Highlights
-
Transporting acute stroke patients to the right hospital in the right time frame to enable treatment is aided by direct paramedic, transport physician and stroke neurologist discussion from the scene – the “rural field consult”.
-
Implementation in Alberta reduced treatment times and extended the delivery of endovascular thrombectomy to rural patients.
Introduction
The optimal approach to triage and transport for acute stroke from rural regions is not necessarily obvious. Speed is the single most important modifier of the treatment effect for both endovascular therapy (EVT) and intravenous thrombolytic drug treatment for acute ischemic stroke. Reference Marler, Tilley and Lu1–Reference Saver, Goyal and van der Lugt3 Faster treatment results in a larger treatment effect by a substantial margin such that minutes matter. Reference Goyal, Almekhlafi and Dippel4 While the inter-patient variance is high, when designing a system of care for stroke, faster average interval times of even 5 min will be associated with measurable improvements in the number of patients achieving favorable outcomes and that in turn will be associated with direct cost avoidance over the 1-year post-stroke period. This relationship between speed and treatment effect size has allowed the use of interval times as quality metrics. Reference Kamal, Benavente and Boyle5 Further, the imperative for fast treatment has resulted in a tremendous focus on prehospital systems of care from stroke onset to hospital arrival.
The problem of getting the patient to the right hospital in the fastest possible time is most challenging in rural areas. In urban areas, under 30-min transport times make decision-making and transport to a stroke hospital relatively straightforward; by contrast, in areas with large rural populations like Canada, this issue is felt severely because of the vast geography and relatively sparse population. Primary stroke centers (PSC) capable of delivering intravenous thrombolysis exist in many small regional and urban cities, but not small towns, but comprehensive stroke centers (CSCs) capable of providing endovascular thrombectomy are necessarily concentrated in metropolitan areas. The best prehospital triage and transport strategy for rural patients with suspected large vessel occlusion ischemic stroke remains controversial.
A randomized trial, conducted in the Catalan region of north-eastern Spain (RACECAT trial) comparing triage to the nearest PSC versus triage to the nearest CSC, did not identify a universally preferred strategy; both approaches resulted in similar outcomes. Reference Perez de la Ossa, Abilleira and Jovin6 In a post hoc analysis, patients with a final diagnosis of intracerebral hemorrhage appeared to have better outcomes when going to a PSC first. Reference Ramos-Pachon, Rodriguez-Luna and Marti-Fabregas7 However, the geography of the Catalan region in Spain is such that distances to CSCs are not long (<60 min) for the majority of the population. While the conclusions of RACECAT may not apply as well to other geographies, modeling data suggest that the trial got it right. Reference Holodinsky, Williamson and Demchuk8,Reference Holodinsky, Kamal, Hill, Perez de la Ossa and Ribo9
Routing of the patient to the closest PSC can paradoxically sometimes bring them geographically further from a CSC. Expertise at a PSC may be limited, and as a result, processes of care may be slower. Slow door-in-door-out (DIDO) times are common globally, and the cumulative delay can result in a lost opportunity to offer successful endovascular thrombectomy for the subset of patients eligible for this treatment. Utilization of a rural emergency medical services (EMS) ambulance and crew for transport directly from the field to the CSC removes an important resource out of the region for the duration of the return trip, often where there are few active ambulances and crews. Thus, transport decisions can affect other patients in the region. Telemedicine in the PSCs and community centers may help in guiding treatment and transfer decisions once the patient arrives at that facility, but it does not effectively solve the problem of long DIDO time and delays in transfers for tertiary care, such as EVT.
Fundamentally, the challenge is that the detailed diagnosis is not known in the field. A severe stroke can be clinically recognized with near certainty, but whether it is an acute ischemic stroke potentially eligible for acute treatment with thrombolysis or EVT or an intracerebral hemorrhage is not discernible on a clinical basis alone. Patients with less severe neurological symptoms may represent stroke mimics. Finally, eligibility for treatment is determined by the medical history (past and current medical conditions, disability and frailty), medications and a detailed understanding of the history of the current event. The summative result is that simple algorithmic rules and clinical scales for identifying potential large vessel occlusion strokes in the field are only moderately sensitive and specific for making triage and transport decisions.
To deal with this reality in Alberta, in 2017, a rural field consultation model was developed and adopted. The approach was novel at the time and arrived at based upon an in-person consensus-based discussion among prehospital care leaders including ground and air transport physicians, emergency physicians, stroke neurologists and acute care nursing leaders. EMS teams with possible stroke would directly initiate a five-way conference call and interact with a stroke physician at the comprehensive center, an EMS transport physician, the air ambulance control center and the nearest PSC emergency department (ED). The overall goal of the rural field consult program is to improve access to early recanalization by facilitating appropriate transfer of the right patient, to the right center, in the right time frame to be able to offer the best acute stroke treatment. The 5-year experience with the rural field consultation model is reported.
Methods
The program was set up in 2017, in the province of Alberta, which is covered by a single health insurance plan so that all 4.5 M residents have access to publicly funded acute stroke care. There are 17 designated stroke centers in Alberta, consisting of 2 comprehensive centers providing EVT (24 h, 7 days a week), 14 rural PSCs and 1 urban PSC providing access to intravenous thrombolytic drug treatment. The province (Figure 1) was divided into three regions: metropolitan urban zones (50 km radius around Calgary and Edmonton, which contain CSCs), small urban zones (30 km radius around the 14 rural PSCs) and rural zones (>30 km from a PSC or CSC, consisting of rural Northern and Southern Alberta). Characteristics of rural PSCs included a CT scan available 24 h × 7 d per week, stroke expertise on-site or via telestroke, intravenous thrombolysis availability and a door-to-CT time of less than 20 min with an EMS pre-notification system. The CSCs were defined as containing a stroke team on-site day and night, providing EVT as well as neurosurgical expertise on-site.
Map of Alberta showing the rural field calls distribution. The green dots represent each call received for a rural field consultation; the blue dots represent a primary stroke center with its corresponding 30 km radius; the red dots represent a comprehensive stroke center in Edmonton (north) and Calgary (south) with its corresponding 50 km radius.
Figure 1 Long description
A map of Alberta showing rural field calls distribution. The map features green dots representing each call received for a rural field consultation. Blue dots indicate primary stroke centers, each with a corresponding 30-kilometer radius. Red dots represent comprehensive stroke centers located in Edmonton to the north and Calgary to the south, each with a corresponding 50-kilometer radius. The distribution of green dots shows a higher concentration around the comprehensive stroke centers, suggesting a focus on faster treatment access in these areas. The map highlights the importance of speed in stroke treatment and the strategic placement of stroke centers to optimize response times. All values are approximated.
The Referral, Access, Advice, Placement, Information & Destination (RAAPID) system is a communications resource staffed by healthcare personnel (principally nurses) that provides a sophisticated communications center to connect all or any healthcare provider(s) with another both within and including third parties from outside Alberta. The system covers the entire province and is integrated into the health system. All calls and connectivity were organized through RAAPID.
The Shock Trauma Air Rescue Service (STARS) is an air ambulance service with three rotary aircraft bases in Calgary, Edmonton and Grande Prairie. Additionally, fixed-wing aircraft are used for longer transport distances from the far north and far south-east of the province. The STARS air ambulance service has the capacity to rendezvous with ground ambulance in selected circumstances. Air ambulance services are limited by weather-related flight safety considerations and daylight hours in the mountainous regions of the province.
The EMS teams notify the Alberta-wide RAAPID system of a potential severe stroke patient at the time of their first assessment in the field. EMS teams utilize the Alberta Heath Service Stroke Screening tool (a modified version of the Cincinnati prehospital scale) Reference Kothari, Pancioli, Liu, Brott and Broderick10 to identify a suspected stroke. Then they use the Los Angeles Motor Score (LAMS) Reference Nazliel, Starkman and Liebeskind11 to assess stroke severity. Patients in rural regions, who have a suspected stroke, are candidates for the rural field consultation. The decision to initiate a call rests with the paramedic and is based upon their acute evaluation of the patient; while typically the call is initiated only for more severe stroke, clinical judgment is encouraged. Depending on whether the call comes from Northern or Southern Alberta, it is directed to the on-call stroke physician at the closest CSC in Edmonton or Calgary. The call can include up to five parties: (1) the paramedic at scene; (2) an emergency physician (online medical control); (3) an air transport emergency physician (STARS or EMS transport physician); (4) the ED physician at the closest PSC; and (5) the stroke neurologist. Expertise in specific weather conditions, transport times, means available (ground or air or a combination of both), the medical stability of the patient for transport and related medical needs are assessed by the transport physicians. The stroke physician interacts with the paramedic crew to determine details about the current neurological presentation, and then pertinent comorbid medical history, medication and details about baseline functioning, disability, frailty and goals of care. Historical or event-based elements, not captured by the Alberta Health Services stroke screen or LAMS score, are used by the stroke physician to evaluate a suspected stroke diagnosis (e.g., intracerebral hemorrhage rather than an ischemic stroke or a stroke mimic) and probable prognosis with treatment, all of which will influence the decision to transfer directly to a PSC rather than a CSC or vice versa. Simple written material training was developed and then provided to each of the component teams involved in prehospital triage and transport.
Administrative data records were collected over 5 years from 2017 to 2022 to evaluate the usage and outcomes. These included the RAAPID system, EMS electronic Patient Care Record (ePCR), the Quality Improvement and Clinical Research (QuICR) registry, the Client Referral Information System (CRIS), ConnectCare (electronic medical record, EPIC), the Discharge Abstract Database (DAD) and the National Ambulatory Care Reporting System (NACRS). Time to first medical contact, the time at first ED, the time of thrombolytic therapy and the time of arterial access for EVT were collected. Clinical outcomes are reported using home time, which is derived from administrative data. Home time is defined as the number of nights spent outside of healthcare institutions within 90 days of the index admission to an acute care hospital, and this outcome is associated with global disability after stroke. Patients who died in the hospital after the index stroke have a home time of zero days. Reference Yu, Rogers and Wang12–Reference Joundi, King and Stang14 Data are reported using simple descriptive statistics.
Results
From April 2017 to March 2022, there were 721 patients who met the definition for a RAAPID call destined for rural field consultation. After filtering for valid personal health numbers, RAAPID calls that originated from hospitals (identified by secondary manual chart review), RAAPID calls that originated within the metro municipalities of Calgary and Edmonton and availability of EMS records, 601 patients were included in the analysis (Figure 2). For context, there were 25,705 community stroke presentations admitted to Alberta hospitals during the same period.
Flowchart of record selection for inclusion in the analysis of the rural field consultation system. RAAPID = Referral, Access, Advice, Placement, Information & Destination; ULI = unique lifetime identifier; EMS = emergency medical services.
Figure 2 Long description
The flowchart illustrates the process of selecting records for the analysis of the rural field consultation system. It starts with 721 records that met the RAAPID cohort definition. After excluding 41 records with invalid unique lifetime identifiers, 680 records remain. Of these, 66 records with no link to emergency medical services are excluded, leaving 614 records. Further, 6 records with Raapid call overlap with emergency department visits and 7 records from the municipalities of Calgary or Edmonton are excluded. Finally, 601 records that linked to emergency medical services and secondary assessment confirm a rural field call.
A majority of calls originated from Alberta North (427/601, 71%) as compared to Alberta South (174/601, 29%). A large majority of patients (541; 90%) were transported by ground ambulance or a combination of ground and air ambulance; 44 (7.3%) were transported by STARS Air Ambulance helicopter (either directly or with a combination of ground and air ambulance), and 16 (2.7%) were transported by the mobile stroke unit, Reference Shuaib and Jeerakathil15 which is based in Edmonton. Patients were routed to a designated stroke center (565/601, 94%) in nearly all cases. A minority had multileg transport to a PSC and then to a CSC (51/601, 8.5%). Most patients had severe deficits defined as a LAMS score of 4 or 5 (425/601, 70.7%); the remainder had missing scores (129/601, 21.5%) or had a LAMS score of 0–3, (47/601, 7.8%). The most common reasons for non-reported LAMS score were medical instability of the patient and EMS crew uncertainty.
Following assessment in the ED, 543/641 (90.3%) were admitted to hospital. At hospital discharge, the most common final diagnosis (Figure 3) was ischemic stroke. Stroke mimics included most frequently a presumed non-organic or unspecified cause of hemiparesis, a delirium presentation or a cerebral neoplasm, migraine, syncope, seizure disorder and acute infection. A majority (430/601, 71.5%) of patients receiving a rural field consultation did not receive acute reperfusion therapy. Intravenous thrombolysis was given to 65 (10.8%) patients, 17 (2.8%) at the PSC and 48 (8.0%) at the CSC. A total of 106 (17.6%) patients underwent EVT, with or without intravenous thrombolysis. Of the patients routed directly to a CSC, a higher proportion of patients (87/404, 21.5%) received EVT, which represented 14.5% of the total with a rural field consultation. The median time from first medical contact to arterial access was 3.2 h in the direct transfers, compared to 6.1 h in patients arriving indirectly to the CSC.
Pie chart showing the final discharge diagnosis for admitted patients (n = 543) following a rural field consultation. A total of 58 of 601 patients were discharged from the emergency department without admission. A majority had ischemic stroke, 8.1% had an intracranial hemorrhage but 18.2% had a final discharge non-stroke diagnosis (stroke mimic). A small proportion were admitted directly to palliative care or rehabilitation, and consistent with the high LAMS scores, only 4.1% had complete resolution, leading to a final diagnosis of transient ischemic attack. AIS = acute ischemic stroke; ICH = intracerebral hemorrhage; TIA = transient ischemic attack; SAH = subarachnoid hemorrhage; Rehab = rehabilitation; Pall care = palliative care.
Patient outcomes are shown using home time (Figure 4). Patients admitted to the hospital had poorer outcomes overall. Nevertheless, one in eight patients who were discharged from the ED after evaluation were deceased at 90 days.
Outcome data. Note that five patients had unknown home time because they were unable to be linked. The top horizontal bar describes the entire cohort, and the two bars below distinguish between those who were admitted to the hospital and those who where were discharged from the emergency department.
Figure 4 Long description
The bar graph compares the percentage of patients with different 90-day home times across three groups: Full, Admitted, and Not admitted. The x-axis represents the percentage, ranging from 0 to 100. The y-axis lists the three groups: Full (n=596), Admitted (n=548), and Not admitted (n=48). Each group has a horizontal bar divided into segments representing different home time intervals: 85 to 90 days, 70 to 84 days, 30 to 69 days, 1 to 29 days, 0 days alive, and 0 days deceased. The Full group shows 21% for 85 to 90 days, 18% for 70 to 84 days, 11% for 30 to 69 days, 6% for 1 to 29 days, 16% for 0 days alive, and 28% for 0 days deceased. The Admitted group shows 17% for 85 to 90 days, 19% for 70 to 84 days, 12% for 30 to 69 days, 6% for 1 to 29 days, 17% for 0 days alive, and 29% for 0 days deceased. The Not admitted group shows 73% for 85 to 90 days, 6% for 30 to 69 days, 6% for 1 to 29 days, 2% for 0 days alive, and 13% for 0 days deceased. The color scheme uses light blue for 85 to 90 days, medium blue for 70 to 84 days, darker blue for 30 to 69 days, and darkest blue for 1 to 29 days, 0 days alive, and 0 days deceased. All values are approximated.
To assess call quality, the population was grouped, post hoc by presentation and outcome, into four different optimal or suboptimal scenarios (Figure 5). The remaining scenarios could not be classified as either optimal or suboptimal, representing 55.6% of rural field consultation subjects. The two most common situations were a direct transfer to a CSC, with no acute reperfusion therapy, and a direct transfer to a PSC first, with no intravenous thrombolytic drug treatment and no transfer to a CSC. There were 36 of 404 patients (8.9% of “direct to CSC” patients) with a primary diagnosis of intracerebral hemorrhage for whom, had this been known in the field, direct transport to a CSC may not have been optimal. Reference Ramos-Pachon, Rodriguez-Luna and Marti-Fabregas7 It is conceivable that basic care (airway, blood pressure management, anti-emetics, etc.) provided earlier may provide both a survival and a disability advantage.
Flowchart and table of possible scenarios from the call up until the final center where the patient was acutely evaluated.
Figure 5 Long description
The table compares various scenarios of stroke treatment and transport options, detailing patient numbers and ranks. It includes columns for comprehensive stroke center (CSC), primary stroke center (PSC), non-stroke hospital, and stroke ambulance, with rows for different treatment and transport scenarios. Each scenario is evaluated based on the use of intravenous thrombolytic drug treatment (IVT) and endovascular therapy (EVT), and whether the patient is transported to a comprehensive stroke center. The table lists the number of patients (n) and the rank of each scenario. Notable trends include the highest patient numbers in scenarios involving direct transport to a CSC with EVT and the lowest ranks for scenarios with no transport to a CSC. The table also includes a description of each scenario, the number and percentage of patients, timelines for medical contact to treatment, and notes on treatment quality.
Qualitatively, the burden of the service on the acute stroke neurology teams was greater in northern Alberta (managed by the University of Alberta Hospital stroke service in Edmonton). However, the number of calls, on average, is slightly less than one per day annually, and this volume was generally well managed within the existing on-call structures. Each call did not last long (<10 min), but the direct communication with the on-scene paramedics necessarily allows for the clinical input of the neurology team to direct patient transport.
Discussion
The rural field consultation system developed in Alberta and implemented in early 2017 impacted approximately 2.3% of all stroke cases in the province over a 5-year period. The yield of the calls, defined by the implementation of acute treatment, is slightly less than one in seven calls, resulting in acute intervention either or both of intravenous thrombolysis or EVT. It is feasible and possible to improve service to rural stroke patients with acceptable specificity.
The qualitative experience was that this system had a positive impact on the care delivered to rural Albertans. The system was perceived to allow for more patients to be routed to a CSC instead of a more limited center in the event of a possible large stroke and allowed for more informed decisions to be made in the field. Direct routing to the CSC undoubtedly provided a faster time to EVT and potentially to recanalization, with the median time from first medical contact to arterial access nearly 3 h faster with direct transfers. Additionally, a collateral benefit can be assumed from this system, provided that more patients with severe neurological presentations were directly routed to a CSC for advanced management of more complex conditions. However, the 36 of 404 patients (8.9% of “direct to CSC” patients) who had intracerebral hemorrhage and were transported for longer periods may have suffered worse outcomes because of the decision to undertake direct transport to the CSC. Reference Ramos-Pachon, Rodriguez-Luna and Marti-Fabregas7
Areas for improvement for the Alberta system were identified. The least desirable scenario was the case of a patient being routed to a PSC or another center after the field consultation with the stroke physician, not receiving intravenous thrombolytic drug treatment at that center, being transferred to a CSC and then receiving EVT. This means that the goal of lowering the time to reperfusion was not achieved. This only affected five patients (0.8%) in the current review, and this low number is encouraging for the performance of the program. Stroke is dynamic, and such situations may still have been appropriate if the reason for transfer to the CSC was because of patient deterioration or even re-embolization to a proximal intracranial artery after they had been routed to the PSC. Nevertheless, the aim is to optimize time efficiency in the system, and by analyzing individual cases, the receiving neurologist can learn about the optimal clinical questions to explore. Further, a high degree of geographical knowledge about the province in relation to the location and capacity of each of the PSCs is important for helping to make optimal triage decisions.
The analysis is empirical and descriptive and has limitations. There is no parallel comparator system of care that uses the same kind of rural triage approach. Overall, the system attempts to provide improved service to the rural population, and the experience encourages consideration of its use in jurisdictions with similar needs.
Author contributions
The primary paper was written by MAP, MDH, MLH and JS. Data analytics were provided by JS. MDH performed the statistical analysis. All authors both contributed to patient care and data collection and provided critical review and revision of the manuscript.
Funding statement
There was no specific funding for this project. The work was done as part of standard job responsibilities under Alberta Health Services.
Competing interests
None of the authors have conflicts of interest for this work. Dr Panzini reports honouraria from lectures, speaking from SSVQ (Société des Sciences Vasculaires du Québec), the CSC (Canadian Stroke Consortium) and the University of Montreal CHUM (Centre Hospitalier de l’Université de Montréal). Ms. Bredenkamp reports stock or stock options in SnapDx Inc. and Let’s Get Proof (Collavidence Inc.) Research collaboration. Ms. Halabi reports a nonfinancial leadership role in the Canadian Stroke Best Practice Recommendation-Advisory Committee. Dr Ganesh reports stock or stock options in SnapDx Inc. and Let’s Get Proof (Collavidence Inc.) Research collaboration. Dr Demchuk reports grant funding from Medtronic LLC; honouraria or consulting fees from AstraZeneca, Medtronic, Roche Canada and Boehringer-Ingelheim; patent authorship licensed to Circle CVI; and chairmanship of the board of directors of the Canadian Stroke Consortium.
Target article
Rural Field Consultation for Remote Acute Stroke Transport Decisions
Related commentaries (1)
Reviewer Comment on Panzini et al. “Rural Field Consultation for Remote Acute Stroke Transport Decisions”