Background: Treatment with tumor necrosis factor antagonists (anti-TNF) has been associated with increased risk of reactivation of tuberculosis (Tb). First national recommendations for the treatment of latent Tb in anti-TNF treatment were published in 2002 by the Society for Research and Education in Rheumatology, in collaboration with some expert specialists in Chest Medicine. These recommendations were updated in the May of 2005.

Objectives: To evaluate the effectiveness of the updated recommendations in improving the risk of reactivation in patients on ant-TNF agents.

Methods: The electronic prescription database of the Pension Fund which provides health insurance to approximately 20% of the Turkish population was searched from Jan 1st of 2000-till June 1st of 2006. A new database was constructed by using all prescription information of all patients who had been prescribed infliximab, etanercept or adalimumab, at least once. Duration of biologic treatment was estimated by using the dates of the first and last prescriptions; but also by considering the number of packs prescribed. Patients whose anti TNF treatments were discontinued following a prescription of anti-Tb drugs compatible with a known active Tb treatment protocol were accepted to have developed Tb. Patients who were prescribed anti-Tb, but continued anti-TNF therapy were considered to have prophylactic treatment for latent Tb. Tb incidence as well as the time of initiation of latent Tb treatment and duration of this treatment, as well were analyzed and compared before and after the updated national consensus recommendations. StatsDirect program was used for statistical analysis.

Results: A total of 1090 adult patients (595 with a diagnosis of rheumatoid arthritis [RA]) who had at least one anti-TNF prescription were detected in the database. 507 patients (629 patient-years) were prescribed anti-TNFagent before the updated recommendations and 583 (233 patient-years) after these recommendations. Development of tuberculosis was detected in 8 patients. In four (3 RA) of these patients, Tb had occurred prior to the initiation of biologics. In the other 4 patients (all RA) Tb infection developed after the commencement of biological therapy (crude incidence: 0.6%; 95% CI 0.2-1.6). All of these four patients had developed Tb, before the release of final national recommendations. None of them had received any treatment for latent Tb. Prophylactic treatment for latent Tb prophylaxis was given to 168 (33%) patients before the last recommendations and to 201 (34%) after these recommendations. Latent Tb treatment was started significantly earlier after the updated recommendation compared to before (mean time for the commencement of latent Tb treatment after and before anti-TNF were 1 and, 194 days respectively; p<0.001). Duration of treatment was slightly longer after last recommendations, but the difference did not reach statistical significance (157±98 and 142±116 days). Isoniaside was the anti-Tb agent of choice for almost all patients.

Conclusion: The risk of Tb reactivation seems to have been decreased after the updated recommendations which have reduced the threshold for treatment of latent tuberculosis. Earlier commencement of latent TB treatment and longer treatment duration may have contributed to the decreased Tb incidence after the update. It is therefore of utmost importance that every effort should be made for the dissemination and implication of recommendations for preventing Tb in patients receiving anti TNF therapy.

---