Every minute blinds: Temporal artery ultrasound in the real-time diagnosis of stroke of the eye

Objective: To evaluate a) whether failure to consider ocular vascular ischemia in patients ≥50 years of age presenting with acute, isolated ocular symptoms leads to missed diagnoses of ‘stroke of the eye’ caused by giant cell arteritis (GCA) and b) to assess the role of temporal artery ultrasound (TAUS) in the diagnosis of GCA in the emergency department (ED). Methods: A retrospective cohort analysis using data from patients aged ≥50 years presenting to the emergency department (ED) with isolated ocular symptoms between 2021 and 2024 was done. Datasets included the University of Kentucky (UKY) Epic (Electronic Health Records EHR) data and Epic Cosmos, a nationwide healthcare data analytics platform developed by Epic Systems, a provider of EHR software (299 million patients nationwide). Patients were stratified into occult GCA, central retinal artery occlusion (CRAO), and ischemic optic neuropathy (ION). The full UKY cohort (n = 787) included patients presenting with occult GCA, CRAO and ION. Multivariable logistic regression was performed using standard statistical software to estimate predictors of GCA diagnosis, adjusting for TAUS, temporal artery biopsy (TAB) and corticosteroid administration prior to a diagnosis. Results: In the UKY cohort of 103 patients with acute and isolated ocular symptoms (occult GCA), 38.8 % had confirmed GCA, 27.2 % were clinically diagnosed, and 34 % were ruled out. Of the confirmed cases, 67.5% underwent TAUS and 17.5% underwent TAB. A total of 66/103 (64 %) received steroids before testing and 46/66 (69.7 %) tested negative for GCA. In the Epic Cosmos data, 62.8 % of 17,372 patients lacked any testing; TAUS-specific data were unavailable. Among patients with CRAO (n = 90) in the UKY cohort, 12.2 % underwent TAUS, with 63.6 % diagnosed with GCA; workup for stroke revealed an embolic stroke 22/90 (24.4 %); 75.5 % received no further workup after stroke evaluation was unrevealing and 1/90 underwent TAB that was negative; with the patient having steroids prior. In Epic Cosmos, of 18,621 patients with CRAO, 561 underwent TAB (3 %), and 5908 (31 %) had ultrasound findings but TAUS-specific data was unavailable. Among 594 UKY patients with ION, 28 had carotid ultrasound or TAUS; 15 had TAUS, with 33 % diagnosed as having GCA; 9 underwent TAB and 33 % had GCA. Additionally, 94.1 % of the UKY cohort had no testing for GCA; In Epic Cosmos, of the 56,554 patients with ION, 2686 (4.7 %) underwent carotid U/S or TAUS, or both. 901(1.6 %) underwent TAB, while 52,967/56554 (93.6 %) had no formal testing for GCA. Conclusions: Recognition of GCA with acute ocular symptoms in the ED remains challenging. Neurologists and ophthalmologists play a central role in the diagnostic workup but delays often occur because no real-time diagnostic tool exists. Integrating TAUS into ED protocols for ocular ischemia may reduce diagnostic delays.