Background and purpose: In patients with acute basilar artery occlusion (BAO) who are undergoing endovascular thrombectomy, whether diffuse pontine infarction (DPI) on baseline imaging study is associated with poor treatment outcomes is unknown. We aimed to investigate the clinical impact of DPI revealed on pretreatment DWI in a large cohort of patients with acute BAO who underwent endovascular thrombectomy.
Methods: We retrospectively analyzed clinical and DWI data from 102 patients who underwent thrombectomy for treatment of acute BAO. DPI was defined as a DWI hyperintense lesion occupying ≥80% of the pons on more than two sections. Association between DPI and extreme poor outcomes (defined as 90-day mRS 5 and 6) was assessed. Binary logistic regression analysis was performed to identify independent predictors of mortality at 90-day. 
Results: DPI was found in 15 patients. Successful reperfusion (m-TICI 2b or 3) was achieved in all patients with DPI. All patients with DPI on pretreatment DWI had mRS 5 (n=6) or mRS 6 (n=9) at 90-day. Patients with DPI on pretreatment DWI had a higher incidence of extreme poor outcomes (100% vs. 24.1%; relative risk 4.413, 95% confidence interval [CI] 2.854-6.013, P<0.001) than those without DPI. 90-day mortality was 17.6% (18/102). On binary logistic regression analysis, DPI was an independent predictor of 90-day mortality (odds ratio 6.787, 95% CI 1.746-26.387, P=0.006). 
Conclusions: DPI on pretreatment DWI was associated with extreme poor outcomes and mortality in patients with acute BAO even if successful reperfusion was achieved with endovascular thrombectomy. Our study suggests that DWI should be obtained before thrombectomy to exclude patients with DPI.

ASSOCIATION OF PRETREATMENT DIFFUSE PONTINE INFARCTION WITH EXTREME POOR OUTCOMES AFTER ENDOVASCULAR THROMBECTOMY IN PATIENTS WITH ACUTE BASILAR ARTERY OCCLUSION