Osteoarthritis of the shoulder in under-50 year-olds: A multicenter retrospective study of 273 shoulders by the French Society for Shoulder and Elbow (SOFEC)

International audienceIntroduction: Osteoarthritis (OA) of the shoulder in under-50 year-olds is rare, and treatment is delicate. Shoulder replacement incurs frequent long-term risk of progression and a high revision rate, making it unsuited to young active patients. The aim of the present study was to determine the epidemiology of shoulder OA in under-50 year-olds and to assess the clinical results of the various treatment options.Hypothesis: The main study hypothesis was that well-conducted non-operative treatment can allow shoulder replacement to be postponed. The secondary hypothesis was that anatomic total shoulder arthroplasty (TSA) is the treatment of choice when other options fail.Materials and methods: A multicenter retrospective study included primary (POA) and post-instability osteoarthritis (PIOA) in patients aged≤50years at symptom onset. Exclusion criteria comprised post-traumatic OA, rheumatoid arthritis and necrosis. Two hundred and sixty-six patients for 273 shoulders were included from 13 shoulder surgery centers: 2 types of non-operative treatment (28 by platelet-rich plasma [PRP] and 88 by viscosupplementation), 73 arthroscopies, and 150 implantations (62 humeral hemiarthroplasties [HA], comprising 10 hemi-metal, 24 hemi-pyrocarbon and 28 hemi-resurfacing; 77 anatomic total prostheses, and 11 reverse prostheses). Minimum follow-up was 12 months for non-operative treatment and 24 months for arthroplasty (some patients having both). Endpoints comprised Constant score, Subjective Shoulder Value (SSV) and number of complications/revision procedures.Results: Mean age at treatment was 43 years (range, 23-65 years), with 75% male predominance. Symptom onset was earlier in PIOA than in POA: 36 vs. 39 years (range, 20-50 years). PRP and viscosupplementation postponed implantation by a mean 3.5 years in 86% of cases, as did arthroscopy in 56%. ER1 restriction was the most negative factor. At 74 months' follow-up for HA and 95 months for TSA, mean Constant score was significantly lower for HA (56 vs. 67; p=0.004), with higher rates of complications (31% vs. 11%) and implant exchange (13% vs. 9%).Discussion/conclusion: PRP, viscosupplementation and arthroscopy allow implantation to be postponed until the shoulder becomes stiff and painful. In case of failure, TSA is the most effective solution in the medium-term.Level of evidence: IV a; therapeutic study - investigating the results of treatment