Telemedicine for Acute Monocular Visual Loss: A Retrospective Large Telestroke Network Experience

Introduction

Telestroke is a well-validated tool to ensure around-the-clock care of acute ischemic stroke patients when limited in-person resources exist, ^1,2 and virtual neurological evaluations have become standard in emergency departments (EDs) nationally ¹ and internationally. ^{3 –5} Telestroke evaluation of patients with acute neurological deficits allows for accurate diagnosis, improves emergent triage, and increases acute reperfusion treatment in eligible patients. ^1,2

Central retinal artery occlusion (CRAO) is a subtype of ischemic stroke, ⁶ a neurovascular emergency often presenting with disabling and permanent monocular visual loss, as well as concomitant cerebral ischemic events. ^7,8 Its age- and sex-adjusted incidence is 1.9 per 100,000 person-years in the United States, ⁹ raising to 10.1/100,000 person-years in those >80 years of age in South Korea. ¹⁰ CRAO is bilateral in 1–2% of cases ¹¹ and has a nonarteritic etiology in most cases. ¹² The Balance, Eye, Face, Arm, Speech, Time (BEFAST) mnemonic was developed and popularized to include sudden visual loss (eyes) as a trigger for acute stroke evaluation, aiming to improve the sensitivity of the traditional Face, Arm, Speech, Time (FAST) prompt. ^13,14

Recently, the American Heart Association/American Stroke Association published a scientific statement supporting the use of intravenous (IV) thrombolysis for nonarteritic CRAO within 4.5 h and intra-arterial alteplase within 6 h of acute vision loss. ⁷ Uptake of clinical practice guidelines is often slow and long gaps can exist between recommended care and real-life actual care. ¹⁵ We aimed to review all telestroke encounters conducted for acute visual loss in the Mayo Clinic telestroke network, focusing specifically on the CRAO presentations and management recommendations. We hypothesized that compared with acute cerebral ischemic strokes, CRAO patients evaluated through telestroke for acute monocular visual loss are less likely to receive thrombolytic therapy.

Methods

This is a retrospective observational cohort study that was approved by the Mayo Clinic Institutional Review Board. Informed consent was waived as data were deidentified. We conducted a retrospective search of all encounters recorded in the Mayo Clinic Arizona Telestroke and Teleneurology database between 2010 and 2017 (6,364 encounters), and Mayo Clinic Enterprise Telestroke database between 2018 and 2021 (3,147 encounters; 32 spokes in 9 U.S. states). We queried the existing databases for telemedicine encounters activated for acute visual loss, by searching the consultation notes for “visual loss” OR “visual disturbance” OR “central retinal artery occlusion” OR “retinal artery occlusion” OR “eye stroke” OR “ischemic optic neuropathy” OR “blurry vision.”

We recorded the ocular diagnosis at the end of the telestroke consultation. For patients with possible or probable CRAO, we further collected the demographics, time from last seen normal to telestroke evaluation, ocular examination, the diagnostic and therapeutic recommendations. We performed a descriptive analysis of the main findings using basic summary statistics for demographic information and average quantification of ages and time to presentation. Details concerning the Mayo Clinic acute care telemedicine workflow, staffing, operations, and technologies have been previously published. ^16,17

Results

From a total of 9,511 audio–video telestroke encounters performed between January 2010 and July 2021, 49 (0.51%) were identified to have an isolated ocular complaint. Table 1 includes the diagnostic categories of visual disturbances that were identified during telestroke consultations. Out of those, five patients (0.05% of the total) had a diagnosis of possible CRAO at the end of the telestroke encounter. These encounters were conducted in 2013, 2014, 2019, 2020, and 2021, respectively (Table 2). Three were female and two were male. Age range was 48–79 years and mean age was 63 years. Two patients reported mild headache, and the others presented with isolated monocular painless visual loss. Visual acuity was not documented in two of five cases and varied between 20/30 to hand motion perception in the remaining. Visual field testing revealed no appreciable deficits in two cases, monocular scotomas in one case, and was not documented in two cases.

Fundoscopic examination was not documented nor commented on in four cases and was normal in one case according to the emergency physician's evaluation. Four CRAO patients (80%) presented within 4.5 h from visual loss' onset (range 1.5–5 h; median 2 h). Head computed tomography was normal in all cases. None received thrombolytic therapy. Antiplatelet therapy was recommended in four of five cases. All telestroke physicians recommended ophthalmology consultation for diagnostic evaluation. In one case that presented within 2 h from severe monocular visual loss, recommendations were made to transfer to a medical center where a hyperacute ophthalmologic evaluation could be performed and subsequent thrombolysis possible, should the diagnosis of nonarteritic CRAO be confirmed in a timely manner. All telestroke physicians recommended a comprehensive stroke evaluation.

Discussion

Our retrospective review of a large multispoke multistate telestroke network found that evaluations for CRAO in routine telemedicine practice were less common than the disease prevalence and the nationally reported admission rates. ¹⁸ In addition, even when patients presented to the rural ED early after visual loss, they were less likely to receive timely ophthalmologic evaluations and thrombolytic therapy. Our observations suggest that education is urgently needed in rural and underserved areas to increase awareness that acute vision loss should prompt 9–1–1 calls for emergent evaluation, as patients with acute nonarteritic CRAO may be candidates for acute reperfusion therapies. ⁷

Even when emergently recognized, this stroke subtype poses specific diagnostic and therapeutic challenges. We noted that the main shortcoming of a telestroke specialist in recommending IV thrombolysis for central retinal artery reperfusion is the lack of timely accompanying ophthalmologic evaluation to confirm the nonarteritic CRAO diagnosis. Remote vascular neurologists lack the appropriate tools to properly diagnose acute retinal ischemia and differentiate it from its mimics, even if the emergency physicians are present at the bedside. ¹⁹ Further efforts should be focused on developing partnerships between telestroke physicians and eye care providers to assure around-the-clock coverage for acute ocular ischemic events. ¹⁹

To our knowledge, this is the first report of the rates of CRAO identification and acute reperfusion therapy utilization in a multicentric telestroke network. A large study based on the National Readmissions Database in the United States reported 2,163 admissions for CRAO between 2013 and 2015. Compared with acute ischemic stroke, CRAO admissions were more often at larger teaching metropolitan hospitals, and had significantly lower rates of thrombolysis (2.9% vs. 8%, p < 0.0001). ¹⁸ We have previously reported rates of IV thrombolysis as high as 40% in our Mayo Clinic telestroke network for acute cerebral ischemic strokes. ^17,20

Formalized protocols for CRAO hyperacute evaluation and subsequent management are not available in most institutions, even though the American Academy of Ophthalmology-endorsed preferred practices highlight that acute symptomatic CRAO should prompt an immediate referral to the nearest stroke center for swift assessment and consideration of acute intervention. ²¹ A 2018 national U.S. survey of academic medical centers found that only 20% of respondents had a formal multidisciplinary protocol in place for CRAO. ²²

A tertiary academic institution affiliated with a comprehensive stroke center reported that patients with acute vision loss are often not evaluated quickly enough by eye care providers resulting in a delay in the diagnosis in the ED, ²³ leading to only 3 out of 181 patients presenting to their institution receiving IV thrombolysis, and none receiving intra-arterial thrombolysis. Implementation of an in-hospital CRAO management guideline led to an increase in thrombolysis use, ²⁴ and the main reason for lack of timely therapies was a delay in prehospital presentation, hence emphasizing the need to increase public awareness that sudden vision loss may herald stroke.

Our report is limited by the lack of immediate follow-up data regarding the ophthalmologic evaluation timeline in the respective community medical center spoke or the accepting hospital (in case a transfer was considered), the definitive ocular diagnosis, treatment plan, and visual outcomes. However, it underscores that telestroke evaluations of patients with visual loss often lack the necessary means to confirm the diagnosis of nonarteritic CRAO, despite acute visual loss mandating stroke-like emergent evaluation and management. The recent COVID-19 pandemic led to an unprecedented growth in telemedicine utilization ² and teleophthalmology resources have become available for practicing eye care providers to help with remote and urgent evaluations in community settings. ^25,26

Similar to teleneurology networks' utility for the rapid and effective evaluation of patients in remote locations requiring neurological care, ²⁷ teleophthalmology carries the promise of having the same clinical outcome as in-person consultations and at least the same diagnostic accuracy. ^{28 –30} In remote areas in Brazil, an average teleophthalmology consultation lasted 8.6 min per patient, and the urgent ocular diagnosis had 81.94% specificity and 92.85% sensitivity compared with the in-hospital ophthalmologist. ³¹ Real-time combined teleophthalmology and telestroke networks could be developed to provide emergent care in remote areas that lack on-site ophthalmologists to care for patients with acute visual complaints. ¹⁹

This collaboration will likely reduce costs of care by preventing unnecessary transfers conducted merely for an ophthalmologic evaluation. The telestroke neurologist can safely recommend acute reperfusion therapies only after the etiological diagnosis of the sudden visual loss is confirmed. Employing ancillary but straightforward ophthalmic testing in the ED, such as nonmydriatic retinal photography, optical coherence tomography and transorbital ultrasonography might help the remote physician (either vascular neurologist or ophthalmologist) to establish a correct and definitive diagnosis in a timely manner. ³²

Conclusion

In conclusion, current assessment of acute visual loss through telestroke is suboptimal. The ongoing efforts of educating the public, ophthalmic providers, and emergency medical services that acute vision loss should be evaluated emergently in specialized stroke centers must be accompanied by redesigning the networks for hyperacute ophthalmologic assessments, especially in remote zones. 24/7 in-person or teleophthalmologic evaluations and advanced ophthalmic equipment in the ED are needed to establish the diagnosis of nonarteritic CRAO and to support the subsequent indication for acute eye stroke reperfusion therapies.