Thrombolysis in cervical artery dissection--data from the Cervical Artery Dissection and Ischaemic Stroke Patients (CADISP) database.
Résumé
OBJECTIVE: To examine whether thrombolysis for stroke attributable to cervical artery dissection (CeAD(Stroke) ) affects outcome and major haemorrhage rates. METHODS: We used a multicentre CeAD(Stroke) database to compare CeAD(Stroke) patients treated with and without thrombolysis. Main outcome measures were favourable 3-month outcome (modified Rankin Scale 0-2) and 'major haemorrhage' [any intracranial haemorrhage (ICH) and major extracranial haemorrhage]. Adjusted odds ratios [OR (95% confidence intervals)] were calculated on the whole database and on propensity-matched groups. RESULTS: Among 616 CeAD(Stroke) patients, 68 (11.0%) received thrombolysis; which was used in 55 (81%) intravenously. Thrombolyzed patients had more severe strokes (median NIHSS score 16 vs. 3; P < 0.001) and more often occlusion of the dissected artery (66.2% vs. 39.4%; P < 0.001). After adjustment for stroke severity and vessel occlusion, the likelihood for favourable outcome did not differ between the treatment groups [OR(adjusted) 0.95 (95% CI 0.45-2.00)]. The propensity matching score model showed that the odds to recover favourably were virtually identical for 64 thrombolyzed and 64 non-thrombolyzed-matched CeAD(Stroke) patients [OR 1.00 (0.49-2.00)]. Haemorrhages occurred in 4 (5.9%) thrombolyzed patients, all being asymptomatic ICHs. In the non-thrombolysis group, 3 (0.6%) patients had major haemorrhages [asymptomatic ICH (n = 2) and major extracranial haemorrhage (n = 1)]. CONCLUSION: As thrombolysis was neither independently associated with unfavourable outcome nor with an excess of symptomatic bleedings, our findings suggest thrombolysis should not be withheld in CeAD(Stroke) patients. However, the lack of any trend towards a benefit of thrombolysis may indicate the legitimacy to search for more efficient treatment options including mechanical revascularization strategies.