Evaluation of Nonfatal Strangulation in Alert Adults

There is a paucity of evidence to guide the diagnostic evaluation of emergency department (ED) patients presenting after nonfatal strangulation (manual strangulation or near hanging). We seek to define the rate of serious injuries in alert strangled patients and determine which symptoms and examination findings, if any, predict such injuries. Using prospectively populated databases and electronic medical record review, we performed a retrospective analysis of alert strangled patients treated in the ED of an academic Level I trauma center. Exclusions were Glasgow Coma Scale (GCS) score less than 13, younger than 16 years, and interhospital transfers. Trained researchers used structured forms to abstract demographics, symptoms, examination findings, radiology and operative findings, and final diagnoses. Injuries requiring greater than 24 hours’ observation or specific treatment (surgery, procedure, specific medication) were considered clinically important. The electronic medical record was searched for 30 days after presentation to identify missed injuries. Advanced imaging (computed tomography or magnetic resonance maging) was obtained in 60%. Injuries were identified in 6 patients (1.7%, 95% CI, 0.7% to 3.6%). Two injuries were clinically important (0.6%, 95% CI, 0.1% to 2.0%). Both were cervical artery dissections with no neurologic deficits, treated with aspirin. No additional injuries were identified within 30 days or at next medical contact. Of 343 uninjured patients, 291 (85%) had documented medical follow up confirming the absence of any new diagnosis of injury or stroke. The small number of injuries precluded analyses of associations. Alert, strangled patients had a low rate of injuries. All patients with neck injuries had concerning findings besides neck pain; specifically, GCS score less than 15 or dysphagia. Our findings suggest, but do not prove, that a selective imaging strategy is safe in alert patients after strangulation findings besides neck pain.