Osteoporosis (OP) occurs more frequently in patients with rheumatoid arthritis (RA) than in healthy individuals. Specific treatments of RA may increase susceptibility to OP, but at the same time decrease inflammatory activity, which is associated with accelerated bone loss. Treatment with TNF-alpha blockers might influence bone metabolism and prevent structural bone damage in RA, in particular at the periarticular level. Our aim was to assess the influence of anti-TNF-alpha therapy on bone metabolism in RA patients. To that end we evaluated a group of 30 RA patients [mean age 50.6 +/- 6.8 years; median disease duration 82 +/- 38 months; median disease activity score (DAS-28) 5.8 +/- 1.2: 70% of whom were positive for the rheumatoid factor IgM (>40 IU/mL)]. Patients were treated with stable therapy of prednisone (7.5 mg/day) and methotrexate (MTX = 10 mg/week). Eleven of these RA patients further received etanercept (25 mg, twice/weekly) and 10 infliximab (3 mg/kg on 0, 2, 6, and every 8 weeks thereafter). A control group included 10 RA patients with stable therapy (prednisone and MTX) and without anti-TNF-alpha therapy. All the patients fulfilled the ACR criteria for the diagnosis of adult RA and were treated for 6 months. Quantitative ultrasound (QUS) bone densitometry was performed at the metaphyses of the proximal phalanges of both hands with a DBM Sonic 1200 QUS device (IGEA, Carpi, Italy). Amplitude-dependent speed of sound (AD-SoS) was evaluated at base line and at 3 and 6 months. Bone mineral density (BMD) of the hip and lumbar spine (L1-L4) was determined by a densitometer (GE Lunar Prodigy, USA) at base line at after 6 months. Soluble bone turnover markers [osteocalcin (BGP) and deoxypyridinoline/creatinine (Dpd/Cr) ratio] were measured in all patients at the same times, using enzyme-linked immunosorbent assay tests. All data were compared using Wilcoxon signed rank test. Results were as follows: AD-SoS values were found increased by 1.3% after 6 months of treatment in the RA patients treated with anti-TNF-alpha therapy. On the contrary, the Ad-SoS levels decreased by 4.6% during the same period in the untreated RA group. BMD increased by 0.2% at lumbar spine and 0.1% at the hip in TNF-alpha-blocker-treated patients and decreased by 0.8% and 0.6% (at lumbar spine and at the hip, respectively) in RA patients without anti-TNF-alpha therapy. However, BMD variations were not significant. In RA patients treated with TNF-alpha blockers, BGP levels were found significantly increased (14.8 +/- 3.8 mg/mL vs. 22.4 +/- 4.2 mg/mL; P < 0.01) and Dpd/Cr levels were found significantly decreased (8.2 +/- 2.1 nM vs. 4.6 +/- 1.8 nM; P < 0.01) at 6 months when compared to base line values. On the contrary, there were no significant differences in the untreated RA patients concerning these latter parameters (BGP = 12.2 +/- 3.1 mg/mL vs. 10.8 +/- 2.8 mg/mL and Dpd/Cr = 8.9 +/- 2.4 nM vs. 10.2 +/- 1.8 nM, respectively). In conclusion, during 6 months of treatment of RA patients with TNF blockers, bone formation seems increased while bone resorption seems decreased. The reduced rate of OP appears to be supported by the same mechanisms involved in the decreased bone joint resorption during anti-TNF-alpha therapy, that is, the marked decrease of the proinflammatory (i.e., TNF-alpha) cytokine effects on bone metabolism.