CIMT as a risk factor for stroke subtype: A systematic review

A Siamese ResNeXt network for predicting carotid intimal thickness of patients with T2DM from fundus images

Objective

To develop and validate an artificial intelligence diagnostic model based on fundus images for predicting Carotid Intima-Media Thickness (CIMT) in individuals with Type 2 Diabetes Mellitus (T2DM).

Methods

In total, 1236 patients with T2DM who had both retinal fundus images and CIMT ultrasound records within a single hospital stay were enrolled. Data were divided into normal and thickened groups and sent to eight deep learning models: convolutional neural networks of the eight models were all based on ResNet or ResNeXt. Their encoder and decoder modes are different, including the standard mode, the Parallel learning mode, and the Siamese mode. Except for the six unimodal networks, two multimodal networks based on ResNeXt under the Parallel learning mode or the Siamese mode were embedded with ages. Performance of eight models were compared via the confusion matrix, precision, recall, specificity, F1 value, and ROC curve, and recall was regarded as the main indicator. Besides, Grad-CAM was used to visualize the decisions made by Siamese ResNeXt network, which is the best performance.

Results

Performance of various models demonstrated the following points: 1) the RexNeXt showed a notable improvement over the ResNet; 2) the structural Siamese networks, which extracted features parallelly and independently, exhibited slight performance enhancements compared to the traditional networks. Notably, the Siamese networks resulted in significant improvements; 3) the performance of classification declined if the age factor was embedded in the network. Taken together, the Siamese ResNeXt unimodal model performed best for its superior efficacy and robustness. This model achieved a recall rate of 88.0% and an AUC value of 90.88% in the validation subset. Additionally, heatmaps calculated by the Grad-CAM algorithm presented concentrated and orderly mappings around the optic disc vascular area in normal CIMT groups and dispersed, irregular patterns in thickened CIMT groups.

Conclusion

We provided a Siamese ResNeXt neural network for predicting the carotid intimal thickness of patients with T2DM from fundus images and confirmed the correlation between fundus microvascular lesions and CIMT.

Association between carotid ultrasonographic parameters and microvascular and macrovascular complications in diabetes: A systematic review and meta-analysis

OBJECTIVE

The present meta-analysis aimed to assess the association between carotid ultrasonographic parameters and diabetic microvascular and macrovascular complications.

METHODS

All published articles were searched in electronic databases including PubMed, Embase, Cochrane Library, and Web of Science databases from the inception to May 27, 2023. Common carotid artery intima-media thickness (CCA-IMT), carotid bifurcation intima-media thickness (CB-IMT), internal carotid artery intima-media thickness (ICA-IMT), carotid plaque, carotid plaque score, plaque number, plaque thickness, carotid atherosclerosis, and resistivity indices (RIs) of ultrasonographic parameters were assessed. The odds ratio (OR), weighted mean difference (WMD), with the 95 % confidence interval (CI) were pooled to estimate the effect. Subgroup analyses were performed in terms of the type of diabetes and study design. Sensitivity analysis was used to evaluate the robustness of the results.

RESULTS

A total of 25 studies involving 12,102 diabetic patients were included in this systematic review and meta-analysis. Our findings suggested the associations between increased CCA-IMT and the risk of diabetic microvascular (WMD: 0.059, 95%CI: 0.026 to 0.091, P < 0.001) and macrovascular complications (WMD: 0.124, 95%CI: 0.061 to 0.187, P < 0.001) including cardiovascular events (OR: 2.362, 95%CI: 1.913 to 2.916, P < 0.001). Subgroup analyses also elaborated an association between CCA-IMT and diabetic microvascular and macrovascular complications. The results of sensitivity analysis show that the association is relatively stable.

CONCLUSION

Our findings revealed associations between carotid ultrasonographic parameters and microvascular and macrovascular complications of diabetes. Application of the carotid ultrasonographic parameters could be a non-invasive method for the early detection of long-term complications of diabetes.

The Applications and Potential Developments of Ultrasound in Oral Cancer Management

Oral cancer is endemic and causes a great burden in Southern Asia. It is preferably treated by surgery with/without adjuvant radiotherapy (RT) or chemoradiation therapy, depending on the stage of the disease. Close or positive resection margin and cervical lymph node (LN) metastasis are important prognostic factors that have been presented to be related to undesirable locoregional recurrence and poor survival. Ultrasound (US) is a simple, noninvasive, time-saving, and inexpensive diagnostic modality. It can depict soft tissues very clearly without the risk of radiation exposure. Additionally, it is real-time and continuous image is demonstrated during the exam. Furthermore, the clinician can perform US-guided fine needle aspiration (FNA) or core needle biopsy (CNB) at the same time. US with/without US-guided FNA/CNB is reported to be of value in determining tumor thickness (TT), depth of invasion (DOI), and cervical LN metastasis, and in aiding the staging of oral cancer. DOI has a relevant prognostic value as reported in the eighth edition of the American Joint Committee on Cancer staging of oral cancer. In the present review, we describe the clinical applications of US in oral cancer management in different phases and potential applications in the future. In the pretreatment and surgical phase, US can be used to evaluate TT/DOI and surgical margins of oral cancer in vivo and ex vivo. The prediction of a malignant cervical LN (nodal metastasis) by the US-based prediction model can guide the necessity of FNA/CNB and elective neck dissection in clinical early-stage oral cancer. In the posttreatment surveillance phase, US with/without US-guided FNA or CNB is helpful in the detection of nodal persistence or LN recurrence, and can assess the possibility and extent of carotid artery stenosis after irradiation therapy. Both US elastography and US swallowing assessment are potentially helpful to the management of oral cancer.

The effect of nutrition on stroke risk: A systematic review

Background: An estimated 33% reduction in cardiovascular events can be achieved when incorporating whole grains, fruits, vegetables, poultry, nuts, and vegetable oils in the diet along with reduced consumption of refined carbohydrates, processed meats, and sugar sweetened beverages. We performed a systematic review to analyze the impact of nutritional intervention on stroke risk, as there is no current consensus concerning dietary recommendation for primary and secondary stroke prevention. Methods: A literature search of the PubMed database from January 2010 to June 2020 was performed using combinations of the following search terms: carotid disease, carotid artery disease, carotid stenosis, carotid intima-media thickness (CIMT), diet, nutrition, micronutrition, embolic stroke, and stroke. We used the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2009 checklist. Results: 28 studies met our inclusion criteria. Multiple studies showed an inverse relationship between consumption of vegetables and fruits and stroke risk. Vitamin B12 or a combination of B Vitamins was the most common supplement studied in stroke prevention. Only one RCT showed the use of B12 (500 micrograms/day) correlated with lower CIMT at follow up in healthy vegetarians. Discussion: The key findings from this systematic review indicate that adopting a diet rich in fruits and vegetables earlier in life may lower stroke risk compared with meats and fat intake. B vitamins also appear to confer some protection against stroke. However, not enough data exists to support the use of multivitamins, calcium, soy products and other supplements for primary or secondary stroke prevention.

Healthy lifestyle changes favourably affect common carotid intima-media thickness: the Healthy Lifestyle Community Programme (cohort 2)

Common carotid intima-media thickness (ccIMT) progression is a risk marker for cardiovascular disease (CVD), whereas healthy lifestyle habits are associated with lower ccIMT. The objective of the present study was to test whether a healthy lifestyle intervention can beneficially affect ccIMT progression. A community-based non-randomised, controlled lifestyle intervention was conducted, focusing on a predominantly plant-based diet (strongest emphasis), physical activity, stress management and social health. Assessments of ccIMT were made at baseline, 6 months and 1 year. Participants had an average age of 57 years and were recruited from the general population in rural northwest Germany (intervention: n 114; control: n 87). From baseline to 1 year, mean ccIMT significantly increased in both the intervention (0⋅026 [95 % CI 0⋅012, 0⋅039] mm) and control group (0⋅045 [95 % CI 0⋅033, 0⋅056] mm). The 1-year trajectory of mean ccIMT was lower in the intervention group (P = 0⋅022; adjusted for baseline). In a subgroup analysis with participants with high baseline mean ccIMT (≥0⋅800 mm), mean ccIMT non-significantly decreased in the intervention group (-0⋅016 [95 % CI -0⋅050, 0⋅017] mm; n 18) and significantly increased in the control group (0⋅065 [95 % CI 0⋅033, 0⋅096] mm; n 12). In the subgroup, the 1-year trajectory of mean ccIMT was significantly lower in the intervention group (between-group difference: -0⋅051 [95 % CI -0⋅075, -0⋅027] mm; P < 0⋅001; adjusted for baseline). The results indicate that healthy lifestyle changes may beneficially affect ccIMT within 1 year, particularly if baseline ccIMT is high.

Evaluation of Intima-Media Thickness and Arterial Stiffness as Early Ultrasound Biomarkers of Carotid Artery Atherosclerosis

Carotid atherosclerosis is a major and potentially preventable cause of ischemic stroke. It begins early in life and progresses silently over the years. Identification of individuals with subclinical atherosclerosis is needed to initiate early aggressive vascular prevention. Although carotid plaque appears to be a powerful predictor of cardiovascular risk, carotid intima-media thickness (CIMT) and arterial stiffness can be detected at the initial phases and, therefore, they are considered important new biomarkers of carotid atherosclerosis. There is a well-documented association between CIMT and cerebrovascular events. CIMT provides a reliable marker in young people, in whom plaque formation or calcification is not established. However, the usefulness of CIMT measurement in the improvement of risk cardiovascular models is still controversial. Carotid stiffness is also significantly associated with ischemic stroke. Carotid stiffness adds value to the existing risk prediction based on Framingham risk factors, particularly individuals at intermediate cardiovascular risk. Carotid ultrasound is used to assess carotid atherosclerosis. During the last decade, automated techniques for sophisticated analysis of vascular mechanics have evolved, such as speckle tracking, and new methods based on deep learning have been proposed with promising outcomes. Additional research is needed to investigate the imaging-based cardiovascular risk prediction of CIMT and stiffness.

A Powerful Paradigm for Cardiovascular Risk Stratification Using Multiclass, Multi-Label, and Ensemble-Based Machine Learning Paradigms: A Narrative Review

Background and Motivation: Cardiovascular disease (CVD) causes the highest mortality globally. With escalating healthcare costs, early non-invasive CVD risk assessment is vital. Conventional methods have shown poor performance compared to more recent and fast-evolving Artificial Intelligence (AI) methods. The proposed study reviews the three most recent paradigms for CVD risk assessment, namely multiclass, multi-label, and ensemble-based methods in (i) office-based and (ii) stress-test laboratories. Methods: A total of 265 CVD-based studies were selected using the preferred reporting items for systematic reviews and meta-analyses (PRISMA) model. Due to its popularity and recent development, the study analyzed the above three paradigms using machine learning (ML) frameworks. We review comprehensively these three methods using attributes, such as architecture, applications, pro-and-cons, scientific validation, clinical evaluation, and AI risk-of-bias (RoB) in the CVD framework. These ML techniques were then extended under mobile and cloud-based infrastructure. Findings: Most popular biomarkers used were office-based, laboratory-based, image-based phenotypes, and medication usage. Surrogate carotid scanning for coronary artery risk prediction had shown promising results. Ground truth (GT) selection for AI-based training along with scientific and clinical validation is very important for CVD stratification to avoid RoB. It was observed that the most popular classification paradigm is multiclass followed by the ensemble, and multi-label. The use of deep learning techniques in CVD risk stratification is in a very early stage of development. Mobile and cloud-based AI technologies are more likely to be the future. Conclusions: AI-based methods for CVD risk assessment are most promising and successful. Choice of GT is most vital in AI-based models to prevent the RoB. The amalgamation of image-based strategies with conventional risk factors provides the highest stability when using the three CVD paradigms in non-cloud and cloud-based frameworks.

Carotid intima-media thickness and risk of mild cognitive impairment: A systematic review and meta-analysis

Mild cognitive impairment (MCI) is an early phase of cognitive decline signalling the beginning of severe neurological diseases. Carotid intima-media thickness (cIMT) has shown some correlation with MCI development. This study was conducted to investigate the impact of elevated cIMT on the risk of MCI in adults. A literature search was conducted in PubMed, EMBASE, Cochrane Library, Scopus, Google Scholar and CINAHL databases till 30 July 2021, with keywords: ('Carotid Intima-Media Thickness' OR 'cIMT' OR 'IMT' AND 'Cognitive Impairment' OR 'Cognition' OR 'Cognitive Decline' AND 'Mild Cognitive Impairment' OR 'MCI'). Pooled standardized mean difference (SMD)/odds ratio (OR) and 95% confidence interval (CI) were determined for factor-disease association using either fixed (when I² <50%) or random effect (when I² >50%) models. Eight studies involving 1,585 MCI cases and 6,700 normal subjects were included in our meta-analysis which showed no significant association of increased cIMT with the risk of MCI [SMD 1.17, 95% CI -0.09 to 2.42]. However, sensitivity analysis revealed an outlier study significantly affecting the effect size. On omitting the outlier study, the re-evaluated meta-analysis revealed a significant association of cIMT with the risk of MCI [SMD 0.52, 95% CI 0.26 to 0.78]. This significant association was also observed during subgroup analysis in Caucasian population [SMD 0.65, 95% CI 0.13 to 1.18] but not in Asian population [SMD 0.39, 95% CI -0.01 to 0.79]. Elevated cIMT poses a potential risk for MCI. However, more population-based studies are required to corroborate these findings.

Influence of Carotid Intima-Media Thickness Levels at Bifurcation on Short-Term Functional Outcomes Among Non-Cardiogenic Ischemic Stroke Patients with and without Type 2 Diabetes Mellitus

PURPOSE

The intima-media thickness (IMT) is broadly reported to have relationships with non-cardiogenic ischemic stroke and with diabetes. But how does IMT affect the short-term prognosis of stroke seems unknown yet. We investigated the influence of the intima-media thickness at carotid bifurcation (IMTbif) on short-term functional outcomes among non-cardiogenic ischemic stroke patients with and without type 2 diabetes mellitus (T2DM).

PATIENTS AND METHODS

A total of 314 patients with non-cardiogenic ischemic stroke (122 with T2DM and 192 without diabetes) were included in this retrospective study. Poor functional outcome was defined as a modified Rankin Scale (mRS) > 2 at 3 months after stroke onset. Group comparisons were done in favorable and poor outcome groups. Linear regression analysis was utilized to verify the associations between IMTbif and mRS in subgroups with and without diabetes, respectively.

RESULTS

The median IMTbif of total patients was 1.40mm. Patients with poor outcomes were significantly older, had higher National Institutes of Health Stroke Scale (NIHSS) scores, lower haemoglobin, higher fasting glucose and higher systolic blood pressure values. Their IMTbif levels were also markedly higher. Among 122 included stroke patients with T2DM, IMTbif levels and NIHSS were independently associated with functional outcomes at 3 months, whereas there was no significant association between IMTbif levels and short-term functional outcomes among patients without diabetes.

CONCLUSION

The IMTbif levels were significantly associated with 3-month functional outcomes in non-cardiogenic ischemic stroke patients with T2DM. The ultrasound detection of the IMTbif therefore suggests a prognostic value among patients with stroke and T2DM.

Characterization of familial hypercholesterolemia in Taiwanese ischemic stroke patients

Familial hypercholesterolemia (FH) is a common genetic disorder characterized by a lifelong elevated low-density lipoprotein cholesterol (LDL-C) level. The relationship between FH and ischemic stroke is still controversial. We enrolled ischemic stroke patients prospectively in our neurological ward, and divided them into two groups according to LDL-C levels with a threshold of 130 mg/dl. Targeted sequencing was performed in all stroke patients for LDLR, APOB, and PCSK9 genes. The fifty-eight high-LDL subjects were older, prevalence of previous myocardial infarction/stroke history was lower, and the first stroke age was older compared with values in the sixty-three low-LDL cases. The prevalence of FH in Han-Chinese stroke patients was 5.0%, and was 10.3% in those with a higher LDL-C level. We identified six carriers, who had higher percentages of large vessel stroke subtype (66.7% vs. 15.4%) and transient ischemic attack (33.3% vs. 3.8%), previous myocardial infarction/stroke history (50.0% vs. 11.5%), statin use (50.0% vs. 11.5%), and increased carotid intima-media thickness (IMT) (0.9-1.2mm vs.0.7-9.0mm) compared with the other hypercholesterolemic patients without pathogenic variants. Ischemic stroke patients carrying FH pathogenic variants seemed to have a higher risk for large artery stroke and transient ischemic attack. The IMT exam could be useful to screen for FH in hypercholesterolemic stroke patients.

The risk of carotid plaque instability in patients with metabolic syndrome is higher in women with hypertriglyceridemia

Background

Metabolic syndrome certainly favors growth of carotid plaque; however, it is uncertain if it determines plaque destabilization. Furthermore, it is likely that only some components of metabolic syndrome are associated with increased risk of plaque destabilization. Therefore, we evaluated the effect of different elements of metabolic syndrome, individually and in association, on carotid plaques destabilization.

Methods

A total of 186 carotid endarterectomies from symptomatic and asymptomatic patients were histologically analysed and correlated with major cardiovascular risk factors.

Results

Metabolic syndrome, regardless of the cluster of its components, is not associated with a significant increase in risk of plaque destabilization, rather with the presence of stable plaques. The incidence of unstable plaques in patients with metabolic syndrome is quite low (43.9 %), when compared with that seen in the presence of some risk factors, but significantly increases in the subgroup of female patients with hypertriglyceridemia, showing an odds ratio of 3.01 (95% CI, 0.25–36.30).

Conclusions

Our data may help to identify patients with real increased risk of acute cerebrovascular diseases thus supporting the hypothesis that the control of hypertriglyceridemia should be a key point on prevention of carotid atherosclerotic plaque destabilization, especially in post-menopausal female patients.

Potential Role of Melatonin as an Adjuvant for Atherosclerotic Carotid Arterial Stenosis

Carotid artery stenosis (CAS) is an atherosclerotic disease characterized by a narrowing of the artery lumen and a high risk of ischemic stroke. Risk factors of atherosclerosis, including smoking, hypertension, hyperglycemia, hyperlipidemia, aging, and disrupted circadian rhythm, may potentiate atherosclerosis in the carotid artery and further reduce the arterial lumen. Ischemic stroke due to severe CAS and cerebral ischemic/reperfusion (I/R) injury after the revascularization of CAS also adversely affect clinical outcomes. Melatonin is a pluripotent agent with potent anti-inflammatory, anti-oxidative, and neuroprotective properties. Although there is a shortage of direct clinical evidence demonstrating the benefits of melatonin in CAS patients, previous studies have shown that melatonin may be beneficial for patients with CAS in terms of reducing endothelial damage, stabilizing arterial plaque, mitigating the harm from CAS-related ischemic stroke and cerebral I/R injury, and alleviating the adverse effects of the related risk factors. Additional pre-clinical and clinical are required to confirm this speculation.