Abstract 128: Yield Of Computed Tomography After Non-Traumatic Out-of-Hospital Cardiac Arrest

Abstract

Background Computed tomography (CT) can identify the etiology of cardiac arrest and injuries related cardiopulmonary resuscitation (CPR). Diagnostic yield of CT has only been characterized in a few small cohort studies with varied imaging practices. Purpose To determine diagnostic yield of CT imaging after out-of-hospital cardiac arrest (OHCA). Methods We included non-traumatic OHCA patients treated at a single center February 2019 to February 2021. Our practice during the study period was to obtain unenhanced head CT in all comatose patients and CT of the cervical spine, chest, abdomen, and pelvis in most cases. We abstracted patient age and sex, CPR duration, and initial neurological examination from our prospective registry. We reviewed clinical records and identified CT imaging obtained within 24 hours of arrival. We reviewed radiology reports, and summarized findings for each body region. Results We included 505 subjects. Mean age was 60 [SD 16] years and 144 (29%) had an initial shockable rhythm. Median CPR duration was 21 [IQR 12-35] minutes and 448 (89%) subjects were comatose on presentation. Almost all subjects had brain imaging (n=481, 95%), of which 33 (7% [95% CI 5-9%]) had intracranial hemorrhage and 157 (33% [95% CI 28-37%]) had cerebral edema. Fewer subjects had a cervical spine CT (n=202, 40%) though 4 (2% [95% CI 0-4%]) had vertebral fractures. Most subjects had a CT of the chest (n=396, 76%), and abdomen and pelvis (n= 347, 69%). Identified chest pathologies included rib or sternal fractures (222, 56% [95% CI 51-61%]), pneumothorax (27, 7% [95% CI 5-10%]), aspiration or pneumonia (300, 76% [95% CI 71-80%]), mediastinal hematoma (18, 5% [95% CI 3-7%]), and pulmonary embolism (13, 3% [95% CI 2-6%]). Significant abdomen and pelvis findings were bowel ischemia (23, 7% [95% CI 4-10%]), and splenic or liver laceration (7, 2% [95% CI 1-4%]). Conclusions Cross sectional imaging identifies clinically important pathology that informs early management.