Article Text
PDF
Abstract
Currently, there is no device capable of detecting acute ischemic stroke (AIS) secondary to emergent large vessel occlusion (ELVO) in the pre-hospital setting. The inability to reliably identify patients that would benefit from primary treatment with endovascular thrombectomy remains an important limitation to optimizing emergency medical services (EMS) triage models and time-to-treatment. Several clinical grading scales that rely solely on clinical examination have been proposed and have demonstrated only moderate predictive ability for ELVO. Consequently, a technology capable of detecting ELVO in the pre-hospital setting would be of great benefit. An analogous scenario existed decades ago, in which pre-hospital detection of acute myocardial infarction (AMI) was unreliable until the emergence of the 12-lead ECG and its adoption by EMS providers. This review details the implementation of pre-hospital ECG (PHECG) for the detection of AMI and explores how early experience with PHECG may be applied to ELVO detection devices, once they become available.
- stroke
- technology
Statistics from Altmetric.com
Introduction
For patients with symptoms suggestive of acute myocardial infarction (AMI), it is recommended that EMS personnel perform a 12-lead ECG either in the field or en route to the hospital.1–5 The benefits of identifying patients with AMI in the pre-hospital setting include early initiation of appropriate medical therapy, ambulance routing to the most suitable hospital for definitive treatment, and early pre-notification/activation of receiving teams in the emergency department (ED) and the cardiac catheterization laboratory.6 Moreover, pre-hospital ECG (PHECG) appears to reduce treatment times and patient mortality rates.7–9
An analogous technology has yet to become available for ELVO detection. Mobile stroke units (MSU) with ambulance-based CT angiography (CTA) capabilities are being tested at selected medical centers, though questions remain regarding their effectiveness and high costs make their availability limited.10 Several ELVO devices are currently under development, including AlphaStroke (Forest Devices Inc., Pittsburgh, PA, USA), Lucid M1 transcranial Doppler Ultrasound System (Neural Analytics Inc., Los Angeles, CA, USA), Volumetric Integral Phase-Shift Spectroscopy (VIPS) (Cerebrotech Medical Systems Inc., Pleasanton, CA, USA), and BrainPulse (Jan Medical Inc., Mountain View, CA, USA). As these devices are finalized and tested, a thorough understanding of the utility and predictive accuracy required of them is of importance to ensure they will be useful in real-world practice. PHECG detection of AMI was the subject of similar consideration decades ago and has since amassed a comprehensive body of literature. By drawing on the studies that led to the incorporation of PHECG into practice, several lessons can be learned and applied to ELVO detection devices, once they become available. In anticipation that these devices will soon emerge onto the market, we examine the history and timeline of PHECG testing and assess how it may be used to inform future ELVO device assessments.
Pre-hospital ECG: historical perspective
The concept of early detection of AMI was first proposed more than five decades ago, when it was suggested that mobile resuscitation teams might help reduce patient mortality.11 The concept was put into practice in the 1970s by equipping ambulances with 12-lead ECG devices.12 However, the benefits were initially limited by the lack of an efficient, high-quality transmission system between mobile ambulances and emergency departments.13 Once transmissions became more reliable, with the advent of cellular telephone networks, evidence began to emerge supporting the potential benefits of PHECG, including shortened time to reperfusion therapy and reduced mortality rates.14 Importantly, early studies emphasized that time spent performing PHECG at the scene was minimal and did not significantly increase the time from first medical contact (FMC) to ED arrival.14
As a result of the gathering evidence in favor of PHECG, the 1996 AHA/ACC guidelines for AMI management were the first to recommend that 12-lead ECG be performed in the pre-hospital setting, if available.15 Subsequent studies showcased the reliable predictive metrics of PHECG and that its readouts could even assist with management in the ED.1 16–18 Investigators also determined that PHECG appears to improve EMS transport efficiency, receiving team preparedness in the ED and catheterization laboratory, in-hospital workflow, and time to reperfusion.8 19 20 As a result, the most recent AHA/ACC guidelines for AMI management, published in 2013, now include a Level of Evidence Class IB recommendation that EMS perform a 12-lead ECG in the field for all patients with symptoms consistent with AMI.2 International guidelines have also echoed this recommendation.5 Despite the guideline recommendations, a favorable body of evidence, and wide availability of the technology, incorporation of PHECG into routine EMS practice has been slow.8 Its utilization rate in the AMI population was reported below 10% from 2000 to 2002, though some regional networks have noticed an increase over time.21–23 Nevertheless, PHECG serves an important role in transport decisions when bypass of centers not equipped for percutaneous coronary intervention (PCI) is indicated and assists in notifying ED teams and catheterization laboratories to prepare for a patient’s arrival.8 19 24 25
The timeline of PHECG testing and implementation into practice provides a blueprint for future ELVO device testing. As with PHECG, reliable ELVO device data transmission to hospital-based physicians may be an important initial step. Communication and data transmissions are not as challenging as they once were, though they may remain a concern in rural areas. Much like it was for PHECG, concern for on-scene time delays may result in preliminary hesitancy to use an ELVO device in the field. Investigators might seek to study these potential time delays early on to allay the concern and gain wider adoption among EMS systems at the outset. Finally, and likely most importantly, hospital and EMS networks that may eventually rely on the device readings for triage decisions and resource activation will need to be provided with evidence supporting reliable predictive metrics and improved time-to-treatment for eligible patients. Altogether, the PHECG historical timeline lays out a useful blueprint that outlines the concerns that may arise and the types of studies that may be needed to successfully integrate an EVLO device into practice.
Expeditious reperfusion therapy
The time required to reach reperfusion is associated with functional outcomes in both ELVO and AMI.26–29 The use of PHECG has been reliably shown to reduce the time-to-reperfusion in AMI.7 8 In one such study, Kahlon and colleagues demonstrated that ST-evelation myocardial infarction (STEMI) detection with PHECG, followed by the decision to bypass a nearer non-PCI center in favor of a more distant PCI-capable center, resulted in a 78 min (50%) reduction in time-to-reperfusion.6 This also translated to a 4.2% reduction in mortality at 1 year, though this number did not reach statistical significance in the study. Nonetheless, these findings showcase the potential benefits of PHECG, particularly for achieving timely reperfusion treatment. Other investigators have replicated these findings and have also demonstrated that pre-hospital notification to prepare receiving teams in the ED and activate catheterization laboratories prior to arrival compounds the reduction in time-to-treatment.19 25 This has translated into improved adherence to guideline-recommended treatment time goals, including door-to-fibrinolysis within 30 min and FMC-to-balloon within 90 min.19 30 31 A recent meta-analysis confirmed these findings in a pooled data sample from 26 studies: patients who received PHECG prior to PCI had a reduction in FMC-to-balloon time of 35 min and were more likely to be treated within the guideline-recommended 90 min timeframe.7
Research efforts in ELVO have focused on many of the same topics. As with AMI, meeting guideline-recommended treatment time goals, including door-to-tPA time under 60 min, onset-to-tPA under 3 to 4.5 hours, and onset-to-IAT under 6 hours, has critical implications for functional outcomes in ELVO.26–29 To meet these time goals, several groups have tested triage protocols for bypassing non-endovascular capable centers (nECC) in favor of endovascular capable centers (ECC).32–37 Preliminary results appear promising, and found that nECC bypass based on clinical symptomatology alone results in substantial reductions in time-to-treatment, despite relative increases in transport time.27 34–37 As a result, some state legislatures have passed laws to promote nECC bypass protocols.38 However, with the lack of a technology analogous to PHECG and the lack of reliable clinical tools to detect ELVO in the pre-hospital setting, the achievable results of nECC bypass based on clinical symptoms alone are inherently limited. In addition to nECC bypass, others have found success meeting guideline-recommended treatment time goals by streamlining ED team workflow and angiography suite activation.32 Therefore, based on experience gathered during PHECG testing and current research efforts in ELVO, future device testing should focus efforts on: 1) time delays to reperfusion therapy, including AMS transport times, inter-hospital transfer delays, ED team workflow, access to CTA imaging, and angiography suite activation efficiency; 2) adherence to guideline-recommended treatment time objectives39 and 3) integration of ELVO devices into nECC bypass protocols currently under development.34–37 40 41
Predictive metrics
Pre-hospital ECG
Numerous studies investigating the predictive metrics of PHECG exist in the literature and an effort to improve PHECG sensitivity and specificity has been a key focus of the field (table 1).42–47 Some have done so by applying computer software interpretation algorithms and deep learning artificial intelligence programs that aim to reduce interpretation error.43 48 Physician interpretation compares favorably, with high sensitivity, specificity, and positive predictive value, though computer interpretation is not far behind.44 47 49 50 In fact, computer interpretations in the pre-hospital setting can help to accurately triage patients to PCI centers with reasonable sensitivity and specificity.44 In the absence of available physician PHECG interpretation, it appears reasonable to rely on computerized PHECG interpretation to guide triage decisions and even direct initial medical therapy (ie, administration of aspirin and sublingual nitroglycerin), though this remains controversial.44–46 49 51 52
ECG testing metrics
Whether PHECG, or a future ELVO detection technology, can be relied on for triage and treatment decisions may depend on the regional prevalence of disease.46 Given the high costs of false-positive downstream ED resource and catheterization laboratory activation, reliance solely on technology should be carefully considered in patients with low pre-test probability.46 However, incidence and prevalence of ELVO and AMI in EMS patient populations is difficult to estimate and highly dependent on the population studied.46 53 54 Triage and treatment decisions may also depend on the rates of false device readings, though reasonable false-positive and false-negative rates are to be expected. False catheterization laboratory STEMI activations occur in approximately 15% of cases.55 56 False readings and associated hospital resource activation should, therefore, be studied carefully in ELVO device investigations. To facilitate such studies, CTA imaging to confirm presence of ELVO will be needed as the reference standard.
Altogether, PHECG shows a sensitivity that ranges from 68.6% to 92.8%, a specificity that ranges from 81.2% to 98.9%, and a positive predictive value that ranges from 51% to 87% (table 1). However, higher-end sensitivity, in particular, is typically only achieved with the addition of supplemental wave criteria and the application of advanced formulas.42 43 57 When relying predominantly on ST-elevation criteria, the primary technique used by paramedics and computerized PHECG algorithms, sensitivity generally occupies the lower range, around 65% to 68%.57 58 Nonetheless, ECG is accepted and encouraged by the AHA/ACC, and used by EMS and physicians to aid field-based triage decisions. Therefore, these ranges may set a preliminary benchmark for predictive metrics that an ELVO detection device should be capable of achieving.
Emergent large vessel occlusion: clinical grading scales
While technology analogous to PHECG has yet to emerge for ELVO, several clinical grading scales have been published to assist EMS personnel in identifying ELVO and guiding triage decisions in the field. These scales include RACE, CPSS, LAMS, FAST-ED, and PASS, each of which were specifically designed to identify ELVO and divert patients to the most appropriate hospital.59–64 The predictive abilities of these scales are moderate at best, with sensitivities ranging from 55% to 85%, specificities that range from 40% to 89%, and positive predictive values that range from 42% to 72% (table 2). Each of the scales includes a different combination of clinical grading criteria, but no single scale has demonstrated clear superiority. As such, modifications to grading criteria are unlikely to achieve significant improvement in reliably detecting ELVO. Proper administration of clinical grading scales requires dedicated training: some authors have demonstrated that careful education can optimize administration accuracy and inter-rater reliability.34 65 However, there exists no consensus regarding which clinical grading scales are most useful, and, therefore, few are used consistently in practice. Clinical grading scales, likely due to their inherently subjective nature, appear to have a lower inter-rater reliability than do human interpretations of ECG readouts.65 66 As such, technology capable of detecting ELVO in the pre-hospital setting will likely be expected to achieve predictive metrics and inter-rater reliabilities that outperform clinical grading scales. Complete replacement of clinical grading scales, however, is unlikely. Rather, much like with PHECG, diagnostic yield and triage decision-making with a pre-hospital ELVO detection device will likely benefit from combining its reading with clinical grading scale results.
Emergent large vessel occlusion clinical grading scale predictive metrics
Discussion
Currently, there exists uncertainty within hospital networks regarding the most efficient routing strategies for patients presenting with symptoms indicative of ELVO, a topic which has become the subject of great interest since the advent of endovascular thrombectomy.40 41 A device with prediction metrics similar to that for ST-elevation detection with PHECG would be of significant benefit, particularly for aiding patient triage in the pre-hospital setting. The ability to accurately distinguish AIS from stroke mimics and ELVO from non-ELVO AIS would fundamentally improve the efficiency of pre-hospital triage and inter-hospital routing protocols. As PHECG did for AMI, this may assist in more effectively delivering patients with ELVO to medical centers with the capability of providing appropriate ELVO treatment and may help in reducing the delay to definitive care.
While drawing on the PHECG experience is undoubtedly helpful and EMS appear to encounter chest pain and neurological complaints with similar frequency,67 68 there are several important considerations that are unique to ELVO. ELVO treatment must be tailored for each and every patient. The robustness of collateral circulation, the presence of hemorrhagic conversion, the size of the infarct core, the volume of salvageable penumbra, and the patient’s premorbid functional status each represent crucial information for decision-making and treatment selection.39 As a result, not all ELVO patients are good candidates for endovascular thrombectomy and, therefore, not all ELVOs are treated by the same method. Because of this, certain pre-hospital protocols that work well for PHECG in AMI may not translate for ELVO.
For example, in-transit decisions to bypass nECCs in favor of ECCs for ELVO treatment may reach their limit in cost-benefit ratio in certain regions.40 41 The AHA/ASA Mission LifeLine Statement, recently recommended a 15 minute upper limit for additional transport time to bypass nECCs,69 just half of the 30 minute recommendation published by the AHA/ACC for non-PCI center bypass in AMI.70 Whether this 15 minute timeframe is appropriate based on the literature is a matter of debate.71 Furthermore, neurointerventional angiography suite activation from the ambulance may not have the same cost-benefit relationship as catheterization laboratory activation does in AMI. Instead, as the evidence suggests, optimization of ED workflow may be a more efficient method of streamlining care delivery in ELVO.32 33
Conclusions
Technology able to detect ELVO in the pre-hospital setting has yet to come to market. However, the methods needed to test ELVO devices, once available, can be drawn from the experience of testing PHECG for AMI detection. The prediction metrics expected of an ELVO device can be estimated from the experience detecting ST-elevation with PHECG, generally with a sensitivity in the 65% to 70% range and a specificity in the 75% to 80% range. In this review, we have provided an outline of the lessons learned from PHECG testing and the considerations to be weighed during the future testing of pre-hospital ELVO detection devices.
References
Footnotes
Contributors All authors contributed to the manuscript through manuscript composition and critical review. All authors provided final approval for publication.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.