Search

Background Hypertension is a stroke risk factor with known disparities in prevalence and management between Black and White patients. We sought to identify if racial differences in presenting blood pressure (BP) during acute ischemic stroke exist. Methods and Results Adults with acute ischemic stroke presenting to an emergency department within 24 hours of last known normal during study epochs 2005, 2010, and 2015 within the Greater Cincinnati/Northern Kentucky Stroke Study were included. Demographics, histories, arrival BP, National Institutes of Health Stroke Scale score, and time from last known normal were collected. Multivariable linear regression was used to determine differences in mean BP between Black and White patients, adjusting for age, sex, National Institutes of Health Stroke Scale score, history of hypertension, hyperlipidemia, smoking, stroke, body mass index, and study epoch. Of 4048 patients, 853 Black and 3195 White patients were included. In adjusted analysis, Black patients had higher presenting systolic BP (161 mm Hg [95% CI, 159–164] versus 158 mm Hg [95% CI, 157–159], P

Background Dysphagia after stroke is common and can impact morbidity and death. The purpose of this population‐based study was to determine specific epidemiological and health risk factors that impact development of dysphagia after acute stroke. Methods and Results Ischemic and hemorrhagic stroke cases from 2010 and 2015 were identified via chart review from the GCNKSS (Greater Cincinnati Northern Kentucky Stroke Study), a representative sample of ≈1.3 million adults from southwestern Ohio and northern Kentucky. Dysphagia status was determined on the basis of clinical assessments and necessity for alternative access to nutrition via nasogastric or percutaneous endoscopic gastrostomy tube placement. Comparisons between patients with and without dysphagia were made to determine differences in baseline characteristics and premorbid conditions. Multivariable logistic regression determined factors associated with increased risk of dysphagia. Dysphagia status was ascertained from 4139 cases (1709 with dysphagia). Logistic regression showed that increased age, Black race, higher National Institutes of Health Stroke Scale score at admission, having a hemorrhagic stroke (versus infarct), and right hemispheric stroke increased the risk of developing dysphagia after stroke. Factors associated with reduced risk included history of high cholesterol, lower prestroke modified Rankin Scale score, and white matter disease. Conclusions This study replicated previous findings of variables associated with dysphagia (older age, worse stroke, right‐sided hemorrhagic lesions), whereas other variables identified were without clear biological rationale (eg, Black race, history of high cholesterol, and presence of white matter disease) and should be investigated in future studies to determine biological relevance and potential influence in stroke recovery.

Background The impact of mobile stroke units (MSUs) on outcomes in patients with large vessel occlusions eligible for endovascular thrombectomy (EVT) has yet to be characterized. Methods We completed a prespecified substudy of patients with EVT‐eligible stroke with anterior and posterior circulation large vessel occlusions on computed tomography and/or computed tomography angiography who were enrolled in BEST‐MSU (Benefits of Stroke Treatment using a Mobile Stroke Unit). Primary outcome was 90‐day utility‐weighted modified Rankin scale. Groups were compared using chi‐square or Fisher's exact tests for categorical variables, and 2‐sample t‐tests for continuous variables. Multiple logistic regression was used to assess the effect of MSU on binary outcomes after adjusting for other baseline factors. Results Of 1515 trial patients, 293 had large vessel occlusions eligible for EVT: 168 in the MSU group and 125 in the emergency medical services group. Baseline characteristics were comparable, with the exception of baseline National Institutes of Health Stroke Scale score (MSU median 19 [interquartile range 13, 23] versus emergency medical services 16 [11, 20], P = 0.002) and study site. The mean (±SD) score on the utility‐weighted modified Rankin scale at 90 days was 0.63±0.39 in MSU group and 0.51±0.41 in emergency medical services group (mean difference 0.13, 95% CI [0.03–0.22]). After adjustment, MSU had significantly higher odds of functional independence (odds ratio 2.60 [95% CI, 1.45–4.77], P = 0.002). Secondary outcomes also favored MSU: early neurologic recovery (30% improvement in National Institutes of Health Stroke Scale score at 24 hours) 68% versus 52%; adjusted odds ratio 1.98 [95% CI, 1.19–3.33]; time of tissue plasminogen activator bolus from symptom onset 65.0 minutes [50.5–92.0] versus 96.0 [79.3–130.0], P≤0.001. The groups had similar onset to arterial puncture (169.0 minutes [133.5, 210.0] versus 162.0 [135.0–207.0], P = 0.83). Conclusions In patients with EVT‐eligible large vessel occlusion stroke, MSU management was associated with better clinical outcomes compared with standard emergency medical services management. MSU management sped thrombolysis but did not expedite EVT treatment times. Future MSU processes should include efforts to capitalize on the potential of MSUs to provide earlier EVT.

Most clinical research stopped during COVID due to possible impact on data quality and personnel safety. We aimed to assess the impact of COVID on acute stroke clinical trial conduct at sites that continued to enroll patients during the pandemic. BEST-MSU is an ongoing study of Mobile Stroke Units (MSU) vs standard management of tPA-eligible acute stroke patients in the pre-hospital setting. MSU personnel include a vascular neurologist via telemedicine, and a nurse, CT technologist, paramedics and emergency medicine technicians on-board. During COVID, consent, 90-day modified Rankin Scale (mRS) and EQ5D were obtained by phone instead of in-person, but other aspects of management were similar to the pre-COVID period. We compared patient demographics, study metrics, and infection of study personnel during intra- vs pre-COVID eras. Five of 6 BEST-MSU sites continued to enroll during COVID. There were no differences in intra- (n = 57) vs pre- (n = 869) COVID enrolled tPA eligible patients' age, sex, race (38.6% vs 38.0% Black), ethnicity (15.8% vs 18.6% Hispanic), or NIHSS (median 11 vs 9). The percent of screened patients enrolled and adjudicated tPA eligible declined from 13.6% to 6.6% (p < .001); study enrollment correlated with local stay-at-home and reopening orders. There were no differences in alert to MSU arrival or arrival to tPA times, but MSU on-scene time was 5 min longer (p = .01). There were no differences in ED door to CT, tPA treatment or thrombectomy puncture times, hospital length of stay, discharge disposition, or remote vs in-person 90-day mRS or EQ5D. One MSU nurse tested positive but did not require hospitalization. Clinical research in the pre-hospital setting can be carried out accurately and safely during a pandemic. tPA eligibility rates declined, but otherwise there were no differences in patient demographics, deterioration of study processes, or serious infection of study staff. Trial registration: NCT02190500.

OBJECTIVE:1) To develop a machine learning approach for detecting stroke cases and subtypes from hospitalization data, 2) to assess algorithm performance and predictors on real-world data collected by a large-scale epidemiology study in the US; and 3) to identify directions for future development of high-precision stroke phenotypic signatures. MATERIALS AND METHODS:We utilized 8,131 hospitalization events (ICD-9 codes 430-438) collected from the Greater Cincinnati/Northern Kentucky Stroke Study in 2005 and 2010. Detailed information from patients' medical records was abstracted for each event by trained research nurses. By analyzing the broad list of demographic and clinical variables, the machine learning algorithms predicted whether an event was a stroke case and, if so, the stroke subtype. The performance was validated on gold-standard labels adjudicated by stroke physicians, and results were compared with stroke classifications based on ICD-9 discharge codes, as well as labels determined by study nurses. RESULTS:The best performing machine learning algorithm achieved a performance of 88.57%/93.81%/92.80%/93.30%/89.84%/98.01% (accuracy/precision/recall/F-measure/area under ROC curve/area under precision-recall curve) on stroke case detection. For detecting stroke subtypes, the algorithm yielded an overall accuracy of 87.39% and greater than 85% precision on individual subtypes. The machine learning algorithms significantly outperformed the ICD-9 method on all measures (P value