# Early surrogates of functional outcome after thrombectomy for MCA-M2 occlusions

## TL;DR

This study finds that NIHSS scores at 24 hours best predict long-term recovery in patients with MCA-M2 strokes after thrombectomy.

## Contribution

The study identifies optimal NIHSS thresholds for predicting outcomes in MCA-M2 occlusion patients.

## Key findings

- 24h NIHSS had the highest predictive accuracy (AUC 0.85) for both good and excellent outcomes.
- Optimal NIHSS thresholds were ≤8 for good and ≤7 for excellent outcomes.
- Age, pre-stroke disability, diabetes, and procedure passes reduced predictive accuracy.

## Abstract

The National Institutes of Health Stroke Scale (NIHSS) at 24h is a strong predictor of long-term independence after mechanical thrombectomy for anterior circulation strokes. Its predictive value in patients with M2 segment of the middle cerebral artery (MCA-M2) occlusions, however, remains unclear. This study compares NIHSS scores at admission and 24h and early neurological improvement (ENI) definitions to predict functional outcomes at 90d for M2 occlusions. Patients with isolated MCA-M2 occlusions from the German Stroke Registry (06/2015–12/2021) were analyzed. NIHSS at admission, 24h and ENI definitions were compared for predicting excellent and good outcome, defined as modified Rankin Scale (mRS) 0-1 and 0-2. Factors impairing predictive power were identified based on multivariable logistic regression. Among 1,268 patients, 24h NIHSS showed the highest predictive accuracy, with an area under the curve (AUC) of 0.85 [95% Confidence Interval: 0.83-0.87] for both good and excellent outcomes. Optimal NIHSS thresholds were ≤8 for good and ≤7 for excellent outcomes. Advanced age, higher pre-stroke mRS, diabetes, and higher procedure passes significantly reduced prognostic accuracy. NIHSS at 24h, with thresholds of ≤8 and ≤7, is the best surrogate for predicting good and excellent functional outcomes in MCA-M2 occlusion patients. Prognostic value diminishes with advanced age, pre-stroke disability, and comorbidities such as diabetes.

The online version contains supplementary material available at 10.1038/s41598-025-34777-8.