DAFNES scale for large vessel occlusion in Ischemic Stroke : development and preliminary internal validation

Early identification of large vessel occlusion (LVO) in acute ischemic stroke is critical, as the benefit of reperfusion therapies diminishes rapidly with time. Although multiple prehospital stroke severity scales exist, most require direct bedside assessment by trained personnel, with limited evidence supporting telephone-based LVO recognition. This study aimed to develop the DAFNES scale and conduct preliminary internal hospital-based validation for LVO identification in patients with suspected acute stroke. A prospective observational study enrolled 593 consecutive patients with suspected stroke admitted to the Emergency Department of a tertiary hospital (January-September 2024). Following exclusion of stroke mimics and intracerebral hemorrhages, 408 patients were analyzed. The DAFNES scale comprises six binary items: oculocephalic deviation, aphasia, facial droop, neglect, hemiparesis, and altered consciousness. Scores were assigned bedside by expert neurologists blinded to CT angiography findings. LVO status was determined by a blinded neurologist based on CT angiography results. Among 408 analyzed patients, 185 (45.3%) presented with LVO and 223 (54.7%) with non-LVO ischemic stroke. The DAFNES scale demonstrated good discriminatory performance with an area under the curve of 0.79 (95% CI 0.75-0.83). At a cutoff score of ≥ 3: sensitivity was 0.78, specificity 0.72, positive predictive value 0.70, and negative predictive value 0.80. Median DAFNES scores were significantly higher in LVO patients compared to non-LVO patients. The DAFNES scale demonstrates promising discriminatory ability for LVO detection in hospital-based cohorts. However, formal prehospital and dispatcher-based validation studies are required before clinical implementation.