Fangjia Tong, Dapeng Li, Xuejie Zhu, Shengkun Zhang, Qifu Chen, Baijing Dong, Yunyan Tang, Lu Tang, Lanlan Wei*, Guobao Li and Ming Chu
Tuberculous meningitis (TBM) is a severe infectious disease in the Central Nervous System (CNS). It’s elusive to differentially diagnose TBM with Bacterial Meningitis (BM). Traditional diagnosis of TBM is based on clinical features, etiological examination, and the biochemistry analysis of cerebrospinal fluid. These conventional methods are time consuming and insensitive, which could lead to a delay in TBM diagnosis. The aim of our study is to develop a diagnosis model which could distinguish TBM from BM rapidly and accurately. A retrospective review of all 191 CNS patients was conducted to determine the differences between TBM (n=145) and BM (n=46) based on clinical and laboratory tests. Logistic regression was used to identify the parameters independently predicting TBM and to develop a diagnosis model. A receiver operator characteristic curve was used to determine the best cutoff for the diagnostic model. Seven parameters were found predictive: Coma, ESR30, FIB, Monocytes%, Lymphocytes%, Neutrophils%, and EOS%. Application of the above seven parameters revealed 89.0% sensitivity and 93.5% specificity. This diagnostic model can help improve the accuracy of the early diagnosis of TBM.
HTML PDFShare this article
Neurological Disorders received 1253 citations as per Google Scholar report
