Erratum to: Clin Rheumatol (2012) 31:1347–1357

DOI 10.1007/s10067-012-2022-4

It is necessary to publish an erratum to article DOI 10.1007/s10067-012-2022-4 to clarify erroneous inclusions in the original article. Mentions of the Cartiscope (ArthroVision, Montreal, Canada) software, as well as references 43 and 44 (Lequesne et al. (1987) and Kellgren et al. (1957), respectively), were included in error.

The corrected and accurate text can be found below. The remainder of the article remains correct at the time of publication.

Methods

Outcomes

Efficacy

Cartilage volume was measured by two trained readers (J Vial and JJ Railhac) blinded to treatment and to MRI examination time points except for baseline. The radiological efficacy variables were measured at baseline (visit 1), weeks 24 and 48 for MRI, and at baseline and week 48 for X-ray: (i) by two readers expert in radiology (J Vial and JJ Railhac). The readers were blinded to treatment and to MRI examination time points except for baseline. Cartilage volume: volumes of tibial, femoral, and patellar cartilage were calculated using a software program (Mac Kesson, San Francisco, CA, USA) after three-dimensional (3D) section and reconstruction of the MRI sections (SIGNA HORIZON LX 1.5 Teslas MRI scanner from General Electrical Medical System, Milwaukee, WI, USA); (ii) description of morphological osteoarticular lesions from the MRI sections (cartilage aspect, ligament, subchondral edema, meniscal lesions, number of cysts, number of osteophytes, presence of foreign bodies, presence of popliteal cysts, subchondral bone, and synovial effusion) and measurement of joint space width (JSW) on X-ray; (iii) intra-reader variability, as assessed by the correlation coefficient of cartilage volume measurements performed on the same views by the same reader 2 weeks apart (MRI of visit 1); (iv) inter-reader variability, as assessed by the correlation coefficient of cartilage volume measurements performed by two different readers (MRI of visit 1).