The pathogenesis of delirium in acute stroke is incompletely understood. The use of medications with anticholinergic (ACH) activity is associated with an increased frequency of delirium. We hypothesized that the intake of medications with ACH activity is associated with delirium in acute stroke patients. Delirium was assessed using the DSM-IV-TR criteria and the Delirium Rating Scale, in a sample of consecutive patients with an acute (< or =4 days) cerebral infarct or intracerebral haemorrhage (ICH). We performed a gender and age matched case-control study. Twenty-two delirious stroke patients (cases) and 52 non-delirious patients (controls) were compared concerning the intake of ACH medications (i) before stroke, (ii) during hospitalization but before the assessment. The variables associated with delirium on bivariate analysis were entered in a stepwise logistic regression analysis. The final regression model (Nagelkerke R2 = 0.65) retained non-neuroleptics ACH medication during hospitalization (OR = 24.4; 95% CI = 2.18-250), medical complications (OR = 20.8; 95% CI = 3.46-125), ACH medication taken before stroke (OR = 17.5; 95% CI = 1.00-333.3) and ICH (OR = 16.9; 95% CI = 2.73-100) as independent predictors of delirium. This preliminary result indicates that drugs with subtle ACH activity play a role in the pathogeneses of delirium in acute stroke. Medication with ACH activity should be avoided in acute stroke patients.