Obesity and radiological severity are associated with viscosupplementation failure in patients with knee osteoarthritis

Viscosupplementation (VS) is still controversial. One of the key points is the lack of well‐identified factors of response. We aimed to identify clinical and radiological factors associated with lack of relevant response after intra‐articular (IA) hyaluronic acid (HA) injections in symptomatic knee osteoarthritis patients. A post hoc analysis of the HAV‐2012 trial, a controlled, multicentre, double‐blind, randomized, non‐inferiority trial comparing 3 weekly IA injections of HA (HAnox‐M or BioHA) for symptomatic tibiofemoral OA was performed. At inclusion, demographic, anthropometric, clinical data (WOMAC score, patient global assessment, presence of knee effusion), and radiological data (OARSI grade, patello‐femoral involvement) were recorded. VS response was defined according to OMERACT–OARSI response criteria at month 6. Predictors of response were investigated in univariate then in multivariate analysis. One hundred and sixty‐six patients with full available data were included. As baseline characteristics and treatment effectiveness were similar between the 2 HA groups, their data were pooled. The mean age was 65.2 [63.7–66.8] years; 101 (60.8%) were women; 73 (44.0%) had severe TF space narrowing. At 6 months, 113 patients (68.1%) were responders. Multivariate analysis showed that obesity (BMI ≥30 kg/m²⁾ and radiological severity (OARSI grade 3) were significantly associated with VS failure (p = 0.001 and p = 0.008, respectively). Moreover, the association of obesity and severe TF space narrowing significantly increased the risk of VS failure. Baseline pain intensity and functional impairment were not associated with VS response. Consequently, IA injection of HA for knee OA should mainly be considered in subjects with low BMI and mild TF space narrowing.