Modest Predictive Value Research Articles

Comparison of machine learning and traditional methods for prediction of adverse clinical events after percutaneous coronary intervention

Abstract Background Accurate prediction of clinical outcomes following percutaneous coronary intervention (PCI) is essential for mitigating risk and periprocedural planning. Traditional risk models have demonstrated modest predictive value. Machine learning (ML) models offer an alternative risk stratification that may provide improved predictive accuracy. We performed a systematic review and meta-analysis comparing ML models and traditional risk scores for prediction of clinical outcomes after PCI. Methods This study was reported according to PRISMA guidelines. PubMed, EMBASE, Web of Science and Cochrane databases were searched until 1st November, 2023 for studies comparing ML models with traditional statistical methods for event prediction after PCI. The primary outcome was comparative discrimination measured by C-statistics with 95% confidence intervals between ML models and traditional methods in estimating the risk of all-cause mortality, major bleeding and the composite outcome major adverse cardiovascular events (MACE). Results Thirty-four models were included across 13 observational studies (4105916 patients). The summary C-statistic all ML models across all clinical endpoints was 0.81 (95% CI, 0.78-0.85), compared to traditional methods 0.79 (95% CI, 0.74-0.85). The difference in C-statistic was 0.02 (95% CI, 0.002-0.04, p=0.56). Of included studies, only 1 model was externally validated and 10 models were calibrated. Conclusion ML models did not outperform traditional risk scores in the discrimination of all-cause mortality, major bleeding or MACE following PCI. While integration of ML algorithms into electronic healthcare systems may improve periprocedural risk stratification, immediate implementation in the clinical setting remains uncertain. Further research is required to overcome methodological and validation limitations.

Artificial intelligence-enhanced electrocardiography predicts 10-year risk of atherosclerotic cardiovascular disease

Abstract Background The ACC/AHA pooled cohort equations (PCE) calculates the 10-year primary risk of atherosclerotic cardiovascular disease (ASCVD), an indication to initiate primary prevention statin therapy in international guidelines. The validity of PCE is limited by its overestimation of risk in high-risk cohorts and an ethnic heterogeneity. Electrocardiography (ECG) has not been incorporated in current ASCVD risk estimation tools. Artificial Intelligence (AI)-based models are capable to capture patterns that are informative for cardiovascular prognosis. Purpose We explored the predictive value of a discrete-time survival model in 10-year ASCVD risk in comparison to PCE. Methods We developed an AI-ECG model to predict the 10-year ASCVD risk in an unselected secondary care population from the USA comprised of 1,163,401 ECGs from 189,539 patients. Data was split into training/validation and hold-out test sets by patient ID. Our AI-ECG model uses deep learning with a residual convolutional neural network with a discrete-time survival loss function to output subject-specific survival curves, accounting for both time to event and censorship (i.e., loss to follow up). We compared the performance of out AI-ECG model using a subset of outpatient data (n=4590) of the BIDMC dataset to PCE and ASCVD risk factors including systolic blood pressure, total cholesterol, HDL cholesterol, hypertension, smoking history, diabetes mellitus, and ethnicity. Performances were evaluated using cox survival analysis. Our AI-ECG model was externally validated in the UKB Biobank (UKB), where we compared the performance of our AI-ECG model to predictions of existing AI-ECG models such as the Stanford Estimator of Electrocardiogram Risk (SEER) model. Results Our AI-ECG successfully predicted future ASCVD events in individuals (C-index 0.721 (0.719-0.723), 5-year AUC 0.761 (0.758-0.763), and differentiated between high and low risk group with an adjusted HR of 2.33 (2.26-2.41). Upon internal validation our AI-ECG model showed a superior performance (C-index 0.679 (0.651-0.708)) than PCE (0.605 (0.577-0.634)) and all risk factors combined (C index 0.642 (0.613-0.672)). When combining AI-ECG predictions with all risk factors, the model showed additional predictive value (0.686 (0.657-0.715)) (Figure 1). Our AI-ECG model showed modest predictive value within the UKB dataset (C-index 0.655 (0.637-0.673)), but significantly outperformed the SEER model (C-index 0.547 (0.527-0.567), p <0.0001 for comparison with AIRE). Conclusion We have shown the potential of AI-ECG models in predicting future ASCVD risks. AI-ECG models demonstrated accurate discrete time survival prediction and good classification performance superior to that of the SEER model, as well as PCE and ASCVD risk factors. . Accurate risk assessment of ASCVD may help prevent adverse events in high risk subjects and save unnecessary pharmacological interventios in low risk subjects.Figure 1

Machine-learning approaches for risk prediction after percutaneous coronary intervention: a systematic review and meta-analysis

Abstract Background Accurate prediction of clinical outcomes following percutaneous coronary intervention (PCI) is essential for mitigating risk and periprocedural planning. Traditional risk models have demonstrated modest predictive value. Machine learning (ML) models offer an alternative risk stratification that may provide improved predictive accuracy. Methods This study was reported according to PRISMA, CHARMS and TRIPOD guidelines. PubMed, EMBASE, Web of Science and Cochrane databases were searched until 1st November, 2023 for studies comparing ML models with traditional statistical methods for event prediction after PCI. The primary outcome was comparative discrimination measured by C-statistics with 95% confidence intervals between ML models and traditional methods in estimating the risk of all-cause mortality, major bleeding and the composite outcome major adverse cardiovascular events (MACE). Results Thirty-four models were included across 13 observational studies (4105916 patients). For all-cause mortality, the pooled C-statistic for top-performing ML models was 0.89 (95% CI, 0.84-0.91), compared to 0.86 (95% CI, 0.80-0.93) for traditional methods (p=0.54). For major bleeding, the pooled C-statistic for ML models was 0.80 (95% CI, 0.77-0.84), versus 0.78 (95% CI, 0.77-0.79) for traditional methods (p=0.02). For MACE, the C-statistic for ML models was 0.83 (95% CI, 0.75-0.91), compared to 0.71 (95% CI, 0.69-0.74) for traditional methods (p=0.007). Out of all included models, only one model was externally validated. Calibration was inconsistently reported across all models. Prediction Model Risk of Bias Assessment Tool (PROBAST) tool demonstrated high risk of bias across all studies. Conclusion ML models marginally outperformed traditional risk scores in the discrimination of all-cause mortality, major bleeding or MACE following PCI. While integration of ML algorithms into electronic healthcare systems has been hypothesised to improve periprocedural risk stratification, immediate implementation in the clinical setting remains uncertain. Further research is required to overcome methodological and validation limitations​.

Clostridioides difficile Infection: Use of Inflammatory Biomarkers and Hemogram-Derived Ratios to Predict Mortality Risk in Hospitalized Patients.

Clostridioides difficile infection (CDI) is a significant cause of mortality, especially in healthcare environments. Reliable biomarkers that can accurately predict mortality in CDI patients are yet to be evaluated. Our study aims to evaluate the accuracy of several inflammatory biomarkers and hemogram-derived ratios in predicting mortality in CDI patients, such as the neutrophil-to-lymphocyte ratio (NLR), the systemic immune-inflammation index (SII), the platelet-to-neutrophil ratio (PNR), the derived neutrophil-to-lymphocyte ratio (dNLR), C-reactive protein (CRP), the platelet-to-lymphocyte ratio (PLR), and procalcitonin (PCT). NLR showed a sensitivity of 72.5% and a specificity of 58.42% with an area under curve (AUC) = 0.652. SII had a sensitivity of 77.5%, a specificity of 54.74%, and an AUC = 0.64. PNR, neutrophils, dNLR, and lymphocytes had lower AUCs which ranged from 0.595 to 0.616, with varied sensitivity and specificity. CRP, leukocytes, and platelets showed modest predictive values with AUCs below 0.6. PCT had a sensitivity of 100%, a low specificity of 7.41%, and an AUC = 0.528. We conducted a retrospective analysis of CDI patients from two different hospital settings in Italy and Romania during the COVID-19 pandemic, from 1 January 2020 to 5 May 2023. Statistical analyses included t-tests, Wilcoxon rank-sum tests, χ2 tests, and multivariate logistic regression to identify predictors of mortality. ROC analysis assessed the accuracy of biomarkers and hemogram-derived ratios. A p value < 0.05 was considered significant. Neutrophils, dNLR, NLR, SII, and PNR are valuable biomarkers for predicting mortality in CDI patients. Understanding these predictors can improve risk stratification and clinical outcomes for CDI patients.

Fostering the use of reflective learning strategies in teaching big data analytics and data science: Enhancing the quality of epidemiological evidence and decision-making

PURPOSEThis study investigated the frequency of use and perceived effectiveness of reflective learning strategies among university educators and practitioners in big data analytics and data science, including epidemiologists. Reflective learning fosters critical thinking and problem-solving skills essential for navigating the complexities of big data. It encourages systematic evaluation of methodologies and the identification and mitigation of biases. These skills are particularly relevant in epidemiology, where the prediction and management of disease outbreaks, as well as public health decisions, increasingly depend on big data analytics. METHODOLOGYA cross-sectional design was employed with a diverse sample of 93 participants from multiple countries, averaging 15 years of teaching and practice experience. Informed by the Diffusion of Innovation Theory and the Theory of Planned Behavior, data were collected in Fall 2023 using an online questionnaire. The questionnaire focused on personal disposition, contextual favorability, and characteristics of reflective learning as predictors of the frequency and perceived effectiveness of reflective learning, analyzed using multiple regression modeling. Acceptable levels of reliability and criterion validity were achieved. This study was approved by the Ethics Panel of Canterbury Christ Church University, UK. RESULTSIn general, reflective learning strategies were perceived as effective for enhancing cognitive skills, such as critical thinking and problem-solving, but less so for technical skills like algorithm and model selection. Preferred methods included real-world applications, problem-based learning, and collaborative projects. Frequency of use was positively correlated with perceived effectiveness. Personal disposition and characteristics of reflective learning scores were significant predictors of frequency of use (R-squared = 0.32), with older participants being less engaged. The characteristics of reflective learning score was the sole significant predictor of perceived effectiveness, though with a modest predictive value (R-squared = 0.09). Contextual favorability did not significantly predict either frequency of use or perceived effectiveness of reflective learning strategies, implying a lesser role for external factors as drivers of engagement. CONCLUSIONSThis study underscores the importance of reflective learning in enhancing analytical and problem-solving skills, and suggests that intrinsic motivation may be a key factor underpinning the use of reflective learning strategies by educators and practitioners in big data analytics and data science. These findings have major implications for curriculum and professional development, particularly for epidemiologists who must continuously adapt to new data and methodologies to effectively address public health challenges.

Application of the international system for reporting serous fluid cytopathology on pleural effusion cytology with paired pleural biopsy: A new insight and novel approach on risk of malignancy.

The risk of malignancy (ROM) remains an area of interest for further evaluation in reporting systems including in International System for reporting serous fluid cytopathology (TIS), which is a standardized system for reporting effusion cytology. Herein, we report our findings in further investigation of ROM in TIS by studying on paired pleural effusion specimens and corresponding pleural biopsies with emphasis on negative for malignancy, and atypia of undetermined significance categories. The Johns Hopkins Hospital pathology database was retrospectively searched for patients with a pleural biopsy (PBX) and a paired pleural effusion (PF) cytology specimens over a 4-year period. We employed the TIS categories. The following statistical parameters were evaluated: sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and ROM. A total of 223 patient cases were included. Effusions TIS reclassification and ROM were as follows: 1.8% non-diagnostic (ROM 75%), 75.8% negative for malignancy (ROM 23%), 4.9% atypical cells of undetermined significance (ROM 45%), 2.2% suspicious for malignancy (ROM 80%), and 15.2% malignant (ROM 100%). Overall accuracy, sensitivity, specificity, PPV and NPV were calculated and were 79.4%, 45%, 97.7%, 91.2% and 77%, respectively. Among, discordant cases diagnosed negative for malignancy on PF and positive for malignancy on PBX, there were significant number of lymphomas, mesotheliomas, and sarcomas. Lung cancer was the most common carcinoma; however, rare types of carcinomas were noted. Cells blocks and immunohistochemistry (IHC) studies were utilized to confirm either malignant conditions or rule out malignancy in both cell blocks and histology biopsies. This study demonstrates the high specificity and ROM for 'malignant' and 'suspicious for malignancy' categories in the TIS reporting system and highlights the modest negative predictive value for the 'negative for malignancy' category. Although Tissue biopsies are usually considered as 'gold standard', any definitive diagnosis of malignancy of body fluid should be considered positive for malignancy in further clinical decision-making.

The Modified Risk of Paradoxical Embolism Score Is Associated with Patent Foramen Ovale in Patients with Ischemic Stroke: A Nationwide US Analysis.

The Risk of Paradoxical Embolism (RoPE) score was developed to identify stroke-related patent foramen ovale (PFO) in patients with cryptogenic stroke. We conducted a retrospective analysis of the 2016 to 2020 National Inpatient Sample to determine the performance of the modified RoPE score in identifying the presence of a PFO in patients with acute ischemic stroke (AIS). A total of 3,338,805 hospital admissions for AIS were analysed and 3.0% had PFO. Patients with PFO were younger compared to those without a PFO (median 63 years vs. 71 years, p < 0.001) and fewer were female (46.1% vs. 49.7%, p < 0.001). The patients with PFO had greater mean modified RoPE scores (4.0 vs. 3.3, p < 0.001). The area under the curve for the RoPE score in predicting PFOs was 0.625 (95%CI 0.620-0.629). The best diagnostic power of the RoPE score was achieved with a cut-off point of ≥4 where the sensitivity was 55% and the specificity was 64.2%. A cut-off point of ≥5 increased the specificity (83.1%) at the expense of sensitivity (35.8%). The strongest predictor of PFOs was deep vein thrombosis (OR 3.97, 95%CI 3.76-4.20). The modified RoPE score had modest predictive value in identifying patients with PFO among patients admitted with AIS.

O-155 Limited utility of mosaicism reporting in PGT-A: independent confirmation from a European multisite, double-blinded analysis including 4293 embryo transfers following the first US experience

Abstract Study question Are results showing limited clinical utility of mosaicism reporting in PGT from a blinded mosaic embryo transfer American study, confirmed in a European setting? Summary answer Even between different patient populations and clinical settings, putative mosaicism reporting has limited clinical utility for embryo selection when co-evaluated with other clinical/embryological factors. What is known already Intermediate chromosomal copy number (ICN) results in NGS-based PGT-A are commonly interpreted as embryo mosaicism. A recent prospective, non-selection study in America showed that reporting mosaicism would not significantly improve embryo selection. At odds with this, results suggesting that embryo selection based on putative mosaicism can improve clinical outcomes continue to be reported, possibly reflecting limitations in accurate mosaicism identification or retrospective analysis biases. This study replicates the design of previously reported American study, using the same PGT-A assay, examining the predictive value of whole-chromosome ICN (wICN) and segmental ICN (sICN) following concurrent copy-number and genotyping-PGT in a European setting. Study design, size, duration A multisite double-blinded study involving 14 European IVF clinics was conducted from April 2022 to July 2023, including 3365 patients and 4293 single embryo transfers. Embryos, reported as negative for uniform aneuploidies, including those with wICN or sICN, were chosen for transfer based solely on standard developmental and morphological evaluation. Primary outcome measure was live birth rate defined as any pregnancy reaching the 24th week of gestation (LBR). Participants/materials, setting, methods Trophectoderm biopsies were analysed using a validated targeted-NGS assay with approximately 5000 loci across the genome, providing both quantitative and genotyping information to support aneuploidy classification. Putative mosaicism was inferred from ICN deviations from the expected two copies for autosomes and partly supported by corresponding SNP B-allele frequency patterns. Related clinical and embryological variables were analysed in a multivariate analysis and a Random Forrest regression used to model the clinical utility of putative mosaicism. Main results and the role of chance Of the embryos transferred, 6.8% (293/4293) were wICN only; 6.2% (269/4293) were sICN only and 0.5% (23/4293) were a combination of wICN and sICN. The detected ICN ranged from 23%-84% with no overrepresentation of any individual chromosomes. The LBR of embryos in the control group (without ICN), sICN and wICN were 46.8% (1736/3708; 95%CI:45.2-48.4%), 36.4% (98/269; 95%CI:30.9-42.3%) and 39.2% (115/293; 95%CI:33.8-44.9%) respectively. The miscarriage rate for control, sICN and wICN were 12.9% (257/1993; 95%CI:11.5-14-4%), 19.7% (24/122; 95%CI:13.6-27-6%) and 16.7% (23/138; 95%CI:11.4-23.8%), respectively. Logistic regression (P = 0.05) showed that the presence of sICN (OR: 0.7; 95%CI: 0.54-0.9) and wICN (OR: 0.62; 95%CI: 0.36-1.07) were modestly but significantly associated with LBR along with other embryological and clinical factors including embryo morphology (OR: 0.89; 95%CI: 0.84-0.96), day of biopsy (OR: 0.65;95%CI: 0.56-0.75), previous ET failures (OR: 0.79; 95%CI: 0.71-0.88) that were more strongly associated. No significant effect was observed of any specific intermediate CN cut-off on the association with LBR. Random Forrest regression, considering all correlating variables, showed a LBR prediction (AUC) equal to 0.578 without putative mosaicism and AUC of 0.583 with putative mosaicism included, a negligible increase considering the low incidence of mosaicism in our clinical setting along with its modest predictive value. Limitations, reasons for caution Follow-up prenatal and post-natal data at this time was not available, hence conclusions about these outcomes wasn’t possible. Despite the overall large sample size, the analysis of specific effects at different ICN values lacked sufficient power. Furthermore, this study’s data is platform-specific and cannot be translated to other PGT-A assays. Wider implications of the findings Although putative mosaicism was associated with modestly impaired pregnancy rate, two independent studies now confirm that reporting mosaicism has limited impact when prospectively selecting embryos for transfer based on standard embryological parameters. Considering the established downsides, including potential embryo disposal, the practice of mosaicism reporting in PGT is questionable. Trial registration number Not applicable

Prognostic Value of the Advanced Lung Cancer Inflammation Index Ratio in Patients with Acute Myocardial Infarction Complicated by Cardiogenic Shock: A Cohort Study.

Acute myocardial infarction (AMI) complicated by cardiogenic shock (CS) carries a high mortality risk. Inflammation and nutrition are involved in the pathogenesis and prognosis of both AMI and CS. The advanced lung cancer inflammation index ratio (ALI) combines the inflammatory and nutritional status. Our present study aimed to explore the prognostic value of ALI in patients with CS following AMI. In total, 217 consecutive patients with AMI complicated by CS were divided into two groups based on the ALI admissions cut-off: 12.69 and 12.69. The primary endpoint of this study was 30-day all-cause mortality. The secondary endpoints were gastrointestinal hemorrhage and major adverse cardiovascular events (MACEs), including 30-day all-cause mortality, atrioventricular block, ventricular tachycardia/ventricular fibrillation, and nonfatal stroke. The association of ALI with the study endpoints was analyzed by Cox regression analysis. During the 30-day follow-up period after admission, 104 (47.9%) patients died and 150 (69.1%) suffered MACEs. The Kaplan-Meier analysis revealed significantly higher cumulative mortality and lower MACE rates in the low-ALI group compared to the high-ALI group (both log-rank p 0.001). The 30-day mortality rate was significantly higher in patients with ALI 12.69 compared to ALI 12.69 (72.1% vs. 22.6%; p 0.001). Furthermore, the incidence of MACEs was higher in patients with ALI 12.69 (85.6% vs. 51.9%; p 0.001). The receiver operating curve showed that ALI had a modest predictive value (area under the curve [AUC]: 0.789, 95% confidence interval [CI]: 0.729, 0.850). After multivariable adjustment, ALI 12.69 was an independent predictor for both 30-day all-cause mortality (hazard ratio [HR]: 3.327; 95% CI: 2.053, 5.389; p 0.001) and 30-day MACEs (HR: 2.250; 95% CI 1.553, 3.260; p 0.001). Furthermore, the addition of ALI to a base model containing clinical and laboratory data statistically improved the predictive value. Assessing ALI levels upon admission can provide important information for the short-term prognostic assessment of patients with AMI complicated by CS. A lower ALI may serve as an independent predictor of increased 30-day all-cause mortality and MACEs.

Lipid-based insulin-resistance markers predict cardiovascular events in metabolic dysfunction associated steatotic liver disease

BackgroundInsulin resistance (IR) is the cornerstone of Metabolic Dysfunction Associated Steatotic Liver Disease (MASLD), pathophysiologically being the key link between MASLD, metabolic disorders, and cardiovascular (CV) diseases. There are no prospective studies comparing the predictive values of different markers of insulin resistance (IR) in identifying the presence of MASLD and the associated risk of cardiovascular events (CVEs).MethodsPost hoc analysis of the prospective Plinio Study, involving dysmetabolic patients evaluated for the presence of MASLD. The IR markers considered were Homeostatic Model Assessment for IR (HOMA-IR), Triglycerides-Glycemia (TyG) index, Triglycerides to High-Density Lipoprotein Cholesterol ratio (TG/HDL-C), Lipid Accumulation Product (LAP) and Visceral Adiposity Index (VAI). Receiver operative characteristic (ROC) analyses were performed to find the optimal cut-offs of each IR marker for detecting MASLD and predicting CVEs in MASLD patients. Logistic and Cox multivariable regression analyses were performed, after dichotomizing the IR markers based on the optimal cut-offs, to assess the factors independently associated with MASLD and the risk of CVEs.ResultsThe study included 772 patients (age 55.6 ± 12.1 years, 39.4% women), of whom 82.8% had MASLD. VAI (Area Under the Curve [AUC] 0.731), TyG Index (AUC 0.723), and TG/HDL-C ratio (AUC: 0.721) predicted MASLD but was greater with HOMA-IR (AUC: 0.792) and LAP (AUC: 0.787). After a median follow-up of 48.7 (25.4–75.8) months, 53 MASLD patients experienced CVEs (1.8%/year). TyG index (AUC: 0.630), LAP (AUC: 0.626), TG/HDL-C (AUC: 0.614), and VAI (AUC: 0.590) demonstrated comparable, modest predictive values in assessing the CVEs risk in MASLD patients.ConclusionIn dysmetabolic patients HOMA-IR and LAP showed the best accuracy in detecting MASLD. The possible use of lipid-based IR markers in stratifying the CV risk in patients with MASLD needs further validation in larger cohorts.

The systemic inflammation response index as a risk factor for hepatic fibrosis and long-term mortality among individuals with metabolic dysfunction-associated steatotic liver disease

Background and aimsThe systemic inflammation response index (SIRI) is associated with various diseases with inflammatory components, but its relationship with the progression of hepatic fibrosis and survival outcomes in patients with metabolic dysfunction-associated steatotic liver disease (MASLD) is still unclear. This study was designed to investigate the potential associations between the SIRI and advanced hepatic fibrosis (AHF) as well as between the SIRI and long-term outcomes in individuals with MASLD. Methods and resultsA prospective cohort study was conducted using data gathered from the National Health and Nutrition Examination Survey (NHANES) spanning from 2005 to 2016. Weighted binary logistic regression, the Cox proportional hazards model, and time-dependent receiver operating characteristic (ROC) analyses were employed to assess the relationships among the SIRI, AHF, and mortality in patients with MASLD. Our study included a total of 5126 patients with MASLD. A higher SIRI was significantly associated with increased odds of AHF (OR 1.55, 95% CI 1.22, 1.96). According to the survival analyses, a higher SIRI was associated with greater all-cause (HR 1.19, 95% CI 1.15, 1.22) and cardiovascular mortality (HR 1.25, 95% CI 1.19, 1.32) after adjustment. The time-dependent ROC analysis indicated that the SIRI had a modest predictive value for discriminating MASLD individuals at higher versus lower mortality risk over 3-year, 5-year, and 10-year follow-up. ConclusionsThe SIRI is a promising tool for identifying MASLD individuals at risk of progressing to AHF and for predicting mortality outcomes.

Heart failure with preserved ejection fraction: diagnostic value of HFA-PEFF score, H₂FPEF score, and the diastolic stress echocardiography.

The aim of our study was to compare 3 diagnostic pathways: diastolic stress echocardiography (DSE) based on the ASE/EACVI 2016 guidelines, the 2018 H₂FPEF score, and the 2019 HFA-PEFF algorithm, in patients suspected of heart failure with preserved ejection fraction (HFpEF). The study group included 80 consecutive patients with a clinical suspicion of HFpEF. The H₂FPEF and HFA-PEFF scores and serum NT-proBNP concentrations were assessed in all the patients before they were sent for DSE. The DSE-based pathway confirmed HFpEF in 17 (21%) patients, the HFA-PEFF algorithm in 43 (54%), and H₂FPEF scoring in 4 (5%) patients. The ROC analysis showed that HFA-PEFF score > 5 predicts a DSE-positive test with a sensitivity of 70.5% and a specificity of 65%, (AUC = 0.711, p = 0.002) with a negative predictive value of 89.1% and positive predictive value of 35.3%. The H₂FPEF score > 3 had a negative predictive value of 90%, a positive predictive value of 29.8%, and predicted positive DSE result with a sensitivity of 82.3% but rather poor specificity of 47.6% (AUC = 0.692, p = 0.004). Both H₂FPEF and HFA-PEFF showed similar predictive values (AUC) in the prediction of positive DSE test (p = ns). The HFA-PEFF score overestimated the rate of HFpEF in comparison to DSE and the H₂FPEF score. The H₂FPEF and HFA-PEFF scores showed only modest predictive values of the positive DSE and had a diagnostic power to rule out the HFpEF.

S2I2N0-3 score predicts short- and long-term mortality and morbidity in HFrEF: a post-hoc analysis of the GUIDE-IT trial.

This study investigated the S2I2N0-3 score, a simple tool comprising stroke history, insulin-treated diabetes, and N-terminal pro-brain natriuretic peptide, for forecasting mortality and morbidity in heart failure (HF) with reduced ejection fraction (HFrEF). Analysing 890 GUIDE-IT HFrEF trial participants, we stratified them by baseline S2I2N0-3 risk score into three risk groups. We examined the score's association with five adverse outcomes over short (90days) and extended periods (median follow-up of 15months) using Cox and competing risk models. Our analysis revealed significant positive associations between the S2I2N0-3 strata and adverse outcomes. When analysed as a continuous variable, each point increment of the S2I2N0-3 score was associated with a higher risk of short- and long-term cardiovascular death [short term: hazard ratio (HR) 1.43, 95% confidence interval (CI) 1.03-1.98; long term: HR 1.18, 95% CI 1.02-1.38], all-cause death (HR 1.52, 95% CI 1.12-2.07; HR 1.18, 95% CI 1.03-1.36), HF hospitalization (HR 1.39, 95% CI 1.20-1.62; HR 1.18, 95% CI 1.06-1.31), any hospitalization (HR 1.19, 95% CI 1.06-1.34; HR 1.09, 95% CI 1.00-1.19), and the composite outcome of cardiovascular death and HF hospitalization (HR 1.39, 95% CI 1.21-1.60; HR 1.17, 95% CI 1.06-1.30). The S2I2N0-3 demonstrated reliable prognostic value, with C-indices ranging from 0.619 to 0.753 across outcomes and time points. When compared with the Meta-Analysis Global Group in Chronic Heart Failure (MAGGIC) score using Z-statistics, net reclassification index, and integrated discrimination improvement, the S2I2N0-3 showed comparable predictive power for all outcomes during both short- and long-term follow-ups. The S2I2N0-3 risk score had modest predictive values for both short- and long-term clinical outcomes in HFrEF patients, offering equivalent performance to the established MAGGIC score.

Association between red blood cell distribution width-to-platelet ratio and post-discharge readmission rate in patients with heart failure: A retrospective cohort study

BackgroundTo date, no studies have investigated the association between red blood cell distribution width (RDW)-to-platelet ratio (RPR) and readmission rates among patients with heart failure (HF). As such, the present study aimed to examine the relationship between RPR and readmission rates in patients with HF. MethodsData for this study were obtained from the Fourth People's Hospital (Zigong, Sichuan Province, China). Patients were diagnosed with HF in accordance with European Society of Cardiology criteria. The primary outcome was the 28-day readmission rate. Various logistic regression models were constructed to explore the association between RPR and the 28-day readmission rate. ResultsThe study comprised 1978 patients with HF, with a 28-day readmission rate of 6.98%. RPR emerged as an independent risk factor for 28-day readmission, evidenced by consistent results across the various regression-adjusted models. The covariate-adjusted propensity score model demonstrated that every 0.1 increase in RPR was associated with an 8.2% increase in 28-day readmission rate (odds ratio [OR] 1.082 [95% confidence interval (CI) 1.012–1.158]; P = 0.0212). Similarly, each 0.1 change in RPR was associated with a 9.8% (OR 1.098 [95% CI 1.014–1.188]) and 7.3% (OR 1.073 [95% CI 0.991–1.161]) increase in 3- and 6-month readmission rates, respectively. However, RPR was not statistically associated with the 6-month readmission rate. Curve fit plots illustrated a nonlinear positive correlation between RPR and 28-day, and 3- and 6-month readmissions. Moreover, the effects of RPR on 28-day, and 3- and 6-month readmission rates remained robust across subgroup variables in stratified analysis. Finally, the effect sizes of pooled multiply imputed data were consistent with the original data, suggesting robust results. ConclusionRPR was an independent risk factor for 28-day readmission among patients with HF and also demonstrated modest predictive value for readmissions at 3 and 6 months, despite being non-significant for the 6-month readmission rate. Early identification of patients with HF with elevated RPR would facilitate management and may confer favorable effects on prognosis.

Electrocardiography score based on the Minnesota code classification system predicts cardiovascular mortality in an asymptomatic low-risk population

Background The use of a single abnormal finding on electrocardiography (ECG) is not recommended for stratifying the risk of cardiovascular (CV) events in low-risk general populations because of its low discriminative power. However, the value of a scoring system containing multiple abnormal ECG findings for predicting CV death has not been sufficiently evaluated. Methods In a prospective community-based cohort study, 8417 participants without atherosclerotic CV diseases (ASCVDs) and any related symptoms were followed for 18 years. The standard 12-lead ECGs were recorded at baseline and the ECG findings were categorized using the Minnesota code classification. CV deaths were defined as death from myocardial infarction (MI), chronic ischemic heart disease, heart failure, fatal arrhythmia, cerebrovascular event, pulmonary thromboembolism, peripheral vascular disease and sudden cardiac arrest and identified using the Korean National Statistical Office (KOSTAT) database. Results In a multivariate Cox proportional hazard (CPH) model, major and minor ST-T wave abnormalities, atrial fibrillation (AF), Q waves in the anterior leads, the lack of Q waves in the posterior leads, high amplitudes of the left and right precordial leads, left axis deviation and sinus tachycardia were associated with higher risks of CV deaths. The ECG score consisted of these findings showed modest predictive values represented by C-statistics that ranged from 0.632 to 760 during the follow-up and performed better in the early follow-up period. The ECG score independently predicted CV death after adjustment for relevant covariates in a multivariate model, and improved the predictive performance of the 10-year ASCVD risk estimator and a model of conventional risk factors including age, diabetes and current smoking. The combined ECG score (Harrell’s C-index: 0.852, 95% confidence interval [CI], 0.828–0.876) composed of the ECG score and the conventional risk factors outperformed the 10-year ASCVD risk estimator (Harrell’s C-index: 0.806; 95% CI, 0.780–0.833) and the model of the conventional risk factors (Harrell’s C-index: 0.841, 95% CI, 0.817–0.865) and exhibited an excellent goodness of fit between the predicted and observed probabilities of CV death. Conclusions The ECG score could be useful to predict CV death independently and may add value to the conventional CV risk estimators regarding the risk stratification of CV death in asymptomatic low-risk general populations.

Developing a DNAm Surrogate to Predict Cognitive Functioning among Older Adults

AbstractBackgroundBlood‐based epigenetic measures are being developed and used to predict aging‐related health outcomes. Specifically, researchers have developed “epigenetic clocks” based on predictions of chronological age and age‐related health declines by DNA methylation. However, these markers are meant to predict aging per se and are not disease specific. Recently, scholars have developed “DNAm surrogates” to predict specific outcomes. This methodological technique has yet to be applied to cognitive functioning despite its potential to improve prediction of cognitive health in later life and provide insight into the role of biological aging.MethodThis study developed a novel DNAm surrogate measure for cognitive functioning (DNAmCogScore) based on an elastic net approach linking using 833,865 CpG sites predicting the Telephone Interview Cognitive Status score (TICS) available in the Health and Retirement Study (HRS), a nationally representative sample of older adults in the United States. Additionally, we evaluate the association of this DNAm surrogate and other neuropathology biomarkers. We used used ridge regression to create DNAmCogScore with 50% of the sample for training and 50% for testing (N = 2009 for each subsample). The Irish Longitudinal Study of Aging (TILDA) will be used for validation (results forthcoming).ResultWe find a positive association and modest predictive value between DNAmCogScore and TICS (see Figure 1, r = .4). Table 1 shows the strong associations with the DNAmCogScore and demographic characteristics. Overall, we find a strong positive coefficient between DNAmCogScore with attenuation after controlling for demographic characteristics, from 1.12 in Model 1 to .20 in Model 3, but remaining statistically significant throughout. Table 3 shows modest negative correlations between DNAmCogScore and NfL, pTau181, and GFAP (ranging from ‐.24 to ‐.36) and no correlation with AB42:40.ConclusionThis study is among the first to create a DNA methylation surrogate measure of cognitive functioning. Our measure is associated with cognitive functioning, even after adjusting for key demographic characteristics. And it has the expected association with other more neuropathology biomarkers. Overall, it suggests that changes in DNA methylation may be a molecular‐level mechanisms by which biological aging impacts cognitive health, and this measure may provide useful prediction of cognitive impairment before the onset of symptoms.

Identification of Mitral Valve Prolapse on Non-electrocardiography-gated Enhanced Chest Computed Tomography.

The primary imaging modality for the diagnosis of mitral valve prolapse (MVP) is echocardiography supplemented by electrocardiography (ECG)-gated cardiac computed tomography (CT) angiography. However, we have recently encountered patients with MVP who were initially identified on non-ECG-gated enhanced chest CT. The purpose of this study is to evaluate the diagnostic accuracy of non-ECG-gated enhanced chest CT to predict the presence of MVP. Of 92 patients (surgically confirmed MVP who underwent non-ECG-gated chest CT), 27 patients were excluded for motion artifact or insufficient surgical correlation, and 65 patients were ultimately included. As a control, 65 patients with dyspnea and without MVP (non-ECG-gated chest CT and echocardiography were performed within 1 month) were randomly selected. We retrospectively analyzed an asymmetric double line sign on axial CT images for the presence of MVP. The asymmetric double line sign was defined as the presence of a linear structure, not located in the plane traversing the mitral annulus. Use of the asymmetric double line sign to predict MVP on non-ECG-gated CT showed modest sensitivity, high specificity, modest negative predictive value, and high positive predictive value of 59% (38/65), 99% (64/65), 70% (64/91), and 97% (38/39), respectively. The asymmetric double line sign on non-ECG-gated enhanced chest CT may be a valuable finding to predict the presence of MVP. Familiarity with this CT finding may lead to prompt diagnosis and proper management of MVP.

PISA reading achievement, literacy motivation, and school burnout predicting Adolescents' educational track and educational attainment

In this four-year longitudinal study, we examined how reading comprehension, as assessed with PISA reading achievement, literacy motivation, and school burnout symptoms at the end of comprehensive school predict the choice of educational track (academic or vocational) and educational attainment in upper secondary education. The sample consisted of 1351 Finnish students whose PISA reading achievement, self-concept of ability in literacy, task value in literacy, and symptoms of school burnout (cynicism and exhaustion) were assessed at the end of comprehensive school. Information concerning their educational track during the first year of upper secondary education was derived from self-reports and information concerning the completion of upper secondary education were collected from school registers. The results indicated that PISA reading achievement, self-concept of ability in literacy, cynicism, and exhaustion were unique determinants of educational track choice after comprehensive school. The results suggest that there is a need, in comprehensive schools, to consider not only the development of basic skills but also literacy motivation and school well-being, as these all can have a long-term impact on students' educational paths. Educational relevance statementThe analyses of the present data showed relatively modest predictive value on the part of PISA performance for adolescents' later educational outcomes. The results of the present longitudinal study suggested that PISA reading score does not predict the completion of upper secondary education but that it is a unique predictor of educational choice after comprehensive school over and above other predictors in the model, suggesting that the explanation for the impact of PISA reading on educational track choice is not due to gender, parental socioeconomic status, or motivational or well-being factors. In addition to the PISA reading test score, a high self-concept in literacy and also symptoms of exhaustion increased the likelihood of attending general upper secondary school, whereas symptoms of cynicism increased the likelihood of attending vocational school. The results suggest that there is a need, in comprehensive schools, to consider not only the development of basic skills but also literacy motivation and school well-being, as these all can have a long-term impact on students' educational paths.

Coronary Plaque Characteristics Associated With Major Adverse Cardiovascular Events in Atherosclerotic Patients and Lesions: A Systematic Review and Meta-Analysis

BackgroundThe clinical value of high-risk coronary plaque characteristics (CPCs) to inform intensified medical therapy or revascularization of non–flow-limiting lesions remains uncertain. ObjectivesThe authors performed a systematic review and meta-analysis to study the prognostic impact of CPCs on patient-level and lesion-level major cardiovascular adverse events (MACE). MethodsThirty studies (21 retrospective, 9 prospective) with 30,369 patients evaluating the association of CPCs with MACE were included. CPCs included high plaque burden, low minimal lumen area, thin cap fibroatheroma, high lipid core burden index, low-attenuation plaque, spotty calcification, napkin ring sign, and positive remodeling. ResultsCPCs were evaluated with the use of intracoronary modalities in 9 studies (optical coherence tomography in 4 studies, intravascular ultrasound imaging in 3 studies, and near-infrared spectroscopy intravascular ultrasound imaging in 2 studies) and by means of coronary computed tomographic angiography in 21 studies. CPCs significantly predicted patient-level and lesion-level MACE in both unadjusted and adjusted analyses. For most CPCs, accuracy for MACE was modest to good at the patient level and moderate to good at the lesion level. Plaques with more than 1 CPC had the highest accuracy for lesion-level MACE (AUC: 0.87). Because the MACE rate was low, estimated positive predictive values for lesion-level MACE were modest. Results were mostly consistent across imaging modalities and clinical presentations, and in studies with prevailing hard outcomes. ConclusionsCharacterization of CPCs identifies high-risk atherosclerotic plaques that place lesions and patients at risk for future MACE, albeit with modest sensitivity and positive predictive value (Coronary Plaque Characteristics Associated With Major Adverse Cardiovascular Events Among Atherosclerotic Patients and Lesions; CRD42021251810)

Clinical-genomic determinants of immune checkpoint blockade response in head and neck squamous cell carcinoma.

BACKGROUNDRecurrent and/or metastatic (R/M) head and neck squamous cell carcinoma (HNSCC) is generally an incurable disease, with patients experiencing median survival of under 10 months and significant morbidity. While immune checkpoint blockade (ICB) drugs are effective in approximately 20% of patients, the remaining experience limited clinical benefit and are exposed to potential adverse effects and financial costs. Clinically approved biomarkers, such as tumor mutational burden (TMB), have a modest predictive value in HNSCC.METHODSWe analyzed clinical and genomic features, generated using whole-exome sequencing, in 133 ICB-treated patients with R/M HNSCC, of whom 69 had virus-associated and 64 had non-virus-associated tumors.RESULTSHierarchical clustering of genomic data revealed 6 molecular subtypes characterized by a wide range of objective response rates and survival after ICB therapy. The prognostic importance of these 6 subtypes was validated in an external cohort. A random forest-based predictive model, using several clinical and genomic features, predicted progression-free survival (PFS), overall survival (OS), and response with greater accuracy than did a model based on TMB alone. Recursive partitioning analysis identified 3 features (systemic inflammatory response index, TMB, and smoking signature) that classified patients into risk groups with accurate discrimination of PFS and OS.CONCLUSIONThese findings shed light on the immunogenomic characteristics of HNSCC tumors that drive differential responses to ICB and identify a clinical-genomic classifier that outperformed the current clinically approved biomarker of TMB. This validated predictive tool may help with clinical risk stratification in patients with R/M HNSCC for whom ICB is being considered.FUNDINGFundación Alfonso Martín Escudero, NIH R01 DE027738, US Department of Defense CA210784, The Geoffrey Beene Cancer Research Center, The MSKCC Population Science Research Program, the Jayme Flowers Fund, the Sebastian Nativo Fund, and the NIH/NCI Cancer Center Support Grant P30 CA008748.