Predictors of mortality in patients with meningeal tuberculosis.

BACKGROUND: Meningeal tuberculosis (TB) has higher mortality compared to other forms of central nervous system TB. However, data on predictors of mortality is limited. AIMS: To determine the predictors of mortality in patients with meningeal TB. MATERIALS AND METHODS: This study retrospectively analyzed the data of patients admitted with a diagnosis of meningeal TB between January 2006 and December 2008. Thwaites' index score of four or less was used for the diagnosis of meningeal TB which is a weighted diagnostic index score for dichotomised clinical variables. Predictors of mortality were analyzed separately for both patients with human immunodeficiency virus (HIV) infection and without. STATISTICAL ANALYSIS: Univariate analysis and multinomial logistic regression was done. RESULTS: Univariate analysis showed age >40 years, Glasgow Coma Scale (GCS) score <8, absence of headache, cerebrospinal fluid (CSF) protein ≤60 mg% and Medical Research Council (MRC) Stage III at presentation to predict in-hospital mortality. In multinomial logistic regression age >40 years was a risk factor for mortality when HIV patients were included (P=0.049) as well as when they were excluded (P=0.048). CSF protein ͳ 60 mg% was found to be a significant risk factor when both HIV seropositive persons (P=0.011) as well as seronegative persons (P=0.004) were included. HIV seropositivity, steroid treatment or delay in treatment did not affect mortality. CONCLUSIONS: Identification of factors predictive of in-hospital mortality will help to prognosticate patients with meningeal TB at the time of admission.

Cognition in type 2 diabetes: Association with vascular risk factors, complications of diabetes and depression.

BACKGROUND: The role of variables like duration of diabetes, diabetic control and microvascular complications in the causation of cognitive decline in patients with type 2 diabetes is not well studied. The contribution of hypertension to the cognitive decline in nondemented diabetic patients is unclear. AIMS: We wanted to see if cognition in patients with type 2 diabetes is associated with the duration of diabetes, control of diabetes, complications of diabetes, vascular risk factors, or depression. We also looked at association of noncompliance with cognition, and depression. SETTINGS AND DESIGN: We recruited ambulant patients with type 2 diabetes who are 55 years or more in age from the weekly diabetic clinic. We excluded patients with past history of stroke. METHODS AND MATERIAL: We selected the time taken for the Trial A test, delayed recall on ten-word list from Consortium to Establish a Registry for Alzheimer's Disease (CERAD), Rowland Universal Dementia Assessment Scale (RUDAS) and Centre for Epidemiologic Studies Depression scale (CES-D) screening instrument to assess these patients. STATISTICAL ANALYSIS USED: We utilized mean, standard deviation, Chi-square test and Pearson's correlation for statistical analysis. We considered P < 0.05 to be significant. RESULTS: RUDAS scores inversely correlated (r = -0.360) with CES-D scores (P = 0.002). Scores of the screening instrument for depression, the CES-D was associated with the duration of diabetes mellitus (P = 0.018), fasting blood glucose (P = 0.029) as well as with 2-hour post prandial blood glucose (P = 0.017). CONCLUSIONS: There is correlation between depression and global cognitive score. Depression seems to be associated with duration of diabetes and control of diabetes.