Is Hospital-Based Outpatient Revision Total Knee Arthroplasty Safe? An Analysis of 2,171 Outpatient Aseptic Revision Procedures.

BACKGROUND

Outpatient primary total knee arthroplasty (TKA) has been well-established as a safe and effective procedure; however, the safety of outpatient revision TKA remains unclear. Therefore, this study utilized a large database to compare outcomes between outpatient and inpatient revision TKA.

METHODS

An all-payor database was queried to identify patients undergoing revision TKA from 2010 to 2022. Patients who had diagnosis codes related to periprosthetic joint infection (PJI) were excluded. Outpatient surgery was defined as a length of stay < 24 hours. Cohorts were matched by age, sex, Elixhauser Comorbidity Index, comorbidities (diabetes, obesity, tobacco use), components revised (1-versus 2-component), and revision etiology. Medical complications at 90 days and surgical complications at 1 and 2 years postoperatively were evaluated through multivariate logistic regression. A total of 4,342 aseptic revision TKAs were included.

RESULTS

No differences in patient characteristics, procedure type, or revision etiologies were seen between groups. The outpatient cohort had a lower risk of PJI (odds ratio (OR): 0.547, 95% confidence interval (CI): 0.337 to 0.869; P = .012), wound dehiscence (OR: 0.393, 95% CI: 0.225 to 0.658; P < .001), transfusion (OR: 0.241, 95% CI: 0.055 to 0.750; P = .027), reoperation (OR: 0.508, 95% CI: 0.305 to 0.822; P = .007), and any complication (OR: 0.696, 95% CI: 0.584 to 0.829; P < .001) at 90 days postoperatively. At 1 year and 2 years postoperatively, outpatient revision TKA patients had a lower incidence of revision for PJI (OR: 0.332, 95% CI: 0.131 to 0.743; P = .011 and OR: 0.446, 95% CI; 0.217 to 0.859; P = .020, respectively) and all-cause revision (OR: 0.518, 95% CI: 0.377 to 0.706; P < .001 and OR: 0.548, 95% CI: 0.422 to 0.712; P < .001, respectively).

CONCLUSIONS

Our findings suggest that revision TKA can be safely performed on an outpatient basis in appropriately selected patients who do not have an increased risk of adverse events relative to inpatient revision TKA. However, we could not ascertain case complexity in either cohort, and despite controlling for several potential confounders, other less tangible differences could exist between groups.