Flow-mediated vasodilation was found to be an independent predictor of changes in the carotid plaque status during a 5-year follow-up

Combinatorial Use of Machine Learning and Logistic Regression for Predicting Carotid Plaque Risk Among 5.4 Million Adults With Fatty Liver Disease Receiving Health Check-Ups: Population-Based Cross-Sectional Study

BACKGROUND

Carotid plaque can progress into stroke, myocardial infarction, etc, which are major global causes of death. Evidence shows a significant increase in carotid plaque incidence among patients with fatty liver disease. However, unlike the high detection rate of fatty liver disease, screening for carotid plaque in the asymptomatic population is not yet prevalent due to cost-effectiveness reasons, resulting in a large number of patients with undetected carotid plaques, especially among those with fatty liver disease.

OBJECTIVE

This study aimed to combine the advantages of machine learning (ML) and logistic regression to develop a straightforward prediction model among the population with fatty liver disease to identify individuals at risk of carotid plaque.

METHODS

Our study included 5,420,640 participants with fatty liver from Meinian Health Care Center. We used random forest, elastic net (EN), and extreme gradient boosting ML algorithms to select important features from potential predictors. Features acknowledged by all 3 models were enrolled in logistic regression analysis to develop a carotid plaque prediction model. Model performance was evaluated based on the area under the receiver operating characteristic curve, calibration curve, Brier score, and decision curve analysis both in a randomly split internal validation data set, and an external validation data set comprising 32,682 participants from MJ Health Check-up Center. Risk cutoff points for carotid plaque were determined based on the Youden index, predicted probability distribution, and prevalence rate of the internal validation data set to classify participants into high-, intermediate-, and low-risk groups. This risk classification was further validated in the external validation data set.

RESULTS

Among the participants, 26.23% (1,421,970/5,420,640) were diagnosed with carotid plaque in the development data set, and 21.64% (7074/32,682) were diagnosed in the external validation data set. A total of 6 features, including age, systolic blood pressure, low-density lipoprotein cholesterol (LDL-C), total cholesterol, fasting blood glucose, and hepatic steatosis index (HSI) were collectively selected by all 3 ML models out of 27 predictors. After eliminating the issue of collinearity between features, the logistic regression model established with the 5 independent predictors reached an area under the curve of 0.831 in the internal validation data set and 0.801 in the external validation data set, and showed good calibration capability graphically. Its predictive performance was comprehensively competitive compared with the single use of either logistic regression or ML algorithms. Optimal predicted probability cutoff points of 25% and 65% were determined for classifying individuals into low-, intermediate-, and high-risk categories for carotid plaque.

CONCLUSIONS

The combination of ML and logistic regression yielded a practical carotid plaque prediction model, and was of great public health implications in the early identification and risk assessment of carotid plaque among individuals with fatty liver.

Relationship between Flow-mediated Endothelial Vasodilation and the Pulse Wave Velocity, and Cervical Carotid Artery Stenosis

Carotid artery stenosis is elicited by atherosclerosis and is the main cause of cerebral thrombosis. Flow-mediated endothelial vasodilation (FMD) can be measured noninvasively to assess vascular endothelial function related to atherosclerosis. The pulse wave velocity (PWV) is used to evaluate the vascular media involved in atherosclerosis. We investigated the relationship between these measurements in 75 consecutive patients with atherosclerotic cerebral thrombosis. They were assigned to three equal groups based on the severity of carotid artery stenosis on ultrasonograms. Group 1 had no stenosis, group 2 manifested moderate stenosis (<60%), and group 3 presented with severe stenosis (≥60%). We compared the FMD and PWV among the three groups. The PWV was significantly lower in group 1 than the other two groups. The FMD was significantly lower in group 3; it was significantly lower in group 2 than group 1. There was an inverse correlation between the FMD and the severity of carotid artery stenosis. Our findings show that for assessing the severity of carotid artery stenosis, the FMD is more useful than the PWV.

Longitudinal association among endothelial function, arterial stiffness and subclinical organ damage in hypertension

OBJECTIVES

To examine the longitudinal mutual association between endothelial dysfunction and arterial stiffness, and also to determine which of the two variables was more closely associated with the progression of subclinical organ damage.

METHODS

The brachial-ankle pulse wave velocity (baPWV), carotid intima-media thickness (CIMT), estimated glomerular filtration rate, microalbuminuria and flow-mediated vasodilatation of the brachial artery (FMD) were measured three times at 1.5-year intervals in 674 Japanese patients receiving antihypertensive treatment.

RESULTS

The change of the baPWV during the study period was larger in the subjects with baseline FMD values in the lowest tertile as compared to those with baseline FMD values in the highest tertile. The change of the CIMT was smaller in the subjects with baseline baPWV values in the lowest tertile than in those with baseline baPWV values in the highest tertile. After the adjustment, the FMD value at the baseline was inversely associated with the baPWV at the end of the study period (beta=-0.07, p=0.01), although, the reverse association was not significant. The baPWV, but not the FMD value, at the baseline was associated with the CIMT (beta=0.06, p=0.04) measured at the end of the study period.

CONCLUSIONS

In hypertension, endothelial dysfunction was associated with the progression of arterial stiffness, although the reverse association was not confirmed. The increased arterial stiffness rather than endothelial dysfunction may be more closely associated with the progression of atherosclerotic vascular damage, and the endothelial dysfunction-arterial stiffness-atherosclerosis continuum may be important in hypertension.

The Effect of Cilostazol on Endothelial Function as Assessed by Flow-Mediated Dilation in Patients with Coronary Artery Disease

Aim: The vascular endothelium plays a key role in the pathophysiology of atherosclerosis. Flow-mediated dilation (FMD) is a novel way of assessing endothelial function. Cilostazol is a unique antiplatelet drug that also has the potential to improve endothelial function. The objective of this present study was to investigate the effects of cilosatzol on endothelial function as assessed by FMD.

Methods: Fifty-one patients with coronary artery disease (CAD) were assigned to one of two groups: the Cilostazol(+) group (with cilostazol) and Cilostazol(−) group (without cilostazol). In addition to conventional dual antiplatelet therapy with aspirin and clopidogrel/ticlopidine, the Cilostazol(+) group (n = 27) was also given cilostazol (100 mg/day). The Cilostazol(−) group (n = 24) did not receive cilostazol. FMD was assessed at enrollment and after 6–9 months.

Results: The FMD of both the Cilostazol(+) and Cilostazol(−) groups remained similar at 5.2 (interquartile range: 3.8–8.5) to 5.4 (interquartile range: 4.2–6.7) (P = 0.29) and 5.0 (interquartile range: 3.6–6.4) to 4.9 (interquartile range: 4.0–7.0) (P = 0.38), respectively. However, the diameters of the baseline and maximal brachial arteries tended to increase in the Cilostazol(−) group (baseline: 4.2 ± 0.7 to 4.4 ± 0.7, P = 0.18; maximal: 4.5 ± 0.7 to 4.6 ± 0.7 P = 0.22), whereas that of the Cilostazol(−) group tended to decrease (baseline: 4.1 ± 0.6 to 3.9 ± 0.5, P = 0.10; maximal: 4.3 ± 0.7 to 4.1 ± 0.5, P = 0.05). The rates of change in the baseline diameter (Cilostazol(+): 3.7 ± 9.8% vs. Cilostazol(−): −3.8 ± 12.2%, P = 0.03) and maximal diameter (Cilostazol(+): +3.1 ± 8.9% vs. Cilostazol(−): −4.4 ± 12.0%, P = 0.02) were significantly different.

Conclusion: Although cilostazol didn't affect the FMD, there was a significant difference in the rates of change in baseline and maximal brachial artery diameter. This may have a beneficial effect in patients with cardiovascular disease.

Vascular Function and Intima-media Thickness of a Leg Artery in Peripheral Artery Disease: A Comparison of Buerger Disease and Atherosclerotic Peripheral Artery Disease

Aim: Both vascular function and structure are independent predictors of cardiovascular events. The purpose of this study was to evaluate vascular function and structure of a leg artery in patients with peripheral artery disease (PAD).

Methods: We measured flow-mediated vasodilatation (FMD) and nitroglycerine-induced vasodilation (NID) as indices of vascular function and intima-media thickness (IMT) as an index of vascular structure of the popliteal artery in 100 subjects, including 20 patients with Buerger disease and 30 patients with atherosclerotic PAD, 20 age- and sex-matched subjects without Buerger disease (control group) and 30 age- and sex-matched patients without atherosclerotic PAD (control group).

Results: IMT was significantly larger in the Buerger group than in the control group (Buerger, 0.63 ± 0.20 mm; control, 0.50 ± 0.07 mm; P = 0.01), whereas there were no significant differences in FMD and NID between the two groups. IMT was significantly larger in the atherosclerotic PAD group than in the control group (atherosclerotic PAD, 0.80 ± 0.22 mm; control, 0.65 ± 0.14 mm; P < 0.01), and FMD and NID were significantly smaller in the atherosclerotic PAD group than in the control group (FMD: atherosclerotic PAD, 3.9% ± 1.1%; control, 5.0% ± 1.8%; P < 0.01; and NID: atherosclerotic PAD, 6.1% ± 2.0%; control, 8.4% ± 2.1%; P < 0.01).

Conclusion: These findings suggest that vascular function is preserved in patients with Buerger disease and that both vascular function and vascular structure are impaired in patients with atherosclerotic PAD.

Endothelial and Metabolic Function Interactions in Overweight/Obese Children

AIM

Although the underlined mechanisms are still unknown, metabolic/coagulation alterations related to childhood obesity can induce vascular impairments. The aim of this study was to investigate the relationship between metabolic/coagulation parameters and endothelial function/vascular morphology in overweight/obese children.

METHODS

Thirty-five obese/overweight children (22 pre-pubertal, mean age: 9.52±3.35 years) were enrolled. Body mass index (BMI), homeostasis model assessment index (HOMAIR), metabolic and coagulation parameters, [adiponectin, fibrinogen, high molecular weight adiponectin (HMW), endothelin-1, and vonWillebrand factor antigen] ultrasound early markers of atherosclerosis [flow-mediated dilatation (FMD), common carotid intima-media thickness (C-IMT), and anteroposterior diameter of infra-renal abdominal aorta (APAO)] were assessed.

RESULTS

APAO was related to anthropometric (age: r=0.520, p=0.001; height: r=0.679, p＜0.001; weight: r=0.548, p=0.001; BMI: r=0.607, p＜0.001; SBP: r=0.377, p=0.026) and metabolic (HOMAIR: r=0.357, p=0.035; HMW: r=－0.355, p=0.036) parameters. Age, height, and systolic blood pressure were positively related to increased C-IMT (r=0.352, p=0.038; r=0.356, p=0.036; r=0.346, p=0.042, respectively). FMD was not related to any clinical and biochemical characteristics of the pediatric population. Age, HOMAIR, fasting glucose levels, and HMW were independent predictors for APAO increase. Each unit decrease in HMW concentrations (1 μg/ml) induced a 0.065 mm increase in APAO.

CONCLUSION

High molecular weight adiponectin is related to cardiovascular risk in overweight/obese children.

Comprehensive assessment of vascular health in patients; towards endothelium-guided therapy

Endothelial function has diagnostic, prognostic and therapeutic significance. A number of non-invasive techniques were introduced for its assessment, including flow-mediated dilation (FMD), finger plethysmography (RH-PAT) and digital thermal monitoring (DTM). All these methods can be performed simultaneously. In addition, various methods for measuring arterial wall stiffness are available such as: pulse wave analysis (PWA), pulse wave velocity (PWV), pulse contour analysis (PCA) and carotid wall distensibility coefficient (DC). Finally, carotid intima-media thickness (cIMT) and ankle brachial index (ABI) are used as surrogate read-outs of atherosclerosis. Here, we briefly describe the advantages, limitations and interrelationships of various methods used for the assessment of endothelial function, arterial stiffness, and present the concept of an integrated evaluation of vascular health based on multiple methods. This strategy may be useful to stratify cardiovascular risk and represents a step towards multiparametric assessment of endothelium for effective endothelium-guided therapy in patients with cardiovascular diseases.