Evaluation of Transient Ischemic Attack and Minor Stroke: A Rapid Outpatient Model for the COVID-19 Pandemic and Beyond

Abstract

The grim circumstances of the COVID-19 pandemic have highlighted the need to refine and adapt stroke systems of care. Patients’ care-seeking behaviors have changed due to perceived risks of in-hospital treatment during the pandemic. In response to these challenges, we optimized a recently implemented, novel outpatient approach for the evaluation and management of minor stroke and transient ischemic attack, entitled RESCUE-TIA. This modified approach incorporated telemedicine visits and remote testing, and proved valuable during the pandemic. In this review article, we provide the evidence-based rationale for our approach, describe its operationalization, and provide data from our initial experience.

Introduction

The 2019 coronavirus (COVID-19) pandemic has profoundly impacted healthcare. Early anecdotes about reductions in emergency department visits for non-COVID-19 related acute complaints such as stroke were soon confirmed by several large studies showing declines in acute stroke evaluations and delayed stroke presentations.^1-3 The observed delay in seeking care signals a need for healthcare systems to meet patients’ new demands and expectations. Fear of infection with SARS-CoV-2 at hospitals and doctor’s offices likely played a major role in patients delaying care; however, late presentation for acute cerebrovascular disease is a problem that predates the COVID-19 pandemic. There is therefore a need for a critical reexamination of stroke systems of care, with a focus on timely and lasting innovation in the outpatient setting for patients with stroke and transient ischemic attack (TIA). At our institution, a novel outpatient care pathway for the Rapid Evaluation of minor Stroke and CerebrovascUlar Events including TIA (RESCUE-TIA) was devised before the COVID-19 pandemic to provide increased access to care for TIA and minor stroke patients. This model’s unique pre-existing infrastructure combined with a transition to a completely telemedicine-based care pathway allowed for an efficient and optimized response to the challenges of the COVID-19 pandemic. In this review, our objectives are to provide an evidence-based rationale for our model, describe its operationalization, and to provide data from our initial experience with the clinic.

Background

Rapid evaluation and management of patients with TIA has been a priority since the high risk of stroke and cardiovascular events after TIA was identified. ⁴ Emergency department-based TIA risk stratification and triaging literature abounds. ⁵ Outpatient care models for TIA and minor stroke in the United States have largely served patients who have already been evaluated in an emergency department, with an aim of reducing hospital admissions. For example, the RAVEN (Rapid Access Vascular Evaluation-Neurology) clinic helps to reduce hospital admissions and costs by facilitating TIA and stroke evaluation in the outpatient setting within 24 hours after discharge from the emergency department.^6,7 Other examples include the TWO-ACES (TIA Work-up as Outpatient Assessment of Clinical Evaluation and Safety) and TIA-TEAM (TIA Triage in Emergency Department using Acute MRI) approaches which utilize similar post-discharge strategies described in detail elsewhere.^8,9

Outpatient models for the evaluation of TIA and minor stroke without initial emergency department triage are largely based on studies from Europe. The EXPRESS (Effect of Urgent Treatment of Transient Ischemic Attack and Minor Stroke On Early Recurrent Stroke) study assessed TIA patients the same afternoon after initial presentation to their primary care physicians and the SOS-TIA (A Transient Ischemic Attack Clinic with Round-The-Clock Access) study directly admitted TIA patients from primary care offices to a short stay hospital-based clinic for “round the clock” TIA evaluation.^10,11 These outpatient programs were associated with low 90-day stroke rates (2.1% and 1.2%, respectively) and established the feasibility and safety of outpatient evaluation of TIA and minor stroke.^10,11 Other countries have now replicated this type of care pathway.^12,13

The safety of outpatient management of TIA and minor stroke established by the EXPRESS and SOS-TIA studies, along with the greater availability of magnetic resonance imaging (MRI) and magnetic resonance angiography (MRA), led us to develop an outpatient TIA and minor stroke clinic that utilizes MRI a great deal more than other models. The RESCUE-TIA (Rapid Evaluation of minor Stroke and CerebrovascUlar Events including TIA) clinic is comprised of an outpatient evaluation by a stroke neurologist paired with rapid (and usually MRI-based) comprehensive outpatient testing without the need for emergency department attendance. Recognizing the utility of our model during the COVID-19 pandemic, we transitioned to a telemedicine-based care pathway given the increased acceptance of telemedicine for stroke and the pressing need for innovative care pathways to reach patients who may have been delaying care during the pandemic. There are limited data describing this type of completely outpatient, telemedicine-based TIA clinic in the United States. In this review, we present our protocol, review the relevant contemporary clinical science and practical considerations for the clinic, and describe our initial experience with the clinic.

Clinic Overview

The RESCUE-TIA clinic provides rapid evaluation and management for patients with transient neurological symptoms consistent with TIA or with minor, non-disabling fixed deficits. The patient or their provider calls the clinic and a trained nurse performs a strict, checklist-guided telephone triage based on the clinic’s inclusion and exclusion criteria. Any patient that does not meet the inclusion criteria or meets an exclusion criteria is referred to the emergency department. Eligible patients are scheduled to see a stroke neurologist or supervised stroke physician assistant in-person or by telemedicine within 24 hours, typically within the same day. The choice of in-office visit or telemedicine visit is largely dictated by COVID-19 prevalence in the surrounding area and by patient choice. After the clinician assessment, the patient has their blood drawn, undergoes an electrocardiogram, and is scheduled for same-day parenchymal and extracranial and intracranial vascular imaging, usually by MRI and MRA. Echocardiography, prolonged Holter monitoring, and other ancillary tests which might be required are arranged to be done in the oncoming few days. There is a low threshold to refer patients to the emergency department at multiple steps in the clinical pathway. Otherwise, results are reviewed promptly, and appropriate secondary prevention and follow-up are implemented.

Resources

A variety of organizational, personnel, and material resources are needed to establish this type of comprehensive clinic (Figure 1). Buy-in from department leadership is of the utmost importance for financial and logistical support. An evidence-based approach to structuring the clinic, as presented here, aids in securing support among key leadership stakeholders who make decisions with limited resources. Close collaboration with the radiology department, cardiology department, and laboratory ensure rapid imaging and laboratory tests are completed reliably and on time. A significant amount of investment into personnel is required as well. Training of nurses in stroke-specific clinical reasoning is essential to ensure safe telephone-based triage and clinic assistants are vital for aiding in insurance authorization for imaging. Where stroke neurologists are unavailable, partnerships with telestroke programs could be utilized. A quality improvement database is required to monitor for deviations from the clinical pathway, safety concerns, and opportunities for optimization. The database should include time metrics to ensure that diagnostic studies are completed reliably and consistently in a timely manner.

Figure 1.

Resources to establish a cerebrovascular evaluation clinic.

Patient Selection Criteria

The RESCUE-TIA clinic serves patients based on strict criteria designed to ensure patient safety and adherence to practice guidelines while minimizing the inclusion of patients with cerebrovascular mimics (Figure 2).

Figure 2.

Inclusion and exclusion criteria for RESCUE-TIA clinic.

Nature of Stroke Deficits

Evaluation in the RESCUE-TIA clinic is permitted if symptoms are resolved or are fixed but minor and non-disabling. In patients with fixed deficits, the deficits must not interfere with activities of daily living or swallowing because organized post-stroke rehabilitation in patients with impaired functioning improves outcomes and early dysphagia screening is associated with reduced risk of post-stroke pneumonia.^14,15 The telephone triage screens specifically for stroke deficits that may result in functional impairment, even if the patient does not endorse loss of independence. The presence of any such disabling symptoms results in triage to the emergency department. At the time of telephone triage, patients are advised that any concern for neurologic deterioration, including fluctuation in symptoms, should prompt immediate care seeking as these patients may be eligible for acute interventions. In-hospital evaluation can address home safety, rehabilitation services, and dysphagia evaluation and treatment.

Time From Symptom Onset

A key criterion, applied to both patients with TIA and minor stroke, is that the symptoms must have begun greater than 24 hours prior to first contact. The reasons for this are different for TIA and minor stroke and warrant separate consideration. The risk of ischemic stroke after a TIA is highest in patients who harbor large artery atherosclerotic lesions, and this risk is highest during the first 24 hours.^16,17 While there is precedent for same-day outpatient evaluation and management of TIA based on the EXPRESS and SOS-TIA studies, we specified this 24-hour window conservatively for patients with TIA to ensure timely access to acute reperfusion therapies if early stroke does occur within 24 hours. Second, we specified this 24-hour window because the POINT and CHANCE trials demonstrated benefit of acute initiation of dual antiplatelet therapy within 12 hours and 24 hours, respectively, for the secondary prevention of stroke after high risk TIA and minor ischemic stroke.^18,19 Third, this 24-hour window was chosen to be parsimoniously consistent with our criteria for patients with minor, non-disabling stroke deficits.

Apart from being an opportunity to initiate dual antiplatelet therapy for secondary prevention, the first 24-hour window after onset of minor stroke symptoms warrants direct presentation to an emergency department or activation of emergency medical services for several reasons. The first reason is that neurological fluctuation and deterioration are common in the first 24-hour period after symptom onset. While patients with minor stroke are less at risk of worsening compared to patients with severe strokes, approximately 25% in the placebo group of the NINDS tPA trial had fluctuation in this time period.^20,21 However, the likelihood of deterioration in TIA and minor stroke is low beyond the 24 hour time window. ²² Thus, a patient with resolved or stable symptoms for 24 hours at time of first contact can be evaluated as an outpatient. Second, patients presenting in the first 24 hours after symptom onset may be eligible for acute reperfusion therapies. Although patients with minor stroke are at less risk for worsening, a patient with initially non-disabling deficits may develop disabling deficits in a timeframe suitable for acute treatment given thrombolysis is recommended for patients with disabling but minor stroke. ²³ Additionally, neurologic deterioration is more likely to occur in the first 24 hours and is associated with the presence of large vessel occlusions. ²⁰ Current guidelines do not recommend endovascular treatment for large vessel occlusion associated with minor stroke. However, there is emerging data that rescue thrombectomy may be beneficial for patients who worsen within 24 hours. ²⁴ Lastly, patients presenting within 24 hours may be eligible for thrombolysis even after 4.5 hours based on a tissue-based approach to treatment, such as the approach used in the WAKE-UP (Efficacy and Safety of MRI-Based Thrombolysis in Wake-Up Stroke) trial. ²⁵ For these reasons, it is important that patients with new symptoms, or symptoms that began in the last 24 hours, be encouraged to activate emergency medical services or go to an emergency department, and that outpatient clinic messaging not counter such public health campaigns. Our clinic seeks to only serve patients with non-disabling, minor stroke deficits after the 24-hour time period as this patient population does not clearly derive benefit from thrombolysis and is at less risk for deterioration. ²⁶

Exclusion of Mimics and High-Risk Conditions

The triage criteria aim to limit the number of stroke mimics evaluated in the clinic. Seizure and toxic metabolic encephalopathy are among common stoke mimics and can be provoked by a large array of underlying medical issues, some of which require emergency department-based care. ²⁷ Therefore, patients with a history of seizures, seizure-like activity at onset of symptoms, or confusion are referred to the emergency department for evaluation. Patients’ symptoms must not be accompanied by persistent or severe headache, as these symptoms may suggest diagnoses of intracranial hemorrhage, subarachnoid hemorrhage, reversible cerebral vasoconstrictive syndrome, cerebral venous sinus thrombosis, and meningoencephalitis. Patients with fever are also referred to the emergency department out of concern for meningoencephalitis, rarer causes of stroke such as vasculitis or endocarditis that portend a poor prognosis if not evaluated promptly, and considering acute infection is associated with neurologic deterioration in TIA and minor stroke patients.^28,29 Patients with recent head trauma are excluded as acute neurological deficits in these patients are more likely to reflect the presence of a traumatic intracranial hemorrhage, such as a subdural hematoma, that may require immediate neurosurgical attention. Lastly, residents of nursing homes and rehabilitation facilities are referred to the emergency department given the high rate of stroke mimics in this population. ³⁰

The triage criteria also seek to appropriately triage high-risk patients who may require multidisciplinary acute care. Any patient with cardiac symptoms is referred for emergent evaluation given the time-sensitive nature of acute coronary syndrome treatment. Pregnant or possibly pregnant patients are referred to the emergency department because ischemic stroke during pregnancy is associated with hypertensive disorders of pregnancy that may need more urgent, multidisciplinary treatment. ³¹ Patients undergoing active chemotherapy or immunotherapy and those with HIV/AIDS are referred to the emergency department in light of the higher probability of occult infections in these patients, including line-associated bacterial endocarditis, as an example.^32,33

Clinic Workflow

Triage

An overview of the RESCUE-TIA clinic workflow is provided in Figure 3. A dedicated telephone line for the RESCUE-TIA clinic is provided to providers, who may then call on behalf of the patient or give the phone number to the patient. A trained, registered nurse performs checklist-based phone triage based on the criteria in Figure 2. A stroke neurologist is available for consultation if the triage nurse deems there is ambiguity. Any patient that does not meet the eligibility criteria is referred immediately to the emergency department. Eligible patients are scheduled for an appointment on the same day. Patients who call late in the day and cannot be seen until the following day are triaged in consultation with the stroke neurologist to either be seen the next day in the clinic or referred to the emergency department. At the end of the telephone call, the patient is instructed to seek emergent care by calling 911 for any new or worsening symptoms that occur before their visit.

Figure 3.

RESCUE-TIA clinic workflow. Abbreviations: ECG, electrocardiogram; ED, emergency department; MRI, magnetic resonance imaging; PA, physician assistant.

Time to Clinic Evaluation

A key component of the RESCUE-TIA clinic is timely neurologic evaluation either same-day or within 24 hours of the patient seeking care. The risk of recurrent stroke is elevated as long as 90 days after TIA, with the highest risk being within the first 48 hours. ⁴ Thus, evaluating patients with minor stroke and TIA as soon as possible, even after the first 24 hours after symptom onset, is important to implement timely secondary prevention measures. For example, the EXPRESS study showed that rapid access (1 vs 3 days after TIA) was associated with a significantly lower 90-day risk of recurrent stroke. ¹⁰ These data support the need for more urgent neurologic evaluation and justify the resources needed to establish a rapid access clinic for patients to be evaluated same-day or within 24 hours of initial contact rather than in a more delayed fashion.

Diagnostic Evaluation

A stroke neurologist, or a trained and supervised physician assistant, evaluates the patient either in office or through telemedicine. The use of telemedicine evaluation is established for stroke evaluation; NIHSS assessment is reliable with telemedicine and stroke mimics can be reliably identified.^34,35 After provider evaluation, diagnostic tests are ordered as indicated (Figure 4). Standard laboratory tests including a lipid panel and a hemoglobin A1c test are sent the same day. Other lab tests such as a hypercoagulable panel are included if clinically indicated. Same-day electrocardiogram and troponin are performed given the high prevalence of cardiac findings, and their prognostic significance, in patients with stroke and TIA.^36,37 Same-day electrocardiography also allows for detection of atrial fibrillation and timely initiation of anticoagulation therapy. Transthoracic echocardiography is performed within 14 days of the visit; the longer time-frame is permitted because results rarely change management in the acute setting. ³⁸ Transesophageal echocardiography or advanced imaging are scheduled if indicated after initial non-invasive evaluation. In patients diagnosed with stroke or TIA with no clear mechanism, prolonged cardiac rhythm monitoring for 4 weeks is initiated.

Figure 4.

Diagnostic tests performed at RESCUE-TIA clinic. Abbreviations: CT, computed tomography; MRI, magnetic resonance imaging; MRA, magnetic resonance imaging. Symbols—*: tests only ordered if deemed clinically indicated by treating physician.

The clinic utilizes rapid MRI and MRA same-day or within 24 hours of the appointment for brain and vessel imaging. Many TIA clinics in the United States assess patients in the clinic only after initial brain imaging is performed in the emergency department to rule out dangerous intracranial pathology.^6,8 The SOS-TIA hospital-based clinic utilized timely MRI with limited sequences; however, only 27% of patients received diffusion-weighted imaging (DWI) within 15 days of evaluation. ¹¹ Rapid MRI performed in the RESCUE-TIA clinic allows for timely detection of dangerous intracranial pathologies. Further, MRI with DWI is more sensitive than computed tomography (CT) for identifying acute ischemic stroke, and MRI is equally sensitive for hemorrhagic stroke. ³⁹ Patients quickly identified with dangerous intracranial pathologies can be triaged quickly for urgent evaluation, and accurately determining infarct size can inform timing of anticoagulation when indicated.

MRA provides advantages compared to other vessel imaging modalities commonly used in TIA clinics, such as carotid duplex ultrasound. There is significant interfacility and interobserver variability for detection of carotid stenosis with carotid duplex ultrasound, and there is evidence it is less reliable for identification of stenosis of 50-70%, a patient population shown to benefit modestly from carotid revascularization.^40-43 MRA is less operator-dependent and therefore can be reliably used for detection of carotid stenosis. Further, for patients with TIA, the use of MR-based imaging facilitates better risk stratification; the ABCD ³ -I score uses DWI findings and carotid artery assessment to improve risk stratification compared to other risk scores. ⁴⁴

Timely identification of patients with symptomatic carotid stenosis with rapid MRA is vital for 2 reasons. First, early identification of this high-risk stroke mechanism is essential in TIA patients given patients with significant carotid stenosis have high stroke risk after TIA irrespective of the ABCD ² score. ⁴⁵ Second, data show the greatest benefit of revascularization when performed within 2 weeks of the initial event. ⁴⁶ Neurointerventional consultation can be quickly pursued once symptomatic carotid stenosis is identified. Additionally, unlike ultrasound-based techniques, MRA allows for rapid, reliable identification of intracranial large vessel occlusion. A minority of patients with minor stroke (NIHSS 0-5) harbor large vessel occlusions ²⁴ and are at increased risk for neurologic worsening. ²⁰ Identification of a large vessel occlusion can prompt direct admission to the stroke service for monitoring because retrospective data shows improved outcomes with thrombectomy in patients with low NIHSS who subsequently worsen. ²⁴ MRA also provides the opportunity to detect intracranial atherosclerosis, the presence of which changes secondary stroke prevention strategies and calls for dual-antiplatelet therapy. ⁴⁷ CT angiography or ultrasound are used if MRI is contraindicated.

Disposition and Secondary Prevention

Disposition is decided after reviewing the same-day results by the stroke neurologist. Patients with symptomatic carotid stenosis or other urgent intracranial pathology can be directly admitted to the stroke service, bypassing the emergency department if appropriate. Patients with fluctuating or disabling deficits, lab abnormalities, or high-risk cardiopulmonary conditions are referred for emergency department evaluation. Otherwise patients are discharged home after same-day testing is complete. Follow-up with a vascular neurologist is scheduled in 2-4 weeks. If a non-neurological condition is diagnosed, the patient is referred to their primary care physician or other care as appropriate.

Personalized secondary stroke prevention therapies are immediately started after results are reviewed, with attention to optimizing antithrombotic regimens, statin treatment, and blood pressure control. Medications are electronically prescribed to patients’ pharmacies for same-day pickup. Clinic assistants ensure timely communication with physicians in the event of medications prescription issues. Medication review is performed at the 2-4 week follow-up appointment to ensure medication adherence. Additionally, the clinic focuses on stroke education and behavioral interventions when appropriate. Review of stroke pathophysiology, individualized risk factor management, and anticipatory guidance is provided for all patients. The COMPASS (COMprehensive Post-Acute Stroke Services) Study and American Heart Association Life’s Simple 7 websites are utilized for patient education materials and risk factor counseling. Nicotine replacement therapies and referral to smoking cessation programs are encouraged for eligible patients.

Response to the COVID-19 Pandemic

The infrastructure of the RESCUE-TIA clinic proved valuable during the COVID-19 pandemic. It was recognized that patients were delaying care and likely avoiding the hospital out of fear of infection with SARS-CoV-2. The RESCUE-TIA clinic’s infrastructure provided a safe outpatient alternative for eligible patients hesitant to seek traditional care at the hospital. Not only did the infrastructure allow for all tests to be completed in the outpatient setting, but in March of 2020, the format was transitioned to telemedicine to enhance safety concerns. Patients and providers connected via Zoom through the Epic application. After telemedicine evaluation, electronic prescriptions for medications and testing were provided and clinic assistants provided patients with instructions on how to proceed with testing, either at our center or at local testing sites. Given some patients chose to complete testing at local centers during the pandemic, ensuring timely follow-up of these results was an essential adaptation to the clinic workflow. Clinic assistants uploaded external results into the medical record and short interval follow-up allowed review of completed testing. This transition to telemedicine combined with the clinic’s existing infrastructure allowed for efficient, nimble innovation to meet the demands of the COVID-19 pandemic.

Our Early Experience

The RESCUE-TIA clinic was established at our institution in 2019. We conducted a retrospective review of patients evaluated in the clinic and summarized patient characteristics and quality data with standard descriptive statistics. The institutional review board at Weill Cornell Medical College approved this study and waived the requirement for informed consent. Deidentified data and the analytic methods that produced the results of this study are available on reasonable request from the corresponding author.

Between December 2019 and August 2020, 21 patients were evaluated in the clinic, including 15 patients during the height of the COVID-19 pandemic in New York. We introduced telemedicine in March 2020 as a result of the pandemic; 67% of patients were seen by telemedicine during the above time period. The median age was 75 years (interquartile range [IQR], 61-82) and 71% were female. The median time from symptom onset to clinic evaluation was 3 days (IQR, 2.5-17.5). Median time from clinic evaluation to laboratory diagnostics was 8 hours (IQR, 2-21), to echocardiogram completion was 8 days (IQR, 0-10), to outpatient telemetry was 5 days (IQR, 1-9), and to imaging completion was 1 day (IQR, 0-5). Among patients evaluated after the transition to telemedicine in March 2020, median time from symptom onset to evaluation was 3 days (IQR, 2-21). Median time to laboratory diagnostics was 1 day (IQR, 0-2), to echocardiogram completion was 5 days (IQR, 2-8), to outpatient telemetry was 5 days (IQR, 1-14), and to imaging completion was 1 day (IQR, 0-7). The median time to imaging in the telemedicine population was similar to the patients seen in-person (1 day, IQR 0-1), but with a larger interquartile range (Figure 5). Final diagnoses, determined by the board-certified vascular neurologist according to AHA/ASA definitions for diagnosing cerebrovascular events, were TIA (n = 12), ischemic stroke (n = 5), transient global amnesia (n = 2), migraine (n = 1), and non-aneurysmal, distal subarachnoid hemorrhage (n = 1). ⁴⁸ There were no documented cases of inappropriate telephone triage. Only one patient was referred to the emergency department for carotid occlusion, which was ultimately determined to be chronic. Recurrent TIA occurred in 1 patient after 67 days, and ischemic stroke occurred in 1 patient 55 days after TIA.

Figure 5.

Box and whisker plots for time from clinic evaluation to imaging. A, All patients. B, Patients evaluated in-person. C, Patients evaluated through telemedicine. Box limits represent 1st and 3rd quartiles. Line in box represents median. Whisker lines represent minimum and maximum values with outliers, if applicable.

Only 21 patients were evaluated over an 8-month period. A possible explanation for this is the clinic’s strict inclusion/exclusion criteria. However, these criteria are necessary to ensure safe triage. Second, while building our program, we did not advertise our program widely, so our referral base was small. Lastly, it is also plausible that not all patients have equal access to technology and internet, limiting the population reached by telemedicine. The interquartile range for time to imaging among patients evaluated through telemedicine was larger than that of patients evaluated in-person. This finding could be due to chance from small sample sizes or because some patients preferred to have testing done at local centers that did not have an agreement with radiology for MRI/MRA within 24 hours, as our center does.

Conclusion

We have developed an outpatient clinic utilizing telemedicine and rapid outpatient MRI for timely, evidence-based evaluation and management of patients with TIA and minor stroke with non-disabling fixed deficits. While the COVID-19 pandemic hopefully wanes, improvements in stroke systems of care as a result of the pandemic may prove beneficial in the long term.

ORCID iD

Paul M. Wechsler, MD https://orcid.org/0000-0002-0314-1867

Neal S. Parikh, MD, MS https://orcid.org/0000-0002-8802-2380