Abstract TP370: Performance of a New “Code Stroke” Process in Hospitalized Patients in a Comprehensive Stroke Center in Minnesota

    loading  Checking for direct PDF access through Ovid

Abstract

Background and Purpose: Literature suggests that recognition and management of stroke of hospitalized patients is difficult, and the morbidity and mortality rates of in-patient strokes exceed those of out of hospital stroke. Timely treatment is an important factor for a favorable prognosis for hospitalized patients suspected of having a stroke. A new protocol for in-patient stroke was implemented as a quality improvement project at our comprehensive stroke center starting January 2017.

Methods: The new protocol included focused nursing education, replacement of 2-step activation process with 1-step process whereby the bedside nurse activates ‘Code Stroke’ using same criteria as used by EMS and ED triage nurses. Code Stroke activates a specialized stroke team (neurologist, ICU physicians, ICU nurses, pharmacy) and clears the CT scanner. Accurate documentation was encouraged and template notes were provided. Expectations were put forth regarding the relevant quality metrics. Feedback was provided in real-time as well as in writing, to participating care team. This analysis was done utilizing data from a prospectively maintained database of inpatient stroke, feedback communications, and chart review. The following metrics were used to examine the performance of the new process: rate of stroke symptoms identification, errors in paging, errors in documentation, time to CT, and outcome of code activation.

Results: In the 6 months prior to the new protocol, a ‘Code Stroke’ was activated 5 times, only one was a true stroke and was treated with thrombectomy. In the 6 months after the new protocol, ‘Code Stroke’ was activated 46 times, with 15 confirmed strokes (14 ischemic, 1 hemorrhagic). A change in care occurred for 13 patients, including IV alteplase (n=2), thrombectomy (n=1), change in medical management (n=9), and decompressive hemicraniectomy (n=1). Mean time for Code Stroke to CT was 26 minutes despite errors in pages (wrong call back number, incorrect code designation).

Conclusion: Our new process increases detection and treatment of in-patient strokes. Elements of success include system-wide organization, simplifying the process, mirroring ED process, and availability of stroke response team.

Related Topics

    loading  Loading Related Articles