Although diagnosing osteoarticular tuberculosis (TB) remains a challenge, a recently developed Mycobacterium tuberculosis-specific ELISPOT assay for diagnosing TB infection has shown promising results. We performed a prospective, blinded, observational study to compare its diagnostic usefulness with those of conventional tests in patients with suspected osteoarticular TB.

All patients presenting at a tertiary hospital between April 2008 and September 2009 with suspected osteoarticular TB were enrolled. In addition to conventional tests for TB, we used ELISPOT assays to measure the IFN-γ response to ESAT-6 and CFP-10 in T-cells in samples of peripheral blood mononuclear cells (PBMC). Patients with suspected osteoarticular TB were classified by diagnostic category.

Of the 65 patients with suspected osteoarticular TB, 5 (8%) were excluded due to inconclusive diagnoses. Of the remaining 60 patients, 23 (38%) were classified as having confirmed TB, 3 (5%) as having probable TB, 2 (3%) as having possible TB, and 32 (53%) as not having active TB. Five (8%) patients with probable or possible TB were excluded from the final analysis. Of the 23 patients with confirmed osteoarticular TB, 15 (65%) had TB spondylitis, 4 (17%) had TB arthritis, 2 (9%) had prosthetic joint infection, and 2 (9%) had extra-spinal TB. The sensitivities of the tuberculin skin test (≥10 mm) and the ELISPOT assay for active osteoarticular TB were 80% (95% confidence interval [CI], 58%–92%) and 100% (95% CI, 85%–100%) (P = 0.04), respectively and their specificities were 68% (95% CI, 51%–81%) and 58% (95% CI, 41%–74%) (P = 0.60), respectively.

A negative ELISPOT assay using PBMC may be a useful test for excluding a diagnosis of active osteoarticular TB.