Christian-Hubert Roux

Effect of discontinuing TNFα antagonist therapy in patients with remission of rheumatoid arthritis

OBJECTIVE The objective of this study was to determine the time to relapse after tumor necrosis factor alpha (TNFalpha) antagonist discontinuation in patients with remission of rheumatoid arthritis (RA). METHODS Among 304 patients taking TNFalpha antagonist therapy for RA, 21 achieved a remission and were taken off the TNFalpha antagonist. Remission was defined as DAS28<2.6 for at least 6 months without nonsteroidal inflammatory drugs or more than 5 mg of prednisone per day but with disease-modifying antirheumatic drug (DMARD) therapy if needed. The same TNFalpha antagonist was restarted in the event of a relapse (DAS28>3.2). RESULTS The 21 patients had a mean age of 61 years, a mean disease duration of 11.3 years, and a mean remission duration at TNFalpha antagonist discontinuation of 19.2 months. The TNFalpha antagonist was infliximab in 2 patients, adalimumab in 5, and etanercept in 14; and 14 patients were taking a concomitant DMARD. The number of patients still in remission after TNFalpha antagonist discontinuation was 9/20 after 6 months and 5/20 after 12 months. Mean time to relapse was 14.7 weeks. While off TNFalpha antagonist therapy, 3 of the 5 relapse-free patients after 12 months were on DMARD therapy, compared to 11 of the 15 patients who relapsed. Compared to the 15 patients who relapsed, the 5 relapse-free patients had a longer time on TNFalpha antagonist therapy (56 months vs. 35 months, P=0.012) and a longer time in remission on TNFalpha antagonist therapy (35 months vs.14.5 months, P=0.04). The 15 patients who relapsed consistently achieved a remission after resuming TNFalpha antagonist therapy; the remission occurred within 2 months in 13 patients. CONCLUSION TNFalpha antagonist discontinuation in patients in remission of RA was followed by a relapse within 12 months in 75% of cases. Relapsing patients responded well to resumption of the same TNFalpha antagonist.

Prevalence of symptomatic hip and knee osteoarthritis: a two-phase population-based survey1

OBJECTIVE Osteoarthritis (OA) epidemiologic data are scarce in Europe. To estimate the prevalence of symptomatic knee and hip OA in a multiregional sample in France. DESIGN A two-phase population-based survey was conducted in six regions in 2007-2009. On initial phone contact using random-digit dialing, subjects 40-75 years old were screened with a validated questionnaire. Subjects screened positive were invited for ascertainment: physical examination and hip and/or knee radiography (Kellgren-Lawrence grade≥2). Multiple imputation for data missing not-at-random was used to account for refusals. RESULTS Of 63,232 homes contacted, 27,632 were eligible, 9621 subjects screened positive, 3707 participated fully in the ascertainment phase, and 1010 had symptomatic OA: 317 hip, 756 knee. Hip OA prevalence according to age class ranged from 0.9% to 3.9% for men and 0.7-5.1% for women. Knee OA ranged from 2.1% to 10.1% for men and 1.6-14.9% for women. Both differed by geographical region. The hip and knee standardized prevalence was 1.9% and 4.7% for men and 2.5% and 6.6% for women, respectively. CONCLUSIONS This confirmed the feasibility of using a screening questionnaire for eliciting population-based estimates of OA. In France, it increases with age and is greater among women above the age of 50. The geographical disparity of hip and knee OA parallels the distribution of obesity. Study registration ID number 906297 at http://www.clinicaltrials.gov/.

Femoro-tibial knee osteoarthritis: One or two X-rays? Results from a population-based study

OBJECTIVE Our objective was to compare the use of both anteroposterior (AP) extended-knee X-ray and semi-flexed X-ray (current gold standard) versus the use of semi-flexed X-ray alone to detect femoro-tibial osteoarthritis (OA). METHODS Individuals 40 to 75 years of age with symptomatic hip and/or knee OA (Kellgren/Lawrence [KL] score≥2) were recruited using a multiregional prevalence survey in France. Both AP and schuss X-rays were performed and read; two years later, the same examiner, blinded to the results of the first reading, performed a second reading of the schuss X-ray. We compared the KL stages of each knee and analyzed osteophyte detection and localization, joint space narrowing (JSN), and the relationship to obesity. RESULTS The analysis included 350 participants with OA of various stages. Comparing the two readings showed that a higher proportion of patients had KL≥2 when the two X-ray views were combined (right knee: P<0.0001; left knee: P<0.001). There were no differences when using the schuss X-ray alone versus in combination with an AP X-ray in terms of detecting JSN, osteophytes. A comparison of schuss X-ray alone versus AP X-ray alone demonstrated the superiority of the schuss view for evaluating JSN (P=0.0001 and P=0.0001) and no difference in osteophyte detection. CONCLUSION Our study shows that the schuss view alone was sufficient for detecting knee osteophytes and JSN. Using one X-ray rather than two will reduce medical costs and irradiation burden. Using two views seems preferable for epidemiological studies.

296 THE “NOTCHED PATELLA” IN THE FRENCH KHOALA COHORT

Methods: Using random digit dialing survey on phone calls, persons > 40 yrs, with pain in the knee or the hip area were selected. Those who agreed were examined in one of the six French investigating centres (Amiens, Brest, Nancy, Nice, Paris, Toulouse), and their knees (AP extension view, Lyon schuss and sky view) and/or their hips (AP pelvis, Lequesne oblique view) were x-rayed. The structural changes of knee and hip OA (Kellgren-Lawrence [KL] ≥ 2) were recorded by centralised reading. Patients fulfilling the inclusion criteria where then proposed to enter the KHOALA cohort (Knee and Hip OsteoArthritis Long-term Assessment). Results: During the first year of recruitment (April 2007 March 2008), 1506 subjects (females: 71%; mean age: 58 yrs) had x-rays: 312 hips, 674 knees, 520 both leading to 832 hip and 1194 knee radiographs (table). For both hips and knees, OA was as often unilateral as bilateral (50/50). In 10% of the cases, a hip/knee prosthesis on one side made these patients classified as “bilateral”. In the presence of bilateral involvement, narrowing of the joint space was at the same location in both sides (96%). In hip OA the narrowing was supero-lateral (64%), supero-medial (22%), inferior/posterior (7%) and global (5%). In knee OA the narrowing was tibio-femoral medial (78%), lateral (11%) or both (7%). Patello-femoral OA was associated in 26% of the cases.