Detecting latent tuberculosis infection during anti-tumour necrosis factor therapy

Background There is little information regarding the reliability of repeat tuberculin skin tests (TSTs) and interferon gamma release assays (IGRAs) in detecting latent tuberculosis infection (LTBI) in people on anti-tumour necrosis factor (TNF) medication. Methods We conducted a prospective, observational study of patients referred to the Saskatoon Tuberculosis (TB) Clinic prior to starting anti-TNF medication. A chest x-ray (CXR), 2-step TST and IGRA (QuantiFERON-TB Gold In-Tube Method) were performed at baseline. Those patients with a baseline TST >= 10 mm and/or a positive IGRA were followed with a clinic visit, CXR,TST and IGRA at 3 and 6 months after starting anti -TNF medication. Results Of 106 potential patients, 91 consented to participate. Twenty-six patients had a positive (>= 10 mm)TST or IGRA at baseline; twelve started and stayed on anti-TNF medication through the 6-month.follow-up and completed both planned follow-up visits. The baseline mean TST measurement for the 12 participants was 13.9 mm (SD 11.4), increasing to a mean of 16.8 mm (SD 9.3) post-booster. At 3 months post-anti-TNF initiation, there was an overall decrease in TST measurement (mean=10.0 mm; SD 9.3; p=0.013), with measurements < 5 mm in 3 of the 12 patients. By the 6-month TST, a response recovery was observed with a mean TST measurement of 14.5 mm (SD 7.7), with 11/12 >= 5 mm. The IGRA was unchanged throughout the study period in all patients. The overall agreement between TST and IGRA was poor (kappa coefficient = 0.180, p=0.020). Conclusion We demonstrated a transient but significant decrease in TST response in the first six months of anti-TNF therapy.