Clinical Outcome and Prognostic Factors of Acute Symptomatic Intracranial Vertebrobasilar Artery Dissection

Abstract

Although intracranial vertebrobasilar dissection (VBD) has been increasingly recognized as a cause of stroke in East Asian populations, its clinical course and treatment method are not yet established. The purpose of this study was to retrospectively evaluate treatment methods, clinical outcome, and prognostic factors of patients with VBD. Two hundred seventy-two patients (M:F=173:99; Mean age, 45 ？ 11 years, range, 21 ？ 81 years) with ruptured (n=84) or unruptured VBD (n=191) were recruited between January 2001 and December 2008. The patients were divided into two groups, ruptured or unruptured, for evaluation. The investigations were focused on clinical outcome (favorable vs. unfavorable; mRS, 0 ？ 1 vs. 2 ？ 5 or death for unruptured VBD and mRS, 0 ？ 2 vs. 3 ？ 5 or death for ruptured VBD) and prognostic factors.

Of the 191 patients with an unruptured VBD, 46 patients (24.1%) underwent endovascular treatment. The remaining 145 patients (75.9%) were medically treated with anticoagulants (n=49), antiplatelets (n=48), or analgesics alone (n=48). Clinical follow-up data were available for 178 patients (102 ischemic and 76 non-ischemic) at 12 to 102 months (median, 46 months). None of the unruptured VBD hemorrhaged. All 76 patients without ischemic presentation had favorable outcomes (mRS, 0 ？ 1). Of the 102 patients with ischemic presentation, outcomes were favorable in 92 and unfavorable in 10 patients. Four patients died; three died of causes unrelated to VBD, and one died as a result of basilar artery (BA) dissection. Older age (OR, 1.101; 95% CI, 1.101 ？ 1.211; p=0.049) and BA involvement (RR, 14.388; 95% CI, 1.983 ？ 104.413; power, 0.97; p=0.008) were independent predictors of unfavorable outcomes in siu-VBD patients with ischemic presentation.

Of the 81 patients with a ruptured VBD, 76 patients underwent endovascular treatments for a ruptured VBD and 5 patients were managed conservatively. Forty-six patients received deconstructive method using internal coil trapping or proximal occlusion of parent artery and thirty received reconstructive treatment using one to three overlapping stents alone or with coiling. Reconstructive treatment was more frequently used in VBD that involved BA (8/8) or posterior inferior cerebellar artery (PICA) origin (20/30) than in other cases (p < 0.01). Clinical outcomes were favorable (modified Rankin Scale, 0 ？ 2) in 59 (77.6%) and unfavorable in 17 patients (22.4%). Nine patients (11.8%) died, and five deaths were due to rebleeding. There were nine recurrences, wherein six had rebleeding and three no rebleeding. PICA origin involvement was the only independent risk factor for recurrence. Outcomes were favorable (mRS, 0 ？ 2) in 59 patients (77.6%) and unfavorable in 17 (22.4%). Rebleeding (RR, 22.717; 95% CI, 1.899 ？ 271.767; power, 0.99; p=0.014), pre-treatment Hunt & Hess grade (HHG) 4 or 5 (RR, 11.778; 95% CI, 1.822 ？ 76.134; power, 0.97; p=0.01), and PICA origin involvement (RR, 4.362; 95% CI, 1.054 ？ 19.431; power, 0.51 p=0.042) were independent predictors of unfavorable outcomes.

In conclusion, clinical outcome of unruptured VBD was favorable in all without ischemic symptom and in most patients with ischemic symptom. Older age and BA involvement are independent predictors of unfavorable outcome in sui-VBD with ischemic symptom. Clinical outcome of ruptured VBD was favorable in majority of patients who underwent endovascular treatment. Rebleeding, HH 4-5 and PICA origin involvement were independent predictors of unfavorable outcome.