Subjects and grouping
A total of 105 hospitalized subjects with ACS were enrolled, including 64 with ACS alone (named as ACS group) and 41 ACS companies with T2DM (named as ACS + T2DM group). This study followed the Helsinki Declaration and was approved by the Research Ethics Review Committee of Ningxia Medical University (Ethics No. 2020-763). All of the subjects provided informed consent in the study.
Diagnostic criteria
The diagnostic criteria were the 2021 AHA/ACC/ASE/CHEST/SAE/ SCCT/SCMR chest pain assessment and diagnostic guidelines for ACS21, and the 2023 ESC Guidelines for the Management of Acute Coronary Syndromes1. T2DM was defined according to the American Diabetes Association (ADA)22.
Inclusion and exclusion criteria
We included patients who met the diagnostic criteria for ACS and T2DM. Exclusion criteria:①History of myocardial infarction or unstable angina or undergoing coronary artery revascularization within 3 months, myocarditis, pericarditis, valvular heart disease, and arrhythmia. ② High blood pressure (systolic blood pressure ≥ 130 mmHg and/or diastolic blood pressure ≥ 80 mmHg) despite treatment with antihypertensive drugs.③Severe cardiac insufficiency, aortic coarctation, or aneurysm.④Chronic obstructive pulmonary disease (COPD), Pulmonary insufficiency, bronchial asthma, pulmonary embolism.⑤Cushing's syndrome, hyperthyroidism, and other endocrine/metabolic disorders. ⑥ Active liver disease, or unexplained elevated serum aminotransferases, or aspartate transaminase (AST), alanine transaminase (ALT) > normal reference values. ⑦ Renal insufficiency with current creatinine clearance < 90mL/min. ⑧ Hematologic disorders, malignancies, autoimmune disorders, and connective tissue disorders. ⑨ Mental or neurological disorders, surgery, severe trauma, severe infectious disease in the last 6 months. ⑩ Other type of diabetes.
Coronary lesions quantitative assessment
All the subjects were administered CAG by certificated and experienced cardiologists to visualize their coronary artery lesions. The CAG results, including the site and degree of stenosis, the lesions of bifurcation, calcification, and diffuse were documented. The segments of lesions were entered into the Syntax scoring system (https://Syntaxscore.org/) by a cardiologist, who remained blinded to the patient information, to calculate the Syntax score I.
Demography, laboratories, and outcomes documentation
The age, gender, body mass index (BMI), systolic and diastolic blood pressure, hypertension and stroke, and medications of all subjects were recorded. Results of the first blood test were documented, uric acid (UA), mainly include serum creatinine (CREA), blood urea nitrogen (BUN), AST, ALT, white blood cell (WBC) and neutrophils (NEUT) counts, triglyceride (TG), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), total cholesterol (TC), myoglobin( Myo), cardiac troponin I (cTnI), creatine kinase MB isoenzyme(CK-MB) brain natriuretic peptide (BNP), FBG, and HbA1c. Left ventricular ejection fraction (LVEF) measured by echocardiogram was also included. The beginning of follow-up was defined as the next day of percutaneous coronary intervention (PCI), and the end was defined as the occurrence of MACE or September 30, 2023. The patients were followed up by phone and/or face-to-face interviews. The median duration of follow-up was 28 months. MACE includes cardiovascular death, acute myocardial infarction, recurrent angina, heart failure, and severe arrhythmias (Supplemental Table S1).
Metabolomics study
Blood was drawn from fasting patients on the morning after admission from the median cubital vein (emergency coronary angiography patients pre-operative extraction) and allowed to settle at 4℃. After centrifugation at 4000 rpm for 10 minutes, the supernatant was aliquoted into 100L volumes and stored at -80℃.
For the detection process, 50μL of serum was mixed with 200μL of extraction solvent (acetonitrile: methanol, 1:1 v/v, with internal standard), vortexed for 30 seconds, sonicated for 10 minutes in an ice-water bath, and incubated at -40℃ for an hour to precipitate proteins. The mixture was then centrifuged at 4℃for 15 minutes at 12,000 rpm (RCF=13,800×g, R=8.6cm), and the supernatant was collected for assessment. Quality control (QC) samples were prepared by pooling equal volumes of each supernatant.
A comprehensive metabolomic investigation was conducted utilizing the high-efficiency UHPLC Vanquish platform (developed by Thermo Fisher Scientific), paired with a UPLC BEH Amide analytical column (dimensions of 2.1mm by 100mm, with particle size 1.7μm) linked to the highly precise Q Exactive HFX Orbitrap mass spectrometer (provided by Thermo). The chromatographic separation employed a mobile phase comprising both 25mmol/L ammonium acetate/ammonia hydroxide (pH=9.75) as solvent A, and pure acetonitrile as solvent B. The auto-sampling device was kept at 4℃, with a set injection volume of 3μL. The state-of-the-art QE HFX mass spectrometry apparatus was operated under the information-dependent acquisition (IDA) mode, which is regulated by Xcalibur software (Thermo). This mode allows for the continuous assessment of the entire MS spectral range in real-time. Settings for the ESI source included a sheath gas flow at a rate of 30 arbitrary units, auxiliary gas at 25 arbitrary units (Arb), capillary heating at 350℃, a full-scan MS resolving power of 60,000, MS/MS resolution set to 7,500, with the collision energy optimized to 10, 30, and 60 for the NCE setting, and ion spray voltages set at 3.6kV for the positive mode and -3.2kV for the negative mode.
The initial mass spectrometry data was transformed into mzXML format using the ProteoWizard tool. Subsequent analysis employed a custom-built software crafted in R, leveraging the capabilities of the XCMS package for peak identification, extraction, alignment, and quantitation processes. To annotate metabolites, we utilized our proprietary MS2 database, BiotreeDB, with an annotation threshold established at 0.3. Data conversion to logarithmic scale and centralization, as well as modeling and analysis, were performed using SIMCA software (version 16.0.2)23. We adopted principal component analysis (PCA) and orthogonal partial least squares-discriminant analysis (OPLS-DA) to discern metabolic patterns. The robustness of our models was determined by the R2X (model generalized X matrix explanatory rate), R2Y (the stability of the model), and Q2 (the predictive accuracy of the model) values. Considering that Q2 values could vary with the permutation of the Y matrix, the Q2 was changed with the order of Y arrangement, so the replacement test could further validate the model validity.
Statistical analysis
The data was processed using SPSS 26.0 statistical software. Continuous variables that followed a normal distribution, were expressed as mean ± standard deviation (`x± SD), otherwise, variables were expressed as median (quartile) [M (Q1, Q3)]. The appropriate statistical tests, either the t-test or the Mann-Whitney U test, were used for comparisons between groups. Categorical variables were expressed as absolute frequencies and percentages. The x2 Test was employed to assess the significance between the two groups. Correlation analyses were performed using Spearman analysis. Risk factors of MACE were analyzed using the COX regression model, and endpoint event curves were plotted using Kaplan-Meier. The time-dependent receiver operating characteristic (ROC) curve was used to assess the predictive value of the differential metabolites for MACE. The significance level was α=0.05, and p<0.05 was considered as statistically significant. The graphical visualization was created using R (version 4.3.2).
Differential metabolites screening between the two groups required the simultaneous fulfillment of ① variable important in the prediction (VIP) >1.0 using the first principal component variable of the OPLS-DA model, ② calibrated p < 0.05 Heatmap and hierarchical cluster analyses were performed using the heatmap package of the R package (https://CRAM.R-project.org/package=pheatmap). Metabolic pathways were identified using the open database Kyoto Encyclopedia of Genes and Genomes (KEGG, http://www.kegg.jp/kegg/pathway.html). Enrichment analysis and topological analysis of the metabolic pathways involving the differential metabolites were conducted using MetaboAnalyst 4.0 (http://www.metaboanalyst.ca/).