Alzheimer's disease (AD) and vascular dementia (VaD) share several features such as overactivation of microglial cells, damage induced by free radicals, glutamate and calcium overload. Propentofylline (HWA 285) has shown beneficial effects on all of these common elements, thus favouring its use in both subtypes of dementia. In a multinational, randomized, 12-month, double-blind, parallel-group study 260 out-patients with mild to moderate AD or VaD received 300 mg propentofylline (n = 129) or placebo (n = 131) three times daily 1 h before meals. The efficacy was tested at four independent rater levels (physician, psychologist, relative and patient) with assessments covering three major domains of dementia (global function, cognitive function and activities of daily living). After 12 months, the total patient population showed statistically significant treatment differences in favour of propentofylline for the global measures of dementia (Gottfries-Bråne-Steen scale, GBS, p = 0.001; Clinical Global Impressions, CGI, item I: p = 0.004, item II: p = 0.072) as well as for the cognitive measures (Syndrome Short Test, SKT, p = 0.002) and Mini-Mental State Examination (p = 0.001). The activities of daily living also showed a significant treatment difference in favour of propentofylline (p = 0.002). No significant treatment differences were found for rating scales performed by the patients. At month 12, VaD patients showed treatment differences in favour of propentofylline for the GBS total score (p = 0.006), CGI item I (p = 0.004), GGI item II (p = 0.044) and SKT (p = 0.028). Treatment differences for AD patients were all in favour of propentofylline and reached statistical significance for the SKT (p = 0.018). Propentofylline showed a good safety profile with respect to adverse events, vital signs, ECG and laboratory changes.