to define the exposures. ROA was defined as radiographic tibiofemoral OA (Kellgren & Lawrence (KL) grade !2) on fixed flexion PA radiographs. SxOA was defined as having both ROA and consistent frequent knee pain in the same knee. We also considered knee replacement as equivalent to ROA and to SxOA. The definition of consistent frequent knee pain was met if a participant reported knee pain on most days during the past month at telephone interview and at the clinic visit that occurred on average one month later. Knee-specific pain severity was measured using the WOMAC pain subscale; the higher of the two ratings was used. WSP was defined as pain above and below the waist, on both sides of the body and axially, using a standard homunculus, excluding knee pain. Incident WSP was defined as presence of WSP at 84 months among those who were free of WSP at 60 months. We assessed the relation of baseline ROA, SxOA, consistent frequent knee pain, and knee pain severity, respectively, to incident WSP using logistic regression, adjusting for baseline age, sex, BMI, comorbidities, physical activity, study site, depressive symptoms, pain catastrophizing, sleep and fatigue. Knee pain severity was also adjusted for in the ROA, SxOA and consistent frequent knee pain models. KL grade was also adjusted for in the consistent frequent knee pain and knee pain severity models. Results: At baseline, 1385 subjects were eligible for analysis (age mean, SD; 67.4, 7.8; BMI 30.5, 6.0 kg/m2, 55% women). Baseline presence of ROA and SxOA in either knee was not associated with incident WSP, respectively (Table). Similarly, baseline unilateral and bilateral ROA or SxOA were not associated with risk of incident WSP. In contrast, baseline presence of consistent frequent knee pain in either knee was significantly associated with a 50% higher incidence of WSP compared with those without consistent frequent knee pain (adjusted OR 1.49 95% CI 1.00, 2.21, p ¼ 0.05). The baseline presence of unilateral or bilateral consistent frequent knee pain was associated with 40% and 65% higher incidence of WSP, respectively, compared with those who did not have consistent frequent knee pain in either knee, albeit not significantly (Table). Knee pain severity was also significantly associated with increased risk of incident WSP, such that each point higher on the WOMAC pain subscale score was associated with a 13% increased risk of incident WSP (adjusted OR 1.13 95% CI 1.08, 1.19 p<0.001). Conclusions: Consistent frequent knee pain and severity of knee pain were significantly associated with incident WSP indicating that regional pain is involved in the development of WSP. In contrast, structural pathology independent of pain severity was not significantly associated with increased risk for incident WSP. These findings call into question the role of peripheral structural pathologic input in the development of WSP, and highlight the importance of pain symptomatology as an important factor that can contribute to the onset of WSP, suggesting a possible role for sensiti...

No abstract

BackgroundThe aim of this study was to investigate the effects of chondroitin sulfate (CS) on the serum levels of Coll2-1 in patients with knee OA.MethodsSeventy two patients with unilateral symptomatic knee OA were involved in a post-authorization open-label study evaluating CS (800 mg/day). The primary outcome was the % relative change in serum Coll2-1 (sColl2-1). The secondary outcomes were the evaluation of pain (VAS) and function (Lequesne’s Index). Responders and non-responders were classified according to OMERACT-OARSI recommendations. Finally, an original cut-off method was applied to categorize patients and interpret individual variations in serum levels of Coll2-1.ResultsPatients showed no difference in the sColl2-1 levels at baseline. When considering responders and non-responders from the ITT population, a significant difference was found for Coll2-1 at 3 months (p = 0.030) and 6 months (p = 0.038). A decrease in pain (VAS) and an improvement in function (LI) were recorded throughout the visits (p < 0.01). Considering an intra-batch cut-off of 21 %, CS decreased Coll2-1 serum levels between baseline and 1-month visit compared to the value of Coll2-1 before treatment (screening visit) which can be interpreted as a drastic reduction of the proportion of patients with an increase of Coll2-1 over 21 % (reduction from 13 to 3 %). It also consisted in a more important proportion of patients with a decrease in Coll2-1 (from 5 to 10 %).ConclusionThis study proposes a new approach for the analysis and the interpretation of the individual variation in biomarker levels and introduces the notion of metabolic responders.Trial registrationID ISRCTN63795830. The trial was retrospectively registered on 2 October, 2015.Electronic supplementary materialThe online version of this article (doi:10.1186/s12891-016-1268-4) contains supplementary material, which is available to authorized users.

No abstract