Background Hand OA is a frequent polyarticular disease. Few is known with respect to its radiological progression over time, which in addition is difficult to assess, considering that no radiographic scoring method has, today, proved being superior to another.
Objectives To assess hand OA radiological progression over 3 years using three validated scoring methods.
Methods Data came from an international 3-year, randomized, placebo-controlled, phase III trial designed to assess the effect of strontium ranelate compared to placebo on the radiographic progression of knee OA which included symptomatic primary knee OA patients (ACR criteria) at a Kellgren-Lawrence (KL) grade II or III, with a minimal joint space width (JSW) between 2.5-5 mm. Baseline and final postero-anterior radiographs of each hand were performed. Hand symptoms were assessed using the functional index for Hand OA (FIHOA; range 0-30) and the AUSCAN-function (0-900 normalized at 100). Two independent readers scored half of the pairs of radiographs obtained each, blinded to treatment and time sequence, using the KL (range 0-128), Kallman (0-204) and Verbruggen anatomical phase (0-218) scoring methods with a good inter-rater reproducibility. Hand OA radiographic progression was studied in the placebo group by looking at 1/ baseline-end changes in global scores, 2/ the numbers of progressors (progression defined for each global score by a change over each reader’s smallest detectable difference (SDD)), 3/ the number of patients with at least 1 new joint affected by OA and 4/ those with at least 1 joint showing a deterioration (worsened joints - improved joints ≥ 1 of at least one grade for KL; of ≥1 phase for Verbruggen score).
Results Of 1669 patients included in the SEKOIA trial, 1360 had hand radiographs. 999 had radiologic hand OA at baseline: 73%. 297 patients out of 472 in the placebo group had baseline and post-baseline radiographs. 71% were female, mean age 64±7 years, body mass index 29.6±5 kg/m², initial knee JSW 3.5±0.8 mm. Baseline hand OA radiologic severity was mild: KL score 21.3±13, Kallman score 24.6±22 and Verbruggen score 13.7±15. FIHOA score was 4±5, Auscan function score was 31.8±28. Mean time interval between baseline and final radiographs was 31.5 months.
Hand OA radiographic progression over 2.6 years, using global scores, was modest with a mean change of 2.2±3.3 for KL score, 3.7±5.3 for Kallman score and 2.0±4.0 for Verbruggen score.
The numbers (%) of progressors (change≥SDD) were 7 (2%), 17 (6%), and 21 (7%) respectively.
The numbers (%) of patients with at least 1 new OA joint were 205 (66%) for KL and 127 (41%) for Verbruggen score.
The numbers (%) of patients with ≥ 1 worsened joint were 160 (52%) for KL and 158 (51%) for Verbruggen score, with respective means of 3.1±2.5 and 2.1±1.5 deteriorated joints.
Conclusions Whatever the radiological scoring method used, and the kind of analysis performed, mild radiographic hand OA patients showed a weak radiological progression over 2.6 years when using global scores. In future structure-modification trials in hand OA, analysing at a joint level the number of patients with at least one new OA joint or one joint worsening could be the most sensitive methods.
Disclosure of Interest E. Maheu Grant/research support from: Servier, C. Cadet Grant/research support from: Servier, F. Carrat: None Declared, Y. Barthe: None Declared, F. Berenbaum Grant/research support from: Servier