TB Manual

3.4.1. Blood-based assays (serology, gene-expression signatures)

For decades, researchers and the industry had pinned their hopes on blood-based assays for the detection of TB. Unfortunately, TB serologic tests are inaccurate,^55,56 prompting the WHO to issue a strong recommendation against their use.⁵⁷ A newer generation of blood-based tests, looking at gene expression signatures, is under development and evaluation. At the time of writing, none has achieved high enough sensitivity to rule out active TB, nor high enough specificity to be used as a stand-alone test for the diagnosis of TB.⁵⁸

Another blood-based assay is the IGRA. As described in Chapter 4: Diagnosis of Tuberculosis Infection, the IGRAs cannot distinguish infection from disease; for this reason, a recent WHO policy on IGRAs has discouraged their use for the diagnosis of TB disease. In children, TST and/or IGRA are used as an indicator of recent infection and can be used to support a diagnosis of TB disease, along with clinical data and radiologic and microbiological investigations (see Chapter 9: Pediatric Tuberculosis).

3.4.2. Fluid-based assays (adenosine deaminase)

Laboratories occasionally receive requests to send pleural, pericardial or peritoneal fluid for adenosine deaminase (ADA), which is a marker of serosal inflammation. A systematic review reported a sensitivity and specific of 90% for these tests in high-burden countries. Given that the test can give false-positive results with empyema and certain types of malignancies, it is not recommended in Canada.⁵⁹ Furthermore, the ADA result can support a diagnosis of TB but cannot inform antibiotic treatment, as there is no DST associated with this result.

3.4.3. Urine-based assays (urinary LAM detection)

Urine-based testing has received increased attention, with studies of patients with advanced HIV/AIDS reporting a more rapid diagnosis of TB and faster initiation of therapy.⁶⁰ These tests are not offered in Canadian TB labs, owing to the availability of other diagnostic modalities and the low burden of HIV-associated TB (see Chapter 10: Treatment of Active Tuberculosis in Special Populations).

Recommendations

• We strongly recommend against the use of serologic TB tests, adenosine deaminase and urine lipoarabinomannan testing for the diagnosis of TB (good evidence).

• We strongly recommend against the use of a tuberculin skin test or interferon-gamma release assay for the diagnosis of active TB in adults (good evidence).