Journal of Orthoaedic Surgery 2014;22(3):378-82 Association of surgeon factors with outcome scores after total knee arthrolasty Sok-Chuen Tan, Yiong-Huak Chan, Hwei-Chi Chong, Pak-Lin Chin, Andy Yew, Shi-Lu Chia, Darren Tay, Ngai-Nung Lo, Seng-Jin Yeo Deartment of Orthoaedic Surgery, Singaore General Hosital, Singaore ABSTRACT Purose. To identify reoerative factors (including surgeon factors) associated with outcome scores after total knee arthrolasty (TKA). Methods. Medical records of 2848 atients (3458 knees) who underwent rimary TKA by 27 orthoaedic secialists were retrieved. Three secialty knee surgeons who had one-year fellowshi in TKA erformed 1930 TKAs, and 24 general orthoaedic surgeons erformed 1528 TKAs. Four of them (including all 3 secialty knee surgeons) were ultrahigh-volume ( 100 TKAs a year), and 21 of them were senior consultants ( 5 years ost residency). At 2 years, 2922 (85%) of knees had comlete followu data. Oxford Knee Score, Knee Society knee and function scores, and SF-36 quality-of-life score were assessed by indeendent hysiotheraists before and after surgery. Outcomes were comared in terms of dichotomised secialty, seniority, and surgical volume of surgeons. Results. Comaring ultra-high-volume ( 100 TKAs er year) secialty knee surgeons with general orthoaedic surgeons, the former achieved better outcomes in terms of the Oxford Knee Score at 6 months, Knee Society knee and function scores at 2 years, and SF-36 scores at 6 months and 2 years. Comaring lower-volume (<100 TKAs er year) secialty knee surgeons with general orthoaedic surgeons, the former still achieved better outcome and quality-of-life scores, excet for SF-36 Mental Comonent Score at 2 years. Conclusion. Secialty training and clinical research in TKA imroved outcome and quality-of-life scores. Key words: arthrolasty, relacement, knee; hositals, high-volume INTRODUCTION Total knee arthrolasty (TKA) is a cost-effective treatment for advanced osteoarthritis of the knee. 1 Its revision rate is <10% after 10 years and <20% after 15 years. 2 Comlication rates were lower for hositals and surgeons that erform >200 and >50 rocedures Address corresondence and rerint requests to: Dr Sok Chuen Tan, Deartment of Orthoaedic Surgery, Singaore General Hosital, Outram Road, 169608, Singaore. Email: tan.sokchuen@gmail.com
Vol. 22 No. 3, December 2014 Association of surgeon factors with outcome scores after TKA 379 er year, resectively. 3 High-volume hositals and surgeons achieve better outcome scores. 4 A systematic review found an association between low-volume surgeons and worse outcomes, but it did not mention about surgeon training in relation to atient outcomes. 5 Increased hosital secialisation is associated with imroved atient outcomes in terms of adverse events. 6 This study aimed to determine whether surgeon factors (esecially surgeon s training) were associated with outcome scores after TKA. MATERIALS AND METHODS This study was aroved by the institutional ethics review board. Medical records of 2848 atients (3458 knees) who underwent rimary TKA by 27 orthoaedic secialists from January 2004 to May 2007 were retrieved from the national joint registry. Physiotheray was standardised and consisted of early range-of-motion exercises and full weightbearing walking. Three secialty knee surgeons who had one-year fellowshi in TKA erformed 1930 TKAs, and 24 general orthoaedic surgeons erformed 1528 TKAs. Four of them (including all 3 secialty knee surgeons) were ultra-high-volume ( 100 TKAs a year), and 21 of them were senior consultants ( 5 years ost residency). At 2 years, 2922 (85%) of knees had comlete follow-u data. The remaining was lost to follow-u owing to financial or social reasons (n=536) Table 1 Patient demograhics and reoerative scores Variable Mean±SD Secialty knee surgeon grou (n=1930) General orthoaedic surgeon grou (n=1528) Patient age (years) 67.4±7.72 66.8±7.47 0.04 Body mass index (kg/m 2 ) 27.9±4.52 27.9±4.48 0.98 Length of hosital stay (days) 6.1±3.99 6.5±3.92 <0.01 Charlson comorbidity index 2.76±1.22 2.72±1.19 0.321 Preo Oxford Knee Score 36.0±8.1 35.5±8.0 0.110 Preo Knee Society knee score 38.2±18.4 41.2±18.9 <0.001 Preo Knee Society function score 48.4±18.6 50.1±18.7 0.010 Preo SF-36 Physical Comonent Score 33.0±11.7 32.5±11.6 0.193 Preo SF-36 Mental Comonent Score 50.7±11.6 51.2±11.1 0.204 Table 2 Association of reoerative factors with SF-36 scores Variable SF-36 Physical Comonent Score SF-36 Mental Comonent Score At 6 months At 2 years At 6 months At 2 years Lower-volume general -2.4 (-3.3 to -1.4) <0.001-1.8 (-2.7 to -0.9) <0.001-1.2 (-2.0 to -0.3) 0.010-0.7 (-1.5 to 0.2) 0.142 orthoaedic surgeons vs. secialty knee surgeons Ultra-high-volume -1.5 (-2.9 to -0.1) 0.042-1.6 (-2.9 to -0.3) 0.015-1.9 (-3.1 to -0.6) 0.005-2.0 (-3.3 to -0.8) 0.001 general orthoaedic surgeons vs. secialty knee surgeons Senior vs. junior consultants 1.0 (-0.5 to 2.6) 0.199-1.0 (-2.6 to 0.5) 0.182 0.5 (-0.9 to 2.0) 0.490 2.2 (0.8 to 3.7) 0.002 Patient age -0.1 (-0.2 to -0.02) 0.016-0.10 (-0.18 to -0.03) 0.005 0.02 (-0.05 to 0.09) 0.504 0.03 (-0.04 to 0.10) 0.360 Body mass index -0.14 (-0.24 to -0.04) 0.004-0.19 (-0.28 to -0.09) <0.001 0.05 (-0.04 to 0.14) 0.257-0.01 (-0.15 to 0.02) 0.143 Males vs. females 2.1 (1.0 to 3.2) <0.001 1.5 (0.5 to 2.5) 0.005 1.1 (0.4 to 2.5) 0.005 1.0 (-0.03 to 2.0) 0.057 Charlson comorbidity index -0.62 (-1.1 to -0.14) 0.012-0.65 (-1.1 to -0.19) 0.005-0.58 (-1.0 to -0.14) 0.010-0.65 (-1.1 to -0.2) 0.004 Corresonding reo scores 0.20 (0.17 to 0.24) <0.001 0.14 (0.10 to 0.17) <0.001 0.40 (0.36 to 0.43) <0.001 0.37 (0.33 to 0.40) <0.001
380 SC Tan et al. Journal of Orthoaedic Surgery Variable Table 3 Association of reoerative factors with Oxford Knee Score and Knee Society knee and function scores Oxford Knee Score At 6 months At 2 years b estimate b estimate Lower-volume general orthoaedic surgeons vs. 1.18 (0.65 to 1.70) <0.001 0.91 (0.45 to 1.36) <0.001 secialty knee surgeons Ultra-high-volume general orthoaedic surgeons vs. 1.30 (0.54 to 2.07) 0.001 0.81 (0.16 to 1.47) 0.015 secialty knee surgeons Senior consultant vs. junior consultant -0.46 (-1.32 to 0.41) 0.300 0.16 (-0.61 to 0.93) 0.689 Patient age 0.02 (-0.02 to 0.07) 0.266 0.07 (0.03 to 0.10) <0.001 Body mass index 0.04 (-0.01 to 0.09) 0.130 0.07 (0.02 to 0.12) 0.003 Males vs. females -0.76 (-1.38 to -0.15) 0.014-0.23 (-0.77 to 0.30) 0.108 Charlson comorbidity index 0.52 (0.26 to 0.79) <0.001 0.32 (0.09 to 0.56) 0.006 Corresonding reo scores 0.22 (0.19 to 0.25) <0.001 0.17 (0.14 to 0.19) <0.001 or had died from various reasons (n=65). Oxford Knee Score, 7,8 Knee Society knee and function scores, 9 and SF-36 quality-of-life score 10 were assessed by indeendent hysiotheraists before and after (6 months and 2 years) surgery. Outcomes were comared in terms of dichotomised secialty (secialty knee surgeons vs. general orthoaedic surgeons), seniority (senior vs. junior consultants), and surgical volume (ultra-high vs. lower) of surgeons. To account for correlation from the same surgeons erforming the surgeries and for atients having multile TKAs, a mixed model analysis was erformed to comare the erformance of the lower-volume secialty knee surgeons with general orthoaedic surgeons. Other factors including atient age, gender, body mass index, Charlson comorbidity index, 11 and corresonding reoerative scores were also analysed for association with outcome scores. The Charlson comorbidity index was adjusted for age and translated from the International Classification of Diseases, Ninth Revision, Clinical Modification codes, according to Deyo et al. 12 Comarisons were made using the Student s t test or the Wilcoxon rank-sum test after checking for normality and homogeneity assumtions. A value of <0.05 was considered statistically significant. RESULTS The secialty knee surgeons and general orthoaedic surgeons were similar in terms of gender, body mass index, Charlson comorbidity index, and reoerative scores, excet that atients in the former grou was older (67.4 vs. 66.8 years, =0.04), had shorter length of hosital stay (6.1 vs. 6.5 days, <0.01), and had worse reoerative Knee Society function score (48.4 vs. 50.1, =0.01) and knee score (38.2 vs. 41.2, <0.001) [Table 1]. Comaring ultra-high-volume ( 100 TKAs er year) secialty knee surgeons with general orthoaedic surgeons, the former achieved better outcomes in terms of the Oxford Knee Score at 6 months, Knee Society knee and function scores at 2 years, and SF-36 scores at 6 months and 2 years. Comaring lower-volume (<100 TKAs er year) secialty knee surgeons with general orthoaedic surgeons, the former still achieved better outcome and quality-of-life scores, excet for SF-36 Mental Comonent Score at 2 years (Tables 2 and 3). Younger atients, lower body mass index, male gender, lower Charlson comorbidity score, and better corresonding reoerative scores were other factors associated with better outcome and quality-of-life scores. DISCUSSION Secialisation of orthoaedic surgeries imroves outcomes after adjusting for atient characteristics and rocedural volume, but the outcome was based on comlications rather than clinical and quality-oflife scores. 6 This hosital erforms a mean of 1200 TKAs er year and is considered ultra-high-volume, according to Katz s definition. 3 In this series, the infection rate was 1.67%, and the revision rate was 0.78% at 3 years. To further imrove outcome, secialty training was indicated. Secialty training imroved outcome
Vol. 22 No. 3, December 2014 Association of surgeon factors with outcome scores after TKA 381 Knee Society knee score Knee Society function score At 6 months At 2 years At 6 months At 2 years b estimate b estimate b estimate b estimate -2.08 (-3.77 to -0.40) 0.023-1.89 (-2.84 to -0.93) <0.001-3.36 (-4.87 to -1.85) <0.001-3.02 (-5.01 to -1.03) 0.009-1.17 (-4.15 to 1.81) 0.341-1.56 (-2.93 to 0.19) 0.025-1.98 (-4.18 to 0.23) 0.079-3.53 (-7.03 to -0.02) 0.049 3.18 (1.15 to 5.21) 0.003 0.17 (-1.44 to 1.79) 0.835 1.56 (-0.92 to 4.02) 0.217 0.28 (-2.45 to 3.00) 0.838 0.04 (-0.05 to 0.12) 0.374 0.03 (-0.04 to 0.11) 0.400-0.41 (-0.53 to -0.28) <0.001-0.53 (-0.66 to -0.41) <0.001 0.11 (0.01 to 0.22) 0.041-0.04 (-0.14 to 0.05) 0.356-0.47 (-0.62 to -0.32) <0.001-0.59 (-0.75 to -0.44) <0.001 0.78 (-0.44 to 2.0) 0.208 0.57 (-0.53 to 1.68) 0.311 4.99 (3.24 to 6.73) <0.001 3.40 (1.65 to 5.14) <0.001-0.60 (-1.13 to -0.07) 0.026-0.50 (-0.99 to -0.02) 0.043-1.48 (-2.24 to -0.72) <0.001-2.04 (-2.81 to -1.28) <0.001 0.03 (0.004 to 0.05) 0.022 0.03 (0.01 to 0.06) 0.004 0.39 (0.35 to 0.43) <0.001 0.31 (0.27 to 0.35) <0.001 and quality-of-life scores. Mixed model analysis showed that secialty knee surgeons achieved better outcome than general orthoaedic surgeons after adjusting for surgeon volume. One robable reason is that secialty surgeons handled TKAs with severe malalignment and thus had better oerative skills and intra-oerative judgement. In this hosital, secialty knee surgeons send u to one year in an overseas fellowshi rogramme in renowned centres. This exoses them to more challenging cases and enables them to learn from highly exerienced surgeons. Secialty knee surgeons focus on knee surgeries and are not involved with other tyes of orthoaedic surgery; it is difficult for general orthoaedic surgeons to be equally cometent in all subsecialty fields. 13,14 Secialty knee surgeons have secial interests in ostoerative ain control and analgesia, which facilitates rehabilitation. 15,16 Secialty knee surgeons may have better atient-hysician communication skills secific to TKA atients. This may lead to better atient selection, communication, and motivation, as comared to high-volume general orthoaedic surgeons who erform assembly-line surgery. Preoerative scores have significant imact on ostoerative outcome. Better reoerative Knee Society knee and function scores are associated with better lower limb strength and thus facilitate rehabilitation. The advantage conferred by better reoerative scores may be neutralised after years of intensive rehabilitation, but elderly atients are not likely to comly with intensive hysiotheray. Carer and home suort may imrove comliance and rehabilitation. More motivated carers may be more effective in encouraging early range-of-motion exercises and ambulation. Patient motivation may also affect outcome. The Charlson comorbidity index incororates age and medical conditions. Age correlates with muscle mass and motivation for rehabilitation and thus outcomes. Diabetes is associated with ostoerative knee stiffness. 17 Comlications such as infection and imlant loosening and wear were not addressed in this study; longer follow-u is necessary. CONCLUSION Secialty training and clinical research in TKA imroved outcome and quality-of-life scores, but there may be other confounders such as choice of rostheses and their design. DISCLOSURE No conflicts of interest were declared by the authors. REFERENCES 1. Losina E, Walensky RP, Kessler CL, Emrani PS, Reichmann WM, Wright EA, et al. Cost-effectiveness of total knee arthrolasty in the United States: atient risk and hosital volume. Arch Intern Med 2009;169:1113 22. 2. Rand JA, Trousdale RT, Ilstru DM, Harmsen WS. Factors affecting the durability of rimary total knee rostheses. J Bone
382 SC Tan et al. Journal of Orthoaedic Surgery Joint Surg Am 2003;85:259 65. 3. Katz JN, Barrett J, Mahomed NN, Baron JA, Wright RJ, Losina E. Association between hosital and surgeon rocedure volume and the outcomes of total knee relacement. J Bone Joint Surg Am 2004;86:1909 16. 4. Katz JN, Mahomed NN, Baron JA, Barrett JA, Fossel AH, Creel AH, et al. Association of hosital and surgeon rocedure volume with atient-centered outcomes of total knee relacement in a oulation-based cohort of atients age 65 years and older. Arthritis Rheum 2007;56:568 74. 5. Lau RL, Perruccio AV, Gandhi R, Mahomed NN. The role of surgeon volume on atient outcome in total knee arthrolasty: a systematic review of the literature. BMC Musculoskeletal Disorders 2012;13:250. 6. Hagen TP, Vaughan-Sarrazin MS, Cram P. Relation between hosital orthoaedic secialisation and outcomes in atients aged 65 and older: retrosective analysis of US Medicare data. BMJ 2010;340:c165. 7. Conaghan PG, Emerton M, Tennant A. Internal construct validity of the Oxford Knee Scale: evidence from Rasch measurement. Arthritis Rheum 2007;57:1363 7. 8. Xie F, Li SC, Lo NN, Yeo SJ, Yang KY, Yeo W, et al. Cross-cultural adatation and validation of Singaore English and Chinese Versions of the Oxford Knee Score (OKS) in knee osteoarthritis atients undergoing total knee relacement. Osteoarthritis Cartilage 2007;15:1019 24. 9. Insall JN, Dorr LD, Scott RD, Scott WN. Rationale of the Knee Society clinical rating system. Clin Ortho Relat Res 1989:248:13 4. 10. Busija L, Osborne RH, Nilsdotter A, Buchbinder R, Roos EM. Magnitude and meaningfulness of change in SF-36 scores in four tyes of orthoedic surgery. Health Qual Life Outcomes 2008;6:55. 11. Charlson M, Szatrowski TP, Peterson J, Gold J. Validation of a combined comorbidity index. J Clin Eidemiol 1994;47:1245 51. 12. Deyo RA, Cherkin DC, Ciol MA. Adating a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Eidemiol 1992;45:613 9. 13. Sarmiento A. Subsecialization in orthoaedics. Has it been all for the better? J Bone Joint Surg Am 2003;85:369 73. 14. Starr A. Subsecialization in orthoaedics: is there really too much? J Bone Joint Surg Am 2003;85:1849 50. 15. Pang HN, Lo NN, Yang KY, Chong HC, Yeo SJ. Peri-articular steroid injection imroves the outcome after unicondylar knee relacement: a rosective, randomised controlled trial with a two-year follow-u. J Bone Joint Surg Br 2008;90:738 44. 16. Ng YC, Lo NN, Yang KY, Chia SL, Chong HC, Yeo SJ. Effects of eriarticular steroid injection on knee function and the inflammatory resonse following Unicondylar Knee Arthrolasty. Knee Surg Sorts Traumatol Arthrosc 2011;19:60 5. 17. Schiavone Panni A, Cerciello S, Vasso M, Tartarone M. Stiffness in total knee arthrolasty. J Ortho Traumatol 2009;10:111 8.