Diagnostic value of T cells spot test of tuberculosis infection on blood and cerebrospinal fluid for tuberculous meningitis

Qingluan Yang,Bingyan Zhang,Qianqian Liu,Wei Zhang,Yuanyuan Liu,Xinhua Weng,Lingyun Shao,Wenhong Zhang,Yan Gao
DOI: https://doi.org/10.3760/cma.j.issn.1000-6680.2018.10.009
2018-01-01
Abstract:Objective To evaluate the diagnostic value of T cells spot test of tuberculosis infection (T-SPOT .TB) on blood and cerebrospinal fluid for tuberculous meningitis (TBM ).Methods One hundred and fifteen adult patients with suspected tuberculous meningitis were retrospectively enrolled from March 2013 to March 2017 in Huashan Hospital affiliated to Fudan University .Among them ,30 were diagnosed with TBM (7 definite ,19 highly probable and 4 possible) ,37 with other infectious meningitis and 29 with non-infectious meningitis .The diagnostic sensitivity ,specificity ,positive predictive values (PPV) and negative predictive values (NPV) of T-SPOT .TB on peripheral mononuclear cells (PBMC) and cerebrospinal fluid mononuclear cells (CSF-MC ) were analyzed using Fisher exact test , and the diagnostic performance was evaluated by using receiver operating characteristic (ROC ) curve and area under the curve (AUC).Results When including the 30 TBM cases and 66 non-TBM cases into analysis , the sensitivities and specificities , PPV and NPV of PBMC and CSF-MC for diagnosing TBM were as follows :93 .1% and 66 .7%,77% and 87 .7%,65 .9% and 71 .4%,95 .9% and 85 .1%,respectively . When including the 30 TBM and 37 other infectious meningitis into analysis , the sensitivities and specificities ,PPV and NPV of the PBMC and CSF-MC for diagnosing TBM were as follows :93 .1% and 66.7%,68 .6% and 86 .5%,71.1% and 80 .0%,92 .3% and 76 .2%,respectively .By ROC curve analysis ,the AUC of blood and CSF were 0 .882 (95% CI :0 .795-0 .969) and 0 .814 (95% CI :0 .704-0 .925) ,respectively .Using a cut-off value of 32 spot forming cells (SFC) per million CSF-MC for T-SPOT .TB on CSF-MC showed a sensitivity of 66 .7%,a specificity of 91 .9%,PPV of 87 .0% and NPV of 77 .3% .The positive likelihood ratio and negative likelihood ratio were 8 .22 and 0 .363 respectively . Conclusions T-SPOT .TB on CSF-MC has a role in diagnosing TBM .And 32 SFC per million CSF-MC might be the optimal cut-off value to differentiate TBM and non-TBM .
What problem does this paper attempt to address?