Life's Essential 8 Cardiovascular Health Score and Cardiorespiratory Fitness in the Community.

BACKGROUND: The relation of cardiorespiratory fitness (CRF) to lifestyle behaviors and factors linked with cardiovascular health remains unclear. We aimed to understand how the American Heart Association's Life's Essential 8 (LE8) score (and its changes over time) relate to CRF and complementary exercise measures in community-dwelling adults. METHODS AND RESULTS: Framingham Heart Study (FHS) participants underwent maximum effort cardiopulmonary exercise testing for direct quantification of peak oxygen uptake (V̇O2). A 100-point LE8 score was constructed as the average across 8 factors: diet, physical activity, nicotine exposure, sleep, body mass index, lipids, blood glucose, and blood pressure. We related total LE8 score, score components, and change in LE8 score over 8 years with peak V̇O2 (log-transformed) and complementary CRF measures. In age- and sex-adjusted linear models (N=1838, age 54±9 years, 54% women, LE8 score 76±12), a higher LE8 score was associated favorably with peak V̇O2, ventilatory efficiency, resting heart rate, and blood pressure response to exercise (all P<0.0001). A clinically meaningful 5-point higher LE8 score was associated with a 6.0% greater peak V̇O2 (≈1.4 mL/kg per minute at sample mean). All LE8 components were significantly associated with peak V̇O2 in models adjusted for age and sex, but blood lipids, diet, and sleep health were no longer statistically significant after adjustment for all LE8 components. Over an ≈8-year interval, a 5-unit increase in LE8 score was associated with a 3.7% higher peak V̇O2 (P<0.0001). CONCLUSIONS: Higher LE8 score and improvement in LE8 over time was associated with greater CRF, highlighting the importance of the LE8 factors in maintaining CRF.

Association of Life's Essential 8 With Cardiovascular Disease and Mortality: The Framingham Heart Study.

BACKGROUND: The association of the American Heart Association's updated cardiovascular health score, the Life's Essential 8 (LE8), with cardiovascular disease (CVD) and death is not described in the FHS (Framingham Heart Study). METHODS AND RESULTS: We evaluated Framingham Offspring participants at examinations 2 and 6 (n=2888 and 1667; and mean age, 44 and 57 years, respectively), free of CVD with information on LE8 components. Using age-sex-adjusted Cox models, we related LE8 and its change (examination 2 to examination 6) with CVD and death risk and compared associations with those of the Life's Simple 7 score. Mean LE8 score at examination 2 was 67 points (minimum, 26 points; maximum, 100 points). At both examinations, participants were reclassified to a different cardiovascular health status, depending on which method (LE8 versus Life's Simple 7) was used (60% of participants in ideal Life's Simple 7 status were in intermediate LE8 category). On follow-up after examination 2 (median, 30 and 33 years for CVD and death, respectively), we observed 966 CVD events, and 1195 participants died. Participants having LE8≥68 (sample median) were at lower CVD and death risk compared with those with LE8<68 (examination 2: CVD hazard ratio [HR], 0.47 [95% CI, 0.41-0.54]; death HR, 0.55 [95% CI, 0.49-0.62]; all P<0.001). Participants maintaining low LE8 scores during life course were at highest CVD and death risk (CVD: HRs ranging from 1.8 to 2.3; P<0.001; death HR, 1.45 [95% CI, 1.13-1.85]; P=0.003 versus high-high group). CONCLUSIONS: Further studies are warranted to elucidate whether the LE8 score is a better marker of CVD and death risk, compared with Life's Simple 7 score.

A machine learning eye movement detection algorithm using electrooculography.

STUDY OBJECTIVES: Eye movement quantification in polysomnograms (PSG) is difficult and resource intensive. Automated eye movement detection would enable further study of eye movement patterns in normal and abnormal sleep, which could be clinically diagnostic of neurologic disorders, or used to monitor potential treatments. We trained a long short-term memory (LSTM) algorithm that can identify eye movement occurrence with high sensitivity and specificity. METHODS: We conducted a retrospective, single-center study using one-hour PSG samples from 47 patients 18-90 years of age. Team members manually identified and trained an LSTM algorithm to detect eye movement presence, direction, and speed. We performed a 5-fold cross validation and implemented a "fuzzy" evaluation method to account for misclassification in the preceding and subsequent 1-second of gold standard manually labeled eye movements. We assessed G-means, discrimination, sensitivity, and specificity. RESULTS: Overall, eye movements occurred in 9.4% of the analyzed EOG recording time from 47 patients. Eye movements were present 3.2% of N2 (lighter stages of sleep) time, 2.9% of N3 (deep sleep), and 19.8% of REM sleep. Our LSTM model had average sensitivity of 0.88 and specificity of 0.89 in 5-fold cross validation, which improved to 0.93 and 0.92 respectively using the fuzzy evaluation scheme. CONCLUSION: An automated algorithm can detect eye movements from EOG with excellent sensitivity and specificity. Noninvasive, automated eye movement detection has several potential clinical implications in improving sleep study stage classification and establishing normal eye movement distributions in healthy and unhealthy sleep, and in patients with and without brain injury.

Quantitative pupillometry and radiographic markers of intracranial midline shift: A pilot study.

Background: Asymmetric pupil reactivity or size can be early clinical indicators of midbrain compression due to supratentorial ischemic stroke or primary intraparenchymal hemorrhage (IPH). Radiographic midline shift is associated with worse functional outcomes and life-saving interventions. Better understanding of quantitative pupil characteristics would be a non-invasive, safe, and cost-effective way to improve identification of life-threatening mass effect and resource utilization of emergent radiographic imaging. We aimed to better characterize the association between midline shift at various anatomic levels and quantitative pupil characteristics. Methods: We conducted a multicenter retrospective study of brain CT images within 75 min of a quantitative pupil observation from patients admitted to Neuro-ICUs between 2016 and 2020 with large (>1/3 of the middle cerebral artery territory) acute supratentorial ischemic stroke or primary IPH > 30 mm3. For each image, we measured midline shift at the septum pellucidum (MLS-SP), pineal gland shift (PGS), the ratio of the ipsilateral to contralateral midbrain width (IMW/CMW), and other exploratory markers of radiographic shift/compression. Pupil reactivity was measured using an automated infrared pupillometer (NeurOptics®, Inc.), specifically the proprietary algorithm for Neurological Pupil Index® (NPi). We used rank-normalization and linear mixed-effects models, stratified by diagnosis and hemorrhagic conversion, to test associations of radiographic markers of shift and asymmetric pupil reactivity (Diff NPi), adjusting for age, lesion volume, Glasgow Coma Scale, and osmotic medications. Results: Of 53 patients with 74 CT images, 26 (49.1%) were female, and median age was 67 years. MLS-SP and PGS were greater in patients with IPH, compared to patients with ischemic stroke (6.2 v. 4.0 mm, 5.6 v. 3.4 mm, respectively). We found no significant associations between pupil reactivity and the radiographic markers of shift when adjusting for confounders. However, we found potentially relevant relationships between MLS-SP and Diff NPi in our IPH cohort (ß = 0.11, SE 0.04, P = 0.01), and PGS and Diff NPi in the ischemic stroke cohort (ß = 0.16, SE 0.09, P = 0.07). Conclusion: We found the relationship between midline shift and asymmetric pupil reactivity may differ between IPH and ischemic stroke. Our study may serve as necessary preliminary data to guide further prospective investigation into how clinical manifestations of radiographic midline shift differ by diagnosis and proximity to the midbrain.

Dexmedetomidine and Other Analgosedatives Alter Pupil Characteristics in Critically Ill Patients.

In critically ill patients with neurologic disease, pupil examination abnormalities can signify evolving intracranial pathology. Analgesic and sedative medications (analgosedatives) target pupillary pathways, but it remains unknown how analgosedatives alter pupil findings in the clinical care setting. We assessed dexmedetomidine and other analgosedative associations with pupil reactivity and size in a heterogeneous cohort of critically ill patients with acute intracranial pathology. DESIGN: Retrospective cohort study. SETTING: Two neurologic ICUs between 2016 and 2018. PATIENTS: Critically ill adult patients with pupil measurements within 60 minutes of analgosedative administration. Patients with a history of intrinsic retinal pathology, extracranial injury, inaccessible brain imaging, or no Glasgow Coma Scale (GCS) data were excluded. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: We used mixed-effects linear regression accounting for intrapatient correlations and adjusting for sex, age, GCS score, radiographic mass effect, medication confounders, and ambient light. We tested the association between an initiation or increased IV infusion of dexmedetomidine and pupil reactivity (Neurologic Pupil Index [NPi]) and resting pupil size (mm) obtained using NeurOptics NPi-200 (NeurOptics, Irvine, CA) pupillometer. Of our 221 patients with 9,897 pupil observations (median age, 60 [interquartile range, 50-68]; 59% male), 37 patients (166 pupil observations) were exposed to dexmedetomidine. Dexmedetomidine was associated with higher average NPi (ß = 0.18 per 1 unit increase in rank-normalized NPi ± 0.04; p < 0.001) and smaller pupil size (ß = -0.25 ± 0.05; p < 0.001). Exploratory analyses revealed that acetaminophen was associated with higher average NPi (ß = 0.04 ± 0.02; p = 0.02) and that most IV infusion analgosedatives including propofol, fentanyl, and midazolam were associated with smaller pupil size. CONCLUSIONS: Dexmedetomidine is associated with higher pupil reactivity (high NPi) and smaller pupil size in a cohort of critically ill patients with neurologic injury. Familiarity with expected pupil changes following analgosedative administration is important for accurate interpretation of pupil examination findings, facilitating optimal management of patients with acute intracranial pathology.

Natural Language Processing of Radiology Reports to Detect Complications of Ischemic Stroke.

BACKGROUND: Abstraction of critical data from unstructured radiologic reports using natural language processing (NLP) is a powerful tool to automate the detection of important clinical features and enhance research efforts. We present a set of NLP approaches to identify critical findings in patients with acute ischemic stroke from radiology reports of computed tomography (CT) and magnetic resonance imaging (MRI). METHODS: We trained machine learning classifiers to identify categorical outcomes of edema, midline shift (MLS), hemorrhagic transformation, and parenchymal hematoma, as well as rule-based systems (RBS) to identify intraventricular hemorrhage (IVH) and continuous MLS measurements within CT/MRI reports. Using a derivation cohort of 2289 reports from 550 individuals with acute middle cerebral artery territory ischemic strokes, we externally validated our models on reports from a separate institution as well as from patients with ischemic strokes in any vascular territory. RESULTS: In all data sets, a deep neural network with pretrained biomedical word embeddings (BioClinicalBERT) achieved the highest discrimination performance for binary prediction of edema (area under precision recall curve [AUPRC] > 0.94), MLS (AUPRC > 0.98), hemorrhagic conversion (AUPRC > 0.89), and parenchymal hematoma (AUPRC > 0.76). BioClinicalBERT outperformed lasso regression (p < 0.001) for all outcomes except parenchymal hematoma (p = 0.755). Tailored RBS for IVH and continuous MLS outperformed BioClinicalBERT (p < 0.001) and linear regression, respectively (p < 0.001). CONCLUSIONS: Our study demonstrates robust performance and external validity of a core NLP tool kit for identifying both categorical and continuous outcomes of ischemic stroke from unstructured radiographic text data. Medically tailored NLP methods have multiple important big data applications, including scalable electronic phenotyping, augmentation of clinical risk prediction models, and facilitation of automatic alert systems in the hospital setting.

Anisocoria and Poor Pupil Reactivity by Quantitative Pupillometry in Patients With Intracranial Pathology.

OBJECTIVES: To describe the prevalence and associated risk factors of new onset anisocoria (new pupil size difference of at least 1 mm) and its subtypes: new onset anisocoria accompanied by abnormal and normal pupil reactivities in patients with acute neurologic injuries. DESIGN: We tested the association of patients who experienced new onset anisocoria subtypes with degree of midline shift using linear regression. We further explored differences between quantitative pupil characteristics associated with first-time new onset anisocoria and nonnew onset anisocoria at preceding observations using mixed effects logistic regression, adjusting for possible confounders. SETTING: All quantitative pupil observations were collected at two neuro-ICUs by nursing staff as standard of care. PATIENTS: We conducted a retrospective two-center study of adult patients with intracranial pathology in the ICU with at least a 24-hour stay and three or more quantitative pupil measurements between 2016 and 2018. MEASUREMENTS AND MAIN RESULTS: We studied 221 patients (mean age 58, 41% women). Sixty-three percent experienced new onset anisocoria. New onset anisocoria accompanied by objective evidence of abnormal pupil reactivity occurring at any point during hospitalization was significantly associated with maximum midline shift (ß = 2.27 per mm; p = 0.01). The occurrence of new onset anisocoria accompanied by objective evidence of normal pupil reactivity was inversely associated with death (odds ratio, 0.34; 95% CI, 0.16-0.71; p = 0.01) in adjusted analyses. Subclinical continuous pupil size difference distinguished first-time new onset anisocoria from nonnew onset anisocoria in up to four preceding pupil observations (or up to 8 hr prior). Minimum pupil reactivity between eyes also distinguished new onset anisocoria accompanied by objective evidence of abnormal pupil reactivity from new onset anisocoria accompanied by objective evidence of normal pupil reactivity prior to first-time new onset anisocoria occurrence. CONCLUSIONS: New onset anisocoria occurs in over 60% of patients with neurologic emergencies. Pupil reactivity may be an important distinguishing characteristic of clinically relevant new onset anisocoria phenotypes. New onset anisocoria accompanied by objective evidence of abnormal pupil reactivity was associated with midline shift, and new onset anisocoria accompanied by objective evidence of normal pupil reactivity had an inverse relationship with death. Distinct quantitative pupil characteristics precede new onset anisocoria occurrence and may allow for earlier prediction of neurologic decline. Further work is needed to determine whether quantitative pupillometry sensitively/specifically predicts clinically relevant anisocoria, enabling possible earlier treatments.

Machine learning and natural language processing methods to identify ischemic stroke, acuity and location from radiology reports.

Accurate, automated extraction of clinical stroke information from unstructured text has several important applications. ICD-9/10 codes can misclassify ischemic stroke events and do not distinguish acuity or location. Expeditious, accurate data extraction could provide considerable improvement in identifying stroke in large datasets, triaging critical clinical reports, and quality improvement efforts. In this study, we developed and report a comprehensive framework studying the performance of simple and complex stroke-specific Natural Language Processing (NLP) and Machine Learning (ML) methods to determine presence, location, and acuity of ischemic stroke from radiographic text. We collected 60,564 Computed Tomography and Magnetic Resonance Imaging Radiology reports from 17,864 patients from two large academic medical centers. We used standard techniques to featurize unstructured text and developed neurovascular specific word GloVe embeddings. We trained various binary classification algorithms to identify stroke presence, location, and acuity using 75% of 1,359 expert-labeled reports. We validated our methods internally on the remaining 25% of reports and externally on 500 radiology reports from an entirely separate academic institution. In our internal population, GloVe word embeddings paired with deep learning (Recurrent Neural Networks) had the best discrimination of all methods for our three tasks (AUCs of 0.96, 0.98, 0.93 respectively). Simpler NLP approaches (Bag of Words) performed best with interpretable algorithms (Logistic Regression) for identifying ischemic stroke (AUC of 0.95), MCA location (AUC 0.96), and acuity (AUC of 0.90). Similarly, GloVe and Recurrent Neural Networks (AUC 0.92, 0.89, 0.93) generalized better in our external test set than BOW and Logistic Regression for stroke presence, location and acuity, respectively (AUC 0.89, 0.86, 0.80). Our study demonstrates a comprehensive assessment of NLP techniques for unstructured radiographic text. Our findings are suggestive that NLP/ML methods can be used to discriminate stroke features from large data cohorts for both clinical and research-related investigations.