Introduction
Screening for latent tuberculosis infection (LTBI) before initiation of biologic is recommended internationally especially in tuberculosis (TB) endemic area like Hong Kong. However, there is no gold-standard and local guideline recommends use of either tuberculin-skin-test (TST) or interferon-gamma-release-assay (IGRA) before starting biologic for rheumatic diseases. Both tests have reduced sensitivity in immunosuppressed patients and a previous local study has demonstrated that the two tests have fair level of agreement only. We conducted this retrospective cohort study to determine whether dual LTBI screening could reduce incidence of TB.
Objectives
This study aims to determine whether dual testing with both TST and IGRA before initiation of biologic agents for rheumatic patients can reduce incidence of TB.
Methodology
This is a retrospective cohort study. Consecutive patients who have received biologic for rheumatic diseases in a regional hospital in Hong Kong from July 2007 to February 2018 were reviewed. All patient underwent LTBI screening, either with single testing by TST/ IGRA or dual testing by both. They were categorized into single or dual test group. Background demographics, concurrent medications and choices of biologic were documented. All patients were followed-up regularly since initiation of biologic agents for at least 6 months. Isoniazid chemoprophylaxis was prescribed if the patient was tested positive for LTBI. The primary outcome was the difference in incidence of TB between two groups. Secondary outcomes included incidence of IGRA and TST test positivity, concordance rates of TST and IGRA, risk factors for the development of TB and adverse events associated with isoniazid chemoprophylaxis.
Results & Outcome
198 patients were included in this study. 119 patients underwent single LTBI testing with either TST or IGRA and 79 patients underwent dual testing. In the single test group, 115 patients had TST only and 4 patients had IGRA only. There is no significant differences in demographic between the two groups. The major indication of biologic agents was rheumatoid arthritis (58% in single test group versus 54% in dual test group). 91.3% in the single test and 84.8% in the dual test group had received at least one anti-tumour-necrosis-factor therapy. TB occurred in 9 out of 119 patients in single test group versus 1 out of 79 patients in dual test group (7.56% versus 1.26%, p=0.048). 35 patients in the single test group and 36 patients in the dual test group were tested positive for LTBI and given isoniazid chemoprophylaxis (28.5% versus 41.1%, p=0.014). The level of agreement between IGRA and TST is 73.4% (kappa value 0.446). However, in patient on prednisolone at screening, kappa value is reduced to 0.384 and further reduced to 0.107 in patients on at least 10mg daily prednisolone. Among all biologic agents, infliximab use was significantly associated with the incidence of TB (p=0.001). Reversible hepatotoxicity occurs in 6 out of 71 courses of isoniazid given, which was not significantly different between the two groups. Conclusion: Dual testing strategy with both TST and IGRA appears to be an effective way to reduce the incidence of tuberculosis in patients on biologic agents for rheumatic diseases. It should be considered especially for patients who are on prednisolone when undergoing LTBI screening.
Co-location with:
