A novel diffusion-weighted imaging-based posterior circulation score for prediction of futile recanalization in acute basilar artery occlusion / 中华神经科杂志

ABSTRACT

Objective:To develop a novel posterior circulation score (PCS) based on pretreatment diffusion-weighted imaging (DWI) for predicting futile recanalization (FR) of acute basilar artery occlusion (BAO) after endovascular therapy (EVT).Methods:A prospectively registered consecutive cohort of BAO patients treated with EVT in Beijing Tiantan Hospital, Capital Medical University during a six-year period was reviewed. This novel DWI-based PCS (DWI-PCS) was calculated according to the characteristics of acute infarction at four sites (pons, midbrain, thalamus and cerebellum), with a normal score of 0 point and a full score of 16 points. FR was defined as the occurrence of poor outcome (modified Rankin Scale score>3) at 90 days despite successful recanalization (modified Thrombolysis In Cerebral Infarction ≥ 2b) at final angiogram after EVT. The independent association of DWI-PCS with FR was evaluated by multivariable Logistic regression, and the predictive discrimination of DWI-PCS was measured by the area under the receiver operating characteristic (ROC) curve. Additionally, the effects of DWI-PCS on FR in different subgroups stratified by age, time window, stroke severity and etiology were explored.Results:Of 109 patients in this study, 48 (44.0%) suffered from FR. The multivariable Logistic analysis showed that DWI-PCS was significantly associated with FR (adjusted OR=1.31, 95% CI 1.07-1.62, P=0.01). The ROC curve analysis demonstrated that the area under the curve of DWI-PCS for predicting FR was 0.74 (95% CI 0.65-0.83), and the optimal cut-off value was ≥ 3 points (sensitivity 0.75, specificity 0.66, accuracy 0.70). Finally, the effects of DWI-PCS on FR were not found to be different across all subgroups ( P>0.10 for all interactions). Conclusions:The novel DWI-PCS may be a valid and reliable predictor of FR in BAO patients treated with EVT. Nevertheless, external validation with blinded outcome is still needed to confirm its performance before clinical application.