Can a prediction model tell me if my endovascular thrombectomy will fail?
Endovascular thrombectomy (EVT) is a highly effective treatment for large vessel occlusion stroke, but in some cases, even when the blocked artery is successfully reopened, patients do not achieve a good functional recovery. This is called futile recanalization. Researchers have developed prediction models that combine clinical and imaging factors to estimate your individual risk of this outcome. While these models can provide useful information, they are not 100% accurate and should be used as a tool to guide discussions with your doctor, not as a definitive answer.
What the research says
A 2025 systematic review and meta-analysis of 51 studies found that lower ASPECTS scores (a measure of early ischemic changes on CT scan) were strongly associated with futile recanalization after thrombectomy 1. Another study developed a prediction model for futile reperfusion in 390 patients who underwent EVT and achieved successful reopening of the vessel 6. The model included nine variables: stroke severity (NIHSS score), CT angiography source image Alberta Stroke Program Early CT Score (ASPECTS), collateral status, serum glucose, neutrophil-to-lymphocyte ratio, homocysteine, age, onset-to-puncture time, and history of atrial fibrillation 6. The model showed good predictive performance, with an area under the curve (AUC) of 0.83, meaning it correctly distinguished between patients who would and would not have futile reperfusion about 83% of the time 6. Additionally, a separate study found that higher levels of the gut metabolite TMAO and a higher neutrophil-to-platelet ratio (NPR) in blood from the blocked vessel were associated with poor outcomes after EVT, and combining these markers improved prediction 5. However, it is important to note that no model is perfect, and individual results can vary.
What to ask your doctor
- What is my ASPECTS score and collateral status on my initial imaging, and how do they affect my risk of futile recanalization?
- Based on my age, stroke severity (NIHSS), and other factors, what does the prediction model estimate for my chances of a good recovery after thrombectomy?
- Are there any blood biomarkers, such as TMAO or inflammatory markers, that might provide additional information about my prognosis?
- How should I interpret the model's prediction when making decisions about my treatment and rehabilitation plan?
- What are the limitations of current prediction models, and how much uncertainty remains in my individual case?
This question is drawn from common patient questions about Pulmonology & Critical Care and answered using cited medical research. We do not provide individualized advice.