Incidence and Risk Factors for Blood Transfusion in Total Joint Arthroplasty: Analysis of a Statewide Database

• Published in: The Journal of arthroplasty 2017-09, Vol.32 (9), p.2684-2687.e1
• Main Authors: Slover, James, MD, MS, Lavery, Jessica A., MS, Schwarzkopf, Ran, MD, Iorio, Richard, MD, Bosco, Joseph, MD, Gold, Heather T., PhD
• Format: Article
• Language: English
• Subjects: Aged; Aged, 80 and over; Arthroplasty, Replacement, Hip - adverse effects; Arthroplasty, Replacement, Knee - adverse effects; Blood Loss, Surgical; blood transfusion; Blood Transfusion - statistics & numerical data; California; Databases, Factual; Female; Humans; Joints; Logistic Models; Male; Middle Aged; Multivariate Analysis; Odds Ratio; Orthopedics; Retrospective Studies; total hip arthroplasty; total knee arthroplasty
• ISSN: 0883-5403; 1532-8406
• DOI: 10.1016/j.arth.2017.04.048

Abstract Background Significant attempts have been made to adopt practices to minimize blood transfusion after total joint arthroplasty (TJA) because of transfusion cost and potential negative clinical consequences including allergic reactions, transfusion-related lung injuries, and immunomodulatory effects. We aimed to evaluate risk factors for blood transfusion in a large cohort of TJA patients. Methods We used the all-payer California Healthcare Cost and Utilization Project data from 2006 to 2011 to examine the trends in utilization of blood transfusion among arthroplasty patients (n = 320,746). We performed descriptive analyses and multivariate logistic regression clustered by hospital, controlling for Deyo-Charlson comorbidity index, age, insurance type (Medicaid vs others), gender, procedure year, and race/ethnicity. Results Eighteen percent (n = 59,038) of TJA patients underwent blood transfusion during their surgery, from 15% with single knee to 45% for bilateral hip arthroplasty. Multivariate analysis indicated that compared with the referent category of single knee arthroplasty, single hip had a significantly higher odds of blood transfusion (odds ratio [OR], 1.76; 95% confidence interval [CI], 1.68-1.83), as did bilateral knee (OR, 3.57; 95% CI, 3.20-3.98) and bilateral hip arthroplasty (OR, 6.17; 95% CI, 4.85-7.85). Increasing age (eg, age ≥80 years; OR, 2.99; 95% CI, 2.82-3.17), Medicaid insurance (OR, 1.36; 95% CI, 1.27-1.45), higher comorbidity index (eg, score of ≥3; OR, 2.33; 95% CI, 2.22-2.45), and females (OR, 1.75; 95% CI, 1.70-1.80) all had significantly higher odds of blood transfusion after TJA. Conclusion Primary hip arthroplasties have significantly greater risk of transfusion than knee arthroplasties, and bilateral procedures have even greater risk, especially for hips. These factors should be considered when evaluating the risk for blood transfusions.