"Code stroke": hospitalized versus emergency department patients.

Stroke rapid-response ("code stroke") teams facilitate the evaluation and treatment of patients presenting to emergency departments (EDs). Little is known about the usefulness of code stroke systems for patients hospitalized primarily for other conditions. We hypothesized that the yield of code stroke evaluations would be lower in hospitalized than in ED patients, and sought to identify potential targets for quality improvement efforts. Diagnoses and management of in-hospital and ED code stroke patients were assessed retrospectively in a Joint Commission-certified primary stroke center over a 1-year period. A total of 93 in-hospital and 204 ED code strokes were identified during this period. Compared with the ED patients, the hospitalized patients were less likely to have had a stroke/transient ischemic attack (26.8% vs 51.4%; P < .0001) and less likely to have been treated with a thrombolytic agent (odds ratio, 0.27; 95% confidence interval, 0.07-0.97: P = .03). Conditions not necessitating immediate neurologic care accounted for 63.4% of in-hospital strokes, compared with 31.3% of ED code strokes (P < .0001). "Altered mental status" was the sole presenting symptom in 48% of the hospitalized patients, compared with only 10% of ED patients (P < .0001), and was the only clinical feature independently associated with a stroke mimic in the hospitalized patients (odds ratio, 63.52; 95% confidence interval, 7.37-547.69; P = .0002). There was no association between a final diagnosis of a stroke mimic and patient age, sex or race-ethnicity or nursing shift. The proportions of patients with acute ischemic stroke and patients treated with thrombolytics after activation of in-hospital code stroke were small, and were lower than those of patients with ED code stroke in the same hospital over the same time period. Developing a standardized assessment protocol for hospitalized patients with altered mental status may improve the efficacy of care. Published by Elsevier Inc.