Background: Thrombectomy devices are increasingly used for intra-arterial recanalization therapy in stroke. We analyzed whether the use of these devices modified the outcome of patients with acute basilar occlusion (BAO) at our institution.
Methods: Between 1998 and 2012, one hundred forty-seven consecutive patients with acute BAO received recanalization therapy. In July 2009, for the first time, a thrombectomy device was used and hence the cohort was split into two chronological groups: BAO-1 (before July 2009) and BAO-2 (after July 2009). All patients were treated at a dedicated neurological ICU following institutional standard operating procedures. A good clinical outcome was defined as a modified Rankin scale score of 0-2 after 3 months. Univariate and multivariate analyses were applied using outcome parameters as dependent variables and baseline variables with a significant p value in univariate tests as independent variables.
Results: One hundred eleven patients (BAO-1) were treated before and 36 were treated after July 2009 (BAO-2). Patients in the BAO-1 and BAO-2 groups had similar neurological deficits on admission as expressed by the Glasgow Coma Scale (BAO-1: median 4, IQR 5, vs. BAO-2: median 4.5, IQR 8, p = 0.41) and the proportion of patients who were presented intubated and ventilated was similar in both groups as well (49.5 vs. 47.7%, p = 0.85). Bridging concepts with intravenous recombinant tissue plasminogen activator (rtPA) were applied in 18.9% (BAO-1) versus 63.9% (BAO-2, p < 0.001) of cases, whereas glycoprotein IIb/IIIa antagonists were used significantly more frequently in the BAO-1 cohort (57.7 vs. 33.3%, p = 0.034). Thrombectomies were performed in 20 patients (55.5%) of the BAO-2 group but in none of the BAO-1 cohort. Complete recanalization (TICI 3) was achieved in 45.1% (BAO-1) versus 66.7% (BAO-2, p = 0.062) of patients. A good clinical outcome was observed in 13.5% of the BAO-1 group and 30.6% of the BAO-2 cohort (p = 0.026); mortality was 57.7% in the earlier group and 36.1% in the later group (p = 0.034). The frequency of symptomatic intracranial hemorrhage was similar in both groups (8.1% BAO-1 vs. 2.8% BAO-2, p = 0.45). Treatment in the BAO-2 cohort was an independent predictor of good clinical outcome (OR 2.56; 95% CI 1.01-6.78) and mortality (OR 0.36; 95% CI 0.15-0.86) in an adjusted logistic regression model.
Conclusion: Our results show improved outcomes in patients in the BAO-2 cohort. The treatment approach in this group was an independent predictor of both good outcome and mortality. Especially in patients with BAO - where endovascular treatment strategies are common clinical practice - bridging protocols with rtPA and modern thrombectomy devices should be used more frequently.