Elizabeth W Paxton1, Maria C S Inacio2, Monti Khatod3, Eric Yue4, Tadashi Funahashi3, Thomas Barber4. 1. Surgical Outcomes and Analysis, Kaiser Permanente, 8954 Rio San Diego Drive, Suite 406, San Diego, CA, 92108, USA. Liz.w.paxton@kp.org. 2. Surgical Outcomes and Analysis, Kaiser Permanente, 8954 Rio San Diego Drive, Suite 406, San Diego, CA, 92108, USA. 3. Department of Orthopaedic Surgery, Southern California Permanente Medical Group, West Los Angeles, CA, USA. 4. Department of Orthopaedic Surgery, The Permanente Medical Group, Oakland, CA, USA.
Abstract
BACKGROUND: Considering the cost and risk associated with revision Total knee arthroplasty (TKAs) and Total hip arthroplasty (THAs), steps to prevent these operations will help patients and reduce healthcare costs. Revision risk calculators for patients may reduce revision surgery by supporting clinical decision-making at the point of care. QUESTIONS/PURPOSES: We sought to develop a TKA and THA revision risk calculator using data from a large health-maintenance organization's arthroplasty registry and determine the best set of predictors for the revision risk calculator. METHODS: Revision risk calculators for THAs and TKAs were developed using a patient cohort from a total joint replacement registry and data from a large US integrated healthcare system. The cohort included all patients who had primary procedures performed in our healthcare system between April 2001 and July 2008 and were followed until January 2014 (TKAs, n = 41,750; THAs, n = 22,721), During the study period, 9% of patients (TKA = 3066/34,686; THA=1898/20,285) were lost to followup and 7% died (TKA= 2350/41,750; THA=1419/20,285). The outcome of interest was revision surgery and was defined as replacement of any component for any reason within 5 years postoperatively. Candidate predictors for the revision risk calculator were limited to preoperative patient demographics, comorbidities, and procedure diagnoses. Logistic regression models were used to identify predictors and the Hosmer-Lemeshow goodness-of-fit test and c-statistic were used to choose final models for the revision risk calculator. RESULTS: The best predictors for the TKA revision risk calculator were age (odds ratio [OR], 0.96; 95% CI, 0.95-0.97; p < 0.001), sex (OR, 0.84; 95% CI, 0.75-0.95; p = 0.004), square-root BMI (OR, 1.05; 95% CI, 0.99-1.11; p = 0.140), diabetes (OR, 1.32; 95% CI, 1.17-1.48; p < 0.001), osteoarthritis (OR, 1.16; 95% CI, 0.84-1.62; p = 0.368), posttraumatic arthritis (OR, 1.66; 95% CI, 1.07-2.56; p = 0.022), and osteonecrosis (OR, 2.54; 95% CI, 1.31-4.92; p = 0.006). The best predictors for the THA revision risk calculator were sex (OR, 1.24; 95% CI, 1.05-1.46; p = 0.010), age (OR, 0.98; 95% CI, 0.98-0.99; p < 0.001), square-root BMI (OR, 1.07; 95% CI, 1.00-1.15; p = 0.066), and osteoarthritis (OR, 0.85; 95% CI, 0.66-1.09; p = 0.190). CONCLUSIONS: Study model parameters can be used to create web-based calculators. Surgeons can enter personalized patient data in the risk calculators for identification of risk of revision which can be used for clinical decision making at the point of care. Future prospective studies will be needed to validate these calculators and to refine them with time. LEVEL OF EVIDENCE: Level III, prognostic study.
BACKGROUND: Considering the cost and risk associated with revision Total knee arthroplasty (TKAs) and Total hip arthroplasty (THAs), steps to prevent these operations will help patients and reduce healthcare costs. Revision risk calculators for patients may reduce revision surgery by supporting clinical decision-making at the point of care. QUESTIONS/PURPOSES: We sought to develop a TKA and THA revision risk calculator using data from a large health-maintenance organization's arthroplasty registry and determine the best set of predictors for the revision risk calculator. METHODS: Revision risk calculators for THAs and TKAs were developed using a patient cohort from a total joint replacement registry and data from a large US integrated healthcare system. The cohort included all patients who had primary procedures performed in our healthcare system between April 2001 and July 2008 and were followed until January 2014 (TKAs, n = 41,750; THAs, n = 22,721), During the study period, 9% of patients (TKA = 3066/34,686; THA=1898/20,285) were lost to followup and 7% died (TKA= 2350/41,750; THA=1419/20,285). The outcome of interest was revision surgery and was defined as replacement of any component for any reason within 5 years postoperatively. Candidate predictors for the revision risk calculator were limited to preoperative patient demographics, comorbidities, and procedure diagnoses. Logistic regression models were used to identify predictors and the Hosmer-Lemeshow goodness-of-fit test and c-statistic were used to choose final models for the revision risk calculator. RESULTS: The best predictors for the TKA revision risk calculator were age (odds ratio [OR], 0.96; 95% CI, 0.95-0.97; p < 0.001), sex (OR, 0.84; 95% CI, 0.75-0.95; p = 0.004), square-root BMI (OR, 1.05; 95% CI, 0.99-1.11; p = 0.140), diabetes (OR, 1.32; 95% CI, 1.17-1.48; p < 0.001), osteoarthritis (OR, 1.16; 95% CI, 0.84-1.62; p = 0.368), posttraumatic arthritis (OR, 1.66; 95% CI, 1.07-2.56; p = 0.022), and osteonecrosis (OR, 2.54; 95% CI, 1.31-4.92; p = 0.006). The best predictors for the THA revision risk calculator were sex (OR, 1.24; 95% CI, 1.05-1.46; p = 0.010), age (OR, 0.98; 95% CI, 0.98-0.99; p < 0.001), square-root BMI (OR, 1.07; 95% CI, 1.00-1.15; p = 0.066), and osteoarthritis (OR, 0.85; 95% CI, 0.66-1.09; p = 0.190). CONCLUSIONS: Study model parameters can be used to create web-based calculators. Surgeons can enter personalized patient data in the risk calculators for identification of risk of revision which can be used for clinical decision making at the point of care. Future prospective studies will be needed to validate these calculators and to refine them with time. LEVEL OF EVIDENCE: Level III, prognostic study.
Authors: Steven M Kurtz; Kevin L Ong; Jordana Schmier; Fionna Mowat; Khaled Saleh; Eva Dybvik; Johan Kärrholm; Göran Garellick; Leif I Havelin; Ove Furnes; Henrik Malchau; Edmund Lau Journal: J Bone Joint Surg Am Date: 2007-10 Impact factor: 5.284
Authors: Robert S Namba; Guy Cafri; Monti Khatod; Maria C S Inacio; Timothy W Brox; Elizabeth W Paxton Journal: J Arthroplasty Date: 2013-08-15 Impact factor: 4.757
Authors: Pasqualina L Santaguida; Gillian A Hawker; Pamela L Hudak; Richard Glazier; Nizar N Mahomed; Hans J Kreder; Peter C Coyte; James G Wright Journal: Can J Surg Date: 2008-12 Impact factor: 2.089
Authors: Ralph B D'Agostino; Ramachandran S Vasan; Michael J Pencina; Philip A Wolf; Mark Cobain; Joseph M Massaro; William B Kannel Journal: Circulation Date: 2008-01-22 Impact factor: 29.690
Authors: John Y S Kim; Alexei S Mlodinow; Nima Khavanin; Keith M Hume; Christopher J Simmons; Michael J Weiss; Robert X Murphy; Karol A Gutowski Journal: Plast Reconstr Surg Glob Open Date: 2015-06-05
Authors: Danny Lee; Ryan Lee; Nikhil B Gowda; William V Probasco; Seth Stake; George Ibrahim; Rajeev Pandarinath Journal: J Clin Orthop Trauma Date: 2019-07-17
Authors: Maaike G J Gademan; Liza N Van Steenbergen; Suzanne C Cannegieter; Rob G H H Nelissen; Perla J Marang-Van De Mheen Journal: Acta Orthop Date: 2021-01-22 Impact factor: 3.717
Authors: Erik Lenguerrand; Michael R Whitehouse; Andrew D Beswick; Andrew D Toms; Martyn L Porter; Ashley W Blom Journal: BMJ Open Date: 2017-07-10 Impact factor: 2.692
Authors: Alana R Cuthbert; Stephen E Graves; Lynne C Giles; Gary Glonek; Nicole Pratt Journal: Clin Orthop Relat Res Date: 2021-02-01 Impact factor: 4.755