Kwak HS, Park JS. Mechanical Thrombectomy in Basilar Artery Occlusion: Clinical Outcomes Related to Posterior Circulation Collateral Score. Stroke. 2020;51(7):2045-2050. doi:10.1161/STROKEAHA.120.029861
Reviewed by Shannon Hextrum, MD
Read the article.*
*You will need to log in to US National Library of Medicine National Institutes of Health to read this article.
Ischemic strokes due to basilar artery occlusion (BAO) are associated with high morbidity and mortality. The American Heart Association/American Stroke Association report uncertain benefit of mechanical thrombectomy in BAO given lack of trial data, though they do support consideration of thrombectomy for BAO within 6 hours from stroke onset (class IIb recommendation). The purpose of this study is to evaluate whether the status of posterior circulation collaterals and the time to recanalization may be used, among other factors, in prognosticating good clinical outcomes after mechanical thrombectomy in BAO.
This retrospective review from a single center analyzed patients with computer tomography angiography (CTA) evidence of BAO from January 2012 to October 2019. To obtain this sample, the investigators identified 908 patients who had undergone endovascular therapy (EVT) for ischemic stroke. Anterior circulation, posterior cerebral artery and vertebral artery orifice occlusions were excluded. CTAs were scored retrospectively based on scoring systems intended to assess collateral status: basilar artery on computed tomography angiography (BATMAN) and posterior circulation collateral score (PC-CS). Both systems are scored out of 10 points, with higher scores representing more patent collateral vessels. Recanalization was achieved using aspiration devices with stent retrievers being used as salvage therapy. Clinical data was recorded from medical records, including the admission modified Rankin Scale (mRS) score and the 3-month mRS score. Two groups were formed based on 3-month outcomes, with favorable defined as mRS ≤2 and poor outcome as mRS>2.
The complete sample of 81 patients had a mean age of 70.3 and median NIHSS of 12. There were no statistically significant differences in mean age or stroke risk factors between the outcome groups. The median NIHSS was lower in the favorable outcome than in the poor outcome group (7.5 vs 15), P<0.001. Compared with the poor outcome group, the favorable outcome group also had a higher percentage of distal basilar artery occlusions (P<0.001).
The favorable outcome group had a higher BATMAN score (6 vs 5; P<0.001), as well as a higher PC-CS (6 vs. 5; P<0.001). A receiver operating characteristic curve analysis for both BATMAN and PC-CS scales yielded a score of 6 as the cutoff value by which to predict a favorable outcome with highest sensitivity and specificity (AUC 0.706 [95% CI, 0.583–0.829]; P=0.002). A multivariate analysis yielded NIHSS < 15 (OR 8.49 [95% CI, 2.01–35.82; P=0.004]), PC-CS >6 (OR 3.79 [ 95% CI 1.05-13.66; p=0.042]) and distal BAO (OR 3.67 [95% CI 1.10-12.26; p=0.035]) as independent predictors of good clinical outcome.
There was no difference between favorable and poor outcome groups with regards to time from symptom onset prior to recanalization, assessed at two intervals: <6 hours and <12 hours. Further analysis of time with intervals <3 hours, 3-6, 6-12, and >12 hours did not reveal a trend in time to recanalization and favorable outcome. Additionally, successful recanalization rate (TICI 2b, 3) was not different between the two outcome groups.
This retrospective review of EVT in BAO showed an association between favorable 3-month outcomes and lower admission NIHSS, distal location of BAO, as well as higher scores on collateral circulation. Notably, the time to recanalization was not correlated with better outcomes, and the rate of successful recanalization was similar between outcome groups.
In terms of favorable prognostic factors, the positive effect of more distal BAO may be a reflection of good collaterals as revealed by a higher BATMAN and PC-CS scores for distal occlusion. The distal BAO group also had lower admission NIHSS than the proximal BAO group, which could similarly reflect good collaterals. While the rate of successful recanalization was similar and high between both outcome groups (83.3% vs. 76.5%), there may be a trend toward better outcomes in patients with complete recanalization (TICI 3). Although the authors do not comment on this, 63.3% of patients with a good outcome had complete recanalization compared to 41.2% with bad outcomes (p=0.055).
The small group sizes and the retrospective nature of the work should be considered while assessing the impact of such findings. The authors suggest that selecting patients for EVT in BAO may be reasonable beyond traditional timeframes, particularly when patients demonstrate other good prognostic factors. Additionally, the study does not address the distribution of tPA administration between the two groups, another factor likely to confound results.
In summary, this is a compelling retrospective study that demonstrates the utility in assessing collateral circulation and warrants further investigation into timeframes for selecting patients for EVT in BAO. Additional considerations evaluating the use of stent retrievers as first-line therapy and utilizing MRI imaging for selecting patients may be worthwhile.