Vulnerable Plaque Is More Prevalent in Male Individuals at High Risk of Stroke: A Propensity Score-Matched Study

Application of machine learning algorithms in predicting carotid artery plaques using routine health assessments

Background

Cardiovascular diseases (CVD) constitute a grave global health challenge, engendering significant socio-economic repercussions. Carotid artery plaques (CAP) are critical determinants of CVD risk, and proactive screening can substantially mitigate the frequency of cardiovascular incidents. However, the unequal distribution of medical resources precludes many patients from accessing carotid ultrasound diagnostics. Machine learning (ML) offers an effective screening alternative, delivering accurate predictions without the need for advanced diagnostic equipment. This study aimed to construct ML models that utilize routine health assessments and blood biomarkers to forecast the onset of CAP.

Methods

In this study, seven ML models, including LightGBM, LR, multi-layer perceptron (MLP), NBM, RF, SVM, and XGBoost, were used to construct the prediction model, and their performance in predicting the risk of CAP was compared. Data on health checkups and biochemical indicators were collected from 19,751 participants at the Beijing MJ Health Screening Center for model training and validation. Of these, 6,381 were diagnosed with CAP using carotid ultrasonography. In this study, 21 indicators were selected. The performance of the models was evaluated using the accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), F1 score, and area under the curve (AUC) value.

Results

Among the seven ML models, the light gradient boosting machine (LightGBM) had the highest AUC value (85.4%). Moreover, age, systolic blood pressure (SBP), gender, low-density lipoprotein cholesterol (LDL-C), and total cholesterol (CHOL) were the top five predictors of carotid plaque formation.

Conclusions

This study demonstrated the feasibility of predicting carotid plaque risk using ML algorithms. ML offers effective tools for improving public health monitoring and risk assessment, with the potential to improve primary care and community health by identifying high-risk individuals and enabling proactive healthcare measures and resource optimization.

Serum neurofilament light chain levels are associated with early neurological deterioration in minor ischemic stroke

Objectives

Patients with minor ischemic stroke (MIS) frequently suffer from early neurological deterioration (END) and become disabled. Our study aimed to explore the association between serum neurofilament light chain (sNfL) levels and END in patients with MIS.

Methods

We conducted a prospective observational study in patients with MIS [defined as a National Institutes of Health Stroke Scale (NIHSS) score 0–3] admitted within 24 h from the onset of symptoms. sNfL levels were measured at admission. The primary outcome was END, defined as an increase in the NIHSS score by ≥2 points within 5 days after admission. Univariate and multivariate analyses were performed to explore the risk factors associated with END. Stratified analyses and interaction tests were conducted to identify variables that might modify the association between sNfL levels and END.

Results

A total of 152 patients with MIS were enrolled, of which 24 (15.8%) developed END. The median sNfL level was 63.1 [interquartile range (IQR), 51.2–83.4] pg/ml on admission, which was significantly higher than that of 40 age- and sex-matched healthy controls (median 47.6, IQR 40.8–56.1 pg/ml; p < 0.001). Patients with MIS with END had a higher level of sNfL (with ND: median 74.1, IQR 59.5–89.8 pg/ml; without END: median 61.2, IQR 50.5–82.2 pg/ml; p = 0.026). After adjusting for age, baseline NIHSS score, and potential confounding factors in multivariate analyses, an elevated sNfL level (per 10 pg/mL) was associated with an increased risk of END [odds ratio (OR) 1.35, 95% confidence interval (CI) 1.04–1.77; p = 0.027). Stratified analyses and interaction tests demonstrated that the association between sNfL and END did not change by age group, sex, baseline NIHSS score, Fazekas' rating scale, hypertension, diabetes mellitus, intravenous thrombolysis, and dual antiplatelet therapy in patients with MIS (all p for interaction > 0.05). END was associated with an increased risk of unfavorable outcomes (modified Rankin scale score ranging from 3 to 6) at 3 months.

Conclusion

Early neurological deterioration is common in minor ischemic stroke and is associated with poor prognosis. The elevated sNfL level was associated with an increased risk of early neurological deterioration in patients with minor ischemic stroke. sNfL might be a promising biomarker candidate that can help to identify patients with minor ischemic stroke at high risk of neurological deterioration, for reaching individual therapeutic decisions in clinical practice.

Sex-specific association between inflammation and endothelial function relevant gene and vulnerable carotid plaque

Objectives: We aimed to explore sex-specific association between genes involved in inflammation and endothelial function and vulnerable carotid plaque, a subclinical precursor of ischemic stroke.

Methods: Carotid plaque and plaque phenotype were assessed by carotid ultrasound in high-risk participants for stroke drawn from a multicenter, cross-sectional survey in southwestern China. We examined 18 single nucleotide polymorphisms (SNPs) in 10 genes related to inflammation and endothelial function. Sex differences in the genotype of the candidate SNPs and risk of vulnerable carotid plaques were assessed. Interaction tests were performed to identify the SNPs that might modify the association between the sex and vulnerable plaques. For SNPs with suggestive evidence for interaction with sex (p for interaction<0.05), stratification analysis by sex was performed to evaluate the sex-specific association between the SNP and vulnerable plaques.

Results: 2,644 high-risk individuals were enrolled, comprising 1,202 (45.5%) men and 1,442 (54.5%) women. Vulnerable carotid plaques were detected in 425 (16.1%) participants. Among candidate SNPs, the genotype frequencies of 5 SNPs (TNFSF4 rs11811788, TNFSF4 rs1234313, IL6R rs4845625, VCAM1 rs2392221, and ITGA2 rs1991013) were significantly different between sex (all p < 0.05). Univariable and multivariable analyses suggested that male individuals had a significantly higher prevalence of vulnerable carotid plaques (20.0% vs. 12.8%, adjusted OR 1.72, 95% CI 1.12–2.66, p = 0.014), while none of the candidate SNPs was significantly associated with vulnerable plaques (all p > 0.05). Interaction tests found the association between sex and vulnerable plaques is affected by the genotype of IL6R rs4845625 (p for interaction = 0.031). Stratification analysis revealed a strong association between IL6R rs4845625 and vulnerable carotid plaque in man (dominant model TT vs. CT + CC: adjusted OR 1.52, 95% CI 1.12–2.07, p = 0.007; codominant model TT vs. CC: adjusted OR 1.50, 95% CI 1.00–2.25, p = 0.048) but not in women (p > 0.05 in all genetic models).

Conclusion: The rs4845625 polymorphism in IL6R has sex-specific effects on vulnerable carotid plaque in Chinese Han high-risk individuals for stroke. Our findings provide a plausible genetic basis underlying the sex difference in carotid plaque vulnerability.