To the Editor: Introduction of conventional synthetic disease-modifying antirheumatic drugs (csDMARDs) has been highly efficacious in reducing the signs and symptoms of rheumatoid arthritis (RA) and in delaying or preventing structural joint damage. Furthermore, the development and clinical use of biological DMARDs (bDMARDs) and targeted synthetic DMARDs (tsDMARDs) have dramatically altered the prognosis of patients with RA, particularly when they are combined with a treat-to-target regimen (T2T).[1] Consequently, over the last two decades, the rates of total hip arthroplasty (THA) and total knee arthroplasty (TKA) remained stable or even decreased in these special patient populations. Nevertheless, compared to patients with osteoarthritis, patients with RA have a higher risk of having worse functional outcomes and higher rates of complications after arthroplasty. This may be attributed to the complexity of the rheumatic disease process, including high disease activity, comorbidities, and immunosuppressive medications.[2] A key to making perioperative medication management decisions is to balance the risk of periprosthetic infection with the risk of a disease flare.[3]

Post-operative flare is a common phenomenon among patients with RA undergoing orthopedic surgery. However, relevant clinical data are scarce. Only Goodman et al[5] evaluated 120 patients with RA who received THA and TKA and reported a flare rate of 63% 6 weeks after surgery. Therefore, this study aimed to prospectively evaluate the rate, clinical characteristics, and risk factors for post-operative flares in patients with RA within 12 weeks after THA or TKA reconstruction at a single, high-volume tertiary care center for musculoskeletal diseases.

This study was approved by the Ethics Committee of Beijing Jishuitan Hospital (No. 202106–57). Between January 2018 and March 2022, 275 patients who underwent primary unilateral THA or TKA and met the 1987 American College of Rheumatology (ACR) classification criteria or the 2010 ACR/European League Against Rheumatism diagnostic criteria were consecutively recruited. All the patients signed informed written consent forms.

Patient demographics, disease-related parameters, patient-reported outcomes (PROs), and laboratory data were collected. According to the clinical and radiographic evaluation of the involved joints, these patients were classified into polyarticular (≥5 joints), oligoarticular (2–4 joints), and monoarticular (one joint). The Disease Activity Score of 28 joints-C reactive protein (DAS28-CRP) and the Simple Disease Activity Index (SDAI) were used to evaluate RA disease activity. PROs were evaluated using the Health Assessment Questionnaire-Disability Index (HAQ-DI) and Short Form-12 outcome score (SF-12), including the physical component score (PCS) and mental component score (MCS).

Medication status, including non-steroidal anti-inflammatory drugs (NSAIDs), csDMARDs, bDMARDs, tsDMARDs, and glucocorticoids (GCs) were recorded, and the patients were considered sustained users if usage was noted in the patient record during half a year or longer before the evaluation period. The csDMARDs were administered throughout the perioperative period without interruption. All bDMARDs and tsDMARDs were discontinued before surgery, which was performed at the end of dosing cycle. These medications were resumed until the stitch was removed. Perioperatively, 71 patients continued their usual daily dose of GCs, whereas a supraphysiological GC stress dose was not administered.

All patients were monitored on an outpatient basis 2, 4, 8, and 12 weeks after surgery. The judgement criteria for post-operative flare of RA were as follows: (1) rapidly progressive polyarticular swelling, tenderness, and morning stiffness combined with elevated DAS-28 CRP (Δvalue ≥ 1.2); (2) rapidly elevated erythrocyte sedimentation rate (ESR) and CRP levels; (3) rapidly progressive swelling, tenderness, and limited range of motion on the operated hip or knee joint; and (4) effective treatment with GCs. Post-operative flare was defined by the presence of (1) + (2) + (4) or (3) + (4).

Descriptive analyses for categorical variables were based on percentages or frequencies and for continuous variables on mean and standard deviation or median with the first and third quartile. The intergroup differences between flare and non-flare group were calculated using independent sample Student's t-tests or Mann–Whitney tests for continuous variables and chi-squared tests for dichotomous variables. A multivariate logistic regression model was built to assess the possible factors identified as significant in the analysis, and the odds ratio (OR) with 95% confidence interval and the associated P value were determined.

Within the 12-week follow-up period, 41 patients were lost to follow-up or provided insufficient clinical information. Finally, 234 patients were enrolled in our study, 84 (35.9%) of whom reported flares (flare group). The remaining 150 patients (64.1%) were allocated to the non-flare group. A total of 70 (83.3%) patients fulfilled the criteria of (1) + (2) + (4) and were regarded as having systemic flares. The other 14 (16.7%) patients fulfilled the criteria of (3) + (4) and were regarded as local flares. Most flares occurred during the first 4 weeks (58 patients, 69.0%).

The baseline characteristics were compared between flare and non-flare group [Table 1]. The percentages of positive family history and degree of global pain (visual analog scale [VAS]) were significantly higher in flare group (P = 0.024; P = 0.002, respectively). The percentage of TKA performed was significantly lower in flare group (P = 0.044). The DAS28-CRP, SDAI, and HAQ-DI were significantly higher in flare group (P < 0.001; P < 0.001; P = 0.001, respectively). The SF-12 MCS was significantly lower in flare group (P = 0.017). The ESR, CRP, rheumatoid factor (RF), and anti-citrullinated protein antibody (ACPA) levels were also significantly higher in flare group (P = 0.015; P = 0.022; P = 0.014; P = 0.036, respectively). The factors identified as significant were introduced into a multivariate logistic regression model. The model revealed that the significant variables for post-operative flares were HAQ-DI (OR = 2.061, P = 0.017) and SDAI (OR = 1.055, P = 0.011).

Table 1 - Baseline characteristics of the patient demographics, disease-related parameters, medication status, and laboratory parameters between the flare group and the non-flare group. Characteristics Total RA patients (n = 234) Flare group (n = 84) Non-flare group (n = 150) P values BMI (kg/m2) 23.1 (20.5, 26.2) 23.4 (20.8, 27.3) 22.9 (20.2, 25.8) 0.113 Male 40 (17.1) 13 (15.5) 27 (18.0) 0.623 Family history 6 (2.6) 5 (6.0) 1 (0.7) 0.024 Smoking history 22 (9.4) 7 (8.3) 15 (10.0) 0.675 Age at disease onset (years) 44.0 (34.0, 51.3) 43.0 (32.3, 54.0) 45.0 (35.0, 51.0) 0.985 Age at arthroplasty (years) 60.0 (51.0, 66.0) 62.0 (48.3, 67.0) 58.5 (51.8, 66.0) 0.471 Diagnosis delay (years) 2.0 (1.0, 3.0) 1.5 (1.0, 4.0) 2.0 (1.0, 3.0) 0.742 Distribution mode of involved joints 0.160  Polyarticular 175 (74.8) 68 (81.0) 107 (71.3)  Oligoarticular 43 (18.4) 10 (11.9) 33 (22.0)  Monoarticular 16 (6.8) 6 (7.1) 10 (6.7) RDCI 1.0 (0, 2.0) 1.0 (0, 2.0) 1.0 (0, 2.0) 0.420 Global pain (VAS) 6.0 (4.0, 7.0) 6.0 (5.0, 8.0) 5.0 (4.0, 7.0) 0.002 Medication status  NSAIDs 76 (32.5) 30 (35.7) 46 (30.7) 0.429  csDMARDs 108 (46.2) 42 (50.0) 66 (44.0) 0.377  bDMARDs 17 (7.3) 7 (8.3) 10 (6.7) 0.638  tsDMARDs 20 (8.5) 7 (8.3) 13 (8.7) 0.930  Glucocorticoids 71 (30.3) 29 (34.5) 42 (28.0) 0.298 Arthroplasty type 0.044  TKA 158 (67.5) 56 (66.7) 102 (68.0)  THA 76 (32.5) 28 (33.3) 48 (32.0) TJC 3.0 (2.0, 4.0) 3.0 (2.0, 4.0) 3.0 (1.0, 4.0) 0.122 SJC 3.0 (2.0, 5.0) 4.0 (2.0, 6.0) 3.0 (2.0, 4.0) <0.001 DAS28-CRP 3.9 (3.4, 4.5) 4.2 (3.6, 5.0) 3.8 (3.3, 4.2) <0.001 DAS28-ESR 4.5 ± 1.0 4.7 ± 1.0 4.3 ± 1.0 0.001 SDAI 17.6 (13.4, 23.0) 21.3 (15.6, 26.7) 16.1 (12.9, 20.9) <0.001 CDAI 16.0 (12.8, 21.0) 18.5 (14.0, 23.8) 15.0 (12.0, 18.0) <0.001 HAQ-DI 0.9 (0.6, 1.3) 1.0 (0.6, 1.6) 0.8 (0.5, 1.1) 0.001 SF-12 PCS 32.3 (23.4, 39.8) 27.3 (22.3, 38.4) 33.4 (25.5, 40.3) 0.032 SF-12 MCS 40.3 (35.3, 45.6) 37.8 (33.7, 44.2) 41.2 (36.1, 45.6) 0.017 ESR (mm/1h) 31.5 (17.0, 61.0) 41.5 (20.5, 67.3) 28.5 (16.0, 54.8) 0.015 CRP (mg/L) 11.0 (3.8, 26.3) 15.8 (4.3, 38.7) 9.5 (3.8, 19.8) 0.022 HGB (g/L) 122.4 ± 15.8 121.7 ± 15.0 122.8 ± 16.2 0.594 ALB (mg/L) 40.4 (37.5, 43.2) 40.1 (37.0, 43.3) 40.8 (37.6, 43.2) 0.679 PLT (mg/L) 251.0 (206.0, 327.5) 271.0 (210.1, 345.8) 246.0 (205.8, 304.3) 0.089 RF (IU/L) 69.2 (16.0, 213.0) 115.5 (22.9, 266.5) 47.6 (14.0, 173.5) 0.014 ACPA (U/L) 204.5 (57.5, 432.0) 285.5 (106.5, 432.0) 139.5 (51.8, 365.5) 0.036

The values of continuous variables are presented as mean ± standard deviation or median and quartile (25th, 75th), and the categorical variables are presented as number plus percentage. ACPA: Anti-citrullinated protein antibody; ALB: Albumin; bDMARDs: Biological synthetic disease-modifying antirheumatic drugs; BMI: Bone mass index; CDAI: Clinical disease activity index; CRP: C-reactive protein; csDMARDs: Conventional synthetic disease-modifying antirheumatic drugs; DAS28-CRP: Disease activity score of 28 joints-C reactive protein; ESR: Erythrocyte sedimentation rate; HAQ-DI: Health assessment questionnaire-disability index; HGB: Hemoglobin; NSAIDs: Non-steroidal anti-inflammatory drugs; PLT: Platelet; RA: Rheumatoid arthritis; RDCI: Rheumatic disease comorbidity index; RF: Rheumatoid factor; SDAI: Simple disease activity index; SF-12 MCS: Short form-12 mental component summary; SF-12 PCS: Short form-12 physical component summary; SJC: Swollen joint count; THA: Total hip arthroplasty; TJC: Tender joint count; TKA: Total knee arthroplasty; tsDMARDs: Targeted synthetic disease-modifying antirheumatic drugs; VAS: Visual analog scale.

During the follow-up period of 12 weeks, we reported an incidence of 35.9% of post-operative flares in our cohort, which was lower than that reported by Goodman et al[4,5]. Two types of post-operative flares of RA were hereby defined: local and systemic. Consequently, the composite criteria containing four domains were used. For patients with polyarticular involvement, post-operative flare of RA must fulfill two criteria: rapidly progressive polyarticular symptoms combined with elevated DAS-28 CRP and rapidly elevated ESR and CRP levels. For patients with oligoarticular or monoarticular involvement, their post-operative flare of RA must fulfill two criteria: rapidly progressive swelling, tenderness, and limited range of motion of the operated joint and effective treatment with GCs. Our judgement criteria were different from those proposed by Goodman et al[4,5]. They identified 63% of patients with post-operative flare by 6 weeks exclusively based on PROs, including the rheumatoid arthritis flare questionnaire. All these flares should be regarded as systemic. In our cohort, the proportions of oligoarticular and monoarticular joint involvement (hip or knee) were as high as 18.4% and 6.8%, respectively. Consequently, we identified 16.7% to be local flares. We concentrated on the short post-operative follow-up period (12 weeks) when the inflammatory response to the stress of surgery remained high. We found that most of the flares occurred during the first 4 weeks (69%). During this period, we discontinued the use of bDMARDs or tsDMARDs.

The baseline characteristics of the flare and non-flare group were significantly different in terms of family history, global pain (VAS), DAS28-CRP, HAQ-DI, SDAI, SF-12 PCS, SF-12 MCS, ESR, CRP, RF, and ACPA. The multivariable logistic regression analysis revealed that baseline HAQ-DI (OR = 2.061) and SDAI (OR = 1.055) predicted post-operative flares in patients with RA after arthroplasty. These findings were consistent with those reported in the previous studies. In studies on flares in non-surgical RA patients, higher disease activity was consistently associated with flares. In a study of RA patients with reconstruction of THA or TKA, Goodman et al[4] reported the baseline DAS28-ESR (OR = 2.11), log-transformed RA disease activity index (OR = 2.97), and CRP (OR = 4.24) as independent predictors of post-operative flares. Our findings support the concept of perioperative treat-to-target (T2T) and emphasize the tight control of disease activity before THA or TKA in RA patients. The perioperative optimal use of antirheumatic drugs, is essential for controlling disease activity and for reducing the rate of adverse events, including flare, delayed wound healing, and deep infection.

In conclusion, flares were common after TKA or THA in patients with RA and were associated with a higher disease activity. However, the use of preoperative antirheumatic medication was not an independent risk factor for flares. Surgeons should consider the concept of T2T in perioperative management strategies.

Conflict of Interest

None.

Funding

This study was supported by a grant of Beijing Jishuitan Hospital Elite Young Scholar Programme (No. XKGG202110).

References 1. Smolen JS, Landewé RBM, Bijlsma JWJ, Burmester GR, Dougados M, Kerschbaumer A, et al. EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2019 update. Ann Rheum Dis 2020; 79:685–699. doi: 10.1136/annrheumdis-2019-216655. 2. Holers VM, Kuhn KA, Demoruelle MK, Norris JM, Firestein GS, James EA, et al. Mechanism-driven strategies for prevention of rheumatoid arthritis. Rheumatol Autoimmun 2022; 2:109–119. doi:10.1002/rai2.12043. 3. Song X, Wang YH, Li MT, Duan XW, Li HB, Zeng XF. Co-authors of CREDIT. Chinese registry of rheumatoid arthritis: IV. Correlation and consistency of rheumatoid arthritis disease activity indices in China. Chin Med J 2021; 134:1465–1470. doi: 10.1097/CM9.0000000000001517. 4. Goodman SM, Bykerk VP, DiCarlo E, Cummings RW, Donlin LT, Orange DE, et al. Flares in patients with rheumatoid arthritis after total hip and total knee arthroplasty: rates, characteristics, and risk factors. J Rheumatol 2018; 45:604–611. doi: 10.3899/jrheum.170366. 5. Goodman SM, Mirza SZ, DiCarlo EF, Pearce-Fisher D, Zhang M, Mehta B, et al. Rheumatoid arthritis flares After total hip and total knee arthroplasty: outcomes at one year. Arthritis Care Res (Hoboken) 2020; 72:925–932. doi: 10.1002/acr.24091.