Search

BackgroundPrehospital automated large vessel occlusion (LVO) detection in Mobile Stroke Units (MSUs) could accelerate identification and treatment of patients with LVO acute ischemic stroke. Here, we evaluate the performance of a machine learning (ML) model on CT angiograms (CTAs) obtained from 2 MSUs to detect LVO.MethodsPatients evaluated on MSUs in Houston and Los Angeles with out-of-hospital CTAs were identified. Anterior circulation LVO was defined as an occlusion of the intracranial internal carotid artery, middle cerebral artery (M1 or M2), or anterior cerebral artery vessels and determined by an expert human reader. A ML model to detect LVO was trained and tested on independent data sets consisting of in-hospital CTAs and then tested on MSU CTA images. Model performance was determined using area under the receiver-operator curve statistics.ResultsAmong 68 patients with out-of-hospital MSU CTAs, 40% had an LVO. The most common occlusion location was the middle cerebral artery M1 segment (59%), followed by the internal carotid artery (30%), and middle cerebral artery M2 (11%). Median time from last known well to CTA imaging was 88.0 (interquartile range, 59.5-196.0) minutes. After training on 870 in-hospital CTAs, the ML model performed well in identifying LVO in a separate in-hospital data set of 441 images with area under receiver-operator curve of 0.84 (95% CI, 0.80-0.87). ML algorithm analysis time was under 1 minute. The performance of the ML model on the MSU CTA images was comparable with area under receiver-operator curve 0.80 (95% CI, 0.71-0.89). There was no significant difference in performance between the Houston and Los Angeles MSU CTA cohorts.ConclusionsIn this study of patients evaluated on MSUs in 2 cities, a ML algorithm was able to accurately and rapidly detect LVO using prehospital CTA acquisitions.

Background: We present results of the development and feasibility testing of CATCH Healthy Smiles, a school-based oral health program, among children in grades K-2 in Houston, Texas. Methods: Study design was cross-sectional (N = 2 schools; N = 125 parent-child dyads; 31 kindergarteners, 42 first graders, and 52 second graders). CATCH Healthy Smiles program was implemented by trained school teachers in the 2016-2017 school year. Trained dentists conducted dental assessments to measure dental caries increment score (d3mfs). Parent-reported 24-hour dietary recalls and surveys assessed child and parent behavioral, environmental, and psychosocial factors. Logistic regression analysis assessed factors associated with caries experience adjusting for covariates. Results: Of the 113 children with complete dental assessments, 54% children in grade K, 62% in first grade, and 73% in second grade had caries experience. Children with caries experience had a higher body weight (AdjOR = 1.13, 95% confidence interval [CI]: 1.02-1.29), were less likely to be girls (AdjOR = 0.22, 95% CI: 0.05-0.82), had greater odds of difficulty drinking hot or cold beverages because of dental problems (AdjOR = 13.13, 95% CI: 1.09-275.14), greater frequency of consuming sugar-sweetened beverages (AdjOR = 11.53, 95% CI: 2.10-87.19), greater odds of receiving government assistance (AdjOR = 14.62, 95% CI: 2.74-119.81), and lower odds of seeing a dental provider (AdjOR = 0.11, 95% CI: 0.02-0.45). Process evaluation showed that 100% of the CATCH Healthy Smiles lessons and activities were taught in the two schools with a high degree of program fidelity and acceptability across the schools, children, and parents. Conclusions: These data will be used to conduct a subsequent fully powered cluster randomized controlled trial.

Although atrial fibrillation (AF) guidelines indicate that pharmacological blockade of the renin-angiotensin system may be considered for primary AF prevention in hypertensive patients, previous studies have yielded conflicting results. We sought to determine whether randomization to lisinopril reduces incident AF or atrial flutter (AFL) compared with chlorthalidone in a large clinical trial cohort with extended post-trial surveillance. We performed a secondary analysis of the ALLHAT (Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial), a randomized, double-blind, active-controlled clinical trial that enrolled hypertensive individuals ≥55 years of age with at least one other cardiovascular risk factor. Participants were randomly assigned to receive amlodipine, lisinopril, or chlorthalidone. Individuals with elevated fasting low-density lipoprotein cholesterol levels were also randomized to pravastatin versus usual care. The primary outcome was the development of either AF or AFL as diagnosed by serial study ECGs or by Medicare claims data. Among 14 837 participants without prevalent AF or AFL, 2514 developed AF/AFL during a mean 7.5±3.2 years of follow-up. Compared with chlorthalidone, randomization to either lisinopril (hazard ratio, 1.04; 95% confidence interval, 0.94-1.15; P=0.46) or amlodipine (hazard ratio, 0.93; 95% confidence interval, 0.84-1.03; P=0.16) was not associated with a significant reduction in incident AF/AFL. Compared with chlorthalidone, treatment with lisinopril is not associated with a meaningful reduction in incident AF or AFL among older adults with a history of hypertension. URL: http://www.clinicaltrials.gov. Unique identifier: NCT00000542.

The effectiveness of COVID-19 vaccines depends on widespread vaccine uptake. Employing a telephone-administered weighted survey with 19,502 participants, we examined the determinants of COVID-19 vaccine acceptance among adults in Texas. We used multiple regression analysis with LASSO-selected variables to identify factors associated with COVID-19 vaccine uptake and intentions to receive the vaccine among the unvaccinated. The prevalence of unvaccinated individuals (22%) was higher among those aged 18–39, males, White respondents, English speakers, uninsured individuals, those facing financial challenges, and individuals expressing no concern about contracting the illness. In a fully adjusted regression model, higher odds of being unvaccinated were observed among males (aOR 1.11), the uninsured (aOR 1.38), smokers (aOR 1.56), and those facing financial struggles (aOR 1.62). Conversely, Asians, Blacks, and Hispanics were less likely to be unvaccinated compared to Whites. Among the unvaccinated, factors associated with stronger intent to receive the vaccine included age (over 65 years), Black and Hispanic ethnicity, and perceived risk of infection. Hispanic individuals, the uninsured, those covered by public insurance, and those facing financial challenges were more likely to encounter barriers to vaccine receipt. These findings underscore the importance of devising tailored strategies, emphasizing nuanced approaches that account for demographic, socioeconomic, and attitudinal factors in vaccine distribution and public health interventions. [ABSTRACT FROM AUTHOR]

Population enrichment designs can play a vital role in detecting treatment benefits, if present, by selecting a study population where the treatment is more likely to show a benefit compared to the unselected population. While population enrichment designs with one primary endpoint have been studied in the current literature, there are limited research on population enrichment with alternative primary endpoints. Many diseases have multiple sequelae and it can be sufficient to show treatment benefits on at least one of these consequences. In this manuscript, we provide the theoretical basis and development of a two-stage adaptive design with two binary alternative primary endpoints that simultaneously evaluates two subgroups for enrichment, selects one or both endpoints for efficacy assessments at the time of the interim analysis using pre-specified decision rules, and maintains an overall type I error rate. Treatment benefits can be declared in one or both endpoints in a selected subgroup or in the overall population at the time of interim or final analyses. We provide stage-wise boundary values for futility and efficacy and rejection probabilities under heterogeneous treatment effects in the subgroups. We further demonstrate the implementation of the proposed design and discuss venues for future development on this topic. [ABSTRACT FROM AUTHOR]

Objective: This study was undertaken to examine averted stroke in optimized stroke systems. Methods: This secondary analysis of a multicenter trial from 2014 to 2020 compared patients treated by mobile stroke unit (MSU) versus standard management. The analytical cohort consisted of participants with suspected stroke treated with intravenous thrombolysis. The main outcome was a tissue‐defined averted stroke, defined as a final diagnosis of stroke with resolution of presenting symptoms/signs by 24 hours attributed to thrombolysis and no acute infarction/hemorrhage on imaging. An additional outcome was stroke with early symptom resolution, defined as a final diagnosis of stroke with resolution of presenting symptoms/signs by 24 hours attributed to thrombolysis. Results: Among 1,009 patients with a median last known well to thrombolysis time of 87 minutes, 159 (16%) had tissue‐defined averted stroke and 276 (27%) had stroke with early symptom resolution. Compared with standard management, MSU care was associated with more tissue‐defined averted stroke (18% vs 11%, adjusted odds ratio [aOR] = 1.82, 95% confidence interval [CI] = 1.13–2.98) and stroke with early symptom resolution (31% vs 21%, aOR = 1.74, 95% CI = 1.12–2.61). The relationships between thrombolysis treatment time and averted/early recovered stroke appeared nonlinear. Most models indicated increased odds for stroke with early symptom resolution but not tissue‐defined averted stroke with earlier treatment. Additionally, younger age, female gender, hyperlipidemia, lower National Institutes of Health Stroke Scale, lower blood pressure, and no large vessel occlusion were associated with both tissue‐defined averted stroke and stroke with early symptom resolution. Interpretation: In optimized stroke systems, 1 in 4 patients treated with thrombolysis recovered within 24 hours and 1 in 6 had no demonstrable brain injury on imaging. ANN NEUROL 2024;95:347–361 [ABSTRACT FROM AUTHOR]