Total knee arthroplasty outcomes in lupus: a study using the US National Inpatient Sample
Abstract Objective To assess whether patients with lupus have more complications and/or health care utilization than patients without lupus after primary total knee arthroplasty (TKA). Methods We used diagnostic codes to identify lupus (710.0) in patients who underwent primary TKA (procedure code, 81.54) in the US National Inpatient Sample from 1998–2014. We performed separate logistic regression analyses to assess whether lupus was associated with each in-hospital post-primary TKA complication (implant infection, implant revision, transfusion, mortality) or health care utilization for the index hospitalization. Results Our study cohort consisted of 8 127 282 primary TKAs with a mean age of 66.4 years, 63% were female, 68% White, 97% with OA, and the Deyo-Charlson comorbidity score was zero in 51%; 30 912 TKAs were performed in people with lupus. In multivariable-adjusted analyses, lupus was independently associated with a significantly higher risk of transfusion, but no significant difference in implant infection, implant revision or mortality, with respective odds ratios [ORs] (95% confidence interval [CI]) of 1.34 (1.25, 1.42), 1.38 (0.73, 2.60), 1.35 (0.83, 2.19) and 0.83 (0.34, 2.03). Lupus was independently associated with a significantly higher risk of total hospital charges above the median, the length of hospital stay >3 days (median) and discharge to an inpatient facility, respective ORs (95% CI) were 1.42 (1.22, 1.66), 1.09 (1.04, 1.15) and 1.34 (1.27, 1.41). Conclusion Our finding of higher transfusion rate and health care utilization in lupus compared with non-lupus patients undergoing primary TKA informs surgeons, patients and policy makers regarding TKA outcomes in lupus, and can also guide appropriate resource allocation.