Madhuri Akella, Étienne Bénard-Séguin, Andrew M Pendley, Ghada Mohamed, Jessica G McHenry, Daniel Adamkiewicz, Andrew Trippiedi, Lennox Xu, Stuart Duffield, Alexis Flowers, Wesley Chan, Matthew Keadey, David W Wright, Fadi Nahab, Nancy J Newman, Valérie Biousse
Despite our Eye-Stroke protocol, facilitated by NFMP-OCT in our ED, only 16% of acute CRAO/BRAO eyes were diagnosed early enough to be considered for intravenous (IV) thrombolysis, which was administered to only 43% of those eligible. Rapid workup found a major cause of CRAO/BRAO in 72%, and 18% had concurrent cerebral infarctions on MRI. The main barrier to delayed diagnosis/care was transfer from other institutions/providers, suggesting that wide deployment of NMFP-OCT for remote diagnosis and treatment via existing telestroke networks is optimal for reducing time to diagnosis, avoiding transfers, and improving patient outcomes.
INTRODUCTION: The diagnosis of acute central retinal artery occlusion (CRAO) and branch retinal artery occlusion (BRAO) is often delayed or missed in emergency departments (EDs) because of limited ocular funduscopic skills and lack of immediate ophthalmology access. We evaluated the clinical impact of our Eye-Stroke protocol using non-mydriatic ocular imaging (non-mydriatic fundus photography and optical coherence tomography [NMFP-OCT]) in our general ED with remote interpretation by ophthalmology on the time to diagnosis and emergent management of acute CRAO/BRAO.
METHODS: Prospective consecutive series of 100 acute CRAO/BRAOs seen within 1 week of vision loss between June 2023 and April 2026 with NMFP-OCT in our ED.
RESULTS: Among 100 CRAO/BRAO eyes seen within 1 week of onset, 16 (16%) had vision loss within 4.5 h, 49 (49%) between 4.5 and 24 h, 35 (35%) between 24 h and 1 week (median times from presentation to NMFP-OCT 28.5 min [IQR, 17.75-57.5 min; range, 10-330 min], 96 min [IQR, 51-180 min; range; 9-317 min], and 134 min [IQR, 96-187 min; range, 30-337 min], respectively). The diagnosis of CRAO/BRAO was made from color photographs and OCT in 77/100 eyes, from OCT only in 19/100 eyes, and 3/100 eyes had uninterpretable imaging. 7/16 eyes presenting within 4.5 h of vision loss received intravenous thrombolysis (median door-to-needle time, 58 min [IQR, 46-78.5 min; range, 41-181 min]). The majority of the patients evaluated within 4.5 h were self-referred to our ED, whereas the majority of the patients presenting later were referred by outside providers or transferred from other EDs. Stroke workup found a major cause of CRAO/BRAO in 72%, and 17/94 (18%) had concurrent cerebral infarctions on brain magnetic resonance imaging (MRI).
CONCLUSION: Despite our Eye-Stroke protocol, facilitated by NFMP-OCT in our ED, only 16% of acute CRAO/BRAO eyes were diagnosed early enough to be considered for intravenous (IV) thrombolysis, which was administered to only 43% of those eligible. Rapid workup found a major cause of CRAO/BRAO in 72%, and 18% had concurrent cerebral infarctions on MRI. The main barrier to delayed diagnosis/care was transfer from other institutions/providers, suggesting that wide deployment of NMFP-OCT for remote diagnosis and treatment via existing telestroke networks is optimal for reducing time to diagnosis, avoiding transfers, and improving patient outcomes.