Mitral regurgitation (MR) is a common valvular disorder and associated with adverse outcomes. Echocardiography is the primary imaging modality for MR assessment; however, standard methods rely on geometric assumptions and single-frame analysis, which are especially inaccurate in the setting of eccentric, multiple or non-holosystolic jets. Recent developments in machine learning have enabled automated quantification of MR, which analyse regurgitant flow throughout systole, account for non-hemispheric orifice area and jet morphology, and demonstrate favourable agreement with cardiac magnetic resonance. In addition, AI capable of detecting and grading MR directly from echocardiographic clips offer potential utility for screening, particularly in low-resource settings where specialist review is limited. Quantification of Mitral Regurgitation: from traditional methods to Artificial Intelligence.

BACKGROUND: Atrial fibrillation (AF) is the most prevalent sustained cardiac arrhythmia and is associated with a markedly increased risk of thromboembolic stroke, primarily originating from the left atrial appendage (LAA). Percutaneous left atrial appendage occlusion (LAAO) is an established alternative for stroke prevention in patients with contraindications to long-term oral anticoagulation. Transesophageal echocardiography (TEE) has traditionally guided LAAO, but intracardiac echocardiography (ICE) is increasingly adopted to facilitate conscious sedation and operator-controlled imaging. Comparative real-world safety and effectiveness data between ICE and TEE remain limited. METHODS: We conducted a retrospective, multicenter cohort study using the TriNetX Research Network, comprising 112 healthcare organizations. Adult patients with atrial fibrillation who underwent percutaneous LAAO guided exclusively by ICE or TEE were identified. The primary endpoint was 30-day all-cause mortality. Secondary endpoints included pericardial effusion, stroke/cerebrovascular accident (CVA), major bleeding, acute kidney injury (AKI), device-related complications, and all-cause hospitalization or emergency department encounters, assessed at 30 and 365 days. Propensity score matching (1:1) was performed using demographics, comorbidities, medications, prior procedures, and laboratory values. Risk ratios, absolute risk differences, Kaplan-Meier analyses, and multivariable Cox proportional hazards models were used to compare outcomes. RESULTS: Among 10,629 eligible patients, 3,930 underwent ICE-guided LAAO and 6,699 underwent TEE-guided LAAO. After propensity score matching, 3,442 patients remained in each cohort with well-balanced baseline characteristics. At 30 days, all-cause mortality was numerically lower with ICE than TEE (0.4% vs. 0.8%; RR 0.577, p = 0.085). ICE guidance was associated with significantly lower rates of pericardial effusion (2.0% vs. 2.8%; RR 0.716, p = 0.040) and reduced all-cause hospitalization or emergency department encounters (21.5% vs. 29.5%; RR 0.731, p < 0.001). Rates of stroke/CVA, major bleeding, and AKI did not differ significantly between groups. In adjusted Cox models, ICE was associated with lower hazards of pericardial effusion, stroke/CVA, and major bleeding at 30 days. At 365 days, all-cause mortality was similar between ICE and TEE (5.5% vs. 6.0%; p = 0.383). Major bleeding and all-cause hospitalization or emergency encounters were significantly lower with ICE, while other outcomes remained comparable. CONCLUSIONS: In this large real-world propensity-matched analysis, ICE-guided LAAO demonstrated comparable short-term and mid-term safety to TEE guidance, with significantly lower early pericardial effusion rates and reduced healthcare utilization. These findings support ICE as a safe and effective alternative to TEE for LAAO in experienced centers, while highlighting the importance of operator expertise and continued evaluation in prospective studies.

Noise (synonyms: variability, imprecision) in clinical data is underappreciated. All measurements are affected by noise and bias. Research studies enrol many patients to average out noise and use randomisation and double-blinding to counter bias. These mitigation techniques are often not available in clinical medicine, where individual patients typically have a variable measured only once.The example of echocardiographic assessment of aortic stenosis severity illustrates how the conventional presentation of data as point estimates masks wide variability in noise between variables, which affects utility. Adding aortic valve area to the quantification of aortic stenosis may result in poorer (noisier) estimates.Every clinical variable reported as a point estimate comes with an invisible cloud of uncertainty (noise) of varying size across variables. Being mindful of how this noise varies across variables should facilitate optimal clinical decisions.

BACKGROUND: Patients with severe left ventricular (LV) dysfunction (ejection fraction [EF] ≤ 35%) undergoing surgical aortic valve replacement (SAVR) are considered high-risk patients. This study evaluated contemporary outcomes in such patients undergoing SAVR for severe aortic stenosis (AS) or aortic regurgitation (AR) and identified clinical and echocardiographic predictors of mortality. METHODS: This retrospective cohort study included 84 patients (> 18 years) with EF ≤ 35% who underwent SAVR between 2015 and 2024. The cohort comprised 53 patients with AS and 31 with AR. Clinical, echocardiographic, and myocardial strain data were collected before and 6 months after SAVR. Mortality data were obtained from national registries, with follow-up through March 2025. Cox proportional hazards models were used to identify mortality predictors. RESULTS: Median EF was 28% in AS and 31.1% in AR. Patients with AR were significantly younger than AS (44 vs. 63 years, p < 0.001). In-hospital mortality for AS was 9.4%, increasing to 13.2% at 1 year and 43.4% at 55 months. AR patients showed no in-hospital deaths and 32.3% mortality at 55 months. Eight-year survival was 38.7% in patients with AS and 43.1% in patients with AR. In AS, creatinine (hazard ratio [HR] = 1.01, p = 0.022) and preoperative peak velocity (HR = 0.52, p = 0.020) predicted mortality. In AR, only length of hospital stays (HR = 1.13, p = 0.017) was predictive. CONCLUSION: SAVR in severe LV dysfunction offers better medium-term survival, with clinical factors guiding prognosis. TRIAL REGISTRATION: Institut Jantung Negara Ethics Committee (IJNREC/779/2025).