Machine Learning Approach to Metabolomic Data Predicts Type 2 Diabetes Mellitus Incidence.

Metabolomics, with its wealth of data, offers a valuable avenue for enhancing predictions and decision-making in diabetes. This observational study aimed to leverage machine learning (ML) algorithms to predict the 4-year risk of developing type 2 diabetes mellitus (T2DM) using targeted quantitative metabolomics data. A cohort of 279 cardiovascular risk patients who underwent coronary angiography and who were initially free of T2DM according to American Diabetes Association (ADA) criteria was analyzed at baseline, including anthropometric data and targeted metabolomics, using liquid chromatography (LC)-mass spectroscopy (MS) and flow injection analysis (FIA)-MS, respectively. All patients were followed for four years. During this time, 11.5% of the patients developed T2DM. After data preprocessing, 362 variables were used for ML, employing the Caret package in R. The dataset was divided into training and test sets (75:25 ratio) and we used an oversampling approach to address the classifier imbalance of T2DM incidence. After an additional recursive feature elimination step, identifying a set of 77 variables that were the most valuable for model generation, a Support Vector Machine (SVM) model with a linear kernel demonstrated the most promising predictive capabilities, exhibiting an F1 score of 50%, a specificity of 93%, and balanced and unbalanced accuracies of 72% and 88%, respectively. The top-ranked features were bile acids, ceramides, amino acids, and hexoses, whereas anthropometric features such as age, sex, waist circumference, or body mass index had no contribution. In conclusion, ML analysis of metabolomics data is a promising tool for identifying individuals at risk of developing T2DM and opens avenues for personalized and early intervention strategies.

Interplay of inflammatory markers and anti-SARS-CoV-2 antibodies in COVID-19 mortality: A prospective cohort study.

OBJECTIVES: Despite high global vaccination coverage, it remains unclear how vaccination and anti-SARS-CoV-2 antibodies affect immune responses and inflammation levels in patients with COVID-19. It is further unclear whether the inflammatory response differs depending on antibody levels and whether the combination of antibody and inflammation levels in COVID-19 patients affects mortality rates. METHODS: We conducted a prospective multicenter cohort study that included 1031 hospitalized COVID-19 patients from five hospitals. Anti-SARS-CoV-2-spike antibodies, interleukin-6 (IL6), and CRP were measured on hospital admission. The prespecified endpoint was all-cause in-hospital mortality. RESULTS: We observed significantly lower levels of CRP (P<0.001) and IL6 (P<0.001) in patients with antibody levels above 1200 BAU/ml. After adjusting for potential confounders, patients with high levels of inflammatory markers (CRP>6 mg/dl or IL6>100 pg/ml) combined with low levels of anti-SARS-CoV-2-spike antibodies (<1200 BAU/ml) were approximately 8 times more likely to die than patients with low inflammatory responses and high antibody levels (CRP: aHR 7.973, 95% CI 2.744-23.169, P<0.001; IL6: aHR 8.973, 95% CI 3.549-22.688, P<0.001). CONCLUSION: Hospitalized COVID-19 patients presenting with high inflammatory markers and low antibody levels exhibited the highest mortality risks. Higher antibody levels are associated with lower levels of inflammation in hospitalized COVID-19 patients.

Ceramides improve cardiovascular risk prediction beyond low-density lipoprotein cholesterol.

Aims: Low-density lipoprotein cholesterol (LDL-C) is the best documented cardiovascular risk predictor and at the same time serves as a target for lipid-lowering therapy. However, the power of LDL-C to predict risk is biased by advanced age, comorbidities, and medical treatment, all known to impact cholesterol levels. Consequently, such biased patient cohorts often feature a U-shaped or inverse association between LDL-C and cardiovascular or overall mortality. It is not clear whether these constraints for risk prediction may likewise apply to other lipid risk markers in particular to ceramides and phosphatidylcholines. Methods and results: In this observational cohort study, we recorded cardiovascular mortality in 1195 patients over a period of up to 16 years, comprising a total of 12 262 patient-years. The median age of patients at baseline was 67 years. All participants were either consecutively referred to elective coronary angiography or diagnosed with peripheral artery disease, indicating a high cardiovascular risk. At baseline, 51% of the patients were under statin therapy. We found a U-shaped association between LDL-C and cardiovascular mortality with a trough level of around 150 mg/dL of LDL-C. Cox regression analyses revealed that LDL-C and other cholesterol species failed to predict cardiovascular risk. In contrast, no U-shaped but linear association was found for ceramide- and phosphatidylcholine-containing markers and these markers were able to significantly predict the cardiovascular risk even after multivariate adjustment. Conclusion: We thus suggest that ceramides- and phosphatidylcholine-based predictors rather than LDL-C may be used for a more accurate cardiovascular risk prediction in high-risk patients.

25 years of lipid-lowering therapy: secular trends in therapy of coronary patients.

BACKGROUND AND AIM: Guidelines on dyslipidemia and lipid-lowering therapy (LLT) over the years recommend lower low-density lipoprotein cholesterol (LDL-C) goals by more intense therapy. Nevertheless, LDL­C has increased in the general population. Real-world trends of LLT medication as well as of LDL­C levels in cardiovascular high-risk patients are unclear. METHODS: From 2158 patients who were referred for elective coronary angiography, lipid medication was analyzed at admission in three cardiovascular observational studies (OS) over the last 25 years: OS1: 1999-2000, OS2: 2005-2008 and OS3: 2022-2023. The three studies were performed at the same cardiology unit of a tertiary care hospital in Austria. RESULTS: The proportion of patients without LLT significantly decreased from OS1 through OS2 to OS3 (49.4%, 45.6%, and 18.5%, respectively, ptrend < 0.001). Moreover, the percentage of patients under high-intensity statin treatment significantly increased from 0% to 5.1%, and 56.5% (ptrend < 0.001). Significantly more patients became treated by more than one compound (OS1: 1.8%, OS2: 1.6%, OS3: 31.2%; ptrend < 0.001). In the latest OS3, a trend to fixed-dose combination of statins with ezetimibe was observed. Mean LDL­C levels decreased from 129 mg/dL over 127 mg/dL to 83 mg/dL, respectively (ptrend < 0.001). Of the patients on high-intensity therapy 34% met the recent ESC/EAS goals (LDL-C < 55 mg/dL), but only 3% on non-intense therapy. CONCLUSION: We conclude that during the observational period of a quarter of a century, treatment intensity increased and LDL­C levels improved considerably. Guidelines apparently matter in this high-risk population and are considered by primary care physicians.

Current Practices of Medication Plans in Austrian Patients Undergoing Coronary Angiography: An In-Depth Analysis.

A complete medication plan (MPlan) increases medication safety and adherence and is crucial in care transitions. Countries that implemented a standardized MPlan reported benefits on patients' understanding and handling of their medication. Austria lacks such a standardization, with no available data on the issue. Objective: This study aimed to investigate the current state of all medication documentations (MDocs) at hospital admission in a population at high risk for polypharmacy in Austria. Methods: We enrolled 512 consecutive patients undergoing elective coronary angiography. Their MDocs and medications were recorded at admission. MDocs were categorized, whereby a MPlan was defined as a tabular list including medication name, dose, route, frequency and patient name. Results: Out of 485 patients, 55.1% had an MDoc (median number of drugs: 6, range 2-17), of whom 24.7% had unstructured documentation, 18.0% physicians' letters and 54.3% MPlans. Polypharmacy patients did not have a MDoc in 31.3%. Crucial information as the patients's name or the originator of the MDoc was missing in 31.1% and 20.4%, respectively. Patients with MDoc provided more comprehensive medication information (p = 0.019), although over-the-counter-medication was missing in 94.5% of MDocs. A discrepancy between the MPlan and current medication at admission existed in 64.4%. In total, only 10.7% of our patient cohort presented an MPlan that was in accordance with their current medication. Conclusion: The situation in Austria is far from a standardized MPlan generated in daily routine. Numerous MPlans do not represent the current medication and could pose a potential risk for the effectiveness and safety of pharmacotherapy.

The value of earlier-in-life systolic and diastolic blood pressure for cardiovascular risk prediction.

Blood pressure (BP) varies over a lifetime. This cardiovascular observation study (OS) compared the predictive value of earlier- and later-in-life blood pressure (BP) in 1,497 cardiovascular disease patients utilizing readings taken during a health survey (HS) and 15 years later from the same subjects at the baseline of this OS. Prediction of the cardiovascular risk during the OS follow-up (21 years) was significantly more effective if the earlier BP readings at HS were used instead of recent OS readings (NRI = 0.30, p < 0.001). For HS readings, each 10 mm Hg increase of systolic and diastolic BP was associated with a 17% and 20% higher risk, respectively. At OS, systolic BP lost significance and diastolic BP reversed its association. Noteworthy, different BP categorizations (European vs. US guidelines) yielded similar results. This study highlights the poor predictive power of BP readings in elderly cardiovascular disease patients but emphasizes the significant prognostic value of earlier-in-life BP.