Comparative Risk of Transfusion and Incremental Total Hospitalization Cost for Primary Unilateral, Bilateral, and Revision Total Knee Arthroplasty Procedures

• Published in: The Journal of arthroplasty Vol. 31; no. 3; pp. 583 - 589.e1
• Main Authors: Nichols, Christine I., MA, MBA, Vose, Joshua G., MD
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.03.2016
• Subjects: Adult; Aged; Aged, 80 and over; Arthroplasty, Replacement, Knee - economics; Arthroplasty, Replacement, Knee - methods; Blood Transfusion; costs; Databases, Factual; Female; Health Resources; Hospital Costs; Hospitalization - economics; Humans; Length of Stay; Logistic Models; Male; Middle Aged; Orthopedics; Reoperation - economics; Retrospective Studies; revision knee arthroplasty; Risk Assessment - methods; simultaneous bilateral knee arthroplasty; total knee arthroplasty; transfusion; total knee arthroplasty; costs; transfusion; simultaneous bilateral knee arthroplasty; revision knee arthroplasty
• ISSN: 0883-5403; 1532-8406
• DOI: 10.1016/j.arth.2015.10.013

Abstract Background This study evaluated the comparative risk of autologous and allogenic blood transfusion, inhospital complications, and incremental total hospitalization costs for primary unilateral, simultaneous bilateral, and revision total knee arthroplasty (TKA) procedures. Methods Using the Premier Perspective database, we identified adults who underwent primary unilateral, simultaneous bilateral, or revision TKA procedures. Logistic regression, controlling for patient and hospital characteristics, was used to determine the risk of autologous or allogeneic blood transfusion. Controlling for the same factors, generalized linear models predicted incremental total hospitalization cost associated with transfusion. Results Between January 2008 and June 2014, 513,558 primary unilateral, 33,977 bilateral, and 32,494 revision TKA patients met selection criteria. The overall percentage receiving a transfusion was 14.1% for unilateral, 36.3% for bilateral, and 20.0% for revision procedures. Logistic regression showed patients aged >65 years, female gender, Northeastern location, large hospitals, and higher Charlson score to be significantly associated with higher transfusion risk. Although overall risk of transfusion decreased over the study period, patients with Charlson score ≥3 were at 2.27 (primary unilateral), 1.88 (bilateral), and 2.44 (revision) greater odds of transfusion compared with healthy controls (Charlson score = 0). Generalized linear models showed an incremental total hospitalization cost among those receiving a transfusion of $2477, $4235, and $8594, respectively, compared with those without transfusion. Conclusions Transfusion risk remains a significant burden in select patient populations and procedures. The incremental cost of receiving a transfusion is significant, including not only direct costs but also staff time and increased hospital resource use.