiofemoral joint space. 19,20 All study sites underwent extensive training to enhance consistency. For this study, the authors used the radiographs obtained at baseline. The Osteoarthritis Initiative investigators assessed baseline radiographs and used the KL and OARSI classification systems to quantify the extent of arthritis of both knees in each radiograph (n5112). For the KL classification system grades (range, 0 to 4), a grade of 0 was normal, 1 indicated doubtful narrowing of joint space and possible osteophytic lipping, 2 indicated definite osteophytes and possible narrowing of joint space, 3 indicated definite narrowing of joint space and some sclerosis and possible deformity of bone ends, and 4 indicated large osteophytes, marked narrowing of joint space, severe sclerosis, and definite deformity of bone ends. 10 For OARSI classification system grades (range, 0 to 3), a grade of 0 was normal, 1 indicated mild narrowing (1%-33% narrowed), 2 indicated moderate narrowing (34%-66%), and 3 indicated severe narrowing (67%-100%). 11,12 The OARSI grades were given for the medial and lateral tibiofemoral compartments (Table 2). For the criterion gold standard measure of osteoarthritis status in the validity analysis, the radiographic scoring data used were provided by OAI researchers. These investigators used an extensive adjudication process to establish the baseline osteoarthritis diagnosis for the knees of each patient. Three central readers from 1 site, either rheumatologists or musculoskeletal radiologists with extensive training and experience with the KL and OARSI classification systems, served Table 1 Baseline Characteristics of Patients Who Subsequently Underwent Total Knee Arthroplasty Variables Mean6SD or No. (%) Demographic Age, y Female 68 (58.6) African American 19 (16.5) Education High school or less 32 (27.6) Some college or graduate 49 (42.2) More than a bachelor s degree 35 (30.2) Divorced or separated 13 (11.2) Annual income <$50,000 45(41.3) General and knee health Comorbidity score.51 (0.86) KL grade, surgical knee 0 2 (1.9) 1 2 (1.9) 2 21 (19.4) 3 35 (32.4) 4 48 (44.4) KL grade, contralateral knee 0 13 (12.0) 1 17 (15.7) 2 44 (40.7) 3 31 (28.7) 4 3 (2.8) WOMAC score Pain 6.9 (4.0) Physical function 21.3 (12.9) Abbreviations: KL, Kellgren-Lawrence; WOMAC, Western Ontario and McMaster Universities Osteoarthritis Index. as adjudicating readers. When 2 readers agreed on independent readings, that score was entered for the patient. When a disagreement was identified, a third expert reviewed scores from both readers. If the third reader agreed with either the first or second reader s scoring, then the agreedon scoring was determined to be final. If the third expert reader did not agree with JANUARY 2013 Volume 36 Number 1 e27

4 either reader, the 3 readers attended an adjudication session in which consensus scoring was obtained. Reliability among the adjudicating readers was substantial to almost perfect, with weighted kappa (kw) coefficients ranging from 0.70 to 0.87 for repeated independent readings separated by 3 to 9 months. 23 Adjudicated readings were available for 108 of the 112 baseline radiographs of patients undergoing TKA during the 3-year study. Four orthopedic surgeons were recruited to make KL and OARSI classification system readings on a total of 112 bilateral knee radiographs of patients (radiographs not available for 4 patients) at baseline and who subsequently underwent TKA during the 3-year study. The authors recruited 2 inexperienced surgeons who recently completed their residency and were beginning their arthroplasty fellowship and 2 experienced surgeons who were attending surgeons at the authors institution. The experienced surgeons had at least 5 years of experience conducting TKA surgeries, and both conducted at least 100 TKAs per year over the prior 5 years. Surgeons were blinded to TKA status or involved side. Table 2 Semiquantitative Radiographic Scoring Systems Used in the Study Kellgren-Lawrence Grading System a Grade 0: No feature of osteoarthritis Grade 1: Doubtful narrowing of joint space and possible osteophytic lipping Grade 2: Definite osteophytes and possible narrowing of joint space Grade 3: Moderate multiple osteophytes, definite narrowing of joint space, and some sclerosis and possible deformity of bone ends Grade 4: Large osteophytes, marked narrowing of joint space, severe sclerosis and definite deformity of bone ends Osteoarthritis Research Society International Grading System for Medial and Lateral Tibiofemoral Joint Space Narrowing (% narrowed) b Grade 0: normal Grade 1: mild (1-33) Grade 2: moderate (34-66) Grade 3: severe (67-100) a Originally described in Kellgren and Lawrence. 21 b Originally described in Altman et al. 22 Surgeons were also unaware of the baseline readings completed by OAI investigators. All surgeons were provided with the KL and OARSI classification systems and radiographic examples illustrating the grading systems. 12 The surgeons practiced by applying the scales individually on a minimum of 10 knee radiographs unrelated to the study. The practice session ended when the surgeons indicated that they were prepared to use the scales to grade the knee radiographs. Statistical Methods Weighted kappa was used to determine the extent of validity between each surgeon s readings and the criterion gold standard readings made by the OAI investigators. Data were reported for the future TKA knee, as well as for the less involved nonsurgical knee separately because of the potential effect of knee osteoarthritis severity on validity. For intrarater reliability, repeated ratings by each examiner, separated by a 2-week timeframe, were compared using kw. Thirty patients were randomly selected for the intrarater reliability analysis. For interrater reliability, the first readings by the inexperienced surgeons were compared, as were the first readings by the experienced surgeons. The entire sample of 112 radiographs was used for the interrater analysis. The authors used kw with quadratic weighting as the optimal chance-corrected agreement coefficient when using ordinal scales and when differences in agreement become more serious as disagreements become greater. 16,17 Reliability and validity by the examiners were judged using the following scale: 05poor agreement, 0.01 to 0.205slight agreement, 0.21 to 0.405fair agreement, 0.41 to 0.605moderate agreement, 0.61 to 0.805substantial agreement, and 0.81 to 15almost perfect agreement. 17 Results Validity coefficients for the KL and OARSI classification systems varied among the 4 surgeons (Table 3). One experienced surgeon (#1) consistently achieved substantial to almost perfect agreement, with kw ranging from 0.80 to 0.92 for all measures obtained on the future surgical knees. One inexperienced surgeon (#3) had similar validity coefficients, with all kw in the substantial agreement range. One experienced surgeon (#2) and 1 inexperienced surgeon (#4) consistently fell below the substantial agreement threshold for measures obtained on the future surgical knees. These scores were generally in the slight to moderate range. Intrarater reliability coefficients for the KL and OARSI classification systems displayed a pattern similar to the validity coefficients in that experienced surgeon #1 and inexperienced surgeon #3 had the highest coefficients (Table 4). Interrater reliability for the experienced surgeons and for the inexperienced surgeons was generally lower than intrarater estimates (Table 5). Reliability and validity estimates for the contralateral knees were consistently below those for the future surgical knees. e28 ORTHOPEDICS Healio.com/Orthopedics

5 Radiographic Knee Osteoarthritis Measures Riddle et al Discussion Radiographic assessments are a key part of the workup for potential TKA candidates, yet the authors found no evidence to indicate whether surgeons radiographic assessment were valid or reliable. The centrally adjudicated OAI radiographic data were used as the gold standard measures to determine whether orthopedic surgeons could make valid determinations of the severity of knee osteoarthritis. Both experienced and inexperienced surgeons obtained valid judgments of knee osteoarthritis severity in future surgical knees for specific tibiofemoral joint space assessment and for generalized knee osteoarthritis severity. However, some surgeon assessments had weak to moderate validity for some measures. Surprisingly, error was not dependent on experience. Reliability evidence was similar in that surgeons who made highly valid judgments were also highly reliable. Interrater reliability fell below that of intrarater reliability compared with surgeons who demonstrated high intrarater reliability. Validity and reliability judgments were consistently lower for contralateral knees than for future surgical knees. The current study had some limitations. Most importantly, reliability was only examined for 4 surgeons, 2 with minimal experience and 2 with substantial experience. Therefore, validity and reliability estimates may have limited generalizability. The training procedures used were minimal in that surgeons were provided summary reviews of the KL and OARSI classification systems and radiographic examples for each scale In addition, surgeons practiced grading at least 10 sample radiographs or until the surgeons reported that they felt prepared to apply the scales to clinical radiographs. This training, although minimal, is easily replicable by others and has been used previously. 24 The findings suggest that this amount of training is adequate, at least for some surgeons, although other surgeons may require more training to achieve acceptable validity and reliability. Future studies should use additional training, which may assist those who continue to have questions regarding scale interpretation. The data also suggest that the amount of training needed is likely unrelated to experience. The scales used in the current study are among the most commonly Table 3 Validity Coefficients Comparing Surgeon Ratings to the Gold Standard Measurement n kw 95% CI for kw Agreement, % Future surgical Experienced surgeon OARSI medial joint space (#1) OARSI lateral joint space (#1) Kellgren-Lawrence grade (#1) OARSI medial joint space (#2) OARSI lateral joint space (#2) Kellgren-Lawrence grade (#2) Inexperienced surgeon OARSI medial joint space (#3) OARSI lateral joint space (#3) Kellgren-Lawrence grade (#3) OARSI medial joint space (#4) OARSI lateral joint space (#4) Kellgren-Lawrence grade (#4) Contralateral knee Experienced surgeon OARSI medial joint space (#1) OARSI lateral joint space (#1) Kellgren-Lawrence grade (#1) OARSI medial joint space (#2) OARSI lateral joint space (#2) Kellgren-Lawrence grade (#2) Inexperienced surgeon OARSI medial joint space (#3) OARSI lateral joint space (#3) Kellgren-Lawrence grade (#3) OARSI medial joint space (#4) OARSI lateral joint space (#4) N/A 10.2 Kellgren-Lawrence grade (#4) Abbreviations: CI, confidence interval; kw, weighted kappa; N/A, not available; OARSI, Osteoarthritis Research Society International. described in the literature, but if surgeons have not used the scales, past experience, or a lack of it, does not appear to assist in learning and applying the scales to future patients. The scales are typically used for different purposes. The KL classification system is used to judge the osteoarthritis status of the entire tibiofemoral joint, JANUARY 2013 Volume 36 Number 1 e29

6 whereas the OARSI classification system describes the extent of osteoarthritis in the medial and lateral compartments separately. Therefore, they are not designed to be used interchangeably. The current results are the first to report validity and reliability estimates for orthopedic surgeons evaluating knee radiographs Table 4 Intrarater Reliability for Surgeon Ratings of Knee OA Status Measurement n kw 95% CI for kw Agreement, % Surgical knee Experienced surgeon OARSI medial joint space (#1) OARSI lateral joint space (#1) Kellgren-Lawrence grade (#1) OARSI Medial joint space (#2) OARSI lateral joint space (#2) Kellgren-Lawrence grade (#2) Inexperienced surgeon OARSI medial joint space (#3) OARSI lateral joint space (#3) Kellgren-Lawrence grade (#3) OARSI medial joint space (#4) OARSI lateral joint space (#4) Kellgren-Lawrence grade (#4) Contralateral knee Experienced surgeon OARSI medial joint space (#1) OARSI lateral joint space (#1) Kellgren-Lawrence grade (#1) OARSI medial joint space (#2) OARSI lateral joint space (#2) Kellgren-Lawrence grade (#2) Inexperienced surgeon OARSI medial joint space (#3) OARSI lateral joint space (#3) Kellgren-Lawrence grade (#3) OARSI medial joint space (#4) , OARSI lateral joint space (#4) Kellgren-Lawrence grade (#4) Abbreviations: CI, confidence interval; kw, weighted kappa; OA, osteoarthritis; OARSI, Osteoarthritis Research Society International. of TKA candidates. Validity estimates compared with the gold standard criterion of experienced OAI rheumatologists and musculoskeletal radiologists highly trained in use of KL and OARSI classification systems were in the substantial to almost perfect range for 1 experienced and 1 inexperienced surgeon. Intra- and interrater reliability estimates for surgeons in the current study are similar to or higher than kw estimates reported for rheumatologists readings of 50 patients with knee osteoarthritis using the KL and OARSI classification systems. 25 The current reliability estimates also appear to be similar to or higher than those reported by Mehta et al, 8 who used a scale similar to the OARSI classification system and reported knee osteoarthritis reliability estimates for patients 5 years after anterior cruciate ligament injury. Intrarater reliability for the current study was also comparable with that reported for the OAI central adjudicating experts. Intrarater KL and OARSI kw estimates for medial joint space narrowing in the current study ranged from 0.57 to 0.93, whereas the OAI experts had kw estimates ranging from 0.70 to These data strongly support the reliability of surgeon estimates in the current study, and more specifically for the 2 surgeons whose reliability estimates were particularly high. No other studies were found that provided validity and reliability estimates for the involved (and more severe) knees vs the contralateral (and lesser involved) knees (Table 1). The evidence suggests that measurements taken on more involved knees are more valid and reliable, particularly for interrater reliability, than those taken from the contralateral knee. The authors suspect this is the case because a greater proportion of involved knees had end-stage disease, which are likely easier to classify. Knees that have less osteoarthritis are more difficult to grade and contribute greater error for validity and reliability. Surgeons should recognize the risk of additional error in osteoarthritis classification when considering interventions for patients with mild or moderate knee osteoarthritis. Conclusion The authors found most validity and reliability estimates of the severity of knee osteoarthritis, as measured by KL and e30 ORTHOPEDICS Healio.com/Orthopedics

7 Radiographic Knee Osteoarthritis Measures Riddle et al OARSI classification systems, to be substantial or better for 2 of the 4 surgeons included in the study. Validity and reliability does not appear to depend on extent of surgeon experience. Rather, it appears some surgeons require additional training with the scales to obtain an acceptable level of validity and reliability. The findings for 2 of the 4 surgeons are generally consistent with literature that has examined psychometric properties of the KL or OARSI classification systems used by either rheumatologists or radiologists. To enhance communication among colleagues and patients, surgeons should consider the routine use of the OARSI classification system to categorize joint space width and the KL classification system to categorize the general osteoarthritis status of the tibiofemoral joint. Some training in the use of the KL and OARSI classification systems would likely assist surgeons in obtaining reliable and valid estimates. Future research should attempt to determine sources of Table 5 Interrater Reliability Comparing Experienced to Inexperienced Surgeons Measurement n kw 95% CI for kw Agreement, % Future surgical knee OARSI medial joint space (experienced) OARSI lateral joint space (experienced) Kellgren-Lawrence grade (experienced) OARSI medial joint space (inexperienced) OARSI lateral joint space (inexperienced) Kellgren-Lawrence grade (inexperienced) Contralateral knee OARSI medial joint space (experienced) OARSI lateral joint space (experienced) Kellgren-Lawrence grade (experienced) OARSI medial joint space (inexperienced) OARSI lateral joint space (inexperienced) Kellgren-Lawrence grade (inexperienced) Abbreviations: CI, confidence interval; kw, weighted kappa; OARSI, Osteoarthritis Research Society International. error for surgeons who are unable to obtain reliable judgments after training similar to that described in the current study. References 1. Cross WW III, Saleh KJ, Wilt TJ, Kane RL. Agreement about indications for total knee arthroplasty. Clin Orthop Relat Res. 2006; (446): NIH Consensus Panel. NIH Consensus Statement on total knee replacement December 8-10, J Bone Joint Surg Am. 2004; 86(6): Valdes AM, Doherty SA, Zhang W, Muir KR, Maciewicz RA, Doherty M. Inverse relationship between preoperative radiographic severity and postoperative pain in patients with osteoarthritis who have undergone total joint arthroplasty. Semin Arthritis Rheum. 2012; 41(4): Cushnaghan J, Bennett J, Reading I, et al. Long-term outcome following total knee arthroplasty: a controlled longitudinal study. Ann Rheum Dis. 2009; 68(5): Trickett RW, Hodgson P, Forster MC, Robertson A. The reliability and accuracy of digital templating in total knee replacement. J Bone Joint Surg Br. 2009; 91(7): The B, Diercks RL, van Ooijen PM, van H Jr. Comparison of analog and digital preoperative planning in total hip and knee arthroplasties. A prospective study of 173 hips and 65 total knees. Acta Orthop. 2005; 76(1): Hirschmann MT, Konala P, Amsler F, Iranpour F, Friederich NF, Cobb JP. The position and orientation of total knee replacement components: a comparison of conventional radiographs, transverse 2D-CT slices and 3D-CT reconstruction. J Bone Joint Surg Br. 2011; 93(5): Mehta VM, Paxton LW, Fornalski SX, Csintalan RP, Fithian DC. Reliability of the international knee documentation committee radiographic grading system. Am J Sports Med. 2007; 35(6): Reichmann WM, Maillefert JF, Hunter DJ, Katz JN, Conaghan PG, Losina E. Responsiveness to change and reliability of measurement of radiographic joint space width in osteoarthritis of the knee: a systematic review. Osteoarthritis Cartilage. 2011; 19(5): Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis. 1957; 16(4): Altman RD, Hochberg M, Murphy WA Jr, Wolfe F, Lequesne M. Atlas of individual radiographic features in osteoarthritis. Osteoarthritis Cartilage. 1995; 3(suppl A): Altman RD, Gold GE. Atlas of individual radiographic features in osteoarthritis, revised. Osteoarthritis Cartilage. 2007; 15(suppl A):A1-A Riddle DL, Jiranek WA, Neff RS, Whitaker D, Hull JR. Extent of tibiofemoral osteoarthritis before knee arthroplasty: multicenter data from the osteoarthritis initiative. Clin Orthop Relat Res. 2012; 470(10): Lester G. Clinical research in OA the NIH Osteoarthritis Initiative. J Musculoskelet Neuronal Interact. 2008; 8(4): Nunnally JC. Psychometric Theory. New York, NY: McGraw-Hill; Sim J, Wright CC. The kappa statistic in reliability studies: use, interpretation, and sample size requirements. Phys Ther. 2005; 85(3): Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977; 33(1): Kothari M, Guermazi A, von IG, et al. Fixedflexion radiography of the knee provides reproducible joint space width measurements in osteoarthritis. Eur Radiol. 2004; 14(9): Brandt KD, Mazzuca SA, Conrozier T, et al. Which is the best radiographic protocol for a clinical trial of a structure modifying drug in patients with knee osteoarthritis? J Rheumatol. 2002; 29(6): Peterfy C, Li J, Zaim S, et al. Comparison of fixed-flexion positioning with fluoroscopic JANUARY 2013 Volume 36 Number 1 e31

8 semi-flexed positioning for quantifying radiographic joint-space width in the knee: test-retest reproducibility. Skeletal Radiol. 2003; 32(3): Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis. 1957; 16(4): Altman RD, Hochberg M, Murphy WA Jr, Wolfe F, Lequesne M. Atlas of individual radiographic features in osteoarthritis. Osteoarthritis Cartilage. 1995; 3(suppl A): Felson DT. Central Reading of Knee X-rays for K-L Grade and Individual Radiographic Features of Knee OA. The Osteoarthritis Initiative Web site. datarelease/default.asp Published February 26, Accessed October 15, Bagge E, Bjelle A, Valkenburg HA, Svanborg A. Prevalence of radiographic osteoarthritis in two elderly European populations. Rheumatol Int. 1992; 12(1): Gossec L, Jordan JM, Mazzuca SA, et al. Comparative evaluation of three semi-quantitative radiographic grading techniques for knee osteoarthritis in terms of validity and reproducibility in 1759 X-rays: report of the OARSI-OMERACT task force. Osteoarthritis Cartilage. 2008; 16(7): e32 ORTHOPEDICS Healio.com/Orthopedics