Arnoud W. J. van 'T Hof

Background and Aims: Patients with acute coronary syndrome (ACS) treated with percutaneous coronary intervention (PCI) have a residual risk of adverse events and all-cause mortality. Machine learning (ML) approaches may help integrate clinical, functional, and lifestyle information for risk assessment. This study aimed to develop and externally evaluate models for predicting all-cause mortality in patients enrolled in cardiac rehabilitation after PCI for ACS. Methods and Results: Data from 1261 patients (mean age 61 ± 11.1 years, 16.6% women) enrolled in cardiac rehabilitation after PCI for ACS were analysed using seven prediction algorithms and repeated stratified cross-validation. External validation was performed using the EU-CaRE dataset of 1386 patients and predictions were explained using the model-agnostic SHapley Additive exPlanation (SHAP) methodology. LightGBM and Random Forest achieved the highest AUC and balanced accuracy (mean AUC > 0.8). SVM also exceeded logistic regression on AUC, though logistic regression remained competitive. LightGBM achieved a balanced accuracy of 0.87, sensitivity of 0.86, and specificity of 0.88. External evaluation yielded sensitivity of 0.78 and specificity of 0.73 in the PCI-treated cohort; specificity was 0.60 in the broader CAD cohort. After omission of predictors unavailable in EU-CaRE, the AUC was 0.816 with seven and 0.812 with six predictors. The highest SHAP values for impact on the model were detected for the variables “diabetes”, “physical exercise capacity (METs achieved during stress test)”, “average daily sitting duration”, “statin use”, “history of family heart disease”, “number of affected vessels”, “revascularization history”, “serum creatinine levels”, and “chronic hepatopathy”. The model was rated as “neutral” using the independent validation screening tool. Conclusions: ML models showed potential for predicting all-cause mortality in patients enrolled in cardiac rehabilitation after PCI for ACS. After prospective validation and assessment of clinical utility, the model may support clinicians in identifying patients at higher risk who could benefit from intensified treatment and lifestyle interventions. The small number of deaths and limitations of the evaluation procedure warrant cautious interpretation.

Background: Chronic total occlusion (CTO) percutaneous coronary intervention (PCI) is known for higher adverse event rates compared with non-CTO PCI. Comparative outcomes with respect to bleeding events remain limited. Objectives: To evaluate bleeding events in patients undergoing CTO PCI compared with non-CTO PCI and to identify patient and procedural characteristics associated with bleeding events. Design and setting: Prospective multicentre study of ongoing collected data registered within the South-East Netherlands Heart Registry (Zuid-Oost Nederland Hart Registratie, ZON-HR). Participants: All consecutive patients with coronary artery disease undergoing clinically indicated PCI are included in the ZON-HR since November 2020. Results: At the time of analyses, the ZON-HR consisted of 8264 included patients with 30-day follow-up between November 2020 and December 2023. A total of 8162 patients were included once in the registry. CTO PCI was performed in 410 (5.0%) patients and non-CTO PCI in 7854 (95.0%) patients. Bleeding events occurred more frequently in patients undergoing CTO PCI compared with non-CTO PCI (9.0% vs 5.4%, p<0.01), driven by a significantly higher rate of access-site bleeding events (7.6% vs 3.2%, p<0.01). After multivariable analysis, CTO was not significantly associated with the occurrence of access-site bleeding events (adjusted HR (aHR) 1.39 (95% CI 0.92 to 2.11)). Access-site bleeding events were associated with female sex (aHR 1.71 (95% CI 1.34 to 2.18)), peripheral artery disease (aHR 1.39 (95% CI 1.02 to 1.89)), oral anticoagulation use (aHR 1.34 (95% CI 1.00 to 1.78)), femoral artery access (aHR 2.90 (95% CI 2.08 to 4.05)) and left main PCI (aHR 1.50 (95% CI 1.04 to 2.14)). Conclusions: The incidence of bleeding events is higher in patients undergoing CTO PCI and particularly driven by access-site bleeding events. The apparent correlation is attributed to non-modifiable and modifiable patient-related and procedure-related characteristics rather than the CTO PCI itself. Trial registration number: NCT06512493 .

Background and Aims: Comprehensive evaluation by invasive coronary function test (CFT) of patients with angina and non-obstructive coronary artery disease is recommended in the current guidelines. The prognosis of patients with CFT-proven coronary vasomotor dysfunction (CVDys) is less well-known. Methods: Patients that provided consent for the previously published NL-CFT registry who underwent clinically indicated invasive CFT between 1 February 2019 and 1 June 2024 were included. Patients with CVDys, defined as an abnormal spasm provocation test and/or microvascular functional assessment, were compared to patients with normal CFT. The primary outcome of major adverse cardiovascular events and emergency room visits for angina (MACE-EA) was defined as a composite of all-cause death, acute coronary syndrome, stroke/transient ischaemic attack, revascularization, repeat angiography without intervention, heart failure hospitalizations, and unplanned hospitalizations or emergency room visits due to angina. Results: In total, 1355 patients were included. In 79.8%, CFT was abnormal, and vasospasm was the most prevalent endotype. In 1252 patients, follow-up of clinical events was complete. During a mean follow-up of 33 ± 13 months, CVDys patients had almost double the number of primary outcome events compared with patients with a normal CFT (31.2% vs 16.8%, P < .001), driven by emergency care visits for angina. Secondary clinical endpoints including death, myocardial infarction, stroke, or revascularization were low and comparable between the groups. Conclusions: The long-term risk of MACE-EA is doubled in ANOCA patients with CFT-confirmed CVDys compared with normal CFT results. Events are dominated by emergency room visits due to angina, illustrating the high disease burden of CVDys. Their (angina-related) health status is significantly worse at follow-up.

Background: Mitral valve regurgitation (MR) is associated with an impaired prognosis. Due to its frequently asymptomatic presentation, MR often leads to undertreatment, which can delay diagnosis. In addition, the optimal timing and choice of intervention remain unclear. The Significant Mitral Insufficiency Limburg Evaluation (SMILE) registry aims to provide a comprehensive real-world characterisation of patients with moderate-to-severe mitral regurgitation. Methods and Results: The SMILE registry is a multicentre prospective registry initiated in 2020, enrolling all consecutive patients with moderate-to-severe MR from all hospitals in Limburg, the Netherlands. Treatment approaches include surgical, catheter-based, and conservative (medical) options, with decision-making from an expert multidisciplinary team. Data are collected via Castor EDC, ensuring compliance with the General Data Protection Regulation using pseudonymisation. The co-primary endpoints are all-cause mortality and heart failure hospitalisations during the five-year follow-up. Secondary objectives include describing management strategies, characterising disease progression and cardiac remodelling, evaluating associations between baseline clinical and echocardiographic characteristics and long-term outcomes, identifying predictors of treatment success, and assessing longitudinal changes in health-related quality of life. Conclusions: The SMILE registry represents an important step towards improving MR management. Its broad data collection, including conservative care and patient-reported outcomes, provides valuable real-world insights beyond procedure-focused studies. The registry may refine intervention timing and personalise treatment strategies to enhance patient outcomes and improve their quality of life.

Background: Among patients with ST segment elevation myocardial infarction (STEMI), higher concentrations of high sensitivity troponin T (hs-cTnT) are associated with larger MI size and predict a worse prognosis. In the 2467 patient CELEBRATE trial, a single subcutaneous injection of the short-acting glycoprotein IIb/IIIa receptor blocker antagonist zalunfiban at first medical contact significantly improved the primary outcome including clinical endpoints. In this study, we assessed the impact of zalunfiban on MI size and association with downstream outcomes remains unclear. Methods: In a prespecified analysis, we studied results among study participants treated with two doses of zalunfiban who had core laboratory measurements concentrations of hs-cTnT. Results: The median concentration of hs-cTnT at presentation was 62 ng/L; at 24 hours it was 1962 ng/L. More elevated hs-cTnT concentrations at presentation were associated with less resolution of ST deviation after revascularization (P =0.006) and more frequent Q wave development (all P <0.001). At coronary angiography more elevated hs-cTnT at presentation was also associated with higher thrombus grade and worse epicardial and myocardial perfusion (all P <0.05). In multivariable analyses, higher hs-cTnT concentrations at 24 hours were associated with greater adjusted risk for all-cause death (odds ratio [OR] 1.83 per log unit increase; P=0.03), cardiovascular death (OR 1.83 per log unit increase; P=0.03), heart failure (OR 2.74 per log unit increase; P <0.001) or the composite of death (or cardiovascular death) and heart failure (P<0.001) by 30 days. At 24 hours, those treated with zalunfiban had lower hs-cTnT compared to placebo (1900 vs 2082 ng/L; P =0.04) and across multiples ≥10 to ≥1000 times elevation, treatment with zalunfiban resulted in smaller hs-cTnT determined MI size. Conclusion: Among patients with STEMI, higher concentrations of hs-cTnT are associated with worse angiographic and ECG measures of reperfusion. More elevated hs-cTnT predicted a higher risk for short-term death or heart failure. A single dose of zalunfiban at first medical contact reduced MI size, as judged by more study participants with lower hs-cTnT concentrations. These results provide a mechanistic basis for the improved clinical outcomes associated with zalunfiban treatment in the CELEBRATE Trial. Study registration: A Phase 3 Study of Zalunfiban in Subjects With ST-elevation MI (CELEBRATE); NCT04825743

Background: Diabetes is a known risk factor of acute coronary syndrome (ACS). However, diabetes de novo and prediabetes are also common in ACS patients. This study explored the prevalence of prediabetes and diabetes de novo in ACS patients, glucose-mediating therapy at discharge, and compared the prevalence of 30-day major adverse cardiac and cerebrovascular events (MACCE) in patients with prediabetes and diabetes de novo with known diabetes. Methods: ACS patients with measured haemoglobin A1c (HbA1c) from the South-East Netherlands Heart Registry, a prospective, multicentre registry of patients undergoing percutaneous coronary intervention (PCI), were analysed. Patients were stratified into two groups: known diabetes, and prediabetes (HbA1c 39–47 mmol/mol) or diabetes de novo (HbA1c ≥ 48 mmol/mol). Outcomes were analysed at 30 days post-PCI. Results: HbA1c was available in 34.1% of ACS patients (n = 1836), of whom 526 (28.7%) had known diabetes, 619 (33.7%) prediabetes, and 180 (9.8%) diabetes de novo. Compared with patients with known diabetes, patients with prediabetes and diabetes de novo had a significantly higher risk of MACCE (HR = 1.81, 95% CI 1.12–2.93, p = 0.016) after multivariable adjustment. At discharge, 59% of patients with diabetes de novo received no insulin, metformin, nor sodium-glucose co-transporter-2 inhibitor, compared with 16% of patients with known diabetes (p < 0.001). Conclusions: Impaired glucose metabolism without known diabetes was observed in nearly 45% of ACS patients and they demonstrated a significantly higher risk of 30-day MACCE compared with patients with known diabetes. Despite clear guideline recommendations, routine screening for hyperglycaemia and the appropriate initiation of glucose-mediating therapy remain underutilised in clinical practice.

BACKGROUND: The universal definition of percutaneous coronary intervention (PCI)–related myocardial infarction (MI) has been substantially updated over the years, including an increase in the biomarker threshold (from 3 to 5 times the upper reference limit) and the introduction of ancillary criteria such as ischemic symptoms and electrocardiographic or angiographic complication. The impact of these changes in patients with acute coronary syndrome (ACS) remains incompletely understood. The objective of this study was to compare prognostic implications of evolving universal definitions of PCI-MI in a large cohort of patients with ACS from the MATRIX trial (Minimizing Adverse Haemorrhagic Events by Transradial Access Site and Systemic Implementation of AngioX). METHODS: Among 6724 patients undergoing PCI in the MATRIX trial, PCI-MI was prospectively adjudicated by the second, third and fourth universal definition of MI (UDMI). The 2 co-primary end points were all-cause and cardiovascular death (from 24 hours to 1 year after PCI) in patients with non–ST-segment–elevation ACS. Hazard ratios and 95% CIs were generated for primary and secondary end points with the use of Cox proportional hazards time-to-event analyses for each MI definition. RESULTS: PCI-MI occurred in 590 patients (9%) with the second UDMI, 193 (3%) with the third UDMI, and 182 (3%) with the fourth UDMI applied in the overall ACS population. Among patients with non–ST-segment–elevation ACS, the corresponding figures were 15%, 5%, and 5%. Only PCI-MI defined by the fourth UDMI in patients with non–ST-segment–elevation ACS was associated with increased risks of all-cause (hazard ratio, 2.08 [95% CI, 1.00–4.30]; P =0.048) and cardiovascular (hazard ratio, 2.62 [95% CI, 1.03–6.65]; P =0.043) death. In patients with ST-segment–elevation myocardial infarction, PCI-MI was uncommon (1% to 4% depending on the working definition) and was not associated with increased mortality. In the absence of objective ancillary criteria (electrocardiographic and angiographic complications), isolated troponin elevations up to 20 times the upper reference limit were not associated with increased mortality risk. CONCLUSIONS: PCI-MI defined according to the fourth UDMI was associated with increased risks of 1-year mortality only in patients with non–ST-segment–elevation ACS. These data support the evolution of the universal definition of PCI-MI. REGISTRATION: URL: http://www.clinicaltrials.gov ; Unique identifier: NCT01433627.

Background: Emergency Medical Services (EMS) patients with chest pain are often suspected of having non-ST-elevation acute coronary syndrome (NSTE-ACS). Current risk stratification protocols for NSTE-ACS have limitations, leading to a lack of a well-organised prehospital diagnostic pathway. Recent studies have demonstrated that using clinical risk scores (CRS) including point-of-care (POC)-troponin in the EMS can improve prehospital diagnostic pathways for suspected NSTE-ACS. The primary aim of this systematic review and individual patient data meta-analysis was to assess safety of low-risk stratification for suspected NSTE-ACS patients in the prehospital setting. Methods: Prospective studies using CRS or POC-troponin for risk stratification in suspected NSTE-ACS patients within the EMS setting were included. Safety was assessed using sensitivity and negative predictive value (NPV) for patients identified as low risk, based on CRS or POC-troponin measurement, for three different endpoints within 30 days: (1) all-cause mortality, (2) composite of mortality and/or acute myocardial infarction (AMI), (3) major adverse cardiac events (MACE). Results: Of 1526 articles screened, 6 were included, comprising 5.239 patients, and all utilised CRS derived from the History, ECG, Age, Risk-factor and Troponin (HEART) score. The summary of low-risk CRS diagnostic performance predicted all-cause mortality with a sensitivity of 93.2% (83.5–98.1) and NPV of 99.8% (99.5–99.9); mortality and/or AMI with a sensitivity of 91.8% (83.0–96.2) and an NPV of 97.3% (89.9–99.3); and MACE with a sensitivity of 92.8% (88.7–95.5) and an NPV of 97.2% (92.1–99.0). Lowering the CRS cut-off value for identifying low-risk patients increased sensitivity and NPV but decreased the proportion of patients classified as low risk. Conclusion: In well-trained EMS systems, where prompt and accurate follow-up of low-risk patients is possible, HEART-derived CRS effectively identify patients with a very low risk of 30-day mortality and MACE. However, implementation in other healthcare systems requires additional validation, given the variations in healthcare structure, risk stratification processes and follow-up capabilities.