Interferon-gamma Levels In Cerebrospinal Fluid Compared With Peripheral Blood T-SPOT. TB For The Diagnosis Of Tuberculous Meningitis

Objectives:Tuberculous meningitis (TBM) was a serious extrapulmonary tuberculosis, which mortality and morbidity were high. Rapid and effective diagnosis and early treatment were the key to reduce the mortality and morbidity. Early and accurate diagnosis of tuberculous meningitis has been a worldwide problem. In recent years, T-SPOT. TB was found and developed gradually, which was a kind of new diagnostic methods for TB and based on enzyme-linked immunosorbent assay. With high sensitivity and specificity, the method in foreign countries has become a major technology for diagnosis of tuberculosis infection, but was rarely applied in tuberculous meningitis. There were significant differences in china and foreign countries on clinical value of peripheral blood T-SPOT. TB diagnosis of tuberculous meningitis, this article will have further study. In recent years, researchers began to study the diagnostic value of cerebrospinal fluid T-SPOT. TB for tuberculous meningitis. The diagnostic test was detecting IFN-gamma-secreting T cells in cerebrospinal fluid by specific antigen stimulation. However, the small amount of cerebrospinal fluid and the number of cells, limiting its application in clinical diagnosis. Compared with the cerebrospinal fluid T-SPOT. TB, detecting unstimulated IFN-gamma in cerebrospinal fluid by enzyme-linked immunosorbent assay was more simple and fast, and required a small amount of cerebrospinal fluid, which may provide a new diagnostic method for tuberculous meningitis. There were no reports on the simultaneous detection of CSF INF-gamma and peripheral blood T-SPOT. TB in patients with tuberculous meningitis. In this study, we applied ELISPOT to detect IFN-gamma-secreting T cells in peripheral blood by tuberculosis-specific antigens stimulation, simultaneously detecting unstimulated IFN-gamma in cerebrospinal fluid by ELISA, and designed to compare the diagnostic value of the two tests and the value of combined detection for the diagnosis of tuberculous meningitis.Methods:Hospitalized cases were enrolled from March2011to March2013in infectious disease department of Huashan hospital affiliated to Fudan university. We collected the medical records of suspected TBM, and screened out of30cases according to the latest diagnostic criteria:the "Definite TBM" and "probable TBM" as the case group (TBM group).51cases of non-tuberculous meningitis patients (non-TBM group) for the control group were enrolled. All patients were collected of1ml cerebrospinal fluid in the first lumbar puncture after admission. A total of10patients of TBM group collected1ml cerebrospinal fluid after anti-tuberculosis treatment for4weeks. All cerebrospinal fluid was centrifuged instantly at3000rpm/minute for10minutes at room temperature. The supernatant was stored at-80℃for detecting unstimulated IFN-gamma in cerebrospinal fluid by ELISA. All patients were collected of5ml anticoagulated blood within two days after admission, then drew intermediate layer of white blood cells after centrifugation for T. SPOT. TB detection. Recorded and analysed all the results of CSF routine, biochemistry, culture and smear examination to TBM group, non-TBM group and four weeks after treatment.Results:1Clinical data analysis:30cases of TBM patients met inclusion criteria,6cases of definite TBM,24cases of probable TBM, including24cases of males and6females, aged from18to79years, an average of45.10+17.65years old. For Non-TBM patients, the control group, a total of51cases, including32cases of males and19females, aged from14to64years with an average of37.73±15.30years old. Compared with the control group, CSF pressure, protein content of TBM group were increased, glucose and chloride content decreased, the differences were statistically significant (P<0.05). Compared with non-meningitis group, CSF pressure, protein content of TBM group were increased, glucose and chloride content were decreased, the differences were statistically significant (P<0.05). Compared with viral meningitis, purulent meningitis and cryptococcal meningitis group, protein content of TBM group was increased, chloride content was decreased, but there was no significant difference in CSF pressure, white blood cell count and glucose content. Compared with TBM group before treatment, CSF pressure, white blood cell count and protein content were lower after the treatment, glucose and chloride content were higher, the differences were statistically significant (P<0.05).2IFN-gamma in CSF:30CSF of TBM group were collected after the first lumbar puncture and then detected by ELISA, The IFN-gamma level in CSF was (0-1688.41)pg/ml, the average content was350.97+372.94pg/ml;In the control group, IFN-gamma levels was (0~954.83)pg/ml, the content of the average was71.30pg/ml. There was significantly difference between TBM group and control group.10patients of30TBM cases were collected of cerebrospinal fluid after anti-tuberculosis treatment for4weeks and then detected by ELISA. In the10patients, IFN-gamma level was (125.18~1688.41) pg/ml before treatment with an average of500.48+504.86pg/ml; IFN-gamma level was (11.60~335.92) pg/ml after4weeks treatment with an average of103.62+92.57pg/ml, the difference was statistically significant (P<0.05). The ROC curve of CSF IFN-gamma for TBM diagnosis was described as Figure4, the area under the curve was0.853, the standard error was0.048,95%confidence interval was [0.759,0.946], the optimal cut-off point was81.36pg/ml, and the corresponding sensitivity was83%, the specificity was88%, Youden index was0.71, the positive predictive value (PPV) was81%, the negative predictive value (NPV) was90%. There were5false-negative patients in TBM group, two of them had been received anti-tuberculosis treatment prior to admission. There were6false-positive patients whose IFN-gamma level was above the cut-off. If2patients who had received anti-tuberculosis treatment prior to admission were removed, the sensitivity of the diagnostic test reached89%, negative predictive value reached94%. Compared with viral meningitis, purulent meningitis, cryptococcal meningitis and non-meningitis group, IFN-gamma content was higher in TBM group, the differences were statistically significant (P<0.05); CSF IFN-gamma in other groups was no significant difference (P>0.05).3Peripheral blood T-SPOT. TB:In the81cases, the sensitivity, specificity, Youden index, positive predictive value, negative predictive value, negative likelihood ratio and positive likelihood ratio of peripheral blood T-SPOT. TB in the diagnosis of TBM were70%(21/30),86%(44/51),0.56,75%,83%,0.35,5.10, respectively. Diagnosis of TBM with T-SPOT. TB, positive rates for TBM and non-TBM group were70%(21/30) and14%(2/15), respectively, the difference was statistically significant (P<0.05). Positive rates for definite TBM and probable TBM were67%(4/6) and71%(17/24), respectively, and there was no significant difference (P>0.05). SFCs distribution of T-SPOT. TB:The numbers of ESAT-6SFCs in PBMC were significantly different between patients with TBM and controls (median number of SFCs was32.5and0out of250,000PBMCs, respectively, P<0.001). The numbers of CFP-10SFCs in PBMC were significantly different between patients with TBM and controls (median number of SFCs was100and0out of250,000PBMCs, respectively, P<0.001).4Mycobacterium tuberculosis culture and acid-fast staining in CSF:Positive rates of Mycobacterium tuberculosis culture in CSF for TBM and non-TBM group were20%(6/30) and0(0/51). The sensitivity, specificity, Youden index, positive predictive value and negative predictive value of Mycobacterium tuberculosis culture in the diagnosis of TBM were20%,100%,0.2,100%,68%, respectively. The acid-fast staining in CSF was negative in all81cases.5Comparison of Mycobacterium tuberculosis culture in CSF, peripheral blood T-SPOT. TB and CSF IFN-gamma:In the three diagnostic tests of peripheral blood T-SPOT. TB, CSF IFN-gamma and Mycobacterium tuberculosis culture, the sensitivity (83%) and negative predictive value (90%) of CSF IFN-gamma were the highest. If we excluded the patients who received anti-tuberculous treatment prior to admission, the sensitivity of CSF IFN-gamma test reached89%, the negative predictive value reached94%. Mycobacterium tuberculosis culture in CSF had the highest specificity (100%) and the lowest sensitivity (only20%).6Combined detection of peripheral blood T-SPOT. TB and CSF IFN-gamma:The sensitivity and specificity of CSF IFN-gamma were higher than peripheral blood T-SPOT. TB, but the differences were not statistically significant (P>0.05). T-SPOT. TB and IFN-gamma at least one positive, the sensitivity of the diagnosis of TBM reached90%, the negative predictive value reached93%, but the specificity(75%) and positive predictive value(68%) were lower. When both of them were positive, the specificity of the diagnosis of TBM reached100%,the positive predictive value reached100%, but the sensitivity was lower (only63%), the negative predictive value was82%. When the two tests were negative, reasonably excluded TBM diagnosis. Both positive might promote the start of empirical antituberculous treatment. Moderate agreement was observed between T-SPOT. TB and IFN-gamma in all participants (κ=0.44). In patients with TBM, agreement between T-SPOT. TB and IFN-gamma was poor (κ=0.27), similarly, in patients with non-TBM, agreement between T-SPOT. TB and IFN-gamma was very poor (κ=-0.15).Conclusions:IFN-gamma-level determination in CSF by ELISA was a fast, cheap and sensitive method for the diagnosis of TBM. Dynamic observation of the content was important for the condition monitoring of TBM. Peripheral blood T-SPOT. TB was also of great significance for the diagnosis of TBM. Joint detection of the two tests could improve the diagnosis efficiency of TBM. At least one positive in the2test were more sensitive than a single test, and both positive were more specific than a single test. Both of the2tests negative helped to exclude TBM, and both positive might promote the diagnosis Of TBM.