Background Treatments based on measuring disease activity and adjusting therapy accordingly optimizes outcomes in rheumatoid arthritis (RA). In patients with long-standing disease, joint surgeries may be preferred due to considerable joint damage and prior treatment failures. However, the impact of joint surgeries on disease activity awaits more thorough study/investigation.
Objectives Using the Japanese National Database, we evaluated the impact of joint surgery on disease activity using two validated disease activity indices with response criteria: the Disease Activity Score (DAS28) and the Routine Assessment of Patient Index Data 3 (RAPID3).
Methods The subjects were patients with RA of 5 years’ duration or longer who had undergone joint surgery for the upper or lower extremities (synovectomy, arthroplasty, joint fusion, or joint replacement) in 2010 and whose records were available in the Japanese National Database. Patients with insufficient data for calculating the DAS28 and RAPID3 scores before and after operation were excluded. The changes in disease activity after joint surgeries were evaluated by the RAPID3 response, the EULAR-DAS28 response, and the Boolean remission, criteria. We also identified patients who achieved a good or moderate response (without adding or changing medication) and assessed the ideal “surgical window of opportunity” by ROC analysis.
Results After excluding subjects with insufficient data, the remaining 93 patients were analyzed. The patients’ age and disease duration were 64.3±11.3 and 20.9±11.8 (yrs). According to the Steinbrocker classification, 64 (72.3%) were stage IV and 16 (18.2%) were Stage III. Fifty-seven (61.3%) had received joint replacement and thirty-six (38.7%) had undergone other procedures (21 arthroplasties, 9 joint fusions and 6 synovectomies). Seventy-three (78.5%) had undergone joint surgeries of the lower extremities. RAPID3 improved from 12.2±6.20 to 10.9±6.46 (p=0.0347) and DAS28 improved from 3.96±1.28 to 3.72±1.30 (p=0.0393). Good and moderate responses were observed in 13 (14.0%) and 20 (21.5%) subjects, respectively, by the EULAR-DAS28 response criteria and in 7 (7.5%) and 21 (22.6%) of the subjects, respectively, by the RAPID3 response criteria (kappa 0.235 p=0.0016, weighted kappa 0.386 p=0.0001). Nine patients (9.7%) achieved the Boolean remission criteria after operation. By a multiple regression analysis, disease duration, adding or changing medications, and types of operation (joint replacement) influenced patients’ responsiveness in DAS28 and RAPID3. The ROC analysis of good and moderate responses in RAPID3 showed that a disease duration of 11 years or less exhibited a sensitivity of 95% and a specificity of 95% (AUC 0.987), while that of DAS28 showed that a disease duration of 15 years or less exhibited a sensitivity of 75% and a specificity of 68% (AUC 0.759).
Conclusions In patients with long-standing RA, joint surgeries, especially if performed within 11-15 years after the diagnosis, may help improve control of disease activity.
Acknowledgements This work was supported in part by grants from the Ministry of Health, Labour and Welfare, Japan.
Disclosure of Interest None Declared