Objective To explore the clinical characteristics of elderly patients with tuberculous meningitis (TBM) and analyze the causes of misdiagnosis.Methods The clinical data of patients aged 60 years and over with TBM in Beijing Chest Hospital Affiliated to Capital Medical University from January 2011 to January 2017 were retrospectively analyzed.The clinical characteristics of the elderly patients with TBM were analyzed and compared between misdiagnosis and non-misdiagnosis groups.Results Among 60 elderly patients with TBM,32 cases (53.3%) were misdiagnosed before hospitalization.Among 32 misdiagnosed cases,11 (34.4%) were misdiagnosed as pulmonary infection,6 (18.8%) as infectious diseases with unspecified site,4(12.5%) as upper respiratory tract infections and 3 (9.4%) as cerebral vascular diseases,and so on.The onset time of TBM,fever prevalence and CSF glucose levels were 42.5 (20.3,60.0) d,96.9% (31 cases),1.5(1.1,2.3)mmol/L in misdiagnosis group respectively,and 15.0 (10.0,20.0) d,75.0%(21 cases),2.3(1.4,3.2)mmol/L in non-misdiagnosis group respectively,which was statistically significantly different between the two groups (all P< 0.05).There were no statistically significant differences in age,gender,occupation,residence,complications,comorbidities,radiographic signs between the two groups (all P> 0.05).Conclusions Elderly patients with TBM have atypical clinical manifestations,many comorbidities,and less specific imaging,and especially,the cerebrospinal fluid pathogen sensitivity is low,which may result in higher misdiagnosis rate.

Objective:To explore the feasibility of U6 and Cel-miR-39 as reference genes for quantitative detection of microRNA (miRNA) in cerebrospinal fluid (CSF) of tuberculous meningitis (TBM), and validate the difference of miRNAs between tuberculous and viral meningitis (VM).Methods:The remaining CSF specimens after routine examination were collected in Beijing Chest Hospital of Capital Medical University. A total of 36 TBM and 34 VM patients were enrolled based on the information in the medical records. Total RNA were extracted from the CSF samples, and Taqman based real-time quantitative PCR (RT-CR) analysis were performed to determine the concentration of the miRNAs in CSF. GeNorm, NormFinder and Bestkeeper software were used for stability analysis of the two reference genes. 2 -ΔCt method was used to determine the relative gene expression. Accordance of repeated tests was analyzed by Pearson correlation test. Continuous variables were compared by the t-test. Results:Among the 70 samples, the average cycle threshold (Ct) value of U6 was 30.40±3.30, while the average Ct value of Cel-miR-39 was 21.49±0.70. The expression level of Cel-miR-39 was higher than that of U6. Correlation analysis showed good accordance of the repeated tests among the reference genes and target genes analysis in the randomly selected 10 samples ( r>0.931, P<0.001). Based on the analyses results of the three software, including GeNorm, NormFinder and Bestkeeper, Cel-miR-39 presented better stability in RT-PCR analysis and was more suitable as a reference gene for miRNA quantitative determination in CSF sample of TBM patients. The relative expression levels of the three target miRNAs were calculated using Cel-miR-39 as the reference gene, and miR-126-3p (1.13±0.41 vs 3.34±0.82, t=2.452, P=0.016), miR-130a-3p (0.56±0.10 vs 2.59±0.70, t=2.960, P=0.004) and miR-151a-3p (0.64±0.25 vs 2.11±0.33, t=3.536, P<0.001) were showed significant lower expression levels in CSF in TBM group than that in VM group. Conclusions:Cel-miR-39 can be used as a reference gene for quantitative detection of miRNAs in CSF of TBM patients. Significant differences were detected in expression level of miR-126-3p, miR-130a-3p and miR-151a-3p between TBM and VM group.