Risk Factors Predict Increased Length of Stay and Readmission Rates in Revision Joint Arthroplasty

• Published in: The Journal of arthroplasty 2016-03, Vol.31 (3), p.603-608
• Main Authors: Keswani, Aakash, BA, Lovy, Andrew J., MD, MS, Robinson, John, MD, Levy, Roger, MD, Chen, Darwin, MD, Moucha, Calin S., MD
• Format: Article
• Language: English
• Subjects: Adult; Aged; Arthroplasty, Replacement, Hip - adverse effects; Arthroplasty, Replacement, Knee - adverse effects; Body Mass Index; complications; Databases, Factual; Diabetes Complications; Female; Heart Diseases - complications; Hemorrhage; Humans; length of stay; Length of Stay - statistics & numerical data; Male; Middle Aged; Multivariate Analysis; Orthopedics; Patient Readmission - statistics & numerical data; Quality Improvement; readmission; Reoperation - adverse effects; Retrospective Studies; revision hip arthroplasty; revision knee arthroplasty; Risk Factors; Smoking; Stroke - complications
• ISSN: 0883-5403; 1532-8406
• DOI: 10.1016/j.arth.2015.09.050

Abstract Background This study aimed to identify risk factors for 30-day readmission and extended length of stay (LOS) in revision total knee (RKA) and hip (RHA) arthroplasty patients. Methods Patients who underwent RKA or RHA from 2011 to 2013 were identified in the National Surgical Quality Improvement Program database. Bivariate and multivariate analyses of risk factors for 30-day readmission and extended LOS (>75th percentile) were assessed using preoperative and intraoperative variables. Results A total of 4977 RKA and 5135 RHA patients were reviewed. The most common causes for revision were mechanical (52% RKA, 52% RHA), infection (13% RKA, 8% RHA), dislocation (6% RKA, 13% RHA), and fracture (1% RKA, 4% RHA). Rate of readmission for RKA patients (6.4%; 318 patients) was lower than for RHA patients (8.0%; 409 patients) ( P = .002). Multivariate analysis identified severe adverse event before discharge, male sex, pulmonary disease, stroke, cardiac disease, and American Society of Anesthesiologists class 3 or 4 as significant predictors of readmission (all P ≤ .03). Surgical complications were the more common cause of readmission for both groups. Multivariate analysis of extended LOS identified infection or fracture etiology relative to mechanical loosening etiology, functional status, body mass index greater than 40 kg/m2, history of smoking, diabetes, cardiac disease, stroke, bleeding-causing disorders, wound class 3 or 4, and American Society of Anesthesiologists class 3 or 4 (all P ≤ .05) as independent predictors. Conclusion Modifiable risk factors should be addressed prior to revision total joint arthroplasty to reduce 30-day readmissions and LOS. Future P4P revision arthroplasty models should incorporate procedural diagnosis as rates of readmission and extended LOS significantly differ across procedural etiologies.