A Retrospective Review of the Two-Step Tuberculin Skin Test in Dialysis Patients

Background: Reactivation of latent Mycobacterium tuberculosis infection (LTBI) is a health concern for patients on dialysis or receiving a kidney transplant, as these patients are often immunosuppressed. The most frequently used test for LTBI screening in this population is the tuberculin skin test (TST). The diagnostic accuracy (sensitivity and specificity) of the TST in a contemporary North American or Western European dialysis population is unknown. Objectives: Our objective was to determine the diagnostic accuracy and clinical utility of the two-step TST in patients receiving dialysis. Design: This is a retrospective cohort study. Setting: This study is set at four tertiary dialysis units across Winnipeg, Manitoba. Patients: There are 483 chronic hemodialysis and peritoneal dialysis patients in the study. Measurements: The measurements are sensitivity and specificity of the TST with respect to abnormal chest X-ray. Methods: All patients received a two-step TST and assessment of risk factors for prior tuberculosis (TB) infection between February 2008 and December 2008. This cohort was retrospectively linked to our tuberculosis registry to ascertain if prophylaxis was received for LTBI. Results: At an induration cutoff of 5 mm, 62 (13 %) patients had a positive two-step TST. Patients with a known Bacillus Calmette-Guérin (BCG) vaccination were more likely to test positive (50 % of those with a positive TST had a BCG versus 34 % with a negative TST, p = 0.05). Using a diagnostic gold standard of an abnormal chest X-ray as a proxy for LTBI, the sensitivity of the TST was only 14 % and the specificity was 88 %. Only 8 of 62 patients with a positive TST (13 %) received prophylaxis for LTBI. None of the patients who tested negative were treated. Limitations: There is a lack of a truly accurate gold standard for LTBI. Conclusions: The TST has limited diagnostic and clinical utility for LTBI screening in patients on dialysis. Further research into the diagnostic accuracy of interferon-gamma release assays and a revision of LTBI screening guidelines in patients on dialysis should be considered