The establishment of a telestroke service using multimodal CT imaging decision assistance: "Turning on the fog lights"

Global stroke statistics 2023: Availability of reperfusion services around the world

BACKGROUND

Disparities in the availability of reperfusion services for acute ischemic stroke are considerable globally and require urgent attention. Contemporary data on the availability of reperfusion services in different countries are used to provide the necessary evidence to prioritize where access to acute stroke treatment is needed.

AIMS

To provide a snapshot of published literature on the provision of reperfusion services globally, including when facilitated by telemedicine or mobile stroke unit services.

METHODS

We searched PubMed to identify original articles, published up to January 2023 for the most recent, representative, and relevant patient-level data for each country. Keywords included thrombolysis, endovascular thrombectomy and telemedicine. We also screened reference lists of review articles, citation history of articles, and the gray literature. The information is provided as a narrative summary.

RESULTS

Of 11,222 potentially eligible articles retrieved, 148 were included for review following de-duplications and full-text review. Data were also obtained from national stroke clinical registry reports, Registry of Stroke Care Quality (RES-Q) and PRE-hospital Stroke Treatment Organization (PRESTO) repositories, and other national sources. Overall, we found evidence of the provision of intravenous thrombolysis services in 70 countries (63% high-income countries (HICs)) and endovascular thrombectomy services in 33 countries (68% HICs), corresponding to far less than half of the countries in the world. Recent data (from 2019 or later) were lacking for 35 of 67 countries with known year of data (52%). We found published data on 74 different stroke telemedicine programs (93% in HICs) and 14 active mobile stroke unit pre-hospital ambulance services (80% in HICs) around the world.

CONCLUSION

Despite remarkable advancements in reperfusion therapies for stroke, it is evident from available patient-level data that their availability remains unevenly distributed globally. Contemporary published data on availability of reperfusion services remain scarce, even in HICs, thereby making it difficult to reliably ascertain current gaps in the provision of this vital acute stroke treatment around the world.

Influence of vascular imaging acquisition at local stroke centers on workflows in the drip-n-ship model: a RACECAT post hoc analysis

BACKGROUND

The influence of vascular imaging acquisition on workflows at local stroke centers (LSCs) not capable of performing thrombectomy in patients with a suspected large vessel occlusion (LVO) stroke remains uncertain. We analyzed the impact of performing vascular imaging (VI+) or not (VI- at LSC arrival on variables related to workflows using data from the RACECAT Trial.

OBJECTIVE

To compare workflows at the LSC among patients enrolled in the RACECAT Trial with or without VI acquisition.

METHODS

We included patients with a diagnosis of ischemic stroke who were enrolled in the RACECAT Trial, a cluster-randomized trial that compared drip-n-ship versus mothership triage paradigms in patients with suspected acute LVO stroke allocated at the LSC. Outcome measures included time metrics related to workflows and the rate of interhospital transfers and thrombectomy among transferred patients.

RESULTS

Among 467 patients allocated to a LSC, vascular imaging was acquired in 277 patients (59%), of whom 198 (71%) had a LVO. As compared with patients without vascular imaging, patients in the VI+ group were transferred less frequently as thrombectomy candidates to a thrombectomy-capable center (58% vs 74%, P=0.004), without significant differences in door-indoor-out time at the LSC (median minutes, VI+ 78 (IQR 69-96) vs VI- 76 (IQR 59-98), P=0.6). Among transferred patients, the VI+ group had higher rate of thrombectomy (69% vs 55%, P=0.016) and shorter door to puncture time (median minutes, VI+ 41 (IQR 26-53) vs VI- 54 (IQR 40-70), P<0.001).

CONCLUSION

Among patients with a suspected LVO stroke initially evaluated at a LSC, vascular imaging acquisition might improve workflow times at thrombectomy-capable centers and reduce the rate of futile interhospital transfers. These results deserve further evaluation and should be replicated in other settings and geographies.

Electrocardiographic Gating and Cerebral Perfusion Computed Tomography Option-Set Prevalence and Utilization Data From 62 Institutions in the United States

OBJECTIVES

To provide the radiology community with data to address the question: "Compared with peer institutions, is my institution efficiently using its electrocardiographic (ECG) gating and cerebral perfusion-capable computed tomography (CT) scanners?"

METHODS

In this retrospective study, we analyze 6 months of scanner utilization data from 62 institutions (299 locations, 507 scanners) to identify scanners capable of performing ECG gating and perfusion CT studies. We report the number of ECG gating/perfusion-capable scanners and locations as a function of the total number of locations and scanners in each institution. We additionally regress the number of ECG-gated and perfusion examinations on (1) the number of locations/scanners capable of performing these examinations and (2) the fraction of the institution's CT examination volume that requires ECG gating or perfusion. We provide look-up tables so an institution can compare its ECG-gated/perfusion examination volume to other institutions with similar ECG-gated/perfusion examination fractions and capable scanners.

RESULTS

We detected an effect of both ECG-gating examination fraction and the number of ECG gating-capable scanners on ECG-gated examination volume ( χ21 = 77.5 [ P < 0.001] and χ21 = 64.2 [ P < 0.001], respectively). Similar results were obtained for perfusion examination fraction and perfusion-capable scanners as they relate to perfusion examination volume ( χ21 = 51.6 [ P < 0.001] and χ21 = 45.2 [ P < 0.001], respectively). The number of ECG gating/perfusion-capable scanners and locations within an institution were found to positively correlate with both the total number of locations and scanners within an institution ( P < 0.001 for all hypothesis tests).

CONCLUSIONS

The study provides multi-institutional data on ECG gating and perfusion examination volumes that can be used to inform CT purchasing decisions.

Diagnostic Utility of Computed Tomography Perfusion in the Telestroke Setting

BACKGROUND

Definitive diagnosis of acute ischemic stroke is challenging, particularly in telestroke settings. Although the prognostic utility of CT perfusion (CTP) has been questioned, its diagnostic value remains under-appreciated, especially in cases without an easily visible intracranial occlusion. We assessed the diagnostic accuracy of routine CTP in the acute telestroke setting.

METHODS

Acute and follow-up data collected prospectively from consecutive suspected patients with stroke assessed by a state-wide telestroke service between March 2020 and August 2021 at 12 sites in Australia were analyzed. All patients in the final analysis had been assessed with multimodal CT, including CTP, which was post-processed with automated volumetric software. Diagnostic sensitivity and specificity were calculated for multimodal CT and each individual component (noncontrast CT [NCCT], CT angiogram [CTA], and CTP). Final diagnosis determined by consensus review of follow-up imaging and clinical data was used as the reference standard.

RESULTS

During the study period, complete multimodal CT examination was obtained in 831 patients, 457 of whom were diagnosed with stroke. Diagnostic sensitivity for ischemic stroke increased by 19.5 percentage points when CTP was included with NCCT and CTA compared with NCCT and CTA alone (73.1% positive with NCCT+CTA+CTP [95% CI, 68.8-77.1] versus 53.6% positive with NCCT+CTA alone [95% CI, 48.9-58.3], P<0.001). No difference was observed between specificities of NCCT+CTA and NCCT+CTA+CTP (98.7% [95% CI, 98.5-100] versus 98.7% [95% CI, 96.9-99.6], P=0.13). Multimodal CT, including CTP, demonstrated the highest negative predictive value (75.0% [95% CI, 72.1-77.7]). Patients with stroke not evident on CTP had small volume infarcts on follow-up (1.2 mL, interquartile range 0.5-2.7mL).

CONCLUSIONS

Acquisition of CTP as part of a telestroke imaging protocol permits definitive diagnosis of cerebral ischemia in 1 in 5 patients with normal NCCT and CTA.

TACTICS - Trial of Advanced CT Imaging and Combined Education Support for Drip and Ship: evaluating the effectiveness of an 'implementation intervention' in providing better patient access to reperfusion therapies: protocol for a non-randomised controlled stepped wedge cluster trial in acute stroke

INTRODUCTION

Stroke reperfusion therapies, comprising intravenous thrombolysis (IVT) and/or endovascular thrombectomy (EVT), are best practice treatments for eligible acute ischemic stroke patients. In Australia, EVT is provided at few, mainly metropolitan, comprehensive stroke centres (CSC). There are significant challenges for Australia's rural and remote populations in accessing EVT, but improved access can be facilitated by a 'drip and ship' approach. TACTICS (Trial of Advanced CT Imaging and Combined Education Support for Drip and Ship) aims to test whether a multicomponent, multidisciplinary implementation intervention can increase the proportion of stroke patients receiving EVT.

METHODS AND ANALYSIS

This is a non-randomised controlled, stepped wedge trial involving six clusters across three Australian states. Each cluster comprises one CSC hub and a minimum of three primary stroke centre (PSC) spokes. Hospitals will work in a hub and spoke model of care with access to a multislice CT scanner and CT perfusion image processing software (MIStar, Apollo Medical Imaging). The intervention, underpinned by behavioural theory and technical assistance, will be allocated sequentially, and clusters will move from the preintervention (control) period to the postintervention period.

PRIMARY OUTCOME

Proportion of all stroke patients receiving EVT, accounting for clustering.

SECONDARY OUTCOMES

Proportion of patients receiving IVT at PSCs, proportion of treated patients (IVT and/or EVT) with good (modified Rankin Scale (mRS) score 0-2) or poor (mRS score 5-6) functional outcomes and European Quality of Life Scale scores 3 months postintervention, proportion of EVT-treated patients with symptomatic haemorrhage, and proportion of reperfusion therapy-treated patients with good versus poor outcome who presented with large vessel occlusion at spokes.

ETHICS AND DISSEMINATION

Ethical approval has been obtained from the Hunter New England Human Research Ethics Committee (18/09/19/4.13, HREC/18/HNE/241, 2019/ETH01238). Trial results will be disseminated widely through published manuscripts, conference presentations and at national and international platforms regardless of whether the trial was positive or neutral.

TRIAL REGISTRATION NUMBER

ACTRN12619000750189; UTNU1111-1230-4161.

The Role of Vascular Imaging atReferral Centers in the Drip and Ship Paradigm

BACKGROUND

In drip-and-ship protocols, non-invasive vascular imaging (NIVI) at Referral Centers (RC), although recommended, is not consistently performed and its value is uncertain. We evaluated the role of NIVI at RC, comparing patients with (VI+) and without (VI-) vascular imaging in several outcomes.

METHODS

Observational, multicenter study from a prospective government-mandated population-based registry of code stroke patients. We selected acute ischemic stroke patients, initially assessed at RC from January-2016 to June-2020. We compared and analyzed the rates of patients transferred to a Comprehensive Stroke Center (CSC) for Endovascular Treatment (EVT), rates of EVT and workflow times between VI+ and VI- patients.

RESULTS

From 5128 ischemic code stroke patients admitted at RC; 3067 (59.8%) were VI+, 1822 (35.5%) were secondarily transferred to a CSC and 600 (11.7%) received EVT. Among all patients with severe stroke (NIHSS ≥16) at RC, a multivariate analysis showed that lower age, thrombolytic treatment, and VI+ (OR:1.479, CI95%: 1.117-1.960, p=0.006) were independent factors associated to EVT. The rate of secondary transfer to a CSC was lower in VI+ group (24.6% vs. 51.6%, p<0.001). Among transferred patients, EVT was more frequent in VI+ than VI- (48.6% vs. 21.7%, p<0.001). Interval times as door-in door-out (median-minutes 83.5 vs. 82, p= 0.13) and RC-Door to puncture (median-minutes 189 vs. 178, p= 0.47) did not show differences between both groups.

CONCLUSION

In the present study, NIVI at RC improves selection for EVT, and is associated with receiving EVT in severe stroke patients. Time-metrics related to drip-and-ship model were not affected by NIVI.

Considerations for the Implementation of a Telestroke Network: A Systematic Review

The application of telestroke has matured considerably since its inception in 1999. The use of telestroke is now recommended in several published guidelines. Consequently, jurisdictions without a telestroke service are seeking practical information on the best approach to implement telestroke. French et al. (2013) reviewed the challenges of implementing a telestroke network including studies between 2000 and 2010. At the time, telestroke networks were largely limited to the UK, USA, Canada and Europe and only one process evaluation had been conducted. Given the prolific expansion of telestroke services since 2010, we conducted a systematic review to determine factors associated with successful establishment, management, and sustainability of a contemporary telestroke services. A comprehensive search of telestroke studies was conducted in July 2021. Empirical studies published between 2010 and 2021 were included if they contained descriptive, evaluation or operational data on the implementation of a telestroke network. Studies were subsequently evaluated using the Consolidated Framework for Implementation Research (CFIR). The initial literature search revealed a total of 7415 potential studies; 38 of which met the inclusion criteria. The past decade of process evaluation studies has enabled a more nuanced investigations into how to implement and sustain a telestroke network. Pre-implementation planning is crucial to ensure clear telestroke processes, governance structures and stakeholder engagement. Sustainability of networks relies on securing long-term investment, providing adequate resources, and maintaining staff motivation and willingness. Recommendations are provided to overcome commonly identified barriers related to technology, staffing, planning and standardisation of processes, evaluation, and sustainability and scale-up. Further research needs to explore how new advancements in stroke care such as endovascular clot retrieval (EVT) and advanced brain imaging can be considered and planned for during the implementation of a new telestroke service.

Improving Access to Care: Telemedicine Across Medical Domains

Over the past 20 years, the use of telemedicine has increased exponentially. Its fundamental aim is to improve access to care. In this review, we assess the extent to which telemedicine has fulfilled this promise across medical domains. Additionally, we assess whether telemedicine has improved related health outcomes. Finally, we determine who has benefited from this novel form of health care delivery. A review of the literature indicates that (a) telemedicine has improved access to care for a wide range of clinical conditions ranging from stroke to pregnancy; (b) telemedicine in select circumstances has demonstrated improved health outcomes; and (c) telemedicine has addressed geographical, but less so social, barriers to care. For telemedicine to fulfill its promise, additional evidence needs to be gathered on health outcomes and cost savings, the digital divide needs to be bridged, and policy changes that support telemedicine reimbursement need to be enacted.

The Need for Structured Strategies to Improve Stroke Care in a Rural Telestroke Network in Northern New South Wales, Australia: An Observational Study

Introduction: A telestroke network in Northern New South Wales, Australia has been developed since 2017. We theorized that the telestroke network development would drive a progressive improvement in stroke care metrics over time.

Aim: This study aimed to describe changes in acute stroke workflow metrics over time to determine whether they improved with network experience.

Methods: We prospectively collected data of patients assessed by telestroke who received multimodal computed tomography (mCT) and were diagnosed with ischemic stroke or transient ischemic attack from January 2017 to July 2019. The period was divided into two phases (phase 1: January 2017 – October 2018 and phase 2: November 2018 – July 2019). We compared median door-to-call, door-to-image, and door-to-decision time between the two phases.

Results: We included 433 patients (243 in phase 1 and 190 in phase 2). Each spoke site treated 1.5–5.2 patients per month. There were Door-to-call time (median 39 in phase 1, 35 min in phase 2, p = 0.18), and door-to-decision time (median 81.5 vs. 83 min, p = 0.31) were not improved significantly. Similarly, in the reperfusion therapy subgroup, door-to-call time (median 29 vs. 24.5 min, p = 0.12) and door-to-decision time (median 70.5 vs. 67.5 min, p = 0.75) remained substantially unchanged. Regression analysis showed no association between time in the network and door-to-decision time (coefficient 1.5, p = 0.32).

Conclusion: In our telestroke network, acute stroke timing metrics did not improve over time. There is the need for targeted education and training focusing on both stroke reperfusion competencies and the technical aspects of telestroke in areas with limited workforce and high turnover.

Test characteristics of a 5-element cortical screen for identifying anterior circulation large vessel occlusion ischemic strokes

OBJECTIVE

Stroke severity screens typically include cortical signs, such as field cut, aphasia, neglect, gaze preference, and dense hemiparesis (FANG-D). The accuracy and reliability of these signs, when assessed by emergency physicians, to identify patients with anterior circulation large vessel occlusion (ACLVO) acute ischemic stroke (AIS) is unknown. We hypothesized that the FANG-D screen applied by emergency physicians would be sensitive and reliable for identifying ACLVO AIS.

METHODS

We conducted a prospective cohort study enrolling consecutive patients with suspected AIS presenting within 4.5 hours of last known well to the emergency department (ED). Emergency physicians performed the FANG-D screen prior to, and blinded to the results of, imaging. The imaging standard was defined as a non-contrast computed tomography (CT) for identifying hemorrhage and CT angiography for identifying large vessel occlusion. ACLVO was defined as an occlusion of the internal carotid artery, the middle cerebral artery, or its first branch. A convenience sample of patients had a duplicate FANG-D screen performed by a second emergency physician to assess interobserver agreement.

RESULTS

We performed 608 FANG-D assessments on 491 patients presenting to the ED, of whom 64 (10%) had an ACLVO. FANG-D had a sensitivity of 91% (confidence interval [CI] = 81%-96%) and a specificity of 35% (CI = 31%-39%) for identifying ACLVO. Interobserver agreement was tested on 133 patients and was found to be substantial, with a Fleiss' kappa of 0.77 (CI = 0.64-0.88).

CONCLUSIONS

The FANG-D screen is a sensitive test for identifying ACLVO when performed by emergency physicians and demonstrates substantial interrater reliability.

Air vs. Road Decision for Endovascular Clot Retrieval in a Rural Telestroke Network

Background and Purpose: Telestroke aims to increase access to endovascular clot retrieval (ECR) for rural areas. There is limited information on transfer workflow for ECR in rural settings. We sought to describe the transfer metrics for ECR in a rural telestroke network with respect to decision making.

Methods: A retrospective cohort study was employed on consecutive patients transferred to the comprehensive stroke center (CSC) for ECR in a rural hub-and-spoke telestroke network between April 2013 and October 2019, by road or air. Key time-based metrics were analyzed.

Results: Sixty-two patients were included. Mean age was 66 years [standard deviation (SD), 14] and median National Institutes of Health Stroke Scale 13 [interquartile range (IQR), 8–18]. Median rural-hospital-door-to-CSC-door (D2D) was 308 min (IQR, 254–351), of which 68% was spent at rural hospitals [door-in-door-out (DIDO); 214 min; IQR, 171–247]. DIDO was longer for air transfers than road (P = 0.004), primarily because of a median 87 min greater decision-to-departure time (Decision-DO, P < 0.001). In multiple linear regression analysis, intubation but not thrombolysis was associated with significantly longer DIDO. The distance at which the extra speed of an aircraft made up for the delays involved in booking an aircraft was 299 km from the CSC.

Conclusions: DIDO is longer for air retrievals compared with road. Decision-DO represents the most important component of DIDO, being longer for air transfers. Systems for rapid transportation of rural ECR candidates need optimization for best patient outcomes, with decision support seen as a potential tool to achieve this.

No Evidence of the "Weekend Effect" in the Northern New South Wales Telestroke Network

Background: Admission outside normal business hours has been associated with prolonged door-to-treatment times and poorer patient outcomes, the so called "weekend effect. " This is the first examination of the weekend effect in a telestroke service that uses multi-modal computed tomography. Aims: To examine differences in workflow and triage between in-hours and out-of-hours calls to a telestroke service. Methods: All patients assessed using the Northern New South Wales (N-NSW) telestroke service from April 2013 to January 2019 were eligible for inclusion (674 in total; 539 with complete data). The primary outcomes measured were differences between in-hours and out-of-hours in door-to-call-to-decision-to-needle times, differences in the proportion of patients confirmed to have strokes or of patients selected for reperfusion therapies or patients with a modified Rankin Score (mRS ≤ 2) at 90 days. Results: There were no significant differences between in-hours and out-of-hours in any of the measured times, nor in the proportions of patients confirmed to have strokes (67.6 and 69.6%, respectively, p = 0.93); selected for reperfusion therapies (22.7 and 22.6%, respectively, p = 0.56); or independent at 3 months (34.8 and 33.6%, respectively, p = 0.770). There were significant differences in times between individual hospitals, and patient presentation more than 4.5 h after symptom onset was associated with slower times (21 minute delay in door-to-call, p = 0.002 and 22 min delay in door-to-image, p = 0.001). Conclusions: The weekend effect is not evident in the Northern NSW telestroke network experience, though this study did identify some opportunities for improvement in the delivery of acute stroke therapies.

Implementation of multimodal computed tomography in a telestroke network: Five-year experience

AIMS

Penumbral selection is best-evidence practice for thrombectomy in the 6-24 hour window. Moreover, it helps to identify the best responders to thrombolysis. Multimodal computed tomography (mCT) at the primary centre-including noncontrast CT, CT perfusion, and CT angiography-may enhance reperfusion therapy decision-making. We developed a network with five spoke primary stroke sites and assessed safety, feasibility, and influence of mCT in rural hospitals on decision-making for thrombolysis.

METHODS

Consecutive patients assessed via telemedicine from April 2013 to June 2018. Clinical outcomes were measured, and decision-making compared using theoretical models for reperfusion therapy applied without mCT guidance. Symptomatic intracranial hemorrhage (sICH) was assessed according to Safe Implementation of Treatments in Stroke Thrombolysis Registry criteria.

RESULTS

A total of 334 patients were assessed, 240 received mCT, 58 were thrombolysed (24.2%). The mean age of thrombolysed patients was 70 years, median baseline National Institutes of Health Stroke Scale was 10 (IQR 7-18) and 23 (39.7%) had a large vessel occlusion. 1.7% had sICH and 3.5% parenchymal hematoma. Three months poststroke, 55% were independent, compared with 70% in the non-thrombolysed group.

CONCLUSION

Implementation of CTP in rural centers was feasible and led to high thrombolysis rates with low rates of sICH.