Predictors of Poor Long-Term Outcomes after Successful Thrombectomy in Large Vessel Acute Ischemic Stroke

Objectives*
Despite advancements in modern endovascular treatment, the burden of poor outcomes in acute ischemic
stroke patients remains high even after successful endovascular thrombectomy (EVT). In this study, we at-
tempted to identify the potential predictors of such outcomes in hospitalized ischemic stroke patients.
Materials and Methods
404 ischemic stroke patients treated with EVT at Pauls Stradiņš Clinical University Hospital, Riga, from 2015 to
August 2024 were selected based on successful EVT TICI scores (2b and 3). Long-term outcomes were assessed
at 90 days after the procedure and identified as good (mRS 0-3) or poor (mRS 4-6).
Results
The following 11 factors were suggested for predictive univariate analysis of poor outcomes after EVT: time to
needle, wake-up stroke, NIHSS at admission, EVT procedure duration, number of EVT passes, use of
intravenous thrombolysis, as well as patient-specific factors such as age, gender, platelet count, presence of
diabetes mellitus and hypertension.
Significant predictors identified in poor vs. good outcomes were:
• Time to needle(TTN) (median 280min [IQR 210-359] vs. 240min [IQR 192-300]; p<0.001);
• NIHSS at admission (median 18 [IQR 14-20] vs. 14 [IQR 9-17]; p<0.001);
• EVT procedure duration (median 40min [IQR 24-59] vs. 30min [IQR 20-45]; p<0.001);
• Number of EVT passes (1 vs. >3;p<0.001);
• Age (median 75y vs.72y;p=0.003).
Most significant predictors of poor outcomes (Se 69.5%, Sp 72.4%, AUC 0.766):
• NIHSS at admission (OR=1.180 per point; p<0.001, 95%CI 1.118-1.246)
• EVT procedure duration (OR=1.021 per minute; p=0.001, 95%CI 1.009-1.033)
• Age (OR=1.033 per year; p=0.007, 95%CI 1.009-1.057)
The logistic regression model achieved 71.0% classification accuracy, with a sensitivity of 69.5%, specificity of
72.4%, and an area under the curve (AUC) of 0.776.
191
Conclusions
Poor outcomes following EVT, despite successful recanalization, are independently associated with age, longer
TTN and EVT procedure with more passes, higher NIHSS scores at admission. Recognizing these predictors can
improve the patient management approach.