Background: Up to 17% of all strokes occur in patients hospitalized for another diagnosis or procedure, and in-hospital strokes complicate up to 0.06% of all admissions. In-hospital strokes carry higher mortality, longer length of stay and greater disability than community-onset strokes. Multiple factors contribute to the worse outcomes of in-hospital stroke. Prompt recognition and treatment of stroke is critical, and the rapid identification of stroke in the hospital is challenging. First, there is an increased likelihood of conditions that mimic stroke, such as delirium. Further, physicians may not be physically available on the hospital unit at all times. Procedural personnel and equipment, including the computed tomography (CT) scanner may be at some distance from the hospital unit, and transportation of hospital patients can be difficult. In the American Heart Association/American Stroke Association national database, time to treatment for in-hospital strokes averaged 100 minutes, with only 1 in 5 patients achieving the goal of 60 minutes from symptoms recognition to treatment.
Purpose: To develop an interdisciplinary nurse driven protocol for rapid diagnosis of in-hospital strokes.
Description: We held a series of process improvement events in effort to develop and implement the Code Stroke protocol for hospitalized patients. First, in 2012, we created a standardized order set and mutually agreed upon responsibilities of various providers during the code stroke response. All in-patient nurses received training on recognition of stroke, specifically facial droop, arm/leg weakness and speech changes, and time of onset, referred to as the FAS symptoms. In the event of a positive FAS finding, the bedside nurse calls a Rapid Response. Subsequently, a critical care nurse evaluates the patient and escalates to an in-house Code Stroke if FAS criteria are confirmed. The Rapid Response Team existed prior to this intervention, but escalation to Code Stroke was a new innovation. Specifically, the initiation of Code Stroke prompts pages to designated practitioners in neurology, diagnostic radiology, interventional radiology, hospital medicine, nursing, and pharmacy. Subsequently, in 2016, the critical care nurses were granted the autonomy to order the non-contrast head computed tomography (CT) and CT angiogram for a Code Stroke patient, contact the neurologist directly, order blood work, and transport the patient to the CT scanner without direct physician involvement. With each of these events, we have noticed significant improvements in the lead time from recognition of stroke symptoms to CT imaging.
Conclusions: A nurse initiated, code stroke protocol that allows for order entry and testing without a physician is an effective way to improve recognition and time to diagnosis of stroke. This approach may be useful in other care settings or patient populations. If successfully implemented, this will decrease time to treatment, and potentially decrease associated complications of stroke.