Nathan Teismann1, Jersey Neilson1, Jeremy Keenan2. 1. Department of Emergency Medicine, University of California-San Francisco, San Francisco, California. 2. Department of Ophthalmology and Francis I. Proctor Foundation, University of California-San Francisco, San Francisco, California.
Abstract
BACKGROUND: Emergency physicians (EPs) frequently evaluate patients at risk for sight-threatening conditions but may have difficulty performing direct ophthalmoscopy effectively. Digital fundus photography offers a potential alternative. OBJECTIVE: We sought to assess the performance of an automated digital retinal imaging platform in a real-world emergency department. METHODS: We performed a prospective, observational study of emergency department patients who were at risk for acute, nontraumatic, posterior segment pathology. Photographs were obtained using an automated digital retinal camera and were subsequently reviewed by an ophthalmologist. We recorded the number of attempts required, total time required, patient comfort, and findings on EP-performed direct ophthalmoscopy, if performed. RESULTS: Of 123 participants completing the study, 93 (75.6%) had ≥1 eye with a diagnostically useful image, while 29 (23.6%) had no photographs of diagnostic value. The mean number of attempts required to obtain images was 1.45 (range 1-3) and the mean elapsed time required to complete photography was 109.6 s. The mean patient comfort score was 4.6 on a 5-point scale, where 5 was the most comfortable. Direct ophthalmoscopy was performed by an emergency department provider for 19 (15.4%) patients. Acute findings were noted in 14 patients during expert review of fundus photographs, though in only 2 of these cases was direct ophthalmoscopy performed by an EP with only 1 finding ultimately identified correctly. CONCLUSIONS: Automated digital imaging of the ocular fundus is rapidly performed, is well tolerated by patients, and can be used to obtain diagnostic quality images without the use of pharmacologic pupillary dilation in most emergency department patients who are at risk for acute posterior segment pathology.
BACKGROUND: Emergency physicians (EPs) frequently evaluate patients at risk for sight-threatening conditions but may have difficulty performing direct ophthalmoscopy effectively. Digital fundus photography offers a potential alternative. OBJECTIVE: We sought to assess the performance of an automated digital retinal imaging platform in a real-world emergency department. METHODS: We performed a prospective, observational study of emergency department patients who were at risk for acute, nontraumatic, posterior segment pathology. Photographs were obtained using an automated digital retinal camera and were subsequently reviewed by an ophthalmologist. We recorded the number of attempts required, total time required, patient comfort, and findings on EP-performed direct ophthalmoscopy, if performed. RESULTS: Of 123 participants completing the study, 93 (75.6%) had ≥1 eye with a diagnostically useful image, while 29 (23.6%) had no photographs of diagnostic value. The mean number of attempts required to obtain images was 1.45 (range 1-3) and the mean elapsed time required to complete photography was 109.6 s. The mean patient comfort score was 4.6 on a 5-point scale, where 5 was the most comfortable. Direct ophthalmoscopy was performed by an emergency department provider for 19 (15.4%) patients. Acute findings were noted in 14 patients during expert review of fundus photographs, though in only 2 of these cases was direct ophthalmoscopy performed by an EP with only 1 finding ultimately identified correctly. CONCLUSIONS: Automated digital imaging of the ocular fundus is rapidly performed, is well tolerated by patients, and can be used to obtain diagnostic quality images without the use of pharmacologic pupillary dilation in most emergency department patients who are at risk for acute posterior segment pathology.
Authors: George He; Hamish P Dunn; Kate E Ahmad; Eloise Watson; Andrew Henderson; Dominique Tynan; John Leaney; Andrew J White; Alex W Hewitt; Clare L Fraser Journal: Eur J Neurol Date: 2022-05-31 Impact factor: 6.288