Evaluation of hand bone loss by digital X-ray radiogrammetry as a complement to clinical and radiographic assessment in early rheumatoid arthritis: results from the SWEFOT trial.

To investigate hand bone loss (HBL) measured by digital X-ray radiogrammetry (DXR) in patients with early rheumatoid arthritis (RA) receiving different treatment regimens, and to evaluate if DXR change rates during the first 12 months correlate with radiological damage after 24 months. From the total SWEFOT trial population, 159 patients had hand radiographs correctly timed and taken with same modality to be analyzed with DXR. All patients started treatment with methotrexate. After 3-4 months, patients with DAS28 > 3.2 were randomized to add sulfasalazine and hydroxychloroquine (triple therapy) or infliximab (MTX + INF). Those with DAS28 ≤3.2 were followed in regular care. Radiographic progression over 24 months was scored according to the Sharp van der Heijde score (SHS) and defined as >5 increase in T-SHS over 24 months. Hand bone mineral density (BMD) was measured by DXR at inclusion and 12 months and a change ≥2.5 mg/cm2/month was used as a cut-off for HBL. In the MTX responders, triple therapy, and MTX + INF groups, the proportions with HBL were 4.1%, 22.2% and 16.4%, respectively (p = 0.01), and the mean (SD) radiological progression in these groups was 3.91 (6.72), 7.40 (14.63) and 2.72 (4.55) respectively (p = 0.06). Patients with HBL had significantly greater risk for radiographic progression, compared with patients without HBL (odds ratio 3.09, 95% CI =1.20-7.79, p = 0.02). Non-responders to MTX had a significantly greater risk of HBL than MTX-responders, despite the add-on therapies. Patients with HBL during the 12 months had greater risk of radiographic progression after 24 months. Evaluation of HBL may help to identify patients who are at risk of radiographic progression.