An interpretable machine learning method for risk stratification of patients with acute coronary syndrome.

Backgrounds: Risk stratification for major adverse cardiovascular events (MACE) within one year in patients with acute coronary syndrome (ACS) undergoing percutaneous coronary intervention (PCI) remains a challenge. Although several predictive models based on machine learning have emerged, they are difficult to understand. This study aimed to develop a machine learning prediction model that is easy to understand and trustworthy by lay people to assess the risk of MACE in ACS patients undergoing PCI within one year of the procedure. Methods: This retrospective cohort study used medical data from 1105 patients to construct a machine-learning model. To ensure thoroughness and multidimensionality of model parsing, Shapley Additive explanations (SHAP) analysis and Local interpretable model-agnostic explanations (LIME) interpretation techniques were used to systematically and deeply interpret the constructed models from a global to a detailed level. Results: The study assessed 12 machine learning methods' prediction models and found that the Random Forest model was the most effective in predicting the risk of MACE in ACS patients after undergoing PCI. The model achieved an AUC value of 0.807 in the validation set, with an accuracy of 0.82, and a stable F1 score of 0.51. SHAP analysis ranked eight key feature variables, such as LVEF, in global importance. The weights of each feature range in the prediction model were revealed using LIME analysis. Conclusion: The machine learning prediction model we developed is capable of accurately predicting the likelihood of patients with ACS experiencing a MACE within one year of surgery.

Prognostic value of Geriatric Nutritional Risk Index and systemic immune-inflammatory index in elderly patients with acute coronary syndromes.

The objective of this study was to evaluate the predictive value of the Geriatric Nutritional Risk Index (GNRI) combined with the Systemic Immunoinflammatory Index (SII) for the risk of major adverse cardiovascular events (MACE) following percutaneous coronary intervention in elderly patients with acute coronary syndrome (ACS). We retrospectively reviewed the medical records of 1202 elderly patients with acute coronary syndromes divided into MACE and non-MACE groups according to whether they had a MACE. The sensitivity analysis utilized advanced machine learning algorithms to preliminarily identify the critical role of GNRI versus SII in predicting MACE risk. We conducted a detailed analysis using a restricted cubic spline approach to investigate the nonlinear relationship between GNRI, SII, and MACE risk further. We constructed a clinical prediction model based on three key factors: GNRI, SII, and Age. To validate the accuracy and usefulness of this model, we compared it to the widely used GRACE score using subject work and recall curves. Additionally, we compared the predictive value of models and GRACE scores in assessing the risk of MACE using the Integrated Discriminant Improvement Index (IDI) and the Net Reclassification Index (NRI). This study included 827 patients. The GNRI scores were lower in the MACE group than in the non-MACE group, while the SII scores were higher in the MACE group (P < 0.001). The multifactorial analysis revealed a low GNRI (OR = 2.863, 95% CI: 2.026-4.047, P = 0.001), High SII (OR = 3.102, 95% CI: 2.213-4.348, P = 0.001). The area under the curve (AUC) for the predictive model was 0.778 (95% CI: 0.744-0.813, P = 0.001), while the AUC for the GRACE score was 0.744 (95% CI: 0.708-0.779, P = 0.001). NRI was calculated to be 0.5569, with NRI + at 0.1860 and NRI- at 0.3708. The IDI was found to be 0.0571, with a P-value of less than 0.001. These results suggest that the newly developed prediction model is more suitable for use with the population in this study than the GRACE score. The model constructed using GNRI and SII demonstrated good standardization and clinical impact, as evidenced by the standard, DCA, and clinical impact curves. The study shows that combining GNRI and SII can be a simple, cost-effective, and valuable way to predict the risk of MACE within one year in elderly acute coronary syndromes.

Comparative analysis of four nutritional scores predicting the incidence of MACE in older adults with acute coronary syndromes after PCI.

To determine the most appropriate nutritional assessment tool for predicting the occurrence of major adverse cardiovascular events (MACE) within 1 year in elderly ACS patients undergoing PCI from four nutritional assessment tools including PNI, GNRI, CONUT, and BMI. Consecutive cases diagnosed with acute coronary syndrome (ACS) and underwent percutaneous coronary intervention (PCI) in the Department of Cardiovascular Medicine of the Air force characteristic medical center from 1 January 2020 to 1 April 2022 were retrospectively collected. The basic clinical characteristics and relevant test and examination indexes were collected uniformly, and the cases were divided into the MACE group (174 cases) and the non-MACE group (372 cases) according to whether a major adverse cardiovascular event (MACE) had occurred within 1 year. Predictive models were constructed to assess the nutritional status of patients with the Prognostic Nutritional Index (PNI), Geriatric Nutritional Risk Index (GNRI), Controlling nutritional status (CONUT) scores, and Body Mass Index (BMI), respectively, and to analyze their relationship with prognosis. The incremental value of the four nutritional assessment tools in predicting risk was compared using the Integrated Discriminant Improvement (IDI) and the net reclassification improvement (NRI). The predictive effect of each model on the occurrence of major adverse cardiovascular events (MACE) within 1 year in elderly ACS patients undergoing PCI was assessed using area under the ROC curve (AUC), calibration curves, decision analysis curves, and clinical impact curves; comparative analyses were performed. Among the four nutritional assessment tools, the area under the curve (AUC) was significantly higher for the PNI (AUC: 0.798, 95%CI 0.755-0.840 P < 0.001) and GNRI (AUC: 0.760, 95%CI 0.715-0.804 P < 0.001) than for the CONUT (AUC: 0.719,95%CI 0.673-0.765 P < 0.001) and BMI (AUC: 0.576, 95%CI 0.522-0.630 P < 0.001). The positive predictive value (PPV) of PNI: 67.67% was better than GNRI, CONUT, and BMI, and the negative predictive value (NPV): of 83.90% was better than CONUT and BMI and similar to the NPV of GNRI. The PNI, GNRI, and CONUT were compared with BMI, respectively. The PNI had the most significant improvement in the Integrated Discriminant Improvement Index (IDI) (IDI: 0.1732, P < 0.001); the PNI also had the most significant improvement in the Net Reclassification Index (NRI) (NRI: 0.8185, P < 0.001). In addition, of the four nutritional assessment tools used in this study, the PNI was more appropriate for predicting the occurrence of major adverse cardiovascular events (MACE) within 1 year in elderly ACS patients undergoing PCI.

Identification of Crucial Genes and Key Functions in Type 2 Diabetic Hearts by Bioinformatic Analysis.

Type 2 diabetes (T2D) patients with SARS-CoV-2 infection hospitalized develop an acute cardiovascular syndrome. It is urgent to elucidate underlying mechanisms associated with the acute cardiac injury in T2D hearts. We performed bioinformatic analysis on the expression profiles of public datasets to identify the pathogenic and prognostic genes in T2D hearts. Cardiac RNA-sequencing datasets from db/db or BKS mice (GSE161931) were updated to NCBI-Gene Expression Omnibus (NCBI-GEO), and used for the transcriptomics analyses with public datasets from NCBI-GEO of autopsy heart specimens with COVID-19 (5/6 with T2D, GSE150316), or dead healthy persons (GSE133054). Differentially expressed genes (DEGs) and overlapping homologous DEGs among the three datasets were identified using DESeq2. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes analyses were conducted for event enrichment through clusterProfile. The protein-protein interaction (PPI) network of DEGs was established and visualized by Cytoscape. The transcriptions and functions of crucial genes were further validated in db/db hearts. In total, 542 up-regulated and 485 down-regulated DEGs in mice, and 811 up-regulated and 1399 down-regulated DEGs in human were identified, respectively. There were 74 overlapping homologous DEGs among all datasets. Mitochondria inner membrane and serine-type endopeptidase activity were further identified as the top-10 GO events for overlapping DEGs. Cardiac CAPNS1 (calpain small subunit 1) was the unique crucial gene shared by both enriched events. Its transcriptional level significantly increased in T2D mice, but surprisingly decreased in T2D patients with SARS-CoV-2 infection. PPI network was constructed with 30 interactions in overlapping DEGs, including CAPNS1. The substrates Junctophilin2 (Jp2), Tnni3, and Mybpc3 in cardiac calpain/CAPNS1 pathway showed less transcriptional change, although Capns1 increased in transcription in db/db mice. Instead, cytoplasmic JP2 significantly reduced and its hydrolyzed product JP2NT exhibited nuclear translocation in myocardium. This study suggests CAPNS1 is a crucial gene in T2D hearts. Its transcriptional upregulation leads to calpain/CAPNS1-associated JP2 hydrolysis and JP2NT nuclear translocation. Therefore, attenuated cardiac CAPNS1 transcription in T2D patients with SARS-CoV-2 infection highlights a novel target in adverse prognostics and comprehensive therapy. CAPNS1 can also be explored for the molecular signaling involving the onset, progression and prognostic in T2D patients with SARS-CoV-2 infection.