The basilar artery (BA) supplies the brain stem, cerebellum, thalami, occipital lobes and medial temporal lobes. If the BA is occluded, symptoms can include decreased consciousness, quadriparesis, pupillary and oculomotor abnormalities, dysarthria, dysphagia or even sudden death.10) BA occlusion accounts for about 20% of ischemic strokes.13) Acute occlusion of this artery produces high morbidity and mortality, with a rate of 85-95%, if left untreated.6)13)21) Without recanalization, the likelihood of good outcome is about 2%.6)

The chief goal in treating acute ischemic stroke is to restore cerebral blood flow as rapidly and safely as possible.1)4)17) Intravenous tissue plasminogen activator (IV-tPA) or endovascular techniques are widely used to produce vascular patency and to improve clinical outcomes. It has become a fairly standard practice in the anterior circulation to set a time restriction to the first 6 hours of symptom onset for endovascular mechanical thrombectomy (EMT) to limit hemorrhagic transformation of infarctions and because of its poor contribution to favorable clinical outcome.15) However, the effective therapeutic time window has not been established for the posterior circulation.16) Therefore, we conducted a study to verify the efficacy and safety of endovascular mechanical thrombectomy for BA occluded patients, even beyond 6 hours of symptom onset.

Medical records of 16 consecutive patients who were treated for symptomatic acute BA occlusion from July 2012 to Feb 2013 were reviewed. To rule out hemorrhagic stroke or old infarction, all patients underwent brain-computed tomography (CT) on arrival. Magnetic resonance images (MRI), such as diffusion-weighted image (DWI), MR angiography and perfusion-weighted image (PWI) were then tracked to detect the occluded vessels, DWI-PWI mismatch, and measure infarct territories. Intravenous thrombolysis (IVT) was started with 0.9 mg/kg IV-tPA when the diagnosis was confirmed within 3 hours of symptom onset. If refractory to IVT or contraindicated or the patients arrived more than 3 hours after symptom onset, EMT was performed. Indications for EMT treatment were a National Institutes of Health Stroke Scale (NIHSS) score of 4 or more, age more than 18 years and arrival at the stroke center beyond 3 hours of symptom onset. Considering these uncertainties and high mortality rate of untreated BA occluded patients, our stroke center adopted a protocol to treat not only patients who came within 6 hours of symptom onset but also who came beyond 6 hours with symptom aggravation, mismatch between DWI and PWI.

All procedures were performed under local anesthesia by an endovascular neurosurgeon. The occlusion status was confirmed by taking angiography after locating the guiding catheter (Guider XF Softip™, Boston Scientific, Plymouth, MN, USA) to the proximal vertebral artery. To prevent procedural thromboembolic events, a mixture of 2,000 IU of heparin and 0.9% normal saline 1,000 ml was administered continuously through the guiding catheter. This was applied even to patients who received IV-tPA. The main endovascular techniques used to restore BA flow were direct thrombectomy using the Penumbra thromboaspiration catheter (Penumbra Inc., Alameda, CA, USA) (Fig. 1) and thrombus retrieval by the Solitaire stent (ev3 Inc., Irvine, CA, USA) (Fig. 2). The Penumbra system with microcatheter (Prowler SELECT™ PLUS, Codman Neurovascular, Raynham, MA, USA) and microwire (Synchro™-14, Stryker Neurovascular, Fremont, CA, USA) were brought into the target vessel and placed at the distal portion of the occlusion site after penetrating the thrombus segment. Using a 50 cc syringe, negative pressure was applied through the Penumbra system. If recanalization failed, the Solitaire stent was used for mechanical thrombectomy. The Solitaire stent was deployed where it could completely cover the occluded segment and left in place for at least 2 minutes before retrieval. The microcatheter and the stent were gently withdrawn together with the guiding catheter while a 50 cc syringe maintained negative pressure through the guiding catheter to aspirate the thrombus and to prevent migration of the emboli. If stenosis of vertebral artery was too severe for catheter entry, balloon angioplasty and stent deployment was performed at the segment and post-balloon was carried out if vascular patency was unsatisfactory. After EMT, the brain was scanned routinely by CT within 1 hour to check for any procedural related intracerebral hemorrhage and by MRI after 24 hours to see any changes in the infarct region.

Neurologic status was recorded based on NIHSS score. The outcome data were collected by: (1) rate of angiographic recanalization; (2) rate of improvement of NIHSS score and mRS at discharge and after 3 months; and (3) rate of cerebral hemorrhagic complications such as symptomatic hemorrhages or parenchymal hematomas, measured to evaluate the safety of EMT. Successful recanalization was defined as achieving TICI II (partial recanalization) or III (complete recanalization).18) Assessment of the modified Rankin Scale (mRS) was performed 90 days after treatment. A mRS of 0 to 2 was defined as a good neurological outcome. To assess the efficacy and safety of EMT performed after 6 hours, patients were assigned to 2 groups. Group A was defined as patients treated within 6 hours of symptom onset and group B as those treated beyond 6 hours. The data from these studies were statistically analyzed using the SPSS version 20. (IBM Corp., Armonk, NY, USA). Statistical procedures included Student's t-test and Pearson's chi-square as appropriate for continuous or categorical variables. Null hypotheses of no difference were rejected if p values were less than 0.05.

Of the 16 consecutive patients with BA occlusion treated mainly with EMT from July 2012 to Feb 2013, 12 were men and 4 were women, with a mean age of 67.8 ± 11 years. The median symptom-to-puncture time was 5.8 hours (range, 3-120). In initial conventional angiography, proximal thrombosis was present in 50% and middle and distal basilar thrombosis in the remaining 50%. Four patients (case number 1, 2, 6, 14) who came to our stroke center within 3 hours of symptom onset were treated with IVT. They presented with gradual aggravation of symptoms and received EMT. The other 12 patients depended on EMT as a first-line treatment (Table 1). The most commonly diagnosed cardiovascular co-morbidities were hypertension (75%), type 2 diabetes (43.8%), previous stroke (25%) and atrial fibrillation (18.8%). The major etiologic risk factors were large artery atherosclerosis (81.2%) followed by cardioembolism (18.8%) (Table 2).

Three patients (18.8%) were unconscious and needed respiratory support to maintain adequate ventilation, while 1 patient died, a mortality rate was 6.2%. Of the 16 patients, revascularization was achieved in 13 (81.3%) (TICI grade IIa/IIB and III). The mean initial NIHSS score was 12.3 ± 8.2 and mean NIHSS score at discharge was 9 ± 11.2. In 7 patients (43.7%), NIHSS scores improved more than 10 or were 0-1 at the time of discharge. At 3 months, 56.3% (9 of 16) of patients had a good neurologic outcome (mRS 0-2) (Table 3). Eight patients who were treated within 6 hours of symptom onset were grouped as A. Of those patients recanalization (TICI II or III) was achieved in 6 patients (75%) and 4 (50%) had good outcomes. The other 8 patients were treated with EMT after 6 hours and were grouped as B. Successful recanalization was achieved in 7 patients (87.5%) and 5 (62.5%) had favorable outcomes. The mean puncture to recanalization time was 97 ± 45.2 minutes. It took a mean of 78.5 ± 29.8 minutes for group A and 115.5 ± 52 minutes for group B. Following EMT, 4 patients (50%) had asymptomatic hemorrhagic transformation in group A and 2 (25%) in group B (Table 4).

If a basilar artery occlusion is left untreated, the prognosis is extremely poor. Therefore, it is important to reverse the occluded artery for a better outcome. Our study suggests that endovascular mechanical thrombectomy has advantage of prompt flow restoration with the potential for a good outcome, even beyond 6 hours of symptom onset. Further studies with larger sample size and a longer duration are needed to support this idea.