Carotid Disease and CEA: Systematic Review with ☸️SAIMSARA.

• Generated: 2026-02-12 22:48:38 CET
• Plan: Pro (expanded craft tokens; source: Semantic Scholar)
• Source: Semantic Scholar
• Scope: All fields
• Keyword Gate: Fuzzy (≥60% of required terms, minimum 2 terms matched in title/abstract)
• Total Abstracts/Papers: 1586
• Downloaded Abstracts/Papers: 1000
• Included original Abstracts/Papers: 229
• Total study participants (naïve ΣN): 253324

• Bias: The included studies exhibit a range of designs, from prospective randomized controlled trials (RCTs) [17, 19, 35, 64, 208] and prospective cohort studies [1, 5, 10, 76, 81, 107, 119, 120, 142, 145, 183, 184, 185, 194, 224] to retrospective cohort studies [4, 8, 32, 37, 39, 41, 42, 45, 63, 67, 68, 69, 72, 73, 74, 75, 77, 82, 85, 86, 87, 89, 93, 94, 96, 98, 101, 102, 104, 105, 106, 110, 121, 122, 125, 126, 128, 129, 130, 132, 133, 134, 137, 138, 139, 140, 143, 150, 152, 153, 154, 159, 161, 164, 168, 169, 170, 171, 172, 173, 174, 175, 179, 182, 185, 186, 187, 188, 189, 191, 192, 193, 198, 199, 200, 201, 203, 204, 205, 209, 210, 213, 215, 216, 218, 219, 222, 223, 225, 228, 229], case series [9] and case reports [22, 23, 26, 136, 214], and cross-sectional studies [83, 190]. Many studies are retrospective, which inherently limits the ability to control for confounding variables and establish causality. Reporting heterogeneity in outcomes and patient populations across studies also introduces variability.

• Vulnerable Plaque Identification and AI: Imaging techniques like duplex ultrasound (DU) and AI-based models show high positive predictive value (PPV) for identifying vulnerable plaque (VP) and >50% stenosis, with AI accuracy increasing from 0.73 to 0.94 in identifying VP [1]. Noninvasive prediction models based on IL-6 and TSHI [43], or multimodal vascular ultrasound parameters and clinical risk factors [211], demonstrate high diagnostic efficacy (AUC 0.959) for VP. Contrast-enhanced ultrasound (CEUS) also shows 86.8% accuracy for vulnerable plaques, correlating with CD147 and MMP-9 [178].
• Perioperative Complications and Risk Factors: Perioperative major adverse cardiac events (MACE) and non-MACE rates for CEA range from 15.8% to 18.8% in CAD patients [4], with MACE occurring in 4.3% of patients overall [47]. Preoperative white matter hyperintensities and lacunes are independently associated with postoperative cerebral hyperperfusion (CH) [11], and intraoperative systolic blood pressure drop >44 mmHg is linked to postoperative cerebrovascular accident (POCVA) [13]. Elevated lipoprotein (a) levels are associated with increased 30-day MACE risk (HR 2.05) [77]. Preoperative anxiety (HADS-A score >6) is associated with higher intra- and postoperative neurological events [76].
• Comparison of Revascularization Strategies (CEA vs CAS vs TCAR): CEA is generally considered the first-line treatment for symptomatic patients with >50% and asymptomatic patients with >60% carotid stenosis [31]. While some studies show comparable perioperative stroke or death rates between CEA, CAS, and TCAR [71, 152, 172], others indicate CEA may be associated with lower odds of post-procedural stroke compared to CAS (OR 0.17) [8], fewer new acute ischemic brain lesions (14.2% CEA vs 47.6% CAS) [51], and a higher stroke-free survival rate (100% CEA vs 93.5% CAS at 12 months) [91]. Covert strokes are 3.75 times lower after CEA (10%) than CAS (38%) [170].
• Concomitant Carotid and Coronary Artery Disease (CABG+CEA): Simultaneous carotid endarterectomy (CEA) and coronary artery bypass grafting (CABG) can be performed with acceptable morbidity and mortality rates [67, 72, 106, 125, 222], though some studies report higher immediate complications and prolonged ventilation compared to staged procedures [40, 93]. Preoperative coronary revascularization is associated with lower odds of perioperative MACE and non-MACE in CAD patients undergoing CEA [4]. Severe left ventricular (LV) dysfunction is a significant risk factor for perioperative myocardial infarction (MI) during CEA in high-risk CAD patients [10].
• Cognitive Outcomes and Cerebral Hemodynamics: CEA can significantly improve cognitive function, cerebral blood flow, energy metabolism, and white matter integrity in patients with carotid artery stenosis (CAS)-induced cognitive impairment [14, 50, 109]. Cerebral hemodynamics are altered by carotid artery disease and improve after CEA [53, 58, 139]. However, preoperative white matter lesions influence brain connectivity [190], and normal BMI in older patients with cardiovascular disease may be a risk factor for postoperative cerebral dysfunction [220].
• Patient Selection and Risk Stratification: Age, comorbidities (e.g., CAD, chronic kidney disease, diabetes, heart failure, peripheral artery disease, COPD), and inflammatory biomarkers are crucial for risk stratification [21, 37, 42, 68, 69, 82, 84, 103, 118, 171, 175, 177, 179, 183, 185, 197, 198, 209, 216]. Elderly patients (>80 years) undergoing CEA or CAS have higher mortality but comparable subsequent vascular events to younger patients [41, 133, 134, 188]. Machine learning models (e.g., XGBoost) can accurately predict 30-day outcomes after CEA (AUC 0.91) [47].
• Surgical Techniques and Anesthesia: CEA under local anesthesia is safe [27, 89, 99], and regional anesthesia is associated with decreased risk of postoperative pneumonia and reduced need for transfusions compared to general anesthesia [182]. Primary closure and patch angioplasty show equivalent complication rates and restenosis at 5 years in selected CEA cases [57]. Modified eversion technique reduces patchplasty and restenosis [117].
• Plaque Characteristics and Pathophysiology: Vulnerable plaque features, including plaque hemorrhage (PH) [75, 121], inflammation (e.g., CCR2 cellular content, enlarged pericarotid lymph nodes, specific ceramides, miR-155, TNK1) [25, 29, 63, 70, 116, 147, 148, 155, 157, 196, 197, 221, 226], and specific molecular signatures (e.g., albumin, CRP, monocyte percentage, CXCL9) [21, 43], are associated with symptomatic disease and adverse outcomes. Women are less likely to have plaque hemorrhage and have longer time to recurrent ischemic events than men [121, 224].
• Long-term Outcomes and Restenosis: Carotid interventions (CEA or CAS) provide durable long-term outcomes [45, 125, 137, 186]. Wall Shear Stress (WSS) topological skeleton features are associated with long-term restenosis after CEA [66]. Dual antiplatelet therapy (DAPT) following CEA is associated with lower rates of late restenosis compared to single antiplatelet therapy (SAPT) [85].

• Patient Selection Refinement: Preoperative assessment should integrate advanced imaging for vulnerable plaque identification [1, 178, 211], along with comprehensive risk stratification considering comorbidities like CAD, diabetes, and chronic kidney disease [4, 10, 82, 185, 198].
• Anesthesia and Surgical Technique: Regional anesthesia for CEA may offer advantages over general anesthesia in reducing postoperative complications like pneumonia and transfusion needs [182]. Standardized techniques like eversion endarterectomy or patch angioplasty demonstrate good safety and efficacy [57, 87, 117].
• Postoperative Monitoring: Close monitoring of cerebral hemodynamics and blood pressure is critical to mitigate risks of cerebral hyperperfusion syndrome and POCVA [11, 13, 184, 189].
• Antiplatelet Therapy: Dual antiplatelet therapy (DAPT) may be beneficial for reducing late restenosis after CEA compared to single antiplatelet therapy (SAPT) [85].
• Concomitant Disease Management: For patients with concomitant carotid and coronary artery disease, the choice between simultaneous or staged procedures remains complex, with both approaches showing acceptable outcomes but varying complication profiles [40, 73, 90, 93].

• Long-term Cognitive Outcomes: Further prospective studies are needed to compare the long-term cognitive impact of CEA versus CAS/TCAR in various patient subgroups [2, 14, 28].
• Standardized Biomarker Panels: Research should focus on validating and integrating novel inflammatory and genetic biomarkers into routine clinical risk prediction models for perioperative and long-term adverse events [43, 44, 63, 68, 179].
• AI Model Validation: Large-scale, external validation of AI models for vulnerable plaque identification and outcome prediction is essential to confirm their generalizability and clinical utility [1, 3, 47, 211].
• Optimal Management of Non-Stenotic Disease: High-quality evidence is needed to define the role of CEA in symptomatic non-stenotic carotid artery disease, particularly regarding long-term stroke prevention [9, 223].
• Comparative Effectiveness in Specific Subgroups: More RCTs are required to compare CEA, CAS, and TCAR outcomes in underrepresented or high-risk populations, such as women, octogenarians, or patients with specific plaque characteristics [41, 80, 121, 173, 224].

• Heterogeneity of Study Designs — The diverse range of study designs (RCTs, cohorts, case reports) limits the ability to draw definitive, generalizable conclusions due to varying levels of evidence and potential for bias.
• Inconsistent Outcome Reporting — Variability in definitions and reporting of perioperative complications (e.g., stroke, death, MACE) across studies complicates direct comparisons and meta-analysis.
• Lack of Direct Comparisons — Many studies compare CEA to CAS or TCAR, but direct head-to-head randomized controlled trials for all permutations and patient subgroups are still limited.
• Focus on Short-Term Outcomes — A significant portion of the literature emphasizes 30-day perioperative outcomes, potentially underrepresenting the long-term efficacy and safety profiles of interventions.
• Selection Bias in Observational Studies — Retrospective and observational studies are prone to selection bias, as patient characteristics often influence the choice between CEA and alternative treatments.

• Large-Scale RCTs — Conduct large-scale, pragmatic RCTs comparing CEA, CAS, and TCAR in diverse patient populations, including asymptomatic and symptomatic subgroups.
• Standardized Outcome Definitions — Develop and adopt standardized definitions for perioperative and long-term outcomes to facilitate more robust comparative effectiveness research.
• AI-Driven Risk Stratification — Further develop and validate AI models for personalized risk assessment, integrating multimodal imaging, clinical data, and biomarkers.
• Biomarker-Guided Therapy — Investigate the utility of novel biomarkers (e.g., inflammatory markers, genetic profiles) to guide patient selection and optimize antiplatelet or medical therapies post-CEA.
• Longitudinal Cognitive Studies — Implement long-term prospective studies to thoroughly evaluate the cognitive impact of different carotid revascularization strategies.