Introduction: Stroke alerts enable rapid detection of acute ischemic stroke and timely access to IV thrombolysis (IVT) and/or endovascular therapy (EVT). Differentiating true stroke from mimics is resource-intensive and constrained by personnel and imaging capacity; overactivation can strain teams and expose patients to unnecessary risk. Appropriate utilization is essential. Emergency departments (EDs) are accustomed to time-critical workflows, whereas inpatient units are heterogeneous and are more likely to face stroke mimics (e.g., isolated encephalopathy). Hypothesis: Compared with ED alerts, inpatient alerts would have lower true stroke rates, fewer acute interventions, and longer treatment times. We also evaluated outcomes among alerts with isolated encephalopathy. Methods: Single-center, retrospective analysis of all stroke alerts in 2024. Alerts were categorized as ED or inpatient. Outcomes: final stroke confirmation; IVT; EVT; composite any acute intervention (IVT or EVT). Door-to-needle (DTN) among IVT recipients was summarized as median (IQR). Proportions were compared by χ2/Fisher with odds ratios (OR, 95% CI); DTN time by Mann–Whitney U. Analyses were performed in Python. Results: We analyzed 1,306 alerts (ED 1,012; inpatient 294). Stroke was confirmed in 46.1% of ED vs 24.5% of inpatient alerts (OR 2.64, 95% CI 1.97–3.54; p < 0.0001). IVT occurred in 7.8% vs 3.1% (OR 2.68, 1.33–5.41; p = 0.0064). EVT occurred in 11.9% vs 3.7% (OR 3.46, 1.84–6.51; p < 0.0001). Any acute intervention occurred in 18.2% vs 6.1% (OR 3.41, 2.06–5.63; p < 0.0001). DTN was 41 min (IQR 34–64, n=79) in the ED vs 81 min (IQR 75–96, n=7) inpatient (p = 0.0016). Isolated encephalopathy alerts comprised 10.3% overall (ED 6.9%, inpatient 21.8%). Among isolated encephalopathy alerts, 14.9% were stroke and 5.2% received any intervention. In inpatient isolated encephalopathy activations, IVT rate was 1/64 (1.6%) and 0% underwent EVT. Conclusions: Under real-world resource constraints, inpatient stroke alerts were lower yield, far less likely to lead to acute intervention, and had substantially longer DTN than ED alerts. Isolated encephalopathy accounted for about 1 in 5 inpatient activations but rarely resulted in treatment. These findings support the need for improved education about and context-aware activation criteria for isolated encephalopathy to optimize the use of limited stroke resources while preserving rapid treatment for high-probability stroke.
Bata et al. (Thu,) studied this question.