Summary

Patients who underwent both primary and revision TKAs at a non-tertiary center had a higher rate of postoperative wound infection and those that had at least one of their primary or revision surgeries at a tertiary center had better functional and pain related outcomes in the short term.

Abstract

Introduction

The establishment of “Centers of Excellence”, specialized centers equipped with fellowship-trained arthroplasty surgeons and enhanced hospital resources, are believed to improve outcomes for patients undergoing complex revision procedures. This study aims to compare postoperative complications and patient-reported outcomes (PROs) between individuals who underwent primary and revision TKA at a tertiary center and those who received surgery at non-tertiary centers within the same health system.

Methods

A total of 963 revision TKAs performed between December 2015 and March 2024 were retrospectively reviewed. Patients were stratified into four groups based on classification of their primary and revision surgery hospitals: 1. BT: both primary and revision at tertiary centers, 2. PNRT: primary at non-tertiary, revision at tertiary, 3. BN: both primary and revision at non-tertiary centers, and 4. PTRN: primary at tertiary, revision at non-tertiary. Complication, mortality, and readmit rates were compared with chi-square and Fisher exact tests. Comparison of preoperative and postoperative PROs were made using Kruskal-Wallis test for non-parametric variables and 1-way analysis of variance (ANOVA) for parametric variables.

Results

Most patients were in group BN (n = 433; 45.0%), followed by group BT (n = 389; 40.4%), group PNRT (n = 114; 11.8%), and group PTRN (n = 27; 2.8%). Age (67.7 ± 10.1 for group BT, 66.4 ± 8.8 for group PNRT, 66.3 ± 9.6 for group BN, 69.7 ± 9.0 for group PTRN, p=0.06), sex (59% female in group BT, 54% in group PNRT, 60% in group BN, 70% in group PTRN, p=0.45), BMI (32.6 ± 8.6 for group BT, 33.7 ± 17.5 for group PNRT, 33.1 ± 6.7 for group BN, 32.1 ± 4.4 for group PTRN, p=0.29), and preoperative PROs were similar between all groups (Table 1). Elixhauser scores differed between groups with highest score 3.0 ± 1.8 in group BN, followed by 2.8 ± 1.9 in group PNRT, 2.6 ± 1.9 in group BT, and 2.0 ± 1.6 in group PTRN(p<0.01). Rates of postoperative wound infection were higher in the non-tertiary BN group (7%) compared with 3% in group BT, 2% in group PNRT, and 0% in group PTRN (p=0.03). Of the PROs, 1-year Knee Injury and Osteoarthritis Outcome Score (KOOS) (p=0.04) and 3-month PROMIS10 Physical (p=0.02) were significantly different between groups (p=0.02). Group PNRT had the highest KOOS score at 1-year, while group PTRN had the highest Patient-Reported Outcomes Measurement Information System (PROMIS10) Physical score at 3-months. All other PROs (p>0.05), 1-year mortality (p=0.30), 7-day readmit (p=1.00), 30-day readmit (p=0.75), 90-day readmit (p=0.64), and complication rates such as subsequent re-revision (p=0.50) were similar among the four groups.

Discussion And Conclusion

The results of this study show that patients who underwent both primary and revision TKAs at a non-tertiary center had a higher rate of postoperative wound infection. Additionally, patients who had at least one of their primary or revision surgeries at a tertiary center had better functional and pain related outcomes in the short term. The establishment of “Centers of Excellence” for revision TKA may improve outcomes in this challenging patient population.