Editorial May 5, 2022

Strategic Opportunities to Improve Stroke Systems of Care

Kori S. Zachrison, Lee H. Schwamm

JAMA. 2022;327(18):1765-1767. doi:10.1001/jama.2022.3820

Restoring reperfusion to the brain as quickly as possible during acute ischemic stroke is important to reduce long-term disability and optimize clinical outcomes for patients, although hospital and resource availability are frequently not well matched with patient location. Accordingly, many patients who may have large-vessel occlusion (LVO) lack rapid access to endovascular thrombectomy (EVT),¹ a highly efficacious procedure with a number needed to treat of approximately 3 to 7 for regaining independence among carefully selected patients.²Therefore, accurate out-of-hospital diagnosis and ensuring that the patient and the neuroendovascular physician speedily arrive at the same location are paramount. In this issue of JAMA, 2 new studies^3,4 report interventions designed to address this mismatch between patient location and time-critical access to stroke expert resources.

Hubert and colleagues³ report results from an observational study that evaluated the outcomes associated with bringing EVT to remotely located patients by flying an interventional team to the patients rather than transferring the patients to the team. The hypothesis was that avoiding lengthy interhospital transfer processes would expedite EVT without increasing adverse events and thereby improve patient outcomes. The study was conducted in the Telemedic Pilot Project for Integrative Stroke Care telestroke network in Bavaria, Germany, among 13 primary stroke centers and multiple EVT-capable hubs, with an alternating on- vs off-week design. Patients treated by the flying intervention team during on weeks (n = 72 patients) compared with patients transferred to an EVT-capable stroke center (n = 85 patients) had a 90-minute shorter median time from the decision to pursue thrombectomy to the start of the procedure (median time, 58 minutes vs 148 minutes, respectively). This faster time to intervention was achieved without any difference in complication rates or safety issues, despite performing the procedure in a local primary stroke center. After 3 months, there was no significant difference in functional outcomes between those who received the flying intervention vs the conventional transfer (median modified Rankin Scale [mRS] score of 3 in both groups; adjusted common odds ratio for improved functional outcome among those who received the flying intervention of 1.91 [95% CI, 0.96-3.88]). However, the study was not powered to demonstrate differences in these outcomes, and further research is needed to assess the clinical benefit to patients, overall safety, cost-benefit ratio, and generalizability.

In another study in this issue of JAMA that aimed to improve access to timely EVT, Pérez de la Ossa and colleagues⁴ report findings from the RACECAT randomized clinical trial that evaluated alternative out-of-hospital transport strategies for patients with suspected LVO. In a nonurban region of Catalonia, Spain, all potential patients with acute stroke were evaluated with an out-of-hospital clinical stroke screen to identify patients with suspected LVO. When the closest hospital was not EVT capable, patients were then randomized to direct transport to an EVT-capable center further away (n = 688 patients) or direct transport to a nearby local primary stroke center (n = 713 patients) for rapid intravenous thrombolysis if eligible, followed by subsequent transfer to an EVT-capable center if LVO was confirmed and they were still clinically eligible for EVT. Among 1401 included patients, there was no significant difference in 90-day functional outcome between the 2 groups (median mRS score of 3 in both groups; adjusted common odds ratio for reduced disability, 1.03 [95% CI, 0.82-1.29]). Notably, patients transported to local stroke centers first had significantly higher odds of receiving intravenous thrombolysis relative to those transported directly to EVT-capable hospitals (60.4% vs 47.5%, respectively), and 16% had resolution of LVO by the time they arrived. In contrast, patients transported directly to EVT-capable hospitals had significantly higher odds of receiving EVT (48.8% vs 39.4%).

The findings of the RACECAT trial could be interpreted as favoring direct transport to EVT-capable hospitals for patients with suspected LVO. This approach did not demonstrate any evidence of harm from delays, thus supporting a strategy for direct triage to the highest level of resources possible. However, transport to the EVT center did not improve outcomes and potentially moved patients farther from their homes and support networks, and possibly could have increased total costs, reduced local hospital stroke patient volume and expertise, prolonged ambulance transports, and unnecessarily contributed to overcrowding at comprehensive stroke centers. Similar outcomes at higher cost equates to lower value in health care, suggesting that transfer to the nearest primary stroke center is still the best-value strategy.

The study by Hubert et al and the study by Pérez de la Ossa sought new ways to strengthen the existing stroke systems of care. The stroke system of care refers to the full spectrum of stroke care, from time of symptom recognition through activation of emergency medical services (EMS), identification of transport destination, hospital presentation, transfer if needed, inpatient care, rehabilitation, and postacute care.⁵ A well-designed stroke system of care includes education, coordination, and facilitation at each of these stages, and patient outcomes are dependent on system investments throughout. Without public education efforts about rapid recognition of stroke symptoms and early activation of EMS, patients may present to hospitals beyond the window of eligibility for reperfusion interventions. Any delays in the acute phase of the stroke system of care may compromise poststroke functional outcomes, lead to increased disability and dependency, and increase total medical expenditures.

Both studies aimed to improve the stroke system of care at the point of interhospital transfer. Patients with stroke are frequently transferred between hospitals, and important efforts have focused on improving care prior to transfer (eg, through telestroke) and care processes such as door-in, door-out times for patients requiring subsequent transfer.⁶ Yet, particularly in the current US context of reduced EMS availability,⁷ on-scene and transfer delays are frequent and have been associated with decreased likelihood of EVT receipt for patients.⁸ The shift to parallel processing of the flying intervention team enabled substantially earlier EVT initiation, which will likely lead to improved patient-centered outcomes including reduced disability and mortality. This strategy has many advantages, but assessment of the value of the intervention also must account for the cost, reliability, and availability of the helicopter. This strategy is similar in some ways to the recent use of mobile stroke units, which have been similarly shown to increase access to timely intravenous thrombolysis within the specific geographic populations studied.⁹ Furthermore, because neuroendovascular specialists are a relatively scarce resource¹⁰ and higher operator case volume is likely associated with improved patient outcomes,¹¹ bringing the proceduralist to the patient may be an important strategy to maintain high levels of procedural competence in regions with low stroke volumes.

The RACECAT randomized clinical trial also sheds additional light on the currently debated question as to whether more rapid thrombectomy access would outweigh delays to intravenous thrombolysis and unnecessary transfers. The expert consensus recommendations¹² of the American Heart Association (AHA) that are endorsed by multiple professional societies are aligned with the RACECAT results as it relates to patients with prolonged transport times. When travel time to an EVT-capable center is greater than 60 minutes (as was the case for more than half of the patients in the RACECAT study), the AHA recommendation is for transport to the nearest acute stroke–ready hospital or primary stroke center where thrombolysis can be administered. When transport times are shorter than 60 minutes, as is often the case in urban or suburban areas, the recommendations support the option of direct transport to an EVT-capable center. However, the optimal transport destination remains unknown when travel distances are short and is highly dependent on the capability and effectiveness of the community-based stroke centers. When these primary stroke centers can provide rapid and safe thrombolysis, as they did in Catalonia, they remain an ideal option, but if they are slower, less safe, or introduce extensive interfacility transport delays, then this strategy may produce worse outcomes than direct triage to an EVT-capable center.

What other opportunities remain to strengthen the stroke system of care? First, innovative methods must continuously be tested to promote public engagement in stroke symptom recognition and activation of EMS. These include school-based programs (eg, Hip-Hop Stroke) or equity-focused programs (eg, Live Chair in African American barbershops, the multilingual Massachusetts Department of Public Health FAST campaign, or other novel public education campaigns¹³), empowered by digital technology such as smartphone applications or crowdsourcing strategies.

Second, once symptoms are recognized and EMS is activated, the system must be a seamless process from well-trained dispatch to rapid patient evaluations and short on-scene times to optimal transport destination. Mobile stroke units are a promising addition to the out-of-hospital care delivery setting,⁹ although their value and utility may depend on particular geographic and population characteristics. To improve transport destination decision-making, better affordable tools are needed for on-scene LVO detection. Current EMS screening examination tools have important limitations and misclassify a substantial proportion of patients.¹⁴ In the RACECAT trial, about one-third of included patients did not have an LVO. New low-cost portable devices based on electroencephalography or near-infrared spectroscopy hold future promise to improve LVO detection, but simulation modeling suggests that the optimal transport destination will still rely on geography, proximity of resources, traffic patterns, and local environmental factors.¹⁵ The ubiquitous availability of smartphones and broadband may eventually support complex algorithms that can be computed on scene to improve reliability of LVO detection and out-of-hospital routing to get the right patient to the right place in the right amount of time. Third, equitable availability of EVT for all patients must be increased, regardless of zip code or other social determinants.

When it comes to stroke, time is brain. A robust, standardized, protocol-driven system response will be important to ensure consistent, high-quality, and equitable care. Achieving this may require different approaches in countries with coordinated, centralized health care planning (such as the countries in which these 2 studies were conducted) vs countries with more decentralized health care delivery. However, the imperative to continually strengthen the stroke system of care and increase the value in the delivery of stroke care remains relevant for all, regardless of location.