A comparison of characteristics of subjects with early knee osteoarthritis based on two different classification criteria

Abstract

Purpose: Osteoarthritis (OA) of the knee is one of the leading causes of pain and disability worldwide. The need to identify subjects at the early stages of the disease process, before any significant structural damage, is of clinical relevance. They could be treated with more targeted management strategies to improve their prognosis, and possibly delay or avoid knee replacement surgery. We have previously reported that an early knee OA population is distinct from a normal control group and established knee OA with regard to a number of parameters. Recently, a new early knee OA classification was proposed to bring this closer to primary care by challenging the necessity of MRI. Our objective was to compare the clinical, functional and structural characteristics of subjects with early knee OA, classified based on these 2 different sets of criteria. Methods: Sixty-nine postmenopausal women with either early or established medial knee OA, as well as 31 asymptomatic controls (all women) were included. The established OA (n = 29) group was classified as defined by the American College of Rheumatology classification criteria: knee pain, age above 50, stiffness less than 30 minutes and crepitus. For the early OA group 2 recent classifications were used: 1) Luyten et al. 2012 early knee OA classification (early OA 2012) and 2) Luyten et al. 2017 early knee OA classification (early OA 2017). The inclusion criteria for the early OA 2012 (n = 40) group were: presence of knee pain, a Kellgren & Lawrence (K&L) grade 0, 1 or 2- (osteophytes only, no joint space narrowing) on radiography and presence of two of four MRI criteria: (1) ≥BLOKS grade 2 for size cartilage loss, (2) ≥BLOKS grade 2 for percentage full-thickness cartilage loss, (3) signs of meniscal degeneration and (4) ≥BLOKS grade 2 for size of bone marrow lesions (BMLs) in any one compartment. Subjects were classified as early OA 2017 (n = 25) if: showed K&L grade 0 or 1, and 2 out of the 4 KOOS subscales scored “positive” (≤85%), and at least 1 clinical sign was present: Joint line tenderness, crepitus. Clinical, functional and structural characteristics of all subjects were assessed and compared between the 4 groups. Results: For the clinical assessment, the 2017 early knee OA group showed significantly lower (isometric and isokinetic) quadriceps and hamstrings strength compared to the controls. The 2012 early OA group, on the other hand, showed only significantly lower isometric hamstrings strength compared to the controls. No significant differences were found between the 4 groups regarding performance-based function as measured with Timed up and Go and Stair Climbing Tests. Looking into the psychosocial profile, pain catastrophizing scale subscores, as well as the total score, were significantly higher in the 2012 early OA group compared to the controls, while this was not the case when comparing the 2017 early OA group with the controls. MRI findings indicate that the 2017 early OA group showed no significant differences compared to the controls regarding the size/number of BMLs, in contrast to the 2012 early OA group. The 2017 early OA group showed more and larger cumulative (not specifically medial compartment) cartilage lesions compared to the controls. Conclusions: Current results identify clinical and functional differences when early knee OA is classified with or without MRI findings. The new early knee OA classification (Luyten et al. 2017), intended for use in primary care, captures patients with more knee pain and symptoms, worse self-reported function, as well as weaker knee musculature compared to the previously introduced early knee OA classification (Luyten et al. 2012). The latter one is partially based on MRI findings and thus potentially more appropriate for second line care. Most importantly, further validation and iterations of these criteria is required to more precisely identify patients at higher risk of knee OA development.