Improving the diagnostic yield of echocardiography in acute ischemic stroke: A quality improvement project in a community hospital in Maryland, USA

BACKGROUND: Echocardiography is routinely ordered in acute ischemic stroke workup. No standardized or structured criteria is used to select or exclude echocardiography in such settings. Moreover, the diagnostic yield of echocardiography in stroke is low in our medical center. This article presents newly proposed selection criteria for echocardiography in ischemic stroke workup. METHODS: A quality improvement project was implemented in a 385-bed community hospital in Maryland, USA. A computerized decision support tool consisting of new criteria for selecting echocardiography in ischemic stroke workup was created. 639 patients hospitalized with ischemic stroke were followed-up over 12 months after intervention, and 686 matched-controls with ischemic stroke were retrospectively analyzed from the 12 months prior to intervention. Cost-effectiveness and diagnostic yield of echocardiography in ischemic stroke were measured before and after intervention. RESULTS: Following intervention, the diagnostic yield of echocardiography in ischemic stroke workup significantly increased by 51% (from 3.8% to 7.8%, odds ratio (OR) 2.1, P= 0.01). The number of echocardiography studies needed to detect and treat one patient with a cardiac source of embolism was reduced from 50 to 25 studies. The overall use of echocardiography in stroke workup significantly decreased (OR 0.4, p < 0.001). Patients with lacunar infarcts or atrial fibrillation had significant reduction in echocardiography (OR 0.2, p < 0.001 and OR 0.4, p < 0.001, respectively). CONCLUSION: The new criteria for echocardiography selection in hospitalized patients with ischemic stroke significantly improved the cost-effectiveness and the diagnostic yield of echocardiography and reduced unnecessary echocardiography in lacunar infarcts or atrial fibrillation.