Background/Purpose: TKR is commonly performed for rheumatoid arthritis (RA) and osteoarthritis (OA). Historically, RA patients were at higher risk of post-operative AEs. The purpose of this study was to evaluate whether this is true in the era of widespread DMARD and biologic use.

Methods: Patients participating in the institution’s TKR registry between 2007 and 2010 were screened for RA by ICD-9 code or self report, and the diagnosis confirmed by chart review. AEs were identified by 6 month questionnaire, and review of office and hospital notes. Self-reported AEs were validated by chart review or phone call. Infection was defined as any surgical site infection as well as systemic infections. Each RA patient was matched to 2 controls by age (+/- 5 years), gender, and primary vs. revision surgery. Baseline characteristics of RA and OA patients were compared by standard statistics and bivariate relationships to AE calculated.

Results: 159 RA TKA patients were identified, and matched to 318 OA controls. Mean age: 63 years (range 22-93), 88% female. 20% of RA were on corticosteroids, and 67% were on DMARDS. Assessment of co-morbidities using Charlson co-morbidity scores did not reveal a significant difference between groups. RA had significantly worse baseline WOMAC pain (47.1 vs. 53.7; p-value < 0.01) and function scores (43.8 vs. 54.2; p-value < 0.01), and lower perceived health status on the EQ-5D (0.59 vs. 0.65; p-value < 0.01), and SF-36 PCS (29.3 vs. 33.3; p-value < 0.01). Operative time (144 minutes RA vs. 146 minutes OA; p-value = 0.66) and length of stay (5 days for both) were no different between groups. There were no deep joint infections in either group, and there was no significant difference between OA and RA rates of superficial infection, i.e. cellulitis or stitch abscesses (9.4% RA vs. 10.1%; p-value = 0.82) or thromboembolism (1.3% RA vs. 0.6%; p-value = 0.60). Re-operation was more common in OA (2.5% vs. 8.8% p-value = .015), largely due to manipulations. 

Conclusion: The belief that RA increases post-operative AEs may be outdated. In spite of worse pre-operative function, and high steroid and DMARD use, infection and wound healing complications were not increased in RA in a high volume orthopedic hospital. RA patients additionally had lower rates of re-operation. These data are reassuring, but further study is needed to see if these trends continue and are generalizable.

Adverse Events	RA cases (N=159)	OA controls (N=318)	P-Value
In hospital (post-op, prior to discharge)
Surgical Site Infection (SSI), N (%)	2 (1.3%)	3 (0.9%)	0.99
Other infection, N (%)	2 (1.3%)	7 (2.2%)	0.72
DVT, N (%)	1 (0.6%)	4 (1.3%)	0.67
PE, N (%)	0	1 (0.3%)	0.99
Pneumonia, N (%)	1 (0.6%)	1 (0.3%)	0.99
In office (post-op, within 6 months)
SSI, N (%)	6 (3.8%)	19 (6.0%)	0.39
Other infection, N (%)	3 (1.9%)	4 (1.3%)	0.69
DVT, N (%)	0	1 (0.3%)	0.99
PE, N (%)	1 (0.6%)	1 (0.3%)	0.99
Pneumonia, N (%)	0	0	—
REOP, N (%)	4 (2.5%)	28 (8.8%)	0.02
REOP manipulation	2 (1.3%)	25 (7.9%)	<0.01
REOP revision	1 (0.6%)	3 (0.9%)	0.99
REOP other	1 (0.6%)	0	0.33
Self-report (at 6 months)
Infection, N (%)	1 (0.72%)	6 (2.0%)	0.44
DVT, N (%)	0	8 (2.6%)	0.06
PE, N (%)	1 (0.72%)	1 (0.3%)	0.52
Pneumonia, N (%)	2 (1.5%)	1 (0.3%)	0.23
6-month Totals*
Infection (SSI and other Infection), N (%)	15 (9.4%)	32 (10.1%)	0.83
DVT, N (%)	1 (0.6%)	10 (3.1%)	0.11
PE, N (%)	2 (1.3%)	2 (0.6%)	0.60
Pneumonia, N (%)	3 (1.9%)	2 (0.6%)	0.34
REOP (manipulation, revision, and other), N (%)	4 (2.5%)	28 (8.8%)	0.02
*The same AE identified by multiple sources was counted only once

---

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/rheumatoid-arthritis-does-not-increase-risk-of-short-term-total-knee-replacement-tkr-adverse-events-ae/