Associations Between Vascular Risk Factor Levels and Cognitive Decline Among Stroke Survivors

Levine, Deborah A.; Chen, Bingxin; Gałecki, Andrzej T.; Gross, Alden L.; Briceño, Emily M.; Whitney, Rachael; Ploutz‐Snyder, Robert; Giordani, Bruno; Sussman, Jeremy B.; Burke, James F.; Lazar, Ronald M.; Howard, Virginia J.; Aparicio, Hugo J.; Beiser, Alexa; Elkind, Mitchell S.V.; Gottesman, Rebecca F.; Koton, Silvia; Pendlebury, Sarah T.; Sharma, Anita; Springer, Mellanie V.; Seshadri, Sudha; Romero, José R.; Hayward, Rodney A.

doi:10.1001/jamanetworkopen.2023.13879

Cited by 4 publications

(7 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Evidence can be found that post-stroke patients with elevated FBG may suffer from accelerated cognitive decline due to cerebral microvascular injury, inflammation, oxidative stress, and neurodegeneration [32].…”

Section: Discussionmentioning

confidence: 99%

Development and validation of a machine learning-based risk prediction model for post-stroke cognitive impairment

Zhong,

Li,

et al. 2024

Preprint

View full text Add to dashboard Cite

Background Machine learning (ML) risk prediction models for post-stroke cognitive impairment (PSCI) are still far from optimal. This study aims to generate a reliable predictive model for predicting PSCI in Chinese individuals using ML algorithms. Methods We collected data on 494 individuals who were diagnosed with acute ischemic stroke (AIS) and hospitalized for this condition from January 2022 to November 2023 at a Chinese medical institution. All of the observed samples were divided into a training set (70%) and a validation set (30%) at random. Logistic regression combined with the least absolute shrinkage and selection operator (LASSO) regression was utilized to efficiently screen the optimal predictive features of PSCI. We utilized seven different ML models (LR, XGBoost, LightGBM, AdaBoost, GNB, MLP, and SVM) and compared their performance for the resulting variables. We used five-fold cross-validation to measure the model's area under the curve (AUC), sensitivity, specificity, accuracy, F1 score and PR values. SHAP analysis provides a comprehensive and detailed explanation of our optimized model's performance. Results PSCI was identified in 58.50% of the 494 eligible AIS patients. The most predictive features of PSCI are HAMD-24, FBG, age, PSQI, and paraventricular lesion. The XGBoost model, among the 7 ML prediction models for PSCI developed based on the best predictive features, demonstrates superior performance, as indicated by its AUC (0.961), sensitivity (0.931), specificity (0.889), accuracy (0.911), F1 score (0.926), and AP value (0.967). Conclusion The XGBoost model developed on HAMD-24, FBG, age, PSQI, and paraventricular lesion performance is exceptional in predicting the risk of PSCI. It provide clinicians with a reliable tool for early screening of patients with cognitive impairment and effective treatment decisions in stroke patients.

show abstract

Section: Discussionmentioning

confidence: 99%

Development and validation of a machine learning-based risk prediction model for post-stroke cognitive impairment

Zhong,

Li,

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“… 10 These cohorts had ≥50 participants with physician-adjudicated acute stroke during follow-up and SBP and cognition measures pre- and post-stroke. 11 …”

Section: Methodsmentioning

confidence: 99%

“…We included participants with self-reported history of stroke at cohort baseline in the primary analysis as their inclusion did not change estimates of an acute stroke’s effect on cognitive trajectories controlling for pre-stroke cognition. 1 , 11 We performed a sensitivity analysis excluding participants with a selfreported history of stroke at cohort baseline.…”

Section: Methodsmentioning

confidence: 99%

“…10 These cohorts had ≥50 participants with physician-adjudicated acute stroke during follow-up and SBP and cognition measures pre-and post-stroke. 11 We included participants ≥18 years with expert-adjudicated acute ischemic or hemorrhagic stroke during follow-up, ≥1 pre-stroke and ≥1 post-stroke cognitive assessments, and ≥1 post-stroke SBP measurement at or before the last post-stroke cognitive assessment. We excluded participants with cohort-defined dementia at or before stroke and those reporting Hispanic ethnicity or race other than Black or White due to cohort design differences.…”

Section: Study Design Participants and Measurementsmentioning

confidence: 99%

See 1 more Smart Citation

Associations Between Stroke Type, Ischemic Stroke Subtypes, and Post-Stroke Cognitive Trajectories

Levine,

Whitney,

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

Background: It is unclear how post-stroke cognitive trajectories differ by stroke type and ischemic stroke subtype. We studied associations between stroke types (ischemic, hemorrhagic), ischemic stroke subtypes (cardioembolic, large artery atherosclerotic, lacunar/small vessel, cryptogenic/other determined etiology), and post-stroke cognitive decline. Methods: This pooled cohort analysis from four US cohort studies (1971-2019) identified 1,143 dementia-free individuals with acute stroke during follow-up: 1,061 (92.8%) ischemic, 82 (7.2%) hemorrhagic, 49.9% female, 30.8% Black. Median age at stroke was 74.1 (IQR, 68.6, 79.3) years. Outcomes were change in global cognition (primary) and changes in executive function and memory (secondary). Outcomes were standardized as T-scores (mean [SD], 50 [10]); a 1-point difference represents a 0.1-SD difference in cognition. Median follow-up for the primary outcome was 6.0 (IQR, 3.2, 9.2) years. Linear mixed-effects models estimated changes in cognition after stroke. Results: On average, the initial post-stroke global cognition score was 50.78 points (95% CI, 49.52, 52.03) in ischemic stroke survivors and did not differ in hemorrhagic stroke survivors (difference, -0.17 points [95% CI, -1.64, 1.30]; P=0.82) after adjusting for demographics and pre-stroke cognition. On average, ischemic stroke survivors showed declines in global cognition, executive function, and memory. Post-stroke declines in global cognition, executive function, and memory did not differ between hemorrhagic and ischemic stroke survivors. 955 ischemic strokes had subtypes: 200 (20.9%) cardioembolic, 77 (8.1%) large artery atherosclerotic, 207 (21.7%) lacunar/small vessel, 471 (49.3%) cryptogenic/other determined etiology. On average, small vessel stroke survivors showed declines in global cognition and memory, but not executive function. Initial post-stroke cognitive scores and cognitive declines did not differ between small vessel survivors and survivors of other ischemic stroke subtypes. Post-stroke vascular risk factor levels did not attenuate associations. Conclusion: Stroke survivors had cognitive decline in multiple domains. Declines did not differ by stroke type or ischemic stroke subtype.

show abstract

“…In the Original Investigation “Associations Between Vascular Risk Factor Levels and Cognitive Decline Among Stroke Survivors,” published May 17, 2023, there were errors in the Methods and Results sections, the Figure, and Table 1. These sections should have stated that all observed glucose levels, rather than fasting glucose levels, were used.…”

mentioning

confidence: 99%

Errors in Methods and Results Sections, Figure, Table 1, and Supplement 1

2024

JAMA Netw Open

View full text Add to dashboard Cite

Associations Between Vascular Risk Factor Levels and Cognitive Decline Among Stroke Survivors

Cited by 4 publications

References 64 publications

Development and validation of a machine learning-based risk prediction model for post-stroke cognitive impairment

Development and validation of a machine learning-based risk prediction model for post-stroke cognitive impairment

Associations Between Stroke Type, Ischemic Stroke Subtypes, and Post-Stroke Cognitive Trajectories

Errors in Methods and Results Sections, Figure, Table 1, and Supplement 1

Contact Info

Product

Resources

About