Conventional Scores Research Articles

QTc interval prolongation impact on in-hospital mortality in acute coronary syndromes patients using artificial intelligence and machine learning.

Prediction of mortality in hospitalized patients is a crucial and important problem. Several severity scoring systems over the past few decades and machine learning models for mortality prediction have been developed to predict in-hospital mortality. Our aim in this study was to apply machine learning (ML) algorithms using QTc interval to predict in-hospital mortality in ACS patients and compare them to the validated conventional risk scores. This study was retrospective, using supervised learning, and data mining. Out of a cohort of 500 patients admitted to a tertiary care hospital from September 2018 to August 2020, who presented with ACS. Prediction models for in-hospital mortality in ACS patients were developed using 3 ML algorithms. We employed the ensemble learning random forest (RF) model, the Naive Bayes (NB) model and the rule-based projective adaptive resonance theory (PART) model. These models were compared to one another and to two conventional validated risk scores; the Global Registry of Acute Coronary Events (GRACE) risk score and Thrombolysis in Myocardial Infarction (TIMI) risk score. Out of the 500 patients included in our study, 164 (32.8%) patients presented with unstable angina, 148 (29.6%) patients with non-ST-elevation myocardial infarction (NSTEMI) and 188 (37.6%) patients were having ST-elevation myocardial infarction (STEMI). 64 (12.8%) patients died in-hospital and the rest survived. Performance of prediction models was measured in an area under the receiver operating characteristic curve (AUC) ranged from 0.83 to 0.93 using all available variables compared to the GRACE score (0.9 SD 0.05) and the TIMI score (0.75 SD 0.02). Using QTc as a stand-alone variable yielded (0.67 SD 0.02) with a cutoff value 450 using Bazett's formula, whereas using QTc in addition to other variables of personal and clinical data and other ECG variables, the result was 0.8 SD 0.04. Results of RF and NB models were almost the same, but PART model yielded the least results. There was no significant difference of AUC values after replacing the missing values and applying class balancer. The proposed method can effectively predict patients at high risk of in-hospital mortality early in the setting of ACS using only clinical and ECG data. Prolonged QTc interval can be used as a risk predictor of in-hospital mortality in ACS patients.

Does different application modes of universal adhesives with universal resin composites affect the microleakage in class V cavities? An in vitro study.

Composite restorations often have gingival margins near the cemento-enamel junction (CEJ), where the microleakage of these margins can significantly contribute to the restoration failure, especially in the cervical lesions. It is important to determine the microleakage is crucial, as it typically occurs through the interfacial gap between the tooth and the restoration. Various resin composites and techniques have been developed to minimize this gap and reduce the risk of microleakage. The aim of this in vitro study was to evaluate the levels of microleakage of different modes of a universal adhesive and two novel resin composites in restoring class V cavities in the central incisors. Sixty-six freshly extracted sound human central incisors of the similar size were randomly assigned to 2 groups (n = 33 per group) according to the brand of resin composite. Each group was further divided into 3 subgroups based on the Scotch Bond Universal (3M ESPE, Saint Paul, MN, USA) application protocol used: (a) total etch, (b) self-etch and (c) selective etch. After composite restoration completed with Omnichroma (Tokuyama Dental Corp., Tokyo, Japan) and Filtek Universal Restorative (3M ESPE, Saint Paul, MN, USA), each tooth was immersed in a 0.5% basic fuchsin dye solution at 37C0 for 24h. After dye penetration, teeth were sectioned and evaluated with conventional (scoring) and digital methods (ImageJ). The intra- and inter-examiner agreement was estimated according to the Kappa statistics and the results were analyzed with the one-way ANNOVA and the Kruskal-Wallis test (p < 0.05). The rates of microleakage of the gingival and incisal margins are statistically similar, regardless of the composite brand and the method of application of the universal adhesive. The microleakage exhibited by current adhesives and resins is independent of the adhesive application mode and measurement method.

Neutrophil-to-lymphocyte ratio (NLR) at diagnosis in essential thrombocythemia: A new promising predictor of thrombotic events.

Myeloproliferative neoplasms represent a heterogeneous group of acquired hematopoietic stem cell diseases in which chronic inflammation is essential for both clonal evolution and thrombotic complications. The neutrophil-to-lymphocyte ratio (NLR), reflecting the imbalance between systemic inflammation and immunity, is emerging as a prognostic biomarker in several diseases, including hematological ones. A total of 473 patients with essential thrombocythemia (ET), the relationship between NLR value at diagnosis and the risk of thrombotic events in the follow-up, in addition to conventional clinical and biological variables, were retrospectively analyzed. A total of 78 thrombotic events were reported for an incidence rate of 1.8 × 100 patients/year. In multivariate analysis, NLR value≥4 at diagnosis was associated with higher cumulative thrombotic risk (hazard ratio [HR], 2.05; 95% CI, 1.29-2.28; p=.0001) as well International Prognostic Score for Thrombosis in Essential Thrombosis score intermediate-high (HR, 2.69; 95% CI, 1.27-5.72; p=.01) and diabetes (HR, 2.49; 95% CI, 1.23-3.05; p=.010). Concerning arterial thrombotic events, in multivariate analysis, NLR value at diagnosis ≥4 was predictive for thrombosis (HR, 2.13; 95% CI, 1.31-4.04; p=.001 as well diabetes (HR, 2.44; 95% CI, 1.05-5.68; p=.04) and hypertension (HR, 2.46; 95% CI, 1.05-5.68; p=.01). About venous thrombotic events, NLR value≥5 was a marker predictive for venous thrombosis (HR, 2.99; 95% CI, 2.45-6.48; p=.01) as well age >60 years old (HR, 2.26; 95% CI, 1.0-5.10; p=.05). NLR value is a simple, cost-effective, and easy-to-obtain inflammatory marker that can predict a diagnosis the risk of thrombosis in ET. Our results suggest that NLR value could be integrated into conventional cardiovascular risk scores, to better classify high-risk patients who are candidates for cytoreductive therapy. Further larger and prospective studies are warranted.

Abstract 4133654: External Validation of the Guha-Stabellini CVD Prostate Cancer-Specific Calculator in Short-Term Follow-Up

Introduction: Conventional cardiovascular disease (CVD) risk scores are inaccurate for prostate cancer patients (PC); the Guha-Stabellini machine learning (GS-ML) score shows promise but lacks external validation. Hypothesis: The GS-ML score is superior to conventional CVD scores in patients with PC. Aim(s): To conduct external validation of the GS-ML score in patients diagnosed with PC. Methods: The validation used holdout data from Seidman Cancer Center’s CAISIS platform (internal validation cohort) and RADICAL PC1 (external validation cohort; a prospective study of men diagnosed with PC within 1 year or starting ADT within 1 month of enrollment), matching covariates with the GS-ML score. With limited follow-up, short-term CVD (2,000 days from PC diagnosis) was the outcome. Performance was assessed via area under the receiver operating characteristic curve (AUC) with assistance from Youden statistic cutoff, comparing metrics with ACC/AHA pooled cohort equations (PCE), SCORE-2, and AHA-PREVENT scores. Atherosclerotic cardiovascular disease (ASCVD) included non-fatal ischemic stroke and myocardial infarction. CVD included ASCVD and heart failure. Results: We included 2,495 patients from RADICAL PC1 and 1,506 from the internal validation cohort ( Table 1 ). In the internal validation cohort, the CVD AUCs were 0.58, 0.49, 0.65, 0,73, and 0.75 for PCE, SCORE2, PREVENT simple, PREVENT enhanced, and GS-ML score, respectively. In RADICAL PC1, the AUCs for CVD were 0.60, 0.43, 0.67, and 0.64 for PCE, SCORE2, PREVENT simple and GS-ML score, respectively. For ASCVD, internal validation cohorts’ AUCs were 0.63, 0.62, 0.66, 0.73, and 0.67 for PCE, SCORE2, PREVENT simple, PREVENT enhanced, and Guha-Stabellini, respectively. The RADICAL PC1 ASCVD AUCs were 0.64, 0.46, 0.65, and 0.61, for PCE, SCORE2, PREVENT simple and GS-ML scores, respectively, Conclusion(s): The GS-ML score was validated for PC patients, showing superior performance to PCE and SCORE2, and similar performance to AHA-PREVENT in predicting CVD. The European Cardio-Oncology guidelines should reconsider using SCORE2 for PC patients. Further improvement and validation with nationally representative datasets are needed to corroborate these findings and enhance the generalizability of the GS-ML score.

Short-term outcome after isolated tricuspid valve surgery: prognostic role of right ventricular strain.

To assess the incremental prognostic value of right ventricular free wall longitudinal strain over conventional risk scores in predicting the peri-operative mortality in patients with severe tricuspid regurgitation (TR) undergoing isolated tricuspid valve (TV) surgery. We retrospectively enrolled 110 consecutive patients with severe TR who underwent isolated TV surgery between November 2016 and July 2022 at San Raffaele Hospital, Milan, Italy. Exclusion criteria were previous TV surgery, urgent surgery, complex congenital heart disease, active endocarditis and inadequate acoustic window. Baseline clinical data were included, as well as laboratory tests and clinical risk score, as TRI-SCORE and MELD-XI. The clinical outcome was peri-operative mortality, defined as all-cause mortality within 30 days. The final cohort included 79 patients. The endpoint occurred in 7 patients (9%), who died within 30-days after isolated TV surgery. ROC curves analysis showed that, among parameters of right ventricular function, right ventricular free wall longitudinal strain was the best parameter to predict peri-operative mortality (AUC: 0.854, 95% CI 0.74-0.96, p = 0.005, sensitivity 68%, specificity 100%) At univariable analysis, left ventricular ejection fraction, diabetes mellitus, creatinine, estimated glomerular filtration rate, serum sodium, MELD-XI, TRI-SCORE, right ventricular areas, right ventricular global longitudinal strain, right ventricular free wall longitudinal strain, fractional area change and the ratio between right ventricular free wall longitudinal strain/pulmonary arterial systolic pressure were significantly associated with the endpoint. The combination of TRI-SCORE and right ventricular Strain, evaluating right ventricular systolic function with speckle-tracking echocardiography, outperformed classic TRI-SCORE in outcome prediction (AUC 0.874 vs 0,787, p value=0.05). Right ventricular free wall longitudinal strain has an incremental prognostic value over conventional parameters and significantly improves the ability of clinical scores to predict peri-operative mortality in patients undergoing isolated TV surgery.

Prognostic value of left ventricular mass measured on coronary computed tomography angiography

BackgroundLeft ventricular (LV) mass is a well-established prognostic indicator for cardiovascular risk. Measurement of LV mass on coronary computed tomography angiography (CCTA) is considered optional. We aimed to assess for associations between LV mass measured on CCTA with all-cause mortality (ACM) risk and to determine age- and sex-specific distributions. MethodsWe evaluated patients without known coronary artery disease (CAD) who underwent CCTA at a single center. We assessed age- and sex-specific distributions (10th, 25th, 50th, 75th, and 90th percentiles) of LV mass index. ACM, the primary endpoint, was recorded over a median period of 5.1 [interquartile range: 1.4–8.4] years. The association between LV mass and mortality risk was assessed using multivariable Cox models adjusted for age, sex, medical history, coronary artery calcium (CAC) score and CCTA stenosis. Results4187 patients (mean age: 61.9 ​± ​11.7, 63 ​% male) were included. Male sex, African American ethnicity, Hypertension, CAC>400, and smoking were independent predictors of increased LV mass index. During the median 5.1 years of study follow, 265 (6.3 ​%) deaths occurred. Increased LV mass index percentiles were associated with increased risk of ACM. The addition of LV mass index percentiles improved discrimination and reclassification for mortality prediction over a model with age, sex, conventional risk factors, CAC score and CCTA stenosis severity (X2 improvement: 22.68, NRI: 28 ​%, both p ​< ​0.001). ConclusionIn a large sample of patients without known CAD who underwent CCTA, increased LV mass index provided independent and incremental prognostic value for all-cause mortality. Assessment of LV mass by CCTA, considering age and gender distribution, can be utilized clinically to identify patients with high myocardial mass.

The added value of deep learning to submaximal exercise electrocardiogram for the 10-year prediction of major adverse cardiovascular and cerebrovascular events

Abstract Background The added value of exercise ECG data to traditional risk factors in predicting cardiovascular events remains unclear. Deep learning of ECG signals has demonstrated state-of-the-art performance in detecting subtle abnormalities indicative of cardiovascular disease. We investigated whether deep learning analysis of exercise ECGs could improve the prediction of major cardiovascular and cerebrovascular events (MACCE) with respect to established risk models in a large population-based cohort. Purpose To assess the added value of submaximal exercise ECG measurements to established cardiovascular risk prediction models. Methods We obtained ECG recordings from 41,076 UK Biobank participants without cardiovascular disease (CVD) who underwent a submaximal exercise ECG test. We obtained ECG recordings from 41,076 UK Biobank participants without cardiovascular disease (CVD) who underwent a submaximal exercise ECG test. We created two deep neural network ECG risk scores: one derived from conventional ECG parameters, measured at rest, peak exercise and late-stage recovery, and another based on the complete ECG (Q-ECG). We assessed the added value of conventional ECG parameters and Q-ECG risk scores to the UK's current prediction algorithm (QRISK3). We estimate the association between ECG parameters and MACCE, using Cox Proportional hazard models, following adjustment for traditional risk factors. All models were internally validated via 5-fold cross-validation and 1000 bootstrap iterations. Predictive performance was evaluated using Harrel's C-index, Net Reclassification Index (NRI) and net benefit. Findings: Incident MACCE was reported in 4,082 (9.9%) individuals in the study population and 3,463(9.7%) individuals with valid ECG parameters over a median follow-up period of 12.5 years. We found combined conventional ECG and Q-ECG scores were independently associated with MACCE, following adjustment for multiple testing: adjusted hazard ratio [HZ] = 1.76 (95% CI:1.63-1.91); HZ = 1.14 (95% CI:1.10-1.18), respectively, per standard deviation increase. Both conventional ECG parameters and Q-ECG were predictive of MACCE, independent of clinical risk factors, C-index = 0.64 (95%CI 0.63-0.65); net benefit = 0.09(95% CI 0.07 - 0.11) and C-index = 0.56 (95% CI 0.55-0.57); net benefit = 0.07(95% CI 0.05 - 0.09). ECG measurement's modestly improved model discrimination over the bassline QRISK3 risk score when combined with QRISK3 risk factors for conventional markers and Q-ECG score, respectively; ΔC-index 0.03 (95% CI: 0.02 – 0.04) and ΔC-index 0.03 (95% CI: 0.02 – 0.03); However, we observed no significant improvements in classification at the current recommended threshold of 10%. Conclusion In individuals without a history of prior cardiovascular disease, ECG measures independently predict the risk of MACCE. When combined with QRISK3, neural-network-derived ECG risk scores marginally improve cardiovascular risk prediction over QRISK3.

Prediction of atrial fibrillation recurrence after ablation: the impact of machine learning and recurrence during the blanking period

Abstract Introduction Pulmonary vein isolation is superior to antiarrhythmic drugs for rhythm control in patients with atrial fibrillation (AF). Nevertheless, recurrences within the first year are still significant. Several clinical risk scores were developed to predict AF recurrence; however, despite good results in development cohorts, the results in external validation are usually poor, limiting the general applicability of these scores. Purpose We aimed to develop machine learning (ML) models to predict AF recurrences within the first year after catheter ablation and compare their performance with conventional risk scores. Methods We used a retrospective dataset of a tertiary hospital center, including all consecutive patients submitted to the first AF ablation between 2017 and 2021. The dataset included 76 features: patient characteristics, medical history, baseline echocardiography, calcium score, procedure variables, and early recurrence (ER) during the 90-day blanking period. The outcome was defined as the occurrence of an electrocardiographically documented late recurrence (LR) of atrial tachycardia/AF &amp;gt; 30s between 3-12 months. Three different supervised ML models (penalized logistic regression, random forests, and XGBoost) were developed on the training cohort, and hyperparameters were tuned using 10-fold cross-validation. A testing dataset was held out to estimate the final performance (25%). The following risk scores were calculated: APPLE, CAAP-AF, and BASE-AF2 (the last includes ER). Areas under the curve (AUC) of receiver operating characteristic curves were compared using DeLong’s test. Results 679 patients were included: 62.7% males, median age of 59±16 years, and 78.2% with paroxysmal AF. The median time from diagnosis to ablation was 2±4 years. 74.2% underwent radiofrequency ablation and the remaining cryoablation. ER occurred in 12.5%, and most of these patients also experienced LR (68.7 vs 19.6% in those without ER, p&amp;lt;0.001). LR was observed in 25.6%. The XGBoost model showed the best performance with an AUC 0.774, 95% confidence interval (CI) 0.688–0.844) outperforming existing scores (picture 1): APPLE score AUC 0.607, 95% CI 0.562–0.653, p-value&amp;lt;0.001; CAAP-AF score AUC 0.622, 95% CI 0.576–0.668, p-value=0.002; BASE-AF2 AUC 0.628, 95% CI 0.581–0.675, p-value= 0.003. Variable importance analysis showed a significant drop in the performance of the models when ER was not considered, indicating its high importance in predicting LR (picture 2). Conclusions Recurrence during the blanking period was the most important predictor of LR in our population. The ML model was superior to conventional risk scores. ML models might be an essential tool to improve the prediction of outcomes and clinical decision-making for optimal follow-up.ROC curvesFeature Importance

Enhancing primary prevention: the incremental predictive value of NTproBNP beyond ASCVD risk assessment

Abstract Background Atherosclerotic cardiovascular disease (ASCVD) is a global health challenge, prompting widespread use of conventional risk assessment tools (SCORE2 in Europe and the ASCVD Risk calculator in the United States) to guide preventive strategies. Emerging evidence indicates that NTproBNP may help refine risk stratification. We investigate whether measuring NTproBNP in primary prevention correlates with mortality and enhances prediction compared to conventional risk score stratification. Methods Utilizing National Health and Nutrition Examination Survey (NHANES) data (1999 to 2004) linked to the National Death Index, our cohort included individuals without a history of CVD at baseline and suitable for assessment of the ASCVD ACC/AHA risk score. Individuals were grouped based on computed 10-year risk (low: &amp;lt;5%; borderline: 5-7.5%; intermediate: 7.5-20%; high: ≥ 20%). NTproBNP levels were measured (Roche assay, cut-off &amp;gt; 125pg/mL considered as positive). Cox regression estimated the association between ASCVD risk categories, with or without NTproBNP, and all-cause and cardiovascular mortality. Results Among 5257 adults, 47% male, mean age 54 ± 9.1 years, the mean 10-year ASCVD risk was 8.17 ± 8.2%, grouped as: low: 53.3%; borderline: 11.5%; intermediate: 23.9%; and high: 11.3%. NTproBNP &amp;gt; 125pg/mL was measured in 12.2%, 16%, 23.2% and 48.6% individuals, respectively (p &amp;lt; 0.001). Over 16.2 ± 3.4 years, all-cause and cardiovascular death was 22.3% and 5.6%, respectively. Compared to low-risk, the hazard ratio (HR) for all-cause death increased to 2.85 (2.2-3.7), 5.6 (4.2-7.5), and 15.3 (11.7-19.9) for borderline, intermediate, and high-risk, respectively. A significant interaction with NTproBNP was observed (low: HR 1.96 (1.33-2.9); borderline: HR 2.5 (1.6-3.9); intermediate: HR 2 (1.6-2.6), high: HR 1.7 (1.4-2.2), p&amp;lt;0.01 for all interactions, see Fig. 1). For cardiovascular death, a significant interaction also occurred across the groups, but of higher magnitude in low and borderline groups (low: HR 3.2 (1.4-7.3); borderline: HR 5.5 (2.5-12.4), p &amp;lt; 0.001 for both). Conclusion Our findings suggest that NTproBNP provides incremental predictive value, approximately doubling the hazard of all-cause death across all risk categories, and more than tripling the hazard of CV death in low/borderline risk groups, compared to ASCVD risk alone. The integration of NTproBNP measurements into risk assessment strategies could refine primary prevention approaches.

Validation of existing risk scores for worsening renal function in patients with newly diagnosed heart failure: a longitudinal population cohort study

Abstract Background Worsening renal function (WRF) is one of the strongest predictors of outcome in patients with heart failure (HF). The presence of WRF often influences the decision to start, up-titrate, or discontinue disease-modifying HF therapies and is one of the key determinants of suboptimal guideline-directed medical therapy. However, there are now potential cardiovascular pharmacotherapies that have favourable cardiorenal outcomes and reduce the risk of renal decline. It is therefore important to identify patients with HF early in their disease trajectory who are at risk of developing WRF. Purpose The aim of this study was to evaluate two existing WRF risk scores (the Forman risk score and the Basel risk score) in a population-based longitudinal cohort diagnosed with incident HF. Methods Data for individuals with incident HF between 2016 to 2021 were extracted and analysed from the NELSON study conducted in Tayside, Scotland. WRF was defined as a composite event of (i) having two consecutive eGFRs declined by 40% or greater; (ii) having an eGFR &amp;lt; 15mL/min/1.73m2; (iii) initiation of sustained dialysis; (iv) development of end-stage kidney disease; or (v) receiving kidney transplantation. The primary outcome was the occurrence of WRF within 180 days of diagnosis of HF. Among patients hospitalised with HF, in-hospital WRF was also assessed. The area under the curve (AUC) was used to assess the discrimination performance of the two risk scores in individuals with incident HF diagnosed as either in-patients or out-patients. Results 4076 individuals (mean age 72.8 ± 13.2; 58% male; 1081 HFrEF, 646 HFmrEF, 1348 HFpEF, and 1001 HF with unknown ejection fraction (EF)) who were WRF-free at incident HF diagnosis were identified. Of these, 2095 (51.4%) were diagnosed as in-patients, and 1981 (48.6%) were diagnosed as out-patients. 285 (7%) patients developed WRF within 180 days post HF diagnosis. Among the 2,095 HF in-patients, 75 (3.6%) developed WRF before discharge. The Forman risk score had an AUC of 0.618 (95% CI: 0.585 – 0.651) in predicting 180-day WRF, and an AUC of 0.650 (0.587 – 0.713) for in-hospital WRF. The Basel risk score showed similar discriminating performance with an AUC of 0.618 (0.585 – 0.651) for 180-day WRF, and 0.656 (0.593 – 0.719) for in hospital WRF. Conclusions In a contemporary population, conventional risk scores such as the Forman or Basel risk scores have limited discrimination in predicting WRF in incident HF. New models with better discrimination for WRF prediction are therefore needed.

Enhancing primary prevention: the incremental predictive value of high-sensitivity cardiac troponin T beyond ASCVD risk assessment

Abstract Background Atherosclerotic cardiovascular disease (ASCVD) is a global health challenge, prompting widespread use of conventional risk assessment tools (SCORE2 in Europe and the ASCVD Risk calculator in the United States) to guide preventive strategies. Emerging evidence indicates that high-sensitivity cardiac troponin (hsTn) may refine risk stratification. Objective We investigate whether measuring hsTnT in primary prevention correlates with mortality and enhances prediction compared to conventional risk score stratification. Methods Utilizing National Health and Nutrition Examination Survey (NHANES) data (1999 to 2004) linked to the National Death Index, our cohort included individuals without a history of CVD at baseline and suitable for assessment of the ASCVD ACC/AHA risk score. Individuals were grouped based on computed 10-year risk (low: &amp;lt;5%; borderline: 5-7.5%; intermediate: 7.5-20%; high: ≥ 20%). hsTroponin T levels were measured (Roche assay, 99 th percentile 22ng/L for men and 17ng/L for women). Cox regression estimated association between ASCVD risk categories, with or without hsTnT, and all-cause and cardiovascular mortality. Results Among 5257 adults, 47% male, mean age 54 ± 9.1 years, the mean 10-year ASCVD risk was 8.17 ± 8.2%, grouped as: low: 53.3%; borderline: 11.5%; intermediate: 23.9%; and high: 11.3%. hsTnT above the 99 th percentile was measured in 1.1%, 2.7%, 4.9% and 13.4%, respectively (p &amp;lt; 0.001). Over 16.2 ± 3.4 years, all-cause and cardiovascular death was 22.3% and 5.6%, respectively. Compared to low-risk, the hazard ratio (HR) for all-cause death increased to 2.85 (2.2-3.7), 5.6 (4.2-7.5), and 15.3 (11.7-19.9) for borderline, intermediate, and high-risk, respectively. A significant interaction with hsTnT above 99 th percentile was observed (low: HR 3.5 (1.3-9.2); borderline: HR 4.75 (2.1-10.8); intermediate: HR 2.6 (1.8-3.7), high: HR 2.2 (1.7-2.9), p&amp;lt;0.01 for all interactions, see Fig. 1). For cardiovascular death, a significant interaction occurred in intermediate (HR 3.3 (2-5.4), p&amp;lt;0.001) and high-risk (HR 3.6 (2.2-6), p&amp;lt;0.001) groups. Conclusion Our findings suggest that hsTnT provides incremental predictive value, at least doubling the hazard of all-cause death across all risk categories, and tripling the hazard of CV death in intermediate and high-risk groups, compared to ASCVD risk stratification alone. The integration of troponin measurements into risk assessment strategies could refine primary prevention approaches.

Prognostic value of MELD-XI score and hemodynamic parameters in patients undergoing transcatheter tricuspid valve intervention

Abstract Background Hepatorenal dysfunction is a strong prognostic predictor in patients with tricuspid regurgitation. Patients are often hospitalized with acute heart failure due to volume overload and peripheral hypoperfusion. End-organ dysfunction is associated with the development of right ventricular failure related to hemodynamic changes such as low cardiac output and passive congestion from increased right ventricular filling pressures. Purpose This study aimed to investigate the prognostic value of hepatorenal dysfunction using the Model for End-stage Liver Disease (MELD) Score, the modified version of the score, MELD-XI, and of parameters determined during right-heart catheterization in patients undergoing transcatheter tricuspid valve intervention. Methods In this single-center prospective cohort study, data from patients undergoing edge-to-edge repair or heterotopic minimal invasive tricuspid valve repair were analyzed. Serum creatinine, INR and bilirubin were measured and the MELD score as well as the MELD XI score was calculated. All patients underwent right heart catheterization before the intervention. Receiver operator characteristics with calculation of the area under the curve (AUC) were performed to determine the predictive value of MELD (XI) score and hemodynamic parameters for the combined endpoint of rehospitalization due to decompensated heart failure and mortality within 3 months. Results Data from 36 consecutive patients were analyzed. 28 patients underwent edge-to-edge repair and 8 patients heterotopic minimal invasive tricuspid valve repair. Mean age of patients was 80.2 ± 5.9 years; 33.3% (12/36) were male. Baseline characteristics are shown in table 1. Rehospitalization due to decompensated heart failure occurred in 9/36 patients (25%). Mortality within the first 3 months was 11% (4/36). The predictive value for the composite endpoint of the modified MELD IX score was AUC 0.82 (95 % CI 0.63-0.99; p=0.009), while the conventional MELD score had an AUC of 0.77 (95 % CI 0.55-0.99; p = 0.059) (figure 1). RV pressure (AUC 0.84, 95% CI 0.66-1.00, p= 0.012) and RA pressure (AUC 0.83, 95% CI 0.67-0.98, p=0.016) showed a good predictive value for the composite endpoint (figure 2). The combination of MELD XI (at admission) and RV or PA pressure improved the AUC (MELD XI + RV: AUC 0.85; MELD XI + PA: AUC 0.84). Conclusion In this pilot-study, the modified MELD XI score, as indicator of hepatorenal dysfunction, was identified as a potential marker for short-term outcome after interventional tricuspid valve treatment.

Artificial intelligence assessment of heart failure and mortality after acute stroke

Abstract Background/Introduction The adverse cardiovascular outcomes following a stroke have shown significant associations with post-stroke mortality. However, studies focusing on predicting heart failure and its impact on long-term mortality are scarce. Purpose To address this gap, we employed an commercially approved artificial intelligence/machine learning enabled software as a medical device (AiTiALVSD version 1.00.00) which uses only 12 lead electrocardiogram data, resulting probability score for predicting left ventricular systolic dysfunction. We derived the AiTiALVSD probability score from initial ECGs at stroke admission, and assess its implications for prediction of heart failure and long-term mortality. Methods We collected data from 335 ischemic stroke patients admitted to a cardiovascular center during acute periods between 2013 and 2019. The AiTiALVSD score was employed to predict heart failure with reduced ejection fraction and heart failure with mild reduced ejection fraction, analyzed through logistic regression. Subsequently, we used the Cox Hazard proportional model, incorporating the AiTiALVSD score and clinical factors, to predict long-term mortality. Results The model incorporating the AiTiALVSD score with clinical variables demonstrated a higher AUC for predicting heart failure (0.905 [95% CI, 0.827-0.927] vs. 0.828 [95% CI, 0.749-0.907], p&amp;lt;0.004) than clinical variables alone. Moreover, for long-term mortality prediction, this model outperformed the conventional mortality score (AUC, 0.884 [95% CI, 0.836-0.933] vs. 0.755 [95% CI, 0.690-0.820], p&amp;lt;0.001). Conclusion Multivariable models which combine the AiTiALVSD score with clinical factors, effectively predicts heart failure and long-term mortality. This model could enhance therapeutic strategy tailoring by providing acute phase prognostic predictions for individual stroke patients.Prediction performance for heart failurePrediction performance for mortality

Long-term outcome of diabetic patients after unprotected left main percutaneous coronary interventions

Abstract Background Diabetes mellitus (DM) represents an elevated risk in terms of percutaneous coronary intervention (PCI), especially in patients with significant left main (LM) disease. Purpose Our aims were to investigate the 60-month mortality and event-free survival of consecutive patients with DM undergoing unprotected LM PCI at our centre and sought to determine the independent predictors of event-free survival. Methods We included all patients undergoing unprotected LM PCI at our tertiary care centre between 1 January 2007 and 31 December 2014. The primary endpoint was the composite of all-cause mortality, myocardial infarction, and target lesion revascularization at 60 months. Survival curves were constructed using the Kaplan-Meier non-parametric estimator. ROC analysis was used to investigate the predictive power of conventional risk scores (SYNTAX, SYNTAX II, EuroSCORE II, additive and logistic EuroSCORE, ACEF, GRACE 2). Independent predictors were determined by univariate and multivariate analysis. Results A total of 513 patients (mean age 68 ± 12 years, 64% male) underwent unprotected LM PCI during this period, among these 310 (60%) were non-diabetic (non-DM) and 203 (40%) were diabetic (DM). In terms of indication, 157 (31%) elective and 356 (69%) acute LM PCI were performed. There was no significant difference between the non-DM and DM groups in terms of acute indication (218 (70%) vs. 138 (68%), p=0.574). Within the DM group, 135 (67%) were treated with oral antidiabetics (OAD) and 68 (33%) were treated with insulin (IDM). The mean survival (34.6 vs. 39.7 vs. 43.1 months, p=0.003) and event-free survival (29.3 vs. 34.7 vs. 39.3 months, p=0.001) were significantly lower in the IDM and OAD groups compared to non-DM, respectively. Of the risk scores, the SYNTAX score was the most predictive in diabetic patients (AUC=0.767). Based on the SYNTAX score, diabetic patients with a low SYNTAX score (&amp;lt;23) had less events (59% vs. 66%, p=0.043) than those having a moderate/high SYNTAX score (≥23). LVEF (HR 0.976; 95% CI, 0.965-0.987, p&amp;lt;0.001), right coronary artery occlusion (HR 2.031; 95% CI, 1.346-3.064, p=0.001), acute coronary syndrome (HR 1.861; 95% CI, 1.252-2.767, p=0.002), use of FFR (HR 0.481; 95% CI, 0.265-0.871, p=0.016), true bifurcation lesion (HR 1.559; 95% CI, 1.068-2.277, p=0.021) and newer DES generation (HR 0.706; 95% CI, 0.568-0.877, p=0.002) were predictors of event-free survival in diabetics by univariate analysis. Glomerular filtration rate (HR 0.982; 95% CI, 0.973-0.991, p&amp;lt;0.001) emerged as an independent predictor according to multivariate analysis. Conclusion Consecutive diabetic patients with significant unprotected LM stenosis have poor long-term outcome after LM PCI. Anatomical complexity of the lesions is a significant predictor of event-free survival. Careful intervention planning and the use of appropriate tools can improve the outcome in this high-risk population.

Machine learning based prediction models for cardiovascular disease risk using electronic health records data: systematic review and meta-analysis

Abstract Background Cardiovascular disease (CVD) remains a major cause of mortality in the UK, prompting the need for improved risk predictive models for primary prevention. Machine learning (ML) models utilizing electronic health records (EHR) offer potential enhancements over traditional risk scores like QRISK3 and ASCVD. Objective To systematically evaluate and compare the efficacy of ML models against conventional CVD risk prediction algorithms using EHR data for medium to long-term (5-10 years) CVD risk prediction. Methods A systematic review and random-effect meta-analysis were conducted according to PRISMA guidelines, assessing studies from 2010 to 2024. We retrieved 32 ML models and 26 conventional statistical models from 20 selected studies, focusing on performance metrics such as Area Under the Curve (AUC) and heterogeneity across models. Results ML models, particularly random forest and deep learning, demonstrated superior performance, with the highest recorded pooled AUCs of 0.865 (95% CI: 0.812-0.917) and 0.847 (95% CI: 0.766-0.927), respectively. These significantly outperformed the conventional risk score of 0.765 (95% CI: 0.734-0.796). However, significant heterogeneity (I² &amp;gt; 99%) and potential publication bias were noted across the studies. Conclusion While ML models show enhanced calibration for CVD risk, substantial variability and methodological concerns limit their current clinical applicability. Future research should address these issues by enhancing methodological transparency and standardization to improve the reliability and utility of these models in clinical settings. Implications This study highlights the advanced capabilities of ML models in CVD risk prediction and emphasizes the need for rigorous validation to facilitate their integration into clinical practice. This review was registered with PROSPERO (CRD42021250058).

Prediction of in-hospital mortality risk for patients with acute ST-elevation myocardial infarction after primary PCI based on predictors selected by GRACE score and two feature selection methods.

Accurate in-hospital mortality prediction following percutaneous coronary intervention (PCI) is crucial for clinical decision-making. Machine Learning (ML) and Data Mining methods have shown promise in improving medical prognosis accuracy. We analyzed a dataset of 4,677 patients from the Regional Vascular Center of Primorsky Regional Clinical Hospital No. 1 in Vladivostok, collected between 2015 and 2021. We utilized Extreme Gradient Boosting, Histogram Gradient Boosting, Light Gradient Boosting, and Stochastic Gradient Boosting for mortality risk prediction after primary PCI in patients with acute ST-elevation myocardial infarction. Model selection was performed using Monte Carlo Cross-validation. Feature selection was enhanced through Recursive Feature Elimination (RFE) and Shapley Additive Explanations (SHAP). We further developed hybrid models using Augmented Grey Wolf Optimizer (AGWO), Bald Eagle Search Optimization (BES), Golden Jackal Optimizer (GJO), and Puma Optimizer (PO), integrating features selected by these methods with the traditional GRACE score. The hybrid models demonstrated superior prediction accuracy. In scenario (1), utilizing GRACE scale features, the Light Gradient Boosting Machine (LGBM) and Extreme Gradient Boosting (XGB) models optimized with BES achieved Recall values of 0.944 and 0.954, respectively. In scenarios (2) and (3), employing SHAP and RFE-selected features, the LGB models attained Recall values of 0.963 and 0.977, while the XGB models achieved 0.978 and 0.99. The study indicates that ML models, particularly the XGB optimized with BES, can outperform the conventional GRACE score in predicting in-hospital mortality. The hybrid models' enhanced accuracy presents a significant step forward in risk assessment for patients post-PCI, offering a potential alternative to existing clinical tools. These findings underscore the potential of ML in optimizing patient care and outcomes in cardiovascular medicine.

Cloud-Enabled Mobile-Based Approach for Enhancing Psychosis Cognitive Assessment.

The study aims to assess the feasibility of developing cognitive tools and integrating the cloud-enabled mobile-based technology into routine clinical practice for psychotic patients. Furthermore, it aims to investigate the correlation between the results obtained using developed tools and established clinical measures, offering valuable insights into tools for enhancing the accuracy, efficiency and ease of administration of cognitive evaluation. A total of 160 participants were recruited (83 outpatients with early course of schizophrenia (SZs) and 77 healthy controls [HCs]). The participants were subjected to cognitive assessments, and the data were collected by cognitive assessment digital smart tool (CADST) and PGI memory scale (PGIMS) to assess attention (ATT) and working memory (WM). Outcome measures of these parameters were digit span (score,time) for ATT and delayed recall (score,time) for WM. The total average score in HCs was significantly higher than in SZs for ATT and WM, and CADST was significantly correlated with PGIMS for evaluating ATT and WM. Furthermore, test completion times for ATT and WM were observed more in SZs although most of SZs had achieved scores as high as that of HCs. The potential of CADST as a valuable addition to the conventional cognitive assessment method is highlighted, showing promising feasibility and strong correlations with the established tool. The importance of integrating time parameters suggests broader implications for understanding cognitive function beyond conventional scoring metrics. It demonstrates the effective and accurate approach for large-scale screening of cognitive parameters in public service settings.

A newly developed, easy-to-use prehospital drug-derived score compared with three conventional scores: A prospective multicenter study.

The use of medications by emergency medical services (EMS) is increasing. Conventional scores are time-consuming and therefore difficult to use in an emergency setting. For early decision-making, an easy-to-use score based on the medications administered by the EMS may have prognostic value. The primary objective of this study was to develop the prehospital drug-derived score (PDDS) for 2-day mortality. A prospective, multicenter, ambulance-based cohort study was conducted in adults with undifferentiated acute diseases treated by EMS and transferred to the emergency department. Demographic data, prehospital diagnosis data, prehospital medication and variables for the calculation of the National Early Warning Score 2 (NEWS2), Rapid Emergency Medicine Score (REMS), and Rapid Acute Physiology Score (RAPS) were collected. The PDDS was developed and validated, establishing three levels of risk of 2-day mortality. The predictive capability of each score was determined by the area under the curve of the receiver operating characteristic curve (AUROC) and compared using the Delong's test (p-value). A total of 6401 patients were included. The PDDS included age and the use of norepinephrine, analgesics, neuromuscular blocking agents, diuretics, antihypertensive agents, tranexamic acid, and bicarbonate. The AUROC of PDDS was .86 (95% CI: .816-.903) versus NEWS2 .866 (95% CI: .822-.911), p = .828; versus REMS .885 (95% CI: .845-.924), p = .311; versus RAPS .886 (95% CI: .846-.926), p = .335, respectively. The newly developed easy-to-use prehospital drug-derived PDDS score has an excellent predictive value of early mortality. The PDDS score was comparable to the conventional risk scores and therefore might serve as an alternative score in the prehospital emergency setting.

Diffusion weighted image-guided transitional zone scoring in the detection of transitional zone prostate cancer: comparison with current PI-RADS v2.1 scoring.

To compare the performance of diffusion-weighted imaging-guided transitional zone (TZ) lesion scoring on T2-weighted imaging (DWI-guided TZ scoring) to conventional PI-RADS TZ scoring. Forty patients carried transition zone prostate cancer (TZPCa), and 40 patients had benign prostatic hyperplasia without TZPCa. A lesion-base, one-to-one correlation between the pathologic mapping sheet and the corresponding MR imaging was conducted by consensus between the genitourinary-specialized radiologist and pathologist. DWI-guided TZ scoring was defined as evaluating the DWI/apparent diffusion coefficient (ADC) images first, identifying the suspicious foci, then correlating the foci with the T2-weighted imaging, and finally assigning the PI-RADS score based on PI-RADS v2.1. Three other radiologists independently recorded the PI-RADS v2.1 scoring for TZ and the DWI-guided TZ scoring, with a time interval of 4weeks. When a PI-RADS score of ≥ 3 was considered a positive lesion, the specificity, PPV, NPV and sensitivity between the DWI-guided TZ scoring and conventional PI-RADS TZ scoring were 0.896 vs. 0.542 (p < .001), 0.764 vs. 0.439 (p < .001), 0.853 vs. 0.759 (p = .001), and 0.687 vs. 0.676 (p = .836), respectively. When PI-RADS scores ≥ 4 was considered cancer-positive, the specificity and PPV were also higher when applying DWI-guided TZ scoring (0.986 vs. 0.944, p = .007; 0.943 vs. 0.810, p = .009, respectively); however, the sensitivity and NPV were not statistically different (0.468 vs. 0.468, p = .998; 0.785 vs. 0.776, p = .537, respectively). The interobserver agreement presented as κ-value was higher in DWI-guided TZ scoring (0.584) than in conventional PI-RADS TZ scoring (0.155) (p = .003). DWI-guided TZ scoring improves the interobserver agreement, specificity, and predictive value without impairing the sensitivity.

Automatic quantification of REM sleep without atonia reliably identifies patients with REM sleep behavior disorder: a possible screening tool?

REM Sleep Behavior Disorder (RBD) is characterized by absence of physiological muscle atonia during REM sleep (REM sleep without atonia, RWA). Nigro-striatal dopaminergic impairment is a feature of Parkinson disease (PD) and can be identified in prodromal stages as well, such as idiopathic RBD (iRBD). Aims of this study are to explore the efficacy of an automatic RWA quantification in identifying RBD patients and the correlation between RWA and nigro-striatal dopaminergic function. Forty-five iRBD, 46 PD with RBD, 24 PD without RBD patients and 11 healthy controls were enrolled in the Genoa Center (group A) and 25 patients with iRBD (group B) were enrolled in the Danish Center. Group A underwent brain [123I]FP-CIT-SPECT and group B underwent brain [18F]PE2I-PET as measures of nigro-striatal dopaminergic function. Chin muscle activity was recorded in all subjects and analyzed by applying a published automatic algorithm. Correlations between RWA and nigro-striatal dopaminergic function were explored. The automatic quantification of RWA significantly differentiated RBD from non-RBD subjects (AUC = 0.86), although with lower accuracy compared with conventional visual scoring (AUC = 0.99). No significant correlation was found between RWA and nigro-striatal dopaminergic function. The automatic quantification of RWA is a reliable tool to identify subjects with RBD and may be used as a first-line screening tool, but without correlations with nigro-striatal dopaminergic functioning.