Search

Introduction: Hospitals have improved stroke thrombolysis times through rigorous assessment of care delays. While mobile stroke units (MSU) expedite intravenous thrombolysis for acute ischemic stroke, the same rigor has not been applied to this novel setting. Methods: We reviewed all cases of intravenous tissue plasminogen activator (tPA) administration onboard our mobile stroke unit in Rochester, NY since its inception in 2019. The dispatch timeline was divided into six intervals: (1) drive time, (2) onboarding time, (3) HCT collection, (4) HCT reading, (5) decision time, and (6) tPA administration. Cases in which tPA was delayed due to medical contraindications or inability to obtain consent were not included in the analysis. Results: The mean ± SD total time between MSU dispatch and tPA was 42 ± 8.4 min (range: 26-60 min; N = 53). The longest interval was (1) drive time (between dispatch and arrival, 12.0 ± 4.7 min). However, the inter-case variation was greatest in (5) decision time (between HCT reading and treatment decision, 6.3 ± 6.2 min). After ranking cases by total time, the interval that had the widest variance between the first and fourth quartiles was decision time (2.45x). The other intervals varied by a factor of 1.50x for (6) tPA administration and less than 1.00x for the other four intervals. Decision time was the interval most closely correlated with total time (Spearman’s ρ = 0.46; p = 0.00051), which was reduced in our second year of operation (38.8 from 44.6 minutes; Mann-Whitney, Z = 2.38, p = 0.02). Decision time was negatively correlated with time since inception of the MSU (Spearman’s ρ = -0.29, p = 0.037), suggesting that improvements in decision time drove improvements in overall thrombolysis time. Conclusion: The largest driver of variation in total time from emergency alert to tPA administration is provider decision time in our MSU program. This observation should be investigated in other settings but may ultimately inform MSU training and staffing.

Introduction: The time from 911 activation (i.e. alarm) to administration of intravenous (IV) alteplase in acute ischemic stroke is associated with functional, patient centered-outcomes. Mobile stroke treatment units (MSTUs) have emerged as a stroke system tool that may hasten treatment times. Optimal workflows on MSTUs remain to fully elucidated. Methods: Retrospective review of a QA database of patients treated on a MSTU with door-to-needle-times (DTN) of Results: In October of 2018 the University of Rochester Medical Center launched a MSTU, initially operating on Monday through Friday in the city of Rochester from 8am until 4 pm. Over the initial 3 months, there were 96 MSTU responses leading to 54 transported patients, 3 of which were treated with IV alteplase. One patient with an initial NIHSS of 17 was treated with IV alteplase within 9 minutes of reaching the MSTU door. Workflow elements felt to hasten treatment included registration of the patient in the electronic health record (EHR) prior to the patient reaching the MSTU and enabling the telestroke provider to listen to the initial history and physical being performed by the MSTU RN at the scene. The later is accomplished by using a telestroke iPhone application that allows for a “3-way-call” between the MSTU RN, the MSTU, and the telestroke provider. The MSTU RN wears a Bluetooth earpiece that captures the conversation with the patient, witnesses on scene, and initial EMS responders and enables the MSTU RN to summarizes key history and exam findings, vitals, and blood glucose results while keeping their hands available for patient care. Simultaneously, the telestroke provider reviews the patient’s chart in the EHR for alteplase contraindications, prior imaging results, and pertinent medical history. Conclusion: Registering the patient in the EHR and integrating the telestroke provider into the initial patient assessment at the scene in the prehospital setting may allow for consistent door to needle times of < 15 minutes on MSTUs, which may further improve patient outcomes.

Introduction: Mechanical thrombectomy (MT) is effective for select acute ischemic strokes due to large vessel occlusion (LVO-AIS). Systems of care need to expeditiously identify, transfer, and treat qualifying LVO-AIS. Data are needed to define which ingredients are most effective when engineering LVO-AIS regional systems of care. Methods: Strong Memorial Hospital (SMH) is a Comprehensive Stroke Center in Rochester, NY serving twenty-two New York State designed stroke centers (NYS-DSC). Arnot Ogden Medical Center (AOMC), Cayuga Medical Center (CMC), and Geneva General Hospital (GGH) are NYS-DSCs located 115, 91, and 50 miles from SMH, respectively. Clinical leaders at each site collaborated to implement an integrated regional system of care for LVO-AIS, dubbed “Code LVO”, which includes the auto-launching of an interfacility transport to the referring hospital for presumed strokes with an NIHSS of >/= 10 and last known well time (LKWT) of /= 6 and proximal anterior circulation LVO on a CTA at the referring hospital to identify the door-in-door-out (DIDO) times, thrombectomy attempt rate, and mortality of patients before and after Code LVO implementation. Wilcoxon Rank-Sum was used to analyze median DIDO times and Fisher’s exact was used to analyze the proportion of DIDO times of < 90 minutes, < 60 minutes, thrombectomy attempt rate, and mortality. Results: There were 51 pre- Code LVO versus 12 post Code-LVO transfers. The median DIDO times were significantly reduced post-Code LVO (80 vs 127 minutes, p=0.001). The proportion of DIDO times < 90 minutes and < 60 minutes were significantly improved (58% vs 16%, p=0.005 and 17% vs 0%, p=0.034, respectively). Mortality was numerically, but not significantly, reduced (17% vs 22%). Median DIDO times were significantly shorter in those undergoing thrombectomy (97 vs 136 minutes, p=0.008) and numerically longer in those who died (138 vs 112 minutes, p=0.24). Conclusions: Auto-launching of interfacility transport within an integrated regional system care for LVO-AIS decreases DIDO times and may improve outcomes. Further study is needed to outline its value, in terms of patient outcomes, resource utilization, and safety.

Introduction: Mechanical thrombectomy (MT) is effective for select acute ischemic strokes due to large vessel occlusion (LVO-AIS). Systems of care need to expeditiously identify, transfer, and treat qualifying LVO-AIS. Data are needed to define which ingredients are most effective when engineering LVO-AIS regional systems of care. Methods: Strong Memorial Hospital (SMH) is a Comprehensive Stroke Center in Rochester, NY serving twenty-two New York State designed stroke centers (NYS-DSC). Arnot Ogden Medical Center (AOMC) is NYS-DSC located in Elmira, NY, 115 miles from SMH. Clinical leaders at SMH and AOMC collaborated to engineer a system of care to expeditiously transfer qualifying MT candidates. The system of care, dubbed Code LVO, included the auto-launching of an interfacility transport team (preferable a medical helicopter) to AOMC and notification to the SMH Vascular Neurologist (VN) on ED arrival for any presumed strokes with an NIHSS of >/= 10 and last known well time (LKWT) of /=6 and an LVO via a cloud-based image sharing platform. We retrospectively reviewed the records of LVO-AIS transfers in a QA database to identify the door-in-door-out (DIDO) time, AOMC door to SMH door (D2D) time, and AOMC door to SMH skin puncture (D2S) time, and the outcomes of patients before and after Code LVO implementation. Results: Over an 18-month period pre- Code LVO, there were seven AOMC to SMH LVO-AIS transfers. None underwent MT due to large or completed infarcts on SMH arrival. Seventy-one percent died or went to hospice. The median DIDO time was 93 minutes (range 66-273) and D2D was 239 minutes (range 112-392). Post- Code LVO, six Code LVO evaluations were requested, and three were transferred. All LVO-AIS underwent successful MT, with a median DIDO time of 45 minutes (range 44-60), D2D time of 95 minutes (range 85-102), and DTS time of 141 minutes (124-158) and went to nursing home or acute rehab. Conclusions: Auto-launching of interfacility transport to referring hospitals for select presumed LVO-AIS may decrease time to MT and improve outcomes. Further study is needed to outline its value, in terms of patient outcomes, resource utilization, and safety.

Introduction: Stroke care in the US is becoming increasingly regionalized. Many patients undergo interhospital transfer to obtain access to specialized, time-sensitive interventions such as mechanical thrombectomy. Methods: We used the stratified survey design of the US Nationwide Inpatient Sample, 2009-2014, to examine nationwide trends in interhospital transfers for ischemic stroke and the relationship between transfers and thrombectomy. International Classification of Disease—Ninth Revision diagnosis and procedure codes were used to identify acute stroke admissions and inpatient procedures (tissue plasminogen activator [tPA] administration and thrombectomy), respectively. The analysis was restricted to hospitals with the capabilities to perform thrombectomy. Trends were assessed using a Cochran-Armitage test. Results: From 2009 to 2014, an estimated 759,648 ischemic stroke admissions were identified within hospitals offering thrombectomy (102,674 in 2009; 154,280 in 2014). The proportion of stroke admissions to these hospitals that underwent interhospital transfer prior to arrival increased from 13.5% (n=13,811) in 2009 to 18.9% (n=19,175) in 2014 ( P trendP P P trend=0.016). Approximately 30% of patients receiving thrombectomy for ischemic stroke were transferred from another acute care facility (n=7,328). Among patients receiving tPA, transferred patients were slightly less likely to receive thrombectomy compared to those arriving by “front door” at thrombectomy performing centers (13.9% vs. 15.4%; P =0.004). Conclusions: From 2009 to 2014, interhospital stroke transfers to thrombectomy performing hospitals more than doubled. For every ~5 additional stroke transfers over the time period examined, one additional transferred patient received thrombectomy. As stroke systems of care continue to evolve across the US, the optimization of stroke transfers presents an opportunity to increase access to meaningful interventions such as thrombectomy.

Background: A subset of stroke patients treated with endovascular recanalization therapy suffers poor outcome despite good arterial recanalization. The aim of this study is to identify whether large ischemic lesion in deep white matter (DWM) on diffusion weighted imaging (DWI) prior to therapy predicts futile recanalization. Method: Consecutive acute stroke patients with anterior circulation ischemia whose artery had been recanalized at Thrombolysis in Myocardial Infarct (TIMI) flow grade 2 or 3 following endovascular therapy were included. All patients underwent CT and MRI before endovascular therapy. We defined large DWI - DWM lesion if DWI showed the hyperintense lesion in the DWM which mostly extended between anterior and posterior horn of the lateral ventricle. We identified Alberta Stroke Program Early CT Scores (ASPECTS) of CT and DWI, and volumes measured on initial DWI and mean transit time of perfusion weighted images. We assessed the prevalence of TIMI grade 3 recanalization and parenchymal hemorrhage type 2, and final ischemic lesion volume after endovascular therapy. Stroke severity was assessed by the National Institutes of Health Stroke Scale (NIHSS) score. Futile recanalization was defined as a 30-day modified Rankin Scale score 3 to 6 despite recanalization at TIMI grade 2 or 3. Univariate and multivariate regression analyses were performed to identify predictors of futile recanalization. Results: Partial or complete recanalization (TIMI grade 2 or 3) was observed in 35 of 46 patients (76 %). Of the 35 patients, futile recanalization was observed in 20 patients (57 %). Patients with futile recanalization had older age (median, 74 versus 58 years old, p = 0.053), higher initial NIHSS score (median, 17 versus 9, p = 0.042), pre-treatment large DWI - DWM lesion (45 versus 9 %, p = 0.022), and larger final ischemic lesion volume (median, 37 versus 9 ml, p = 0.003). ASPECTS of CT (median, 9 versus 9, p = 0.527) and DWI (median, 8 versus 8, p = 0.943) and initial ischemic lesion volumes on DWI (median, 21 versus 14 ml, p = 0.230) and on mean transit time (median, 189 versus 192 ml, p = 0.976) did not demonstrate statistically significance between patients with and without futile recanalization. There were also no significant differences between the outcome groups in times from stroke onset to imaging or endovascular therapy. Logistic regression analysis demonstrated that large DWI - DWM lesion was an independent predictor of futile recanalization (OR 10.40, 95 % CI 1.03 to 105.21, p = 0.047). Conclusion: Patients who have large pre-treatment DWI - DWM lesion may be poor candidates for endovascular therapy.

Background and Purpose— The introduction of balloon remodeling has revolutionized the approach to coiling of wide-neck aneurysms. We studied the effects of balloon inflation during coil embolization on ischemic complications. Methods— A retrospective review was undertaken of the most recent 147 patients undergoing balloon remodeling for unruptured intracranial aneurysm coil embolization at a single institution (81 balloon, 66 unassisted). All underwent postprocedural MRI. Results— Among patients in the “balloon” group, the mean total inflation time was 18 minutes (range, 1–43), a mean number of inflations of 4 (range, 1–9), a mean maximum single inflation time of 7 minutes (range, 1–19), a mean reperfusion time of 2.2 minutes between inflations, and an average procedure time of 2 hours and 10 minutes. Asymptomatic diffusion-weighted imaging abnormalities were detected on postprocedural MRI in 21.5% of patients and symptomatic lesions were identified in 3.8%. Both silent and symptomatic ischemic rates were similar in the internal control group. Patients with ischemic findings were older and more likely have diabetes; no differences were found with respect to total balloon inflation time, number of inflations, maximum inflation time, or reperfusion times. Conclusions— We found no significant relationship between balloon inflation practices and ischemic events. Older and diabetic patients were more likely to have ischemic events develop.