Effectiveness of thrombolysis in the Rhône region, France: a prospective population-based study

Although thrombolysis with intravenous alteplase significantly improves stroke outcomes (1), overall effectiveness of acute stroke management is often unknown because most studies use stroke units or stroke registries perspective (2). Our aim was to observe in a defined French population management of stroke from diagnostic suspicion. AVC69 study was a population-based prospective study founded by the Ministry of Health. Over seven-months, all consecutive patients with a suspected acute stroke admitted to any emergency department (ED) or stroke unit of the Rhone area (1·7 millions inhabitants) were included. Definitive diagnosis was based on cerebral imaging (computed tomography scan or magnetic resonance imaging). Among the 1206 patients included in this comprehensive cohort (Fig. 1), only 108 were directly hospitalized in a stroke unit (9%), whereas 1098 (91%) were admitted in an ED, among which 101 (9%) were subsequently transferred to a stroke unit. Among the 193 patients eligible to thrombolysis admitted in ED, only 8% were thrombolysed, although 50% arrived within three-hours from onset. Median time from arrival to imaging was 2h20 for patients admitted in the ED, six times longer than for those admitted directly in the stroke unit (P < 0·0001). Despite numerous campaigns designed to increase stroke symptoms awareness and the need to call emergency medical services, at the population level, 91% patients with a suspected stroke were admitted to an ED in 2007. These results led us to develop an ongoing randomized stepped wedge trial to assess the effectiveness of an educational program designed to increase awareness and skills of ED professionals. This program associated with the effect of time-window extension, and creation of three stroke units in the region is expected to significantly increase thrombolysis rate (3,4). Our study will be repeated in 2013 as the effectiveness of thrombolysis for acute ischaemic stroke in assessment should rely on population-based studies (5).