Latent Tuberculosis Infection Testing in People with Compromised Immunity Prior to Biologic Therapy: A Review of Diagnostic Accuracy, Clinical Utility, and Guidelines

Key Message

Evidence regarding the diagnostic accuracy of the tuberculin skin test for patients with compromised immunity prior to initiating biologic treatment was available from one systematic review with meta-analyses (that included 16 relevant primary studies) and three additional diagnostic test accuracy studies. The results of these studies were inconsistent. The findings of the systematic review suggested that the coherence between the tuberculin skin test and interferon-gamma release assays was relatively high at 85%; however, the authors of the three diagnostic test accuracy studies concluded that the agreement between the two tests in their study populations was poor or fair. One relevant non-randomized study was identified regarding the clinical utility of using the tuberculin skin test for patients with compromised immunity prior to initiating biologic treatment. This study was non-comparative and was descriptive in nature. The authors noted that the frequency of active tuberculosis infection was 6.8% during 55 months of follow-up after patients initiated biologic therapy, despite being screened for latent tuberculosis using the tuberculin skin test (those who tested positive for the tuberculin skin test were given prophylaxis with isoniazid). Six guidelines that provided recommendations regarding the testing for latent tuberculosis infection in patients with compromised immunity prior to initiating biologic therapy were identified. All six of these guidelines recommended in favour of screening for latent tuberculosis in patients who are candidates for biologic therapy due to their increased risk for developing active tuberculosis. The tuberculin skin test, interferon-gamma release assays, chest x-ray, and chest computed tomography were proposed as viable diagnostic techniques; however, recommendations did not clearly indicate one was preferred over the others in all clinical scenarios. The underlying evidence informing these recommendations was generally of low quality or the recommendations were based solely on the expert opinion of the guideline development groups. The limitations of the included literature (e.g., the lack of an established gold standard for detecting latent TB infection, the unclear generalizability to Canadian settings) should be considered when interpreting the findings of this report.