L'activité physique régulière préserve-t-elle l'articulation

[Nutrimuscle-Conseils]

Association of daily walking with the risk of total knee replacement over 5 years: an observational study
H. Master Osteoarthritis and Cartilage VOLUME 26, SUPPLEMENT 1, S237-S238, APRIL 01, 2018

Purpose: The presence of functional limitation, such as difficulty getting up from a chair or out of bed, is a reason adults elect for a total knee replacement (TKR), the definitive treatment for end-stage knee osteoarthritis (OA). Regular participation in physical activity, such as walking ≥6000 steps/day, reduces the risk of functional limitation in people with knee OA. However, it is not known whether walking more and in particular walking ≥6000 steps/day are associated with less risk of TKR. Moreover, given that functional limitation is more common in older (≥65 years) versus younger (<65 years) adults, it is not known whether the association of daily walking with TKR may be different for older versus younger adults. Therefore, the purpose of our study was to determine the association of daily walking and risk of TKR over 5 years. We also separately investigated this association in older and younger adults with or at high risk of knee OA to identify to whom daily walking may be most important for reducing the risk of TKR.

Methods: Using data from the Osteoarthritis Initiative (OAI), we included participants who did not have TKR at or before the 48-month follow-up visit, which we considered our study baseline. Physical activity was measured with an accelerometer (Actigraph GT1M) worn during waking hours over the right hip during the baseline and quantified as steps/day. Time to TKR was quantified in months from the baseline visit date to the date of TKR through the 108-month OAI follow-up visit, i.e., 5 years later. The date for TKR was confirmed through adjudicated medical records whenever it was available during the course of OAI study. Participants without TKR at the 5-year follow-up were censored. Death and loss to follow-up in those without a TKR were considered competing events to TKR when they occurred between baseline and the 5-year follow-up. Hazard ratios (HR) with 95% confidence intervals (CI) were calculated from the Fine and Gray sub-distribution hazard model to assess the strength of the association between daily walking and risk of TKR in presence of competing events while adjusting for potential confounders (baseline age, BMI, sex, race, education, comorbidities, frequent knee pain, previous knee injuries or surgeries, and radiographic OA status). Daily walking was quantified as a continuous (i.e., per 1000 steps/day) and a categorical (< or ≥6000 steps/day) exposure measure in separate models. To examine if the association for daily walking and risk of TKR may be different by age groups, we performed subgroup analyses stratified by older (≥65 years) and younger (<65 years) adults.

Results: Of the 1816 participants without TKR at baseline and who wore the monitor for >4 days (age [mean ± sd] 65.0 ± 9.1 years, BMI [mean ± sd] 28.4 ± 4.8 kg/m2, 55% female), 108 (6%) participants had TKR over 5 years. There was no significant association between daily walking (i.e., either continuous or dichotomous measures) and risk of TKR over 5 years in the overall sample or in younger adults (table). However, older adults were 12% less likely to have TKR for each additional 1000 steps/day walked (adjusted HR 0.88, 95% CI [0.79–0.99]). Moreover, older adults who walked > 6000 steps/day were 52% less likely to have TKR (adjusted HR 0.48, 95% CI [0.26–0.89]) compared to people who walked <6000 steps/day (table).

Conclusions: Taking more steps/day and particularly walking >6000 steps/day may moderately reduce the risk for TKR in older (>65 years), but not younger (<65 years) adults. Our findings highlight the potential benefit of walking for older adults with or at risk of knee OA. A moderate reduction in risk of TKR may be an added benefit of walking more for older adults with or at risk of knee OA.

[Nutrimuscle-Conseils]

[b]The effects of exercise on synovitis and bone marrow lesions in knee osteoarthritis: secondary outcomes from a randomized controlled trial[/b]
E. Bandak Osteoarthritis and Cartilage VOLUME 26, SUPPLEMENT 1, S315-S316, APRIL 01, 2018

Purpose: Synovitis and bone marrow lesions (BMLs) are markers of inflammation in knee osteoarthritis (OA) and associated with knee OA pain and a higher risk of radiographic disease progression. Therefore, synovitis and BMLs are of particular interest as a target for treatments of knee OA. Exercise has beneficial effects on knee OA pain and function, and is recommended as a first line treatment. However, the pain-relieving mechanisms of exercise are elusive and it is unknown if exercise affects synovitis and BML to an extent that can explain the exercise-related pain reduction in knee OA. In a randomised trial, we compared the effects of a 12-week therapeutic exercise program on synovitis and BMLs assessed by dynamic contrast enhanced MRI (DCE-MRI) in knee OA with a no-attention control intervention. Further, we explored if changes in synovitis and BMLs were associated with changes in symptoms.

Methods: A secondary outcome analyses of a randomized controlled trial with per-protocol analyses (ClinicalTrials.gov: NCT01545258). Participants with knee OA were randomized (1:1) to either 12 weeks of supervised exercise therapy (ET) 3 times weekly, or a no attention control group (CG). Only participants who adhered to the protocol AND who had valid MRI data were analyzed. From DCE-MRI of the knee, synovitis was quantified as perfusion of enhancing synovium in the following regions of the knee: Anterior synovium, Posterior synovium, and synovium related to Hoffa's fat pad (Fig. 1A). Furthermore BMLs were outlined if present (Fig. 1A). For each of the regions the DCE-MRI variable ‘Initial Rate of Enhancement Composite Score’ (IRExNvoxel) was calculated as the average Initial Rate of contrast Enhancement (IRE) multiplied by the number of Highly Perfused Voxels defined as the number of voxels with a high degree of perfusion, interpreted as a marker of synovitis (Fig. 1B-C). Static non-CE-MRI of the knee was used to assess BMLs and effusion-synovitis as recommended in the MRI in Osteoarthritis Knee Score (MOAKS). Furthermore, static CE-MRI of the knee was used to assess the whole-knee synovitis score, ‘CE-Synovitis’ as proposed by Guermazi and colleagues. Pain was assessed using the Knee injury and Osteoarthritis Outcome Score (KOOS). Analysis of covariance was used to determine group differences in changes from baseline to follow-up adjusting for baseline values and baseline differences in demographic variables. Spearman's rho correlations were used to explore associations between changes in perfusion variables and KOOS pain.

Results: Of 60 participants randomized, 33 adhered to the protocol and had complete DCE-MRI data (ET, n = 16, CG, n = 17). At follow-up there was a statistically significant group difference in IRExNvoxel in the region of Anterior synovium in favour of the CG, indicating a relative increase in synovitis associated with exercise. Although not statistically significant, the results suggest similar trends in synovitis in the other regions and BML (both DCE-MRI and static MRI variables)(Table 1). There was a statistically significant and clinically relevant group difference in changes from baseline in KOOS pain in favor of ET (-11.7 KOOS points (95%CI: -20.1 to -3.4)). No statistically significant correlations were found between changes in synovitis or BMLs and changes in KOOS pain (P>0.05).

Conclusions: This study demonstrated that a 12 week exercise program associates with a clinical benefit on pain and a possible detrimental effect on synovitis in the anterior part of the knee when compared to a no attention control group in knee OA patients. The pain-reducing effect of exercise was not associated with the changes in synovitis or BMLs. These results suggest potential undesirable pro-inflammatory effects of exercise that needs to be examined further in larger studies.

[Nutrimuscle-Conseils]

The effects of exercise on synovitis and bone marrow lesions in knee osteoarthritis: secondary outcomes from a randomized controlled trial
E. Bandak Osteoarthritis and Cartilage VOLUME 26, SUPPLEMENT 1, S315-S316, APRIL 01, 2018

Purpose: Synovitis and bone marrow lesions (BMLs) are markers of inflammation in knee osteoarthritis (OA) and associated with knee OA pain and a higher risk of radiographic disease progression. Therefore, synovitis and BMLs are of particular interest as a target for treatments of knee OA. Exercise has beneficial effects on knee OA pain and function, and is recommended as a first line treatment. However, the pain-relieving mechanisms of exercise are elusive and it is unknown if exercise affects synovitis and BML to an extent that can explain the exercise-related pain reduction in knee OA. In a randomised trial, we compared the effects of a 12-week therapeutic exercise program on synovitis and BMLs assessed by dynamic contrast enhanced MRI (DCE-MRI) in knee OA with a no-attention control intervention. Further, we explored if changes in synovitis and BMLs were associated with changes in symptoms.

Methods: A secondary outcome analyses of a randomized controlled trial with per-protocol analyses (ClinicalTrials.gov: NCT01545258). Participants with knee OA were randomized (1:1) to either 12 weeks of supervised exercise therapy (ET) 3 times weekly, or a no attention control group (CG). Only participants who adhered to the protocol AND who had valid MRI data were analyzed. From DCE-MRI of the knee, synovitis was quantified as perfusion of enhancing synovium in the following regions of the knee: Anterior synovium, Posterior synovium, and synovium related to Hoffa's fat pad (Fig. 1A). Furthermore BMLs were outlined if present (Fig. 1A). For each of the regions the DCE-MRI variable ‘Initial Rate of Enhancement Composite Score’ (IRExNvoxel) was calculated as the average Initial Rate of contrast Enhancement (IRE) multiplied by the number of Highly Perfused Voxels defined as the number of voxels with a high degree of perfusion, interpreted as a marker of synovitis (Fig. 1B-C). Static non-CE-MRI of the knee was used to assess BMLs and effusion-synovitis as recommended in the MRI in Osteoarthritis Knee Score (MOAKS). Furthermore, static CE-MRI of the knee was used to assess the whole-knee synovitis score, ‘CE-Synovitis’ as proposed by Guermazi and colleagues. Pain was assessed using the Knee injury and Osteoarthritis Outcome Score (KOOS). Analysis of covariance was used to determine group differences in changes from baseline to follow-up adjusting for baseline values and baseline differences in demographic variables. Spearman's rho correlations were used to explore associations between changes in perfusion variables and KOOS pain.

Results: Of 60 participants randomized, 33 adhered to the protocol and had complete DCE-MRI data (ET, n = 16, CG, n = 17). At follow-up there was a statistically significant group difference in IRExNvoxel in the region of Anterior synovium in favour of the CG, indicating a relative increase in synovitis associated with exercise. Although not statistically significant, the results suggest similar trends in synovitis in the other regions and BML (both DCE-MRI and static MRI variables)(Table 1). There was a statistically significant and clinically relevant group difference in changes from baseline in KOOS pain in favor of ET (-11.7 KOOS points (95%CI: -20.1 to -3.4)). No statistically significant correlations were found between changes in synovitis or BMLs and changes in KOOS pain (P>0.05).

Conclusions: This study demonstrated that a 12 week exercise program associates with a clinical benefit on pain and a possible detrimental effect on synovitis in the anterior part of the knee when compared to a no attention control group in knee OA patients. The pain-reducing effect of exercise was not associated with the changes in synovitis or BMLs. These results suggest potential undesirable pro-inflammatory effects of exercise that needs to be examined further in larger studies.

[Nutrimuscle-Diététique]

Traduction de l'étude

Les effets de l'exercice sur la synovite et les lésions de la moelle osseuse dans l'arthrose du genou: résultats secondaires d'un essai contrôlé randomisé
E. Bandak Osteoarthritis and Cartilage VOLUME 26, SUPPLEMENT 1, S315-S316, 01 AVRIL 2018

Objectif: La synovite et les lésions de la moelle osseuse (BML) sont des marqueurs de l'inflammation dans l'arthrose du genou (OA) et sont associées à une douleur arthrosique du genou et à un risque plus élevé de progression radiographique de la maladie. Par conséquent, la synovite et les BML présentent un intérêt particulier en tant que cible pour les traitements de l'arthrose du genou. L'exercice a des effets bénéfiques sur la douleur et la fonction de l'arthrose du genou, et est recommandé comme traitement de première intention. Cependant, les mécanismes de soulagement de la douleur de l'exercice sont insaisissables et on ne sait pas si l'exercice affecte la synovite et la BML dans une mesure qui peut expliquer la réduction de la douleur liée à l'exercice dans l'arthrose du genou. Dans un essai randomisé, nous avons comparé les effets d'un programme d'exercices thérapeutiques de 12 semaines sur la synovite et les BML évalués par IRM dynamique par contraste amélioré (DCE-IRM) dans l'arthrose du genou avec une intervention de contrôle sans attention. De plus, nous avons exploré si les modifications de la synovite et des BML étaient associées à des modifications des symptômes.

Méthodes: Une analyse des résultats secondaires d'un essai contrôlé randomisé avec des analyses par protocole (ClinicalTrials.gov: NCT01545258). Les participants souffrant d'arthrose du genou ont été randomisés (1: 1) soit à 12 semaines de thérapie par l'exercice supervisé (ET) 3 fois par semaine, soit à un groupe de contrôle sans attention (CG). Seuls les participants ayant adhéré au protocole ET disposant de données IRM valides ont été analysés. À partir de l'ECD-IRM du genou, la synovite a été quantifiée comme une perfusion de la synoviale rehaussée dans les régions suivantes du genou: synovie antérieure, synoviale postérieure et synoviale liée au coussinet adipeux de Hoffa (Fig. 1A). En outre, les BML ont été décrites si elles étaient présentes (figure 1A). Pour chacune des régions, la variable DCE-MRI 'Initial Rate of Enhancement Composite Score' (IRExNvoxel) a été calculée comme le taux initial moyen d'amélioration du contraste (IRE) multiplié par le nombre de voxels hautement perfusés défini comme le nombre de voxels avec un haut degré de perfusion, interprété comme un marqueur de synovite (Fig. 1B-C). L'IRM statique non-CE du genou a été utilisée pour évaluer les BML et la synovite à épanchement comme recommandé dans l'IRM dans le score de l'arthrose du genou (MOAKS). En outre, une CE-IRM statique du genou a été utilisée pour évaluer le score de synovite du genou entier, «CE-Synovite», comme proposé par Guermazi et ses collègues. La douleur a été évaluée à l'aide de la blessure au genou et du score de résultat de l'arthrose (KOOS). L'analyse de la covariance a été utilisée pour déterminer les différences de groupe dans les changements entre le départ et le suivi en ajustant les valeurs de base et les différences de base dans les variables démographiques. Les corrélations rho de Spearman ont été utilisées pour explorer les associations entre les changements dans les variables de perfusion et la douleur KOOS.

Résultats: Sur 60 participants randomisés, 33 ont adhéré au protocole et avaient des données DCE-IRM complètes (ET, n = 16, CG, n = 17). Au suivi, il y avait une différence de groupe statistiquement significative dans IRExNvoxel dans la région de la synovie antérieure en faveur du CG, indiquant une augmentation relative de la synovite associée à l'exercice. Bien que non statistiquement significatifs, les résultats suggèrent des tendances similaires de la synovite dans les autres régions et de la LMB (variables DCE-IRM et IRM statiques) (tableau 1). Il y avait une différence de groupe statistiquement significative et cliniquement pertinente dans les changements par rapport à la valeur initiale de la douleur KOOS en faveur de l'ET (-11,7 points KOOS (IC à 95%: -20,1 à -3,4)). Aucune corrélation statistiquement significative n'a été trouvée entre les modifications de la synovite ou des BML et les modifications de la douleur KOOS (P> 0,05).

Conclusions: Cette étude a démontré qu'un programme d'exercice de 12 semaines est associé à un bénéfice clinique sur la douleur et à un effet néfaste possible sur la synovite dans la partie antérieure du genou par rapport à un groupe témoin sans attention chez les patients arthrosiques du genou. L'effet de réduction de la douleur de l'exercice n'a pas été associé aux modifications de la synovite ou des BML. Ces résultats suggèrent des effets pro-inflammatoires indésirables potentiels de l'exercice qui doivent être examinés plus en détail dans des études plus larges.