RA flares frequent and severe after arthroplasty, and predicted by higher baseline disease activity

Flares may be frequent in patients with rheumatoid arthritis (RA) who undergo arthroplasty, with significant increases in risk for those with higher baseline disease activity, according to a recent study published in The Journal of Rheumatology. 

“This is the first study, to our knowledge, to prospectively assess postoperative patient-reported flares of RA in patients undergoing total hip arthroplasty (THA) and total knee arthroplasty (TKA) using the Rheumatoid Arthritis Flare Questionnaire (RA-FQ), a validated patient-reported outcome instrument for identifying RA flares, in addition to other patient-reported measures,” wrote the authors, led by Sarah M. Goodman, MD, Department of Rheumatology, Hospital for Special Surgery, New York, NY.

“Our study demonstrates that RA flares are frequent and severe after arthroplasty. While patients with higher disease activity report more flares, preoperative antirheumatic medication withdrawal is not an independent risk factor for flares,” they added.

Dr. Goodman and colleagues recruited 120 patients (mean age: 62 years; median RA duration: 14.8 years; 83% female; 81% Caucasian) with RA before they underwent elective TKA (n=67; 56%) and THA (n=53; 44%).

At 0-2 weeks pre-operatively and 6 weeks post-operatively, researchers evaluated clinical RA characteristics in these patients. For 6 weeks, the patients responded to questions about disease activity, including self-reported joint counts and flare status.

The majority of patients (82%) met criteria from the 2010/1987 American College of Rheumatology/European League Against Rheumatism. About half (51%) were taking biologics. All treatment with biologics was discontinued pre-operatively, but glucocorticoids (GCs) and methotrexate (MTX) were continued.

By week six, 63% of patients had flared, with a median time to flare of 2 weeks, a median flare severity of 7 on a scale of 1-10, and a median duration of 4-7 days. Most flares (52%) occurred by week 2, and were rated as severe by 36% of flarers.

Flares were managed by decreasing activity (23%) or activity avoidance (19%), while 17% of patients increased analgesic or anti-inflammatory medications, and 5% increased GCs. Only 4% of flarers sought help from their rheumatologist.

At baseline, patients experiencing flares had significantly higher disease activity, erythrocyte sedimentation rates (ESR), C-reactive protein levels (CRP), and pain levels. Also, significantly higher in flarers than in nonflarers were the number of tender joints (median 2.0 vs 1.0; P=0.013), RADAI Joint Score (median 10 vs 5.0; P < 0.001), and the number of swollen joints (median 4.0 vs 3.0; P=0.04).

More flarers met both 1987 and 2010 RA criteria compared with those who did not flare (49% vs 24%, respectively), and flarers had significantly worse baseline MD Health Assessment Questionnaire function (mean 4.1 vs 3.4; P=0.009). This was also true at 6 weeks (median 3.7 vs 2.7; P=0.002). The change from baseline to 6 weeks, however, was similar between flarers and nonflarers (mean -0.5 vs -0.5; P=0.97).

Higher baseline disease activity predicted flares by 6 weeks (OR: 2.12; P=0.02). Those with 28-joint Disease Activity Score (DAS28)-ESR values exceeding 5.1 were more than 25 times more likely to report a flare over 6 weeks compared with those with low disease activity or remission (DAS28-ESR ≤ 2.6; OR: 25.59 P=0.003). Elevated Routine Assessment of Patient Index Data (RAPID)-3 was also associated with increased risk of flare (OR: 1.18; P < 0.001).

In patients with CRP levels ≤ 1.5 mg/dL and ≤ 2.0 mg/dL, researchers found a decreased risk of flare (OR: 0.29; P=0.006; and OR: 0.17; P=0.002, respectively). Those with normal ESR levels also had a lower risk of flare (OR: 0.30; P=0.03).

More patients who flared were being treated with biologics compared with those who did not flare (57% vs 42%; P=0.014).

Upon multivariate analysis, researchers found that baseline DAS28 and log-transformed Rheumatoid Arthritis Disease Activity Index (RADAI) joint scores were independent predictors of physician-validated patient reports of flares.

Increased CRP levels increased risk of flares, but disease duration, medication use, and discontinuation of biologic treatment did not. Finally, neither GC (flarers 88% vs nonflarers 86%) nor MTX use (flarers 51% vs nonflarers 53%) at the time of surgery protected against the risk of flares.

"Given the high rate of disease flares after THA and TKA, the notion that patients with established RA undergoing THA and TKA have ‘burnt out’ disease lends further credence to the spreading idea that RA is an ongoing disease over a lifetime, at best controlled by medication,” these researchers concluded.

This study was supported by a grant from the US National Institutes of Health Accelerating Medicine Partnership program, the Weill Cornell Clinical Translational Science Center, and the Block Family Foundation.