The aim of the present work is to compare drug survival and safety of infliximab, etanercept, and adalimumab (tumor necrosis factor [TNF] antagonists) in spondylarthritis (SpA) with those of rheumatoid arthritis (RA). To this purpose, we analysed the data in BIOBADASER (2000-2005), a drug registry launched in 2000 for long-term follow-up of the safety of these biologics in rheumatic diseases. The rates of drug discontinuation and adverse events (AEs) in SpA (n = 1,524) were estimated and compared with those of RA (n = 4,006). Cox regression analyses were used to adjust for independent factors. Total exposure to TNF antagonists for SpA was 2,430 patient-years and 7,865 for RA. Drug survival in SpA was significantly greater than in RA at 1, 2, and 3 years. The hazard ratio (HR) for discontinuation in SpA compared with RA was 0.66 (95% confidence interval [CI], 0.57-0.76) after adjustment for age, gender, and use of infliximab. The difference remained after controlling for the individual medication and its place in the sequence of treatment. There were fewer SpA patients with AEs (17%) than RA patients (26%; p < 0.001). The HR for AEs in SpA was 0.80 (95% CI, 0.70-0.91) compared with RA after adjustment for age, disease duration, and use of infliximab. In conclusion, due in part to a better safety profile, survival of TNF antagonists in SpA is better than in RA. TNF antagonists are at present a safe and effective therapeutic option for long-term treatment of patients with SpA failing to respond to traditional drugs. Because chronic therapy is necessary, continual review of this issue is necessary.