Development of a Risk Prediction Model for Major Adverse Cardiovascular Events After PCI in Patients with Coronary Heart Disease

Development of a Risk Prediction Model for Major Adverse Cardiovascular Events After PCI in Patients with Coronary Heart Disease | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Development of a Risk Prediction Model for Major Adverse Cardiovascular Events After PCI in Patients with Coronary Heart Disease Yijie Lin, Xinru Guo, Shiting Mi, Tielong Chen This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6994881/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract OBJECTIVE: To identify predictors of major adverse cardiovascular events (MACE) after percutaneous coronary intervention (PCI) in coronary heart disease (CHD) patients and construct a risk prediction model to identify high-risk patients and optimize postoperative management. METHODS: A single-center retrospective study enrolled 500 CHD patients who underwent PCI at Hangzhou TCM Hospital between April 2021 and October 2022. Data on demographics, laboratory results, imaging parameters, and postoperative outcomes were collected. Variables were selected using LASSO regression, and a predictive model was built with the Cox proportional hazard model. Model performance was assessed with AUC, Brier score, sensitivity (TP), specificity (TN), positive predictive value (PPV), and negative predictive value (NPV), and visualized using a column-line plot. RESULTS: The 1-, 2-, and 3-year MACE rates were 32.8%, 37.0%, and 37.4%, respectively. Eleven independent predictors were identified, and the AUCs for 1-, 2-, and 3-year MACE predictions in the test set were 0.754 (95% CI: 0.661-0.847), 0.747 (0.630-0.864), and 0.771 (0.546-0.996), outperforming traditional scores. The model effectively stratified risk (log-rank P<0.05). Calibration curves showed high agreement between predicted and actual risks (Brier score<0.25), and decision curve analysis (DCA) indicated significant clinical benefit. CONCLUSION: This study provides robust evidence for the accurate management of post-PCI patients, enhancing predictive efficacy, risk stratification, and clinical applicability through multidimensional data integration, advanced variable selection, and visualization tools. risk prediction model major adverse cardiovascular events (MACE) percutaneous coronary intervention (PCI) coronary heart disease (CHD) LASSO regression Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 1. Background Cardiovascular disease (CVD) is one of the major chronic non-communicable diseases threatening global public health, and its incidence has been increasing worldwide, posing a serious threat to the healthy life expectancy of the population, and the worsening disease burden not only has a serious impact on the quality of life of individuals, but also poses a serious challenge to the health care system, making it an urgent public health problem that needs to be prevented and controlled. It is a public health problem that needs to be prevented and controlled 1 . Coronary heart disease (CHD) is a pathologic change that occurs when coronary arteries undergo atherosclerotic lesions, leading to narrowing of the lumen of the vessel, functional spasm, or complete occlusion, which ultimately leads to ischemia and hypoxia in myocardial tissue 2 . Epidemiological surveys have shown that CHD cases account for about one-third of the total number of CVD patients, and their hospitalization costs account for up to 43.9% of total CVD expenditures, which has become a key factor contributing to the population's health damages and healthcare resource consumption 3 . As one of the clinical interventions for CHD, percutaneous coronary intervention (PCI) is a minimally invasive procedure that significantly reduces the area of myocardial cell necrosis during the acute phase of the disease by catheterizing the diseased vessel and restoring effective perfusion with the help of balloon dilatation or stent implantation 4 , 5 . This minimally invasive procedure can significantly reduce the area of myocardial cell necrosis in the acute phase of the disease, reduce in-hospital mortality by approximately 40%-60% 6 , and improve the long-term survival of patients by improving cardiac function, which is of significant clinical benefit 7 . However, the incidence of Major Adverse Cardiovascular Events (MACE) after PCI remains high 8 , and studies have shown that MACE (including recurrent myocardial infarction, target vessel revascularization, heart failure, and cardiac death) occurs in about 10%-20% of patients within 1 year after the procedure 9 . Therefore, the early and accurate identification of high-risk patients and the optimization of post-operative risk management are of great importance 10 . Clinical prediction models are common and simple tools for identifying high-risk patients, but existing tools still have significant limitations in predicting the risk of MACE after PCI 11 , and the currently widely used GRACE score 12 and SYNTAX score 13 provide initial risk stratification, but their predictive efficacy may have a "ceiling effect" 14 . The GRACE score is mainly based on clinical indicators, whereas the SYNTAX score focuses on coronary anatomy. In this study, we innovatively integrated variables of multiple dimensions, such as metabolism (AIP), inflammation (hs-CRP), anatomy (GS score), and behavior (history of alcohol consumption), to make the model more comprehensively reflect the potential risk factors of MACE after PCI, and improve the accuracy and applicability of the prediction. Traditional risk prediction models mostly use single-factor or multifactor regression analysis, which may be affected by the problem of variable covariance, leading to a decrease in prediction accuracy 15 . In this study, LASSO regression was introduced to screen the key variables, and the Cox proportional risk model was combined to construct the final prediction model, which not only avoids the problem of high-dimensional data redundancy, but also improves the model's robustness and generalization ability, which makes it more applicable in different populations. In summary, this study aims to construct a risk prediction model for MACE after PCI in patients with CHD, develop a visual risk stratification tool to help clinicians quickly identify high-risk patients, reduce the incidence of MACE, achieve individualized follow-up and precise intervention, and improve the short and long-term prognosis of the patients, as well as to provide a new way of thinking and a practical path for the precise management of cardiovascular diseases. 2 Research Subjects and Methods 2.1 Ethical Statement and Data Source This study was a single-center, retrospective study that included patients (N = 500) who were admitted to the Department of Cardiovascular Medicine of Hangzhou Hospital of Traditional Chinese Medicine between April 2021 and October 2022 and were diagnosed with CHD according to clinical guidelines 16 , 17 and underwent percutaneous coronary angioplasty (PCI). The study was approved by the Ethics Committee of Hangzhou Traditional Chinese Medicine Hospital and adheres to the ethical standards outlined in the Declaration of Helsinki and international medical ethics guidelines (Ethics Application Number: 2024KL1216). Due to the retrospective nature of the study, the ethics committee granted an exemption from obtaining informed consent. We are committed to maintaining the confidentiality of all patient information that is collected from the electronic medical record database. The reporting of this study followed the STROBE guidelines to ensure standardized reporting of observational epidemiological studies 18 . 2.2 Enrollment and randomization With the following inclusion criteria: Patients who had been diagnosed with suspected coronary atherosclerotic heart disease (ASD) due to the presence of chest pain and tightness, as well as changes in electrocardiography (ECG) indicative of ASD; All patients underwent PCI and were diagnosed with ASD based on the findings of the angiogram; Patients aged ≥ 30 years. The test set and validation set were randomly divided proportionally (70%:30%) using R software. The flow chart of the study is detailed in Fig. 1 . The study received ethical exemption from the Ethics Committee of Hangzhou Hospital of Traditional Chinese Medicine (Approval No. 2024KL1216), and informed consent was not required for any patient. The retrospective data set encompassed a wide range of information, including but not limited to PCI imaging data, gender, age, BMI, antecedent history, laboratory indices, and ECG and ultrasound reports. The primary endpoints were extracted by physicians through hospital records, clinical records, and laboratory reports, which were licensed by the review board of Hangzhou Hospital of Traditional Chinese Medicine. 2.3 Data collection The primary outcome in this study was MACE, defined as a composite of all-cause death, recurrent myocardial infarction, stroke (including ischemic and hemorrhagic stroke), heart failure, and target vessel revascularization. Secondary outcomes were the incidence of MACE at 1 year, the incidence of MACE at 2 years, and the number of MACE events during follow-up (categorized as: ≤1 and > 1). Data in this study included:(i) Demographic characteristics: Sex, Age, BMI;(ii) acute coronary syndromes(ACS, 0-absent; 1-ST-elevation myocardial infarction; 2-non-ST-elevation myocardial infarction; 3-unstable angina pectoris), chronic coronary syndromes (CCS, included stable angina pectoris, ischemic cardiomyopathy, occult coronary artery disease), lesion type (lipid-1; fiber-2; calcification-3) and lesion location (1-LAD left anterior descending branch; 2-RCA right coronary artery; 3-LCX left circumflex coronary artery); (iii) medical and personal history: Hypertension, Diabetes Mellitus, Cerebral Infarction (CI), Chronic Kidney Disease (CKD); History of Smoking, History of Alcohol Consumption; (iv) Neutrophils, Lymphocytes, Leukocytes, High Sensitive C-Reactive Protein (hsCRP), Brain Natriuretic Peptide (BNP), D-Dimer (DD), Albumin, Uric Acid, Total Cholesterol, Low Density Cholesterol (LDL-C), High-Density Cholesterol (HDL-C), Homocysteine, Triglycerides, Fasting Blood Glucose (FBG), Plaque Burden (PB), Atherogenic Index of Plasma (AIP), Gensini Score (GS), (v) Length of Hospital Stay (LOS) and Follow-Up Time (FUT). For the above laboratory tests, variables were deleted if more than 20% of the variable was missing, and imputed using multiple interpolation if less than 20% was missing 19 . AIP = Log (triglycerides/HDL cholesterol) 20 . GS 21 : Based on the imaging results, the degree of stenosis of each coronary vessel lesion was quantitatively assessed, and the degree of stenosis was based on the most severe point, where a stenosis diameter < 25% was scored as 1 point, 25% ≤ diameter < 50% was scored as 2 points, 50% ≤ diameter < 75% was scored as 4 points, 75% ≤ diameter < 90% was scored as 8 points, 90% ≤ diameter < 90% was scored as 8 points, 90% ≤ diameter < 90% was scored as 4 points. 25% was scored as 1 point, 25% ≤ diameter < 50% was scored as 2 points, 50% ≤ diameter < 75% was scored as 4 points, 75% ≤ diameter < 90% was scored as 8 points, 90% ≤ diameter < 99% was scored as 16 points, and ≥ 99% was scored as 32 points. According to the different coronary branches, the above scores were multiplied by the corresponding coefficients: left main stem lesion, score *5; left anterior descending branch proximal segment *2.5, mid-segment score *1.5, distal segment score *1; the first diagonal branch *1, the second diagonal branch *0.5; left coronary branch proximal segment *25, distal segment and posterior descending branch *1, posterior lateral branch *0.5; right coronary branch proximal, mid-segment, distal segment and posterior descending branch *1. The total score for each diseased branch is as follows: the sum of the scores of each lesion branch is the total score of the patient's coronary artery lesion stenosis degree. 2.4 Statistical methods R 4.1.2 was used to statistically analyze the extracted data. The included research subjects were randomly divided into 70% training set and 30% test set, and the construction of the model was carried out in the training set, and the evaluation of the model was carried out in the test set and external validation set. For quantitative data, quantitative data that conformed to normal distribution after the Shapiro-Wilk test were expressed as \(\:\stackrel{-}{x}\) ± s, and t-test was used for differences between groups, and quantitative data that did not conform to normal distribution were expressed as M [ P 25 , P 75 ], and Wilcoxon rank sum test was used for differences between groups. For classified information, expressed as frequencies and percentages, differences between groups were tested using the chi-square test and Fisher's exact probability method. Survival differences between groups were depicted by plotting Kaplan-Meier curves (K-M), and differences in survival were tested by the log-rank method. The covariance between variables was assessed by the variance inflation factor VIF, and when the value of VIF ≤ 5, it can be assumed that there is no covariance between variables. In order to avoid the impact of covariance between variables in the model on the prediction results, Lasso (Least absolute shrinkage and selection operator) regression with L1 regularization was introduced in the training set. Screening variables, the screened variables were initially included in the Cox regression model, and the backward method was used to further determine the variables to be included in the model based on the AIC Bare Pool information criterion, in order to determine the final model to be constructed. Differentiation and calibration were used as model evaluation indexes, and differentiation indexes included Area under the receiver operating characteristic curve (AUC), Sensitivity (TP), Specificity (TN), F1 score; calibration was evaluated by Brier Score, positive predictive value (PPV), negative predictive value (NPV) and calibration curve. Clinical applicability of the model was analyzed by clinical decision curve analysis (DCA), and the more the DCA curve of the model deviated from the two reference lines, the better the clinical applicability of the model. In order to comprehensively demonstrate the accuracy and stability of the model prediction, this study used Bootstrap sampling to conduct multiple putative back sampling in the training and test sets to generate multiple datasets with the same sample size as the original dataset, from which the means and 95% confidence intervals of the differentiation and calibration metrics were obtained. At the same time, the risk stratification ability of the model was evaluated by categorizing patients into high-risk and low-risk groups based on the risk of MACE occurrence predicted by the model. Column line plots (Nomograms) were drawn using the rms package of R 4.1.2 to visualize the risk indices of the Cox regression model to guide individualized treatment regimens. All test levels for this study were set at two-sided α = 0.05. 3 Results 3.1 Comparison of Baseline Characteristics of the Study Population A total of 500 CAD patients who had undergone PCI were included in this study, with a mean age of 70 (61.00, 78.25) years, predominantly male (73.6%), with a mean BMI of 23.87 ± 3.20, patients with a mean BMI of 23.87 ± 3.20 kg/m², more than half of the patients had both hypertension (82.8%) and CCS（44% had diabetes mellitus, 33.8% had CKD, and the proportion of smokers (25%) was higher than the proportion of patients who consumed alcohol (14.8%). The mean FUT of the patients was 436.00 [150.50,836.00] days, the mean LOS was 7.00 [4.00,11.00] days, the mean GS score was 46.00 [28.00, 84.00], the NOL was predominantly fibrotic (64.0%), and the LL was in the left circumflex coronary artery of the LCX in 23% of the patients. Throughout the follow-up period, MACE recurred in 37.4% of patients, with 32.8%, 37.0% and 37.4% having recurrence within one, two and three years, respectively. The patients were randomized into a training set (N=350) and a test set (N=150) in a ratio of 7:3, with good balance between the two datasets ( Table1 ), and the K-M curves showed no survival difference between the two populations (log-rank p>0.05, Figure2 ). Table 1 Comparison of the characteristics between the training set and the test set Variables Overall (N=500) Training set (N=350) Test set (N=150) p-value Age (years) 70.00 [61.00,78.25] 70.50 [62.00,79.00] 68.50 [60.00,76.00] 0.126 Sex (male), n (%) 368 (73.6) 256 (73.1) 112 (74.7) 0.808 BMI (kg/m 2 ) 23.87 (3.20) 23.91 (3.19) 23.79 (3.25) 0.699 Hypertension, n (%) 414 (82.8) 294 (84.0) 120 (80.0) 0.339 Diabetes, n (%) 220 (44.0) 154 (44.0) 66 (44.0) 1.000 Smoking, n (%) 125 (25.0) 86 (24.6) 39 (26.0) 0.822 Drinking, n (%) 74 (14.8) 53 (15.1) 21 (14.0) 0.847 CKD, n (%) 169 (33.8) 123 (35.1) 46 (30.7) 0.386 CI, n (%) 93 (18.6) 77 (22.0) 16 (10.7) 0.004 Tumor, n (%) 31 (6.2) 25 (7.1) 6 (4.0) 0.257 ACS, n (%) 0.481 0 320 (64.0) 225 (64.3) 95 (63.3) 1 52 (10.4) 36 (10.3) 16 (10.7) 2 69 (13.8) 52 (14.9) 17 (11.3) 3 59 (11.8) 37 (10.6) 22 (14.7) CCS, n (%) 312 (62.4) 218 (62.3) 94 (62.7) 1.000 Neutrophil (× 10 9 /L) 4.04 [3.22, 5.22] 4.02 [3.19, 5.14] 4.10 [3.29, 5.41] 0.539 Lymphocyte (× 10 9 /L) 1.38 [1.05,1.83] 1.30 [1.03,1.81] 1.47 [1.09,1.88] 0.059 Platelet (× 10 9 /L) 195.00 [159.75,232.25] 197.00 [160.25,233.00] 191.50 [160.00,231.75] 0.530 White Blood Cell (× 10 9 /L) 6.44 [5.30, 7.80] 6.36 [5.28,7.74] 6.62 [5.34,7.97] 0.397 NLR 2.91 [2.04,4.35] 2.92 [2.04,4.44] 2.89 [1.99,4.26] 0.595 PLR 142.10 [103.82,186.02] 143.82 [106.75,204.68] 136.01 [99.57,172.85] 0.057 SII 569.05 [366.31,897.56] 559.62 [364.69,940.39] 571.01 [372.56,799.59] 0.479 hsCRP (mg/l) 2.33 [0.91,9.85] 2.34 [0.90,9.37] 2.23 [0.92,10.94] 0.603 QTC (ms) 441.00 [420.00,458.00] 439.50 [417.25,459.00] 444.00 [426.00,458.00] 0.275 BNP (pg/ml) 259.00 [76.75,911.00] 255.00 [76.50,911.00] 274.00 [77.00,885.00] 0.911 DD (mg/l) 0.46[0.25,0.93] 0.46[0.27,0.97] 0.39 [0.22,0.89] 0.066 Albumin (g/l) 36.40 [33.30, 39.10] 36.35 [33.45,39.00] 36.60 [33.20, 39.27] 0.763 Uric Acid (m mol/l) 347.00 [283.75,419.25] 353.00 [285.25,433.75] 326.50 [279.25,405.75] 0.028 Total Cholesterol (mmol/l) 3.58 [2.99,4.41] 3.59 [2.97,4.42] 3.58 [3.00,4.38] 0.948 LDL_C (mmol/l) 2.16 [1.62,2.86] 2.15 [1.64,2.87] 2.19 [1.55,2.79] 0.632 HDL_C (mmol/l) 0.87 [0.76,1.01] 0.86[0.75,1.00] 0.87 [0.76,1.02] 0.633 AIP 0.15[-0.02,0.35] 0.16[-0.01,0.35] 0.13[-0.04,0.35] 0.444 triglyceride (mmol/l) 1.24 [0.88,1.76] 1.24 [0.88,1.80] 1.21 [0.87,1.73] 0.705 FBG (mmol/l) 5.68 [4.81,7.23] 5.70 [4.85,7.17] 5.65 [4.71,7.43] 0.523 Homocysteine (m mol/l 13.97 [11.14,18.01] 14.64 [11.49,18.68] 12.69 [10.54,15.77] 0.003 PB (%) 0.76[0.69,0.81] 0.76[0.69,0.82] 0.76[0.69,0.81] 0.603 MACE, n (%) 187 (37.4) 138 (39.4) 49 (32.7) 0.183 FUT (day) 436.00 [150.50,836.00] 432.50 [146.75,855.50] 444.50 [178.25,775.00] 0.860 LOS (day) 7.00 [4.00,11.00] 7.00 [4.00,12.00] 6.00 [5.00,10.00] 0.443 GS 46.00 [28.00,84.00] 47.00 [28.00,84.00] 42.00 [27.25,79.75] 0.403 NOL, n (%) 0.979 1 79 (15.8) 55 (15.7) 24 (16.0) 2 320 (64.0) 225 (64.3) 95 (63.3) 3 101 (20.2) 70 (20.0) 31 (20.7) LL=3, n (%) 115 (23.0) 80 (22.9) 35 (23.3) 1.000 Abbreviations: BMI, body mass index; CKD, Chronic Kidney Disease; CI, Cerebral Infarction; ACS, Acute Coronary Syndrome; CCS, Chronic Coronary Syndrome; NLR, neutrophil-lymphocyte ratio; PLR, platelet-lymphocyte ratio; SII, systemic immune inflammation index. NLR, neutrophil-lymphocyte ratio; PLR, platelet-lymphocyte ratio; SII, systemic immune inflammation index; hsCRP, high sensitive C-reactive protein; QTC, Corrected QT Interval; BNP, B-type Natriuretic Peptide; DD, D-Dimer; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; AIP, Atherogenic Index of Plasma; FBG, Fasting Blood Glucose; PB, plaque burden; MACE, major adverse cardiac events; FUTA, major adverse cardiac events; MACE, major adverse cardiovascular events; FUTA, major adverse cardiovascular events. major adverse cardiac events; FUT, Follow-up Time; LOS, Length of Stay; GS, Gensini Score; NOL, Nature of Lesion (Lipid-1; Fiber-2; Calcified-3); LL, Lesion Location (1- LAD, left anterior descending branch; 2-RCA, right coronary artery; 3-LCX, left circumflex coronary artery) 3.2 Lasso regression initial screening variable results After repeated iterations of ten-fold cross-validation using Lasso regression, a total of 15 variables were screened for Sex, BMI, CI, Platelet Count, hsCRP, drinking, smoking, QTC, DD, LDL-C, AIP, FBG, Homocysteine, LOS, and GS. ( Figure3 ) 3.3 COX regression to construct a prediction model The variables screened by Lasso regression were further included in Cox regression analysis ( Table 2 ), and a total of 11 variables were screened to be included in the final model when the AIC value was the smallest, and the VIF values of all variables were less than 5, indicating that there was no covariance between variables ( Figure4 ). Among them, as the values of BMI, hsCRP, QTC, AIP, Homocysteine, LOS, and GS increased, the risk of recurrent MACE in patients increased (HR>1, P<0.05), and the risk of MACE in patients who consumed alcohol was 1.699 times higher than that of patients who didn't consume alcohol (P<0.05). Table 2 COX regression analysis results Characteristics HR (95% CI) p-value BMI 1.066 [1.005, 1.131] 0.035 Cerebral Infarction 0.531 [0.332, 0.850] 0.008 Platelet Count 0.997 [0.994, 1.000] 0.034 hsCRP 1.009 [1.002, 1.017] 0.012 Drinking 1.699 [1.085, 2.662] 0.021 QTC 1.006 [1.001, 1.011] 0.011 DD 0.862 [0.737, 1.008] 0.063 AIP 2.107 [1.114, 3.985] 0.022 Homocysteine 1.003 [1.000, 1.007] 0.070 LOS 1.032 [1.013, 1.051] 0.001 GS 1.011 [1.008, 1.015] <0.001 3.4 Model evaluation 3.4.1 Results of model differentiation and calibration for training and test sets The prediction results of the model in the training and test sets are detailed in Figure5 . In terms of differentiation, the AUC of the model was higher than 0.7 at different time points in the test and training sets, and slightly lower in the test set than in the training set ( Figure6 ), and its AUCs at the three time nodes were 0.754 (95% CI 0.661,0.847), 0.747 (95% CI 0.630,0.864) and 0.771 (95%CI 0.546,0.996), and from the time-dependent changes in AUC values, the model predicted poorly at the beginning of the follow-up period, but the predictive efficacy of the model stabilized with the extension of the follow-up period ( Figure7 ,8 ). The TP of the three time points in the test set were 0.722 (95% CI 0.441,0.993), 0.795 (95% CI 0.573,0.918) and 0.863 (95% CI 0.614,0.911); the TN were 0.737 (95% CI 0.555,0.919), 0.669 (95% CI 0.493,0.845) and 0.688 (95% CI 0.264,0.912); F1 scores were 0.721 (95% CI 0.581,0.861), 0.745 (95% CI 0.625,0.865), and 0.795 (95% CI 0.645,0.945), respectively; in terms of calibration, the calibration of the training set was better than that of the test set ( Figure9 , 10 ), and the BS of the test set at the three time nodes were 0.192 (95% CI 0.170,0.214), 0.205 (95% CI 0.184,0.225), and 0.205 (95% CI 0.184,0.225), respectively; and PPV were 0.739 (95% CI 0.660,0.818), 0.709(95%CI 0.621,0.798) and 0.757(95%CI 0.521,0.994) respectively; NPV were 0.741(95%CI 0.585,0.898), 0.776(95%CI 0.626,0.926) and 0.857(95%CI 0.656,0.959), which indicate that the calibration of the model is gradually decreasing with time ( Figure9 , 10 ). 3.4.2 Results of analyzing the risk stratification ability and clinical decision curves of the models in the training and test sets The model divided the patients in the training and test sets into high-risk and low-risk groups according to the risk index, and the KM curves showed that there was a significant difference between the survival curves of the high-risk group and the low-risk group in both datasets (log rank P<0.05), indicating that the model has a better risk capability ( Figure11 ). The clinical decision curve of the model was also better than the two reference lines, indicating that the model has some clinical applicability in application with theoretical patient health gains ( Figure12 ) . 3.5 Model Visualization Column Line Charts The column chart allows for personalized interpretation of the patient, indexing the corresponding scores according to the values of the variables, summing the scores of all the variables to obtain the total score, and obtaining the probability of the patient's occurrence of MACE at different time points based on the total score. The probability of a patient experiencing MACE at different time points can be obtained by entering the values of the different indicators for the patient. Subject id=10666500, as shown in Figure13 , the patient was a female with a BMI of 15.22 kg/m², did not suffer from a heart attack, did not consume alcohol, had a blood count of 252, an ultrasensitive C-reactive protein of 9.42, a QTC of 466, a DD of 0.31, an AIP of 0.02, a homocysteine of 24.59, a hospitalization length of 10 days, and a GS score was 54, and the probability of MACE in the patient was 53.68% obtained by including the above indicators in the column chart calculator, and the patient had another MACE 23 days after re-discharge. 3.6 External validation of the model The robustness and extrapolation of the model constructed in this study were assessed using the external validation set. In the external validation set, the model performed well in terms of differentiation in terms of AUC at the three nodes, with the AUC and its 95% CI at 1 year, 2 years, and 3 years, respectively, being 0.792 (0.741,0.844), 0.805 (0.749,0.861), and 0.727 (0.659, 0.796); in terms of calibration, the BS of the model at all three time nodes is less than 0.5, indicating that the model is also well calibrated in the external validation set. In terms of clinical applicability, the model yields a net benefit when the threshold is set at 55% ( Figure14 ). 4. Discussion Based on the clinical data of MACE after PCI in patients with CHD, this study constructed a risk prediction model by integrating multidimensional variables to provide clinical decision support. In terms of model prediction efficacy, the prediction model constructed in this study demonstrated excellent performance in the training set, test set, and external validation set. The AUCs of 1-year, 2-year, and 3-year MACE prediction in the test set are 0.754, 0.747, and 0.771, respectively, which are significantly better than the traditional GRACE scores (C-statistic 0.65-0.72) 12 and the SYNTAX scores (C-statistic 0.68-0.75) 14 , demonstrating that the model of the present research is superiority in risk prediction. In addition, the calibration performance of this study was equally robust, with the test set Brier Score of 0.192 (95% CI 0.170-0.214) at 1 year and 0.205 (95% CI 0.184-0.225) at 3 years, reflecting a better calibration, indicating that the predicted probability was highly consistent with the actual risk. Meanwhile, DCA further validated the utility of the model, suggesting that it can effectively avoid over-intervention in low-risk patients and accurately identify high-risk groups requiring close follow-up. Furthermore, the time-dependent ROC curve showed that the predictive efficacy of the model stabilized over the follow-up period (3-year AUC fluctuation range ± 0.05), indicating that it still has a strong ability to capture long-term risks, which is superior to the existing tools focusing on the short-term or a single event only 22 , 23 . In this study, 11 independent risk factors were screened by LASSO-Cox combination, covering four major dimensions: metabolic, inflammatory, anatomical and behavioral, reflecting the complex pathophysiological mechanisms of MACE. Among the metabolic-related indexes, AIP(HR=2.107) was found to be important in predicting MACE. AIP quantifies the ratio of triglycerides to HDL-C, which may reflect the degree of lipid metabolism disorders, and has been confirmed to be closely related to the instability of atherosclerotic plaques 24 - 26 . This finding is consistent with the study by Zhao et al 8 on the association between AIP and coronary artery plaque vulnerability, suggesting that AIP can be used as an important monitoring index for post-PCI patients. Inflammation indicators hsCRP (HR=1.009) is a marker of chronic inflammatory state and endothelial dysfunction 27 , which may promote thrombosis by activating the monocyte-endothelial cell adhesion pathway 28 . The present study demonstrated that each unit of its elevation increased the risk of MACE by 0.9%, which further supports the important role of inflammation in the development of cardiovascular events after PCI. The anatomical parameter GS (HR=1.011) portrays coronary complexity more finely than the conventional SYNTAX score by systematically quantifying the number of lesion branches, degree of stenosis, and calcification characteristics 29 , and the optimization of the weighting of the left main stem and bifurcation lesions in particular improves the predictive specificity 30 . The results of this study support the importance of GS scores in MACE risk prediction, suggesting that clinicians should individualize the assessment by incorporating anatomical features of the coronary artery. Notably, the behavioral factor history of alcohol consumption (HR=1.699) was included in the prediction system, which may be due to the ethanol metabolite acetaldehyde inducing oxidative stress and sympathetic excitation, thus exacerbating myocardial ischemic injury 31 , however, the effect of alcohol consumption on cardiovascular health remains controversial, and its dose effect and racial differences need to be further verified 32 . In addition, prolonged QTC interval (HR=1.006) was also included in the final model. Prolonged QTC interval may indicate autonomic dysfunction and ventricular repolarization abnormality, which may increase the risk of malignant arrhythmia and sudden death 33 , which is echoed by Verevkin et al 9 on the mechanism of arrhythmia complication after PCI. The main strength of this study lies in multidimensional data integration and methodological optimization. Compared with traditional models that focus on a single data type, this study systematically integrates demographic characteristics (eg, gender, BMI), laboratory indicators (eg, hs-CRP, Homocysteine), imaging parameters (eg, GS score), and postoperative dynamics (eg, hospitalization length). It comprehensively captures the risk signals at different stages of "preoperative- intraoperative-postoperative", which improves the prediction accuracy and clinical applicability of the model. In terms of methodology, LASSO regression effectively screens key variables and reduces the covariance problem of high-dimensional data, while Bootstrap resampling and external validation further confirm the robustness and generalization ability of the model, making its applicability in different populations better than the traditional tools relying on single-center retrospective data. In addition, the model adopts a column-line graph visualization tool, which facilitates clinicians to visually assess the MACE risk of individual patients, and assists in the formulation of precise risk stratification and individualized intervention strategies. However, this study still has some limitations. First, the data in this study were obtained from a single medical center, which may have sample selection bias and affect the generalization ability of the model. The stability and applicability of the model needs to be further verified by multicenter prospective studies in the future. Second, the present study is still deficient in the coverage of dynamic indicators. Despite the inclusion of length of hospitalization, a postoperative variable, antiplatelet drug responsiveness, inflammatory factor fluctuation trends, or wearable device monitoring data (eg, heart rate variability) were not integrated, whereas previous studies have demonstrated that postoperative DAPT compliance and CYP2C19 gene polymorphisms are strongly associated with MACE risk 34 , 35 . In addition, emerging biomarkers such as high-sensitivity troponin, lipoprotein a, and imaging histological features (eg, plaque vulnerability texture analysis) were not included in this study, which may limit the model's ability to parse complex lesions to some extent 36 . Future studies may construct more accurate cross-omics prediction models through multimodal data fusion (integrating information from genomics, proteomics, and imaging genomics) and develop a dynamic monitoring platform to collect real-time postoperative physiological parameters and medication data, which can be combined with machine learning algorithms (eg, XGBoost) to achieve real-time early warning and intervention optimization of individualized MACE risk. 5. Conclusion The post-PCI MACE risk prediction model constructed in this study has significantly improved its predictive efficacy and clinical utility through multidimensional data integration and methodological optimization, and its line graph tool provides a quantitative basis for individualized risk management. In the future, the model should be further improved by expanding the sample size and incorporating dynamic and emerging biomarkers to promote the practice of precision medicine in the cardiovascular field, and ultimately reduce the burden of disease and improve the long-term prognosis of patients. Declarations Acknowledgments The authors sincerely thank all the individuals who contributed to this study. Funding Sources This study were supported by Hangzhou Biomedical and Health Industry Development Support Technology Special Project (2021WJCY002), Clinical Medical Research Project of Zhejiang Medical Association (2018ZYC-A39), the Zhejiang Administration Bureau of Traditional Chinese Medicine (2023ZR040), the Zhejiang Medical Association Fund (2023ZYC-A13), the Hangzhou TCM Hospital Affiliated to Zhejiang Chinese Medical University Hospital-Level Fund (YJ202305), and the Construction Fund of Medical Key Disciplines of Hangzhou (2020SJZDXK06). Disclosures The authors report no conflicts of interest in this work. Data Sharing Statement The datasets used and/or analyzed during this study are available from the corresponding authors on reasonable request. Ethical Approval and Consent to Participate All procedures in studies involving human participants were conducted in accordance with the ethical standards of the institutional and national research committee and with the Helsinki Declaration of 1964 and its subsequent amendments or comparable ethical standards.Due to the retrospective nature of the study, the Ethics Committee of Hangzhou Hospital of Traditional Chinese Medicine (Approval No. 2024KL1216) waived the need of obtaining informed consent. Clinical trial number Not applicable. Consent for Publication This was a retrospective cohort study, and an exemption from informed consent was approved by the Ethics Committee of Hangzhou Traditional Chinese Medicine Hospital (Ethical Application Ref: 2024KL1216). Author Contributions Conceptualization, X.G. and T.C.; Data curation, Y.L. and S.M.; Formal analysis, X.G.; Funding acquisition, T.C.; Methodology, X.G., Y.L. and S.M.; Project administration, T.C.; Software, X.G., Y.L. and S.M.; Supervision, X.G., Y.L., S.M. and T.C.; Validation, T.C.; Writing—original draft, X.G., Y.L. and S.M.; Writing—review and editing, T.C. All authors have read and agreed to the published version of the manuscript. References Cardiovascular diseases (CVDs). Accessed March 20. 2025. https://www.who.int/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds ). Henderson A. Coronary heart disease: Overview[J]. Lancet. 1996;348(Suppl 1):s1–4. Zhu KF, Wang YM, Zhu JZ, et al. National prevalence of coronary heart disease and its relationship with human development index: a systematic review[J]. Eur J Prev Cardiol. 2016;23(5):530–43. Bhatt DL. Percutaneous coronary intervention in 2018[J]. JAMA. 2018;319(20):2127–8. Doenst T, Haverich A, Serruys P, et al. Pci and cabg for treating stable coronary artery disease: jacc review topic of the week[J]. J Am Coll Cardiol. 2019;73(8):964–76. Feng S, Li M, Fei J, et al. Ten-year outcomes after percutaneous coronary intervention versus coronary artery bypass grafting for multivessel or left main coronary artery disease: a systematic review and meta-analysis[J]. J Cardiothorac Surg. 2023;18(1):54. Rinfret S, Baron SJ, Cohen DJ. Percutaneous coronary intervention: finally mature enough[J]. J Am Coll Cardiol. 2015;65(23):2508–10. Zhao B, Zhang J, Li Y, et al. Prevalence, predictors, and clinical presentation of acute pericardial effusion following percutaneous coronary intervention[J]. Front Cardiovasc Med. 2021;8:759164. Verevkin A, Von Aspern K, Leontyev S, et al. Early and long-term outcomes in patients undergoing cardiac surgery following iatrogenic injuries during percutaneous coronary intervention[J]. J Am Heart Assoc. 2019;8(1):e010940. Liu Y, Wang LF, Yang XC, et al. In-hospital outcome of primary pci for patients with acute myocardial infarction and prior coronary artery bypass grafting [J]. J Thorac Dis. 2021;13(3):1737–45. Cheng Y, Fang Z, Zhang X, et al. Association between triglyceride glucose-body mass index and cardiovascular outcomes in patients undergoing percutaneous coronary intervention: a retrospective study[J]. Cardiovasc Diabetol. 2023;22(1):75. Collet JP, Thiele H, Barbato E, et al. 2020 esc guidelines for the management of acute coronary syndromes in patients presenting without persistent st- segment elevation[J]. Eur Heart J. 2021;42(14):1289–367. Sianos G, Morel MA, Kappetein AP, et al. The syntax score: an angiographic tool grading the complexity of coronary artery disease[J]. EuroIntervention. 2005;1(2):219–27. Zhang C, Li M, Liu L, et al. Systemic immune-inflammation index as a novel predictor of major adverse cardiovascular events in patients undergoing percutaneous coronary intervention: a meta-analysis of cohort studies[J]. BMC Cardiovasc Disord. 2024;24(1):189. Vatcheva KP, Lee M, McCormick JB, Rahbar MH. Multicollinearity in Regression Analyses Conducted in Epidemiologic Studies. Epidemiol (Sunnyvale). 2016;6(2):227. 2023 AHA/ACC/ACCP/ASPC/NLA/PCNA Guideline for the Management of Patients With Chronic Coronary Disease: A Report of the American Heart Association/American College of Cardiology Joint Committee on Clinical Practice Guidelines-All Databases. Accessed March 17, 2025. https://wvpn.zcmu.edu.cn/https/77726476706e69737468656265737421e7f2439321236b597b068aa9d6562f34899051d9fc85a85327 /wos/alldb/full-record/WOS:001057951100001 2025 ACC/AHA/ACEP/NAEMSP/SCAI Guideline for the Management of Patients With Acute Coronary Syndromes: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines.-All Databases. Accessed March 17. 2025. https://wvpn.zcmu.edu.cn/https/77726476706e69737468656265737421e7f2439321236b597b068aa9d6562f34899051d9fc85a85327 /wos/alldb/full-record/MEDLINE:40014670 von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370(9596):1453–7. Beesley LJ, Bondarenko I, Elliot MR, et al. Multiple imputation with missing data indicators[J]. Stat Methods Med Res. 2021;30(12):2685–700. Fernández-Macías JC, Ochoa-Martínez AC, Varela-Silva JA, Pérez-Maldonado IN. Atherogenic Index of Plasma: Novel Predictive Biomarker for Cardiovascular Illnesses. Arch Med Res. 2019;50(5):285–94. Rampidis GP, Benetos G, Benz DC, et al. A guide for gensini score calculation[J]. Atherosclerosis. 2019;287:181–3. Song J, Liu Y, Wang W, et al. A nomogram predicting 30-day mortality in patients undergoing percutaneous coronary intervention[J]. Front Cardiovasc Med. 2022;9:897020. Song JJ, Liu YP, Wang WY, et al. Development and validation of a nomogram predicting one-year mortality in patients undergoing percutaneous coronary intervention[J]. J Geriatr Cardiol. 2022;19(12):960–9. Lioy B, Webb RJ, Amirabdollahian F. The association between the atherogenic index of plasma and cardiometabolic risk factors: a review[J]. Healthc (Basel), 2023, 11(7). Rabiee Rad M, Ghasempour Dabaghi G, Darouei B, et al. The association of atherogenic index of plasma with cardiovascular outcomes in patients with coronary artery disease: a systematic review and meta-analysis[J]. Cardiovasc Diabetol. 2024;23(1):119. Al Shawaf E, Al-Ozairi E, Al-Asfar F et al. Atherogenic index of plasma (aip) a tool to assess changes in cardiovascular disease risk post laparoscopic sleeve gastrectomy[J]. J Diabetes Res, 2020, 2020: 2091341. From C-R. Protein to Interleukin-6 to Interleukin-1 Moving Upstream To Identify Novel Targets for Atheroprotection-All Databases. Accessed March 20, 2025. https://wvpn.zcmu.edu.cn/https/77726476706e69737468656265737421e7f2439321236b597b068aa9d6562f34899051d9fc85a85327/wos/alldb/full-record/WOS:000368455200018 Rizo-Téllez SA, Sekheri M, Filep JG. C-reactive protein: a target for therapy to reduce inflammation[J]. Front Immunol. 2023;14:1237729. Boyraz B, Peker T. Comparison of syntax and gensini scores in the decision of surgery or percutaneous revascularization in patients with multivessel coronary artery disease[J]. Cureus. 2022;14(2):e22482. Wang N, Liang C. Relationship of gensini score with retinal vessel diameter and arteriovenous ratio in senile chd[J]. Open Life Sci. 2021;16(1):737–45. Rosoff DB, Davey Smith G, Mehta N, et al. Evaluating the relationship between alcohol consumption, tobacco use, and cardiovascular disease: a multivariable mendelian randomization study[J]. PLoS Med. 2020;17(12):e1003410. Larsson SC, Burgess S, Mason AM, et al. Alcohol consumption and cardiovascular disease: a mendelian randomization study[J]. Circ Genom Precis Med. 2020;13(3):e002814. El Amrawy AM, Abd El Salam S, Ayad SW, et al. Qtc interval prolongation impact on in-hospital mortality in acute coronary syndromes patients using artificial intelligence and machine learning[J]. Egypt Heart J. 2024;76(1):149. Czarny MJ, Nathan AS, Yeh RW, et al. Adherence to dual antiplatelet therapy after coronary stenting: a systematic review[J]. Clin Cardiol. 2014;37(8):505–13. Kheiri B, Osman M, Abdalla A, et al. Cyp2c19 pharmacogenetics versus standard of care dosing for selecting antiplatelet therapy in patients with coronary artery disease: a meta-analysis of randomized clinical trials[J]. Catheter Cardiovasc Interv. 2019;93(7):1246–52. Zhou Z, Gao Y, Zhang W, et al. Deep learning-based prediction of percutaneous recanalization in chronic total occlusion using coronary ct angiography[J J]. Radiology. 2023;309(2):e231149. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 04 Aug, 2025 Reviewers agreed at journal 04 Aug, 2025 Reviewers invited by journal 04 Aug, 2025 Editor invited by journal 08 Jul, 2025 Editor assigned by journal 04 Jul, 2025 Submission checks completed at journal 04 Jul, 2025 First submitted to journal 27 Jun, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6994881","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":496685911,"identity":"83044f36-7bb4-47af-93e4-50fd239bfd9c","order_by":0,"name":"Yijie Lin","email":"","orcid":"","institution":"Jiangxi University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Yijie","middleName":"","lastName":"Lin","suffix":""},{"id":496685912,"identity":"5748581d-61e4-46ef-a2c9-fd84b5170d08","order_by":1,"name":"Xinru Guo","email":"","orcid":"","institution":"Zhejiang Chinese Medical University","correspondingAuthor":false,"prefix":"","firstName":"Xinru","middleName":"","lastName":"Guo","suffix":""},{"id":496685913,"identity":"6cefcec8-64bf-4302-86db-56175fbfdc0f","order_by":2,"name":"Shiting Mi","email":"","orcid":"","institution":"Zhejiang Chinese Medical University","correspondingAuthor":false,"prefix":"","firstName":"Shiting","middleName":"","lastName":"Mi","suffix":""},{"id":496685914,"identity":"bc69f44f-62a2-4658-be67-cf6eacb05c31","order_by":3,"name":"Tielong Chen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAqElEQVRIiWNgGAWjYFAC5mYgYcPDz99ArA42xmaGAwxpMpIzDpCm5bCNQUMCkTr45zc2G3/4c57HgOEA44ePOURokTjG2JxwgOc2jzlzA7PkzG1EaDEAOuzAAYnbPJYNB9iYeYnXYnCOx+BAAglaEg4AnUa8Foljic0GZw4k80jOONhMnF/4mw8flqj4Y2fPz9988MNHYrQgAcYG0tSPglEwCkbBKMANAHLdNWyt9bhdAAAAAElFTkSuQmCC","orcid":"","institution":"Hangzhou TCM Hospital Affiliated to Zhejiang Chinese Medical University","correspondingAuthor":true,"prefix":"","firstName":"Tielong","middleName":"","lastName":"Chen","suffix":""}],"badges":[],"createdAt":"2025-06-28 01:38:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6994881/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6994881/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":88526071,"identity":"39bf7312-a284-4a0a-acde-5b4b32878e52","added_by":"auto","created_at":"2025-08-07 10:30:58","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":141771,"visible":true,"origin":"","legend":"\u003cp\u003eStudy flowchart.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-6994881/v1/f22286e8aaddbf407e02d98d.png"},{"id":88527449,"identity":"23ecf92b-60f8-4783-9939-484e98511281","added_by":"auto","created_at":"2025-08-07 10:38:59","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":360301,"visible":true,"origin":"","legend":"\u003cp\u003eKM curves for the training and the test set.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-6994881/v1/d4042db1a549405e04d3766d.png"},{"id":88525075,"identity":"4a745ce7-97c9-4917-9bd5-6b8a2f31acc1","added_by":"auto","created_at":"2025-08-07 10:22:58","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":192824,"visible":true,"origin":"","legend":"\u003cp\u003eTen folds cross validation of lasso regression.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-6994881/v1/180877540fe2ea341ad1f1dd.png"},{"id":88525078,"identity":"905ec9e9-9a72-4557-8d7d-670819fcec91","added_by":"auto","created_at":"2025-08-07 10:22:59","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":113889,"visible":true,"origin":"","legend":"\u003cp\u003eDiagnosis results of collinearity between variables in Cox model.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-6994881/v1/0a6b759a42aff70061ca6284.png"},{"id":88527451,"identity":"abc6e46f-0edf-4833-b804-ebf953fa72b3","added_by":"auto","created_at":"2025-08-07 10:38:59","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":491204,"visible":true,"origin":"","legend":"\u003cp\u003eResults of model differentiation and calibration index.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-6994881/v1/90a93c4250214f16d5230226.png"},{"id":88525080,"identity":"8876c1f7-7474-44f6-b19c-40a2919b398b","added_by":"auto","created_at":"2025-08-07 10:22:59","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":165359,"visible":true,"origin":"","legend":"\u003cp\u003eResults of model differentiation index (A-training set; B-test set).\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-6994881/v1/25f647a6df323763a46008cd.png"},{"id":88527921,"identity":"b1a8d0b4-9a8e-4db6-a9b0-bcf426068cfd","added_by":"auto","created_at":"2025-08-07 10:46:59","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":174510,"visible":true,"origin":"","legend":"\u003cp\u003eROC curves of prediction model (A-training set; B-test set).\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-6994881/v1/d50e363f8014bbfe0d52c41e.png"},{"id":88526075,"identity":"13725c1d-5113-4e5e-b459-f1de7f32de95","added_by":"auto","created_at":"2025-08-07 10:30:59","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":269640,"visible":true,"origin":"","legend":"\u003cp\u003eAUC value of different time (red-training set; blue-test set).\u003c/p\u003e","description":"","filename":"image8.png","url":"https://assets-eu.researchsquare.com/files/rs-6994881/v1/d6454d6a14dc7eb1c7b81fbe.png"},{"id":88527457,"identity":"19a2e0f4-74cf-4003-b62b-6f8ca7ccecdf","added_by":"auto","created_at":"2025-08-07 10:38:59","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":151313,"visible":true,"origin":"","legend":"\u003cp\u003eResults of model calibration index (A-training set; B-test set).\u003c/p\u003e","description":"","filename":"image9.png","url":"https://assets-eu.researchsquare.com/files/rs-6994881/v1/18b97dc6f6d94615f8df7532.png"},{"id":88526076,"identity":"99165d56-02e4-4bba-b8cd-7791229bca8a","added_by":"auto","created_at":"2025-08-07 10:30:59","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":160385,"visible":true,"origin":"","legend":"\u003cp\u003eCalibration curve of model prediction efficiency (A-training set; B-test set).\u003c/p\u003e","description":"","filename":"image10.png","url":"https://assets-eu.researchsquare.com/files/rs-6994881/v1/0b62ce6320526e8cced0525d.png"},{"id":88525106,"identity":"e9eccf5b-b655-49f1-9d1c-32ac5d95b571","added_by":"auto","created_at":"2025-08-07 10:22:59","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":228823,"visible":true,"origin":"","legend":"\u003cp\u003eSurvival curves of high-risk group and low-risk group (A-training set; B-test set).\u003c/p\u003e","description":"","filename":"image11.png","url":"https://assets-eu.researchsquare.com/files/rs-6994881/v1/770f0162f67c1c0d7a74da1f.png"},{"id":88525085,"identity":"35d8ed42-7445-462a-be6b-bd3e51580376","added_by":"auto","created_at":"2025-08-07 10:22:59","extension":"png","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":127697,"visible":true,"origin":"","legend":"\u003cp\u003eClinical decision curves of model (A-training set; B-test set).\u003c/p\u003e","description":"","filename":"image12.png","url":"https://assets-eu.researchsquare.com/files/rs-6994881/v1/c6dc59b5a110219e585d9ffb.png"},{"id":88526078,"identity":"c524b102-f2d9-467b-a7db-8445c07390d3","added_by":"auto","created_at":"2025-08-07 10:30:59","extension":"png","order_by":13,"title":"Figure 13","display":"","copyAsset":false,"role":"figure","size":688245,"visible":true,"origin":"","legend":"\u003cp\u003eModel visualization of the nomogram.\u003c/p\u003e","description":"","filename":"image13.png","url":"https://assets-eu.researchsquare.com/files/rs-6994881/v1/1304ef705c8eda9d603f87e5.png"},{"id":88526090,"identity":"c7c8864c-09c1-4661-b88d-2f38c4c84846","added_by":"auto","created_at":"2025-08-07 10:30:59","extension":"png","order_by":14,"title":"Figure 14","display":"","copyAsset":false,"role":"figure","size":287424,"visible":true,"origin":"","legend":"\u003cp\u003eThe predictive performance of the model in an external validation set.\u003c/p\u003e","description":"","filename":"image14.png","url":"https://assets-eu.researchsquare.com/files/rs-6994881/v1/05027f956ba507b7b29bda5b.png"},{"id":89062813,"identity":"bbd8721d-c904-4193-9e77-0a3cd415cbd5","added_by":"auto","created_at":"2025-08-14 09:45:30","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4293910,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6994881/v1/aec73142-cd6e-4773-bf1e-4ffe503a5fdd.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Development of a Risk Prediction Model for Major Adverse Cardiovascular Events After PCI in Patients with Coronary Heart Disease","fulltext":[{"header":"1. Background","content":"\u003cp\u003eCardiovascular disease (CVD) is one of the major chronic non-communicable diseases threatening global public health, and its incidence has been increasing worldwide, posing a serious threat to the healthy life expectancy of the population, and the worsening disease burden not only has a serious impact on the quality of life of individuals, but also poses a serious challenge to the health care system, making it an urgent public health problem that needs to be prevented and controlled. It is a public health problem that needs to be prevented and controlled \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. Coronary heart disease (CHD) is a pathologic change that occurs when coronary arteries undergo atherosclerotic lesions, leading to narrowing of the lumen of the vessel, functional spasm, or complete occlusion, which ultimately leads to ischemia and hypoxia in myocardial tissue \u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. Epidemiological surveys have shown that CHD cases account for about one-third of the total number of CVD patients, and their hospitalization costs account for up to 43.9% of total CVD expenditures, which has become a key factor contributing to the population's health damages and healthcare resource consumption \u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. As one of the clinical interventions for CHD, percutaneous coronary intervention (PCI) is a minimally invasive procedure that significantly reduces the area of myocardial cell necrosis during the acute phase of the disease by catheterizing the diseased vessel and restoring effective perfusion with the help of balloon dilatation or stent implantation \u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. This minimally invasive procedure can significantly reduce the area of myocardial cell necrosis in the acute phase of the disease, reduce in-hospital mortality by approximately 40%-60% \u003csup\u003e6\u003c/sup\u003e, and improve the long-term survival of patients by improving cardiac function, which is of significant clinical benefit \u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. However, the incidence of Major Adverse Cardiovascular Events (MACE) after PCI remains high \u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e, and studies have shown that MACE (including recurrent myocardial infarction, target vessel revascularization, heart failure, and cardiac death) occurs in about 10%-20% of patients within 1 year after the procedure \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. Therefore, the early and accurate identification of high-risk patients and the optimization of post-operative risk management are of great importance \u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eClinical prediction models are common and simple tools for identifying high-risk patients, but existing tools still have significant limitations in predicting the risk of MACE after PCI \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e, and the currently widely used GRACE score \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e and SYNTAX score \u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e provide initial risk stratification, but their predictive efficacy may have a \"ceiling effect\" \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. The GRACE score is mainly based on clinical indicators, whereas the SYNTAX score focuses on coronary anatomy. In this study, we innovatively integrated variables of multiple dimensions, such as metabolism (AIP), inflammation (hs-CRP), anatomy (GS score), and behavior (history of alcohol consumption), to make the model more comprehensively reflect the potential risk factors of MACE after PCI, and improve the accuracy and applicability of the prediction. Traditional risk prediction models mostly use single-factor or multifactor regression analysis, which may be affected by the problem of variable covariance, leading to a decrease in prediction accuracy \u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. In this study, LASSO regression was introduced to screen the key variables, and the Cox proportional risk model was combined to construct the final prediction model, which not only avoids the problem of high-dimensional data redundancy, but also improves the model's robustness and generalization ability, which makes it more applicable in different populations.\u003c/p\u003e\u003cp\u003eIn summary, this study aims to construct a risk prediction model for MACE after PCI in patients with CHD, develop a visual risk stratification tool to help clinicians quickly identify high-risk patients, reduce the incidence of MACE, achieve individualized follow-up and precise intervention, and improve the short and long-term prognosis of the patients, as well as to provide a new way of thinking and a practical path for the precise management of cardiovascular diseases.\u003c/p\u003e"},{"header":"2 Research Subjects and Methods","content":"\u003cp\u003e\u003cem\u003e2.1 Ethical Statement and Data Source\u003c/em\u003e\u003c/p\u003e\u003cp\u003eThis study was a single-center, retrospective study that included patients (N\u0026thinsp;=\u0026thinsp;500) who were admitted to the Department of Cardiovascular Medicine of Hangzhou Hospital of Traditional Chinese Medicine between April 2021 and October 2022 and were diagnosed with CHD according to clinical guidelines\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e and underwent percutaneous coronary angioplasty (PCI). The study was approved by the Ethics Committee of Hangzhou Traditional Chinese Medicine Hospital and adheres to the ethical standards outlined in the Declaration of Helsinki and international medical ethics guidelines (Ethics Application Number: 2024KL1216). Due to the retrospective nature of the study, the ethics committee granted an exemption from obtaining informed consent. We are committed to maintaining the confidentiality of all patient information that is collected from the electronic medical record database. The reporting of this study followed the STROBE guidelines to ensure standardized reporting of observational epidemiological studies \u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Enrollment and randomization\u003c/h2\u003e\u003cp\u003eWith the following inclusion criteria: Patients who had been diagnosed with suspected coronary atherosclerotic heart disease (ASD) due to the presence of chest pain and tightness, as well as changes in electrocardiography (ECG) indicative of ASD; All patients underwent PCI and were diagnosed with ASD based on the findings of the angiogram; Patients aged\u0026thinsp;\u0026ge;\u0026thinsp;30 years. The test set and validation set were randomly divided proportionally (70%:30%) using R software. The flow chart of the study is detailed in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. The study received ethical exemption from the Ethics Committee of Hangzhou Hospital of Traditional Chinese Medicine (Approval No. 2024KL1216), and informed consent was not required for any patient. The retrospective data set encompassed a wide range of information, including but not limited to PCI imaging data, gender, age, BMI, antecedent history, laboratory indices, and ECG and ultrasound reports. The primary endpoints were extracted by physicians through hospital records, clinical records, and laboratory reports, which were licensed by the review board of Hangzhou Hospital of Traditional Chinese Medicine.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.3 Data collection\u003c/h2\u003e\u003cp\u003eThe primary outcome in this study was MACE, defined as a composite of all-cause death, recurrent myocardial infarction, stroke (including ischemic and hemorrhagic stroke), heart failure, and target vessel revascularization. Secondary outcomes were the incidence of MACE at 1 year, the incidence of MACE at 2 years, and the number of MACE events during follow-up (categorized as: \u0026le;1 and \u0026gt;\u0026thinsp;1). Data in this study included:(i) Demographic characteristics: Sex, Age, BMI;(ii) acute coronary syndromes(ACS, 0-absent; 1-ST-elevation myocardial infarction; 2-non-ST-elevation myocardial infarction; 3-unstable angina pectoris), chronic coronary syndromes (CCS, included stable angina pectoris, ischemic cardiomyopathy, occult coronary artery disease), lesion type (lipid-1; fiber-2; calcification-3) and lesion location (1-LAD left anterior descending branch; 2-RCA right coronary artery; 3-LCX left circumflex coronary artery); (iii) medical and personal history: Hypertension, Diabetes Mellitus, Cerebral Infarction (CI), Chronic Kidney Disease (CKD); History of Smoking, History of Alcohol Consumption; (iv) Neutrophils, Lymphocytes, Leukocytes, High Sensitive C-Reactive Protein (hsCRP), Brain Natriuretic Peptide (BNP), D-Dimer (DD), Albumin, Uric Acid, Total Cholesterol, Low Density Cholesterol (LDL-C), High-Density Cholesterol (HDL-C), Homocysteine, Triglycerides, Fasting Blood Glucose (FBG), Plaque Burden (PB), Atherogenic Index of Plasma (AIP), Gensini Score (GS), (v) Length of Hospital Stay (LOS) and Follow-Up Time (FUT). For the above laboratory tests, variables were deleted if more than 20% of the variable was missing, and imputed using multiple interpolation if less than 20% was missing\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eAIP\u0026thinsp;=\u0026thinsp;Log (triglycerides/HDL cholesterol) \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. GS \u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e : Based on the imaging results, the degree of stenosis of each coronary vessel lesion was quantitatively assessed, and the degree of stenosis was based on the most severe point, where a stenosis diameter\u0026thinsp;\u0026lt;\u0026thinsp;25% was scored as 1 point, 25% \u0026le; diameter\u0026thinsp;\u0026lt;\u0026thinsp;50% was scored as 2 points, 50% \u0026le; diameter\u0026thinsp;\u0026lt;\u0026thinsp;75% was scored as 4 points, 75% \u0026le; diameter\u0026thinsp;\u0026lt;\u0026thinsp;90% was scored as 8 points, 90% \u0026le; diameter\u0026thinsp;\u0026lt;\u0026thinsp;90% was scored as 8 points, 90% \u0026le; diameter\u0026thinsp;\u0026lt;\u0026thinsp;90% was scored as 4 points. 25% was scored as 1 point, 25% \u0026le; diameter\u0026thinsp;\u0026lt;\u0026thinsp;50% was scored as 2 points, 50% \u0026le; diameter\u0026thinsp;\u0026lt;\u0026thinsp;75% was scored as 4 points, 75% \u0026le; diameter\u0026thinsp;\u0026lt;\u0026thinsp;90% was scored as 8 points, 90% \u0026le; diameter\u0026thinsp;\u0026lt;\u0026thinsp;99% was scored as 16 points, and \u0026ge;\u0026thinsp;99% was scored as 32 points. According to the different coronary branches, the above scores were multiplied by the corresponding coefficients: left main stem lesion, score *5; left anterior descending branch proximal segment *2.5, mid-segment score *1.5, distal segment score *1; the first diagonal branch *1, the second diagonal branch *0.5; left coronary branch proximal segment *25, distal segment and posterior descending branch *1, posterior lateral branch *0.5; right coronary branch proximal, mid-segment, distal segment and posterior descending branch *1. The total score for each diseased branch is as follows: the sum of the scores of each lesion branch is the total score of the patient's coronary artery lesion stenosis degree.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.4 Statistical methods\u003c/h2\u003e\u003cp\u003eR 4.1.2 was used to statistically analyze the extracted data. The included research subjects were randomly divided into 70% training set and 30% test set, and the construction of the model was carried out in the training set, and the evaluation of the model was carried out in the test set and external validation set. For quantitative data, quantitative data that conformed to normal distribution after the Shapiro-Wilk test were expressed as\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\stackrel{-}{x}\\)\u003c/span\u003e\u003c/span\u003e \u0026plusmn; s, and t-test was used for differences between groups, and quantitative data that did not conform to normal distribution were expressed as \u003cem\u003eM\u003c/em\u003e [\u003cem\u003eP\u003c/em\u003e\u003csub\u003e25\u003c/sub\u003e, \u003cem\u003eP\u003c/em\u003e\u003csub\u003e75\u003c/sub\u003e], and Wilcoxon rank sum test was used for differences between groups. For classified information, expressed as frequencies and percentages, differences between groups were tested using the chi-square test and Fisher's exact probability method. Survival differences between groups were depicted by plotting Kaplan-Meier curves (K-M), and differences in survival were tested by the log-rank method.\u003c/p\u003e\u003cp\u003eThe covariance between variables was assessed by the variance inflation factor VIF, and when the value of VIF\u0026thinsp;\u0026le;\u0026thinsp;5, it can be assumed that there is no covariance between variables. In order to avoid the impact of covariance between variables in the model on the prediction results, Lasso (Least absolute shrinkage and selection operator) regression with L1 regularization was introduced in the training set. Screening variables, the screened variables were initially included in the Cox regression model, and the backward method was used to further determine the variables to be included in the model based on the AIC Bare Pool information criterion, in order to determine the final model to be constructed.\u003c/p\u003e\u003cp\u003eDifferentiation and calibration were used as model evaluation indexes, and differentiation indexes included Area under the receiver operating characteristic curve (AUC), Sensitivity (TP), Specificity (TN), F1 score; calibration was evaluated by Brier Score, positive predictive value (PPV), negative predictive value (NPV) and calibration curve. Clinical applicability of the model was analyzed by clinical decision curve analysis (DCA), and the more the DCA curve of the model deviated from the two reference lines, the better the clinical applicability of the model.\u003c/p\u003e\u003cp\u003eIn order to comprehensively demonstrate the accuracy and stability of the model prediction, this study used Bootstrap sampling to conduct multiple putative back sampling in the training and test sets to generate multiple datasets with the same sample size as the original dataset, from which the means and 95% confidence intervals of the differentiation and calibration metrics were obtained. At the same time, the risk stratification ability of the model was evaluated by categorizing patients into high-risk and low-risk groups based on the risk of MACE occurrence predicted by the model.\u003c/p\u003e\u003cp\u003eColumn line plots (Nomograms) were drawn using the rms package of R 4.1.2 to visualize the risk indices of the Cox regression model to guide individualized treatment regimens. All test levels for this study were set at two-sided α\u0026thinsp;=\u0026thinsp;0.05.\u003c/p\u003e\u003c/div\u003e"},{"header":"3 Results","content":"\u003cp\u003e\u003cem\u003e3.1 Comparison of Baseline Characteristics of the Study Population\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eA total of 500 CAD patients who had undergone PCI were included in this study, with a mean age of 70 (61.00, 78.25) years, predominantly male (73.6%), with a mean BMI of 23.87 \u0026plusmn; 3.20, patients with a mean BMI of\u0026nbsp;23.87 \u0026plusmn; 3.20\u0026nbsp;kg/m\u0026sup2;, more than half of the patients had both hypertension (82.8%) and CCS（44% had diabetes mellitus, 33.8% had CKD, and the proportion of smokers (25%) was higher than the proportion of patients who consumed alcohol (14.8%). The mean FUT of the patients was\u0026nbsp;436.00 [150.50,836.00] days, the mean LOS was 7.00 [4.00,11.00] days, the mean GS score was 46.00 [28.00, 84.00], the NOL was predominantly fibrotic (64.0%), and the LL was in the left circumflex coronary artery of the LCX in 23% of the patients. Throughout the follow-up period, MACE recurred in 37.4% of patients, with 32.8%, 37.0% and 37.4% having recurrence within one, two and three years, respectively. The patients were randomized into a training set (N=350) and a test set (N=150) in a ratio of 7:3, with good balance between the two datasets (\u003cstrong\u003eTable1\u003c/strong\u003e), and the K-M curves showed no survival difference between the two populations (log-rank p\u0026gt;0.05,\u003cstrong\u003eFigure2\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable\u003c/strong\u003e\u003cstrong\u003e1\u003c/strong\u003e Comparison of the characteristics between the training set and the test set\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eVariables\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003eOverall\u003c/p\u003e\n \u003cp\u003e(N=500)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003eTraining set (N=350)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003eTest set\u003c/p\u003e\n \u003cp\u003e(N=150)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e\u003cem\u003ep-value\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eAge (years)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e70.00 [61.00,78.25]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e70.50 [62.00,79.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e68.50 [60.00,76.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.126\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eSex (male), n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e368 (73.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e256 (73.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e112 (74.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.808\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eBMI (kg/m\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e23.87 (3.20)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e23.91 (3.19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e23.79 (3.25)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.699\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eHypertension, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e414 (82.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e294 (84.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e120 (80.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.339\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eDiabetes, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e220 (44.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e154 (44.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e66 (44.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 26px;\"\u003e\n \u003cp\u003eSmoking, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e125 (25.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e86 (24.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e39 (26.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e0.822\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 26px;\"\u003e\n \u003cp\u003eDrinking, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e74 (14.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e53 (15.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e21 (14.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e0.847\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eCKD, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e169 (33.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e123 (35.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e46 (30.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.386\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eCI, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e93 (18.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e77 (22.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e16 (10.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.004\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eTumor, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e31 (6.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e25 (7.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e6 (4.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.257\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eACS, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.481\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e320 (64.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e225 (64.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e95 (63.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e52 (10.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e36 (10.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e16 (10.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e69 (13.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e52 (14.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e17 (11.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e59 (11.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e37 (10.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e22 (14.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eCCS, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e312 (62.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e218 (62.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e94 (62.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eNeutrophil (\u0026times; 10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e4.04 [3.22, 5.22]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e4.02 [3.19, 5.14]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e4.10 [3.29, 5.41]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.539\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eLymphocyte (\u0026times; 10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e1.38 [1.05,1.83]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e1.30 [1.03,1.81]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e1.47 [1.09,1.88]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.059\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003ePlatelet (\u0026times; 10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e195.00 [159.75,232.25]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e197.00 [160.25,233.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e191.50 [160.00,231.75]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.530\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 26px;\"\u003e\n \u003cp\u003eWhite Blood Cell\u0026nbsp;(\u0026times; 10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e6.44 [5.30, 7.80]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e6.36 [5.28,7.74]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e6.62 [5.34,7.97]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e0.397\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eNLR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e2.91 [2.04,4.35]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e2.92 [2.04,4.44]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e2.89 [1.99,4.26]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.595\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003ePLR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e142.10 [103.82,186.02]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e143.82 [106.75,204.68]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e136.01 [99.57,172.85]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.057\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eSII\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e569.05 [366.31,897.56]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e559.62 [364.69,940.39]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e571.01 [372.56,799.59]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.479\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003ehsCRP (mg/l)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e2.33 [0.91,9.85]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e2.34 [0.90,9.37]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e2.23 [0.92,10.94]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.603\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eQTC (ms)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e441.00 [420.00,458.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e439.50 [417.25,459.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e444.00 [426.00,458.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.275\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eBNP (pg/ml)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e259.00 [76.75,911.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e255.00 [76.50,911.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e274.00 [77.00,885.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.911\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eDD (mg/l)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e0.46[0.25,0.93]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e0.46[0.27,0.97]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e0.39 [0.22,0.89]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.066\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eAlbumin (g/l)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e36.40 [33.30, 39.10]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e36.35 [33.45,39.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e36.60 [33.20, 39.27]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.763\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eUric Acid (m\u0026nbsp;mol/l)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e347.00 [283.75,419.25]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e353.00 [285.25,433.75]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e326.50 [279.25,405.75]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.028\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eTotal Cholesterol (mmol/l)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e3.58 [2.99,4.41]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e3.59 [2.97,4.42]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e3.58 [3.00,4.38]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.948\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eLDL_C (mmol/l)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e2.16 [1.62,2.86]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e2.15 [1.64,2.87]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e2.19 [1.55,2.79]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.632\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eHDL_C (mmol/l)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e0.87 [0.76,1.01]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e0.86[0.75,1.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e0.87 [0.76,1.02]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.633\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 26px;\"\u003e\n \u003cp\u003eAIP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e0.15[-0.02,0.35]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e0.16[-0.01,0.35]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e0.13[-0.04,0.35]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e0.444\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003etriglyceride (mmol/l)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e1.24 [0.88,1.76]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e1.24 [0.88,1.80]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e1.21 [0.87,1.73]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.705\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 26px;\"\u003e\n \u003cp\u003eFBG (mmol/l)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e5.68 [4.81,7.23]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e5.70 [4.85,7.17]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e5.65 [4.71,7.43]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e0.523\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eHomocysteine (m\u0026nbsp;mol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e13.97 [11.14,18.01]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e14.64 [11.49,18.68]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e12.69 [10.54,15.77]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.003\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003ePB (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e0.76[0.69,0.81]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e0.76[0.69,0.82]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e0.76[0.69,0.81]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.603\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eMACE, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e187 (37.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e138 (39.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e49 (32.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.183\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eFUT (day)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e436.00 [150.50,836.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e432.50 [146.75,855.50]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e444.50 [178.25,775.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.860\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eLOS (day)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e7.00 [4.00,11.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e7.00 [4.00,12.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e6.00 [5.00,10.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.443\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eGS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e46.00 [28.00,84.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e47.00 [28.00,84.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e42.00 [27.25,79.75]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.403\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eNOL, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e0.979\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e79 (15.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e55 (15.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e24 (16.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e320 (64.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e225 (64.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e95 (63.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e101 (20.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e70 (20.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e31 (20.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003eLL=3, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e115 (23.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21px;\"\u003e\n \u003cp\u003e80 (22.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 21px;\"\u003e\n \u003cp\u003e35 (23.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eAbbreviations: BMI, body mass index; CKD, Chronic Kidney Disease; CI, Cerebral Infarction; ACS, Acute Coronary Syndrome; CCS, Chronic Coronary Syndrome; NLR, neutrophil-lymphocyte ratio; PLR, platelet-lymphocyte ratio; SII, systemic immune inflammation index. NLR, neutrophil-lymphocyte ratio; PLR, platelet-lymphocyte ratio; SII, systemic immune inflammation index; hsCRP, high sensitive C-reactive protein; QTC, Corrected QT Interval; BNP, B-type Natriuretic Peptide; DD, D-Dimer; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; AIP, Atherogenic Index of Plasma; FBG, Fasting Blood Glucose; PB, plaque burden; MACE, major adverse cardiac events; FUTA, major adverse cardiac events; MACE, major adverse cardiovascular events; FUTA, major adverse cardiovascular events. major adverse cardiac events; FUT, Follow-up Time; LOS, Length of Stay; GS, Gensini Score; NOL, Nature of Lesion (Lipid-1; Fiber-2; Calcified-3); LL, Lesion Location (1- LAD, left anterior descending branch; 2-RCA, right coronary artery; 3-LCX, left circumflex coronary artery)\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e3.2 Lasso regression initial screening variable results\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAfter repeated iterations of ten-fold cross-validation using Lasso regression, a total of 15 variables were screened for Sex, BMI, CI, Platelet Count, hsCRP, drinking, smoking, QTC, DD, LDL-C, AIP, FBG, Homocysteine, LOS, and GS. (\u003cstrong\u003eFigure3\u003c/strong\u003e)\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e3.3 COX regression to construct a prediction model\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe variables screened by Lasso regression were further included in Cox regression analysis (\u003cstrong\u003eTable 2\u003c/strong\u003e), and a total of 11 variables were screened to be included in the final model when the AIC value was the smallest, and the VIF values of all variables were less than 5, indicating that there was no covariance between variables (\u003cstrong\u003eFigure4\u003c/strong\u003e). Among them, as the values of BMI, hsCRP, QTC, AIP, Homocysteine, LOS, and GS increased, the risk of recurrent MACE in patients increased (HR\u0026gt;1, P\u0026lt;0.05), and the risk of MACE in patients who consumed alcohol was 1.699 times higher than that of patients who didn\u0026apos;t consume alcohol (P\u0026lt;0.05).\u003c/p\u003e\n\u003cp\u003eTable 2 COX regression analysis results\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"99%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 39px;\"\u003e\n \u003cp\u003eCharacteristics\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 41px;\"\u003e\n \u003cp\u003eHR (95% CI)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003ep-value\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 39px;\"\u003e\n \u003cp\u003eBMI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 41px;\"\u003e\n \u003cp\u003e1.066 [1.005, 1.131]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e0.035\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 39px;\"\u003e\n \u003cp\u003eCerebral Infarction\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 41px;\"\u003e\n \u003cp\u003e0.531 [0.332, 0.850]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e0.008\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 39px;\"\u003e\n \u003cp\u003ePlatelet Count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 41px;\"\u003e\n \u003cp\u003e0.997 [0.994, 1.000]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e0.034\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 39px;\"\u003e\n \u003cp\u003ehsCRP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 41px;\"\u003e\n \u003cp\u003e1.009 [1.002, 1.017]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e0.012\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 39px;\"\u003e\n \u003cp\u003eDrinking\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 41px;\"\u003e\n \u003cp\u003e1.699 [1.085, 2.662]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e0.021\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 39px;\"\u003e\n \u003cp\u003eQTC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 41px;\"\u003e\n \u003cp\u003e1.006 [1.001, 1.011]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e0.011\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 39px;\"\u003e\n \u003cp\u003eDD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 41px;\"\u003e\n \u003cp\u003e0.862 [0.737, 1.008]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e0.063\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 39px;\"\u003e\n \u003cp\u003eAIP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 41px;\"\u003e\n \u003cp\u003e2.107 [1.114, 3.985]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e0.022\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 39px;\"\u003e\n \u003cp\u003eHomocysteine\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 41px;\"\u003e\n \u003cp\u003e1.003 [1.000, 1.007]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e0.070\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 39px;\"\u003e\n \u003cp\u003eLOS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 41px;\"\u003e\n \u003cp\u003e1.032 [1.013, 1.051]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 39px;\"\u003e\n \u003cp\u003eGS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 41px;\"\u003e\n \u003cp\u003e1.011 [1.008, 1.015]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cem\u003e3.4 Model evaluation\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e3.4.1 Results of model differentiation and calibration for training and test sets\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe prediction results of the model in the training and test sets are detailed in \u003cstrong\u003eFigure5\u003c/strong\u003e\u003cstrong\u003e.\u0026nbsp;\u003c/strong\u003eIn terms of differentiation, the AUC of the model was higher than 0.7 at different time points in the test and training sets, and slightly lower in the test set than in the training set (\u003cstrong\u003eFigure6\u003c/strong\u003e), and its AUCs at the three time nodes were 0.754 (95% CI 0.661,0.847), 0.747 (95% CI 0.630,0.864) and 0.771 (95%CI 0.546,0.996), and from the time-dependent changes in AUC values, the model predicted poorly at the beginning of the follow-up period, but the predictive efficacy of the model stabilized with the extension of the follow-up period (\u003cstrong\u003eFigure7\u003c/strong\u003e\u003cstrong\u003e,8\u003c/strong\u003e). The TP of the three time points in the test set were 0.722 (95% CI 0.441,0.993), 0.795 (95% CI 0.573,0.918) and 0.863 (95% CI 0.614,0.911); the TN were 0.737 (95% CI 0.555,0.919), 0.669 (95% CI 0.493,0.845) and 0.688 (95% CI 0.264,0.912); F1 scores were 0.721 (95% CI 0.581,0.861), 0.745 (95% CI 0.625,0.865), and 0.795 (95% CI 0.645,0.945), respectively; in terms of calibration, the calibration of the training set was better than that of the test set (\u003cstrong\u003eFigure9\u003c/strong\u003e\u003cstrong\u003e,\u003c/strong\u003e\u003cstrong\u003e10\u003c/strong\u003e), and the BS of the test set at the three time nodes were 0.192 (95% CI 0.170,0.214), 0.205 (95% CI 0.184,0.225), and 0.205 (95% CI 0.184,0.225), respectively; and PPV were 0.739 (95% CI 0.660,0.818), 0.709(95%CI 0.621,0.798) and 0.757(95%CI 0.521,0.994) respectively; NPV were 0.741(95%CI 0.585,0.898), 0.776(95%CI 0.626,0.926) and 0.857(95%CI 0.656,0.959), which indicate that the calibration of the model is gradually decreasing with time (\u003cstrong\u003eFigure9\u003c/strong\u003e\u003cstrong\u003e,\u003c/strong\u003e\u003cstrong\u003e10\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e3.4.2 Results of analyzing the risk stratification ability and clinical decision curves of the models in the training and test sets\u003c/p\u003e\n\u003cp\u003eThe model divided the patients in the training and test sets into high-risk and low-risk groups according to the risk index, and the KM curves showed that there was a significant difference between the survival curves of the high-risk group and the low-risk group in both datasets (log rank P\u0026lt;0.05), indicating that the model has a better risk capability (\u003cstrong\u003eFigure11\u003c/strong\u003e). The clinical decision curve of the model was also better than the two reference lines, indicating that the model has some clinical applicability in application with theoretical patient health gains (\u003cstrong\u003eFigure12\u003c/strong\u003e\u003cstrong\u003e)\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e3.5 Model Visualization Column Line Charts\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe column chart allows for personalized interpretation of the patient, indexing the corresponding scores according to the values of the variables, summing the scores of all the variables to obtain the total score, and obtaining the probability of the patient\u0026apos;s occurrence of MACE at different time points based on the total score. The probability of a patient experiencing MACE at different time points can be obtained by entering the values of the different indicators for the patient.\u003c/p\u003e\n\u003cp\u003eSubject id=10666500, as shown in \u003cstrong\u003eFigure13\u003c/strong\u003e, the patient was a female with a BMI of 15.22 kg/m\u0026sup2;, did not suffer from a heart attack, did not consume alcohol, had a blood count of 252, an ultrasensitive C-reactive protein of 9.42, a QTC of 466, a DD of 0.31, an AIP of 0.02, a homocysteine of 24.59, a hospitalization length of 10 days, and a GS score was 54, and the probability of MACE in the patient was 53.68% obtained by including the above indicators in the column chart calculator, and the patient had another MACE 23 days after re-discharge.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e3.6 External validation of the model\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe robustness and extrapolation of the model constructed in this study were assessed using the external validation set. In the external validation set, the model performed well in terms of differentiation in terms of AUC at the three nodes, with the AUC and its 95% CI at 1 year, 2 years, and 3 years, respectively, being 0.792 (0.741,0.844), 0.805 (0.749,0.861), and 0.727 (0.659, 0.796); in terms of calibration, the BS of the model at all three time nodes is less than 0.5, indicating that the model is also well calibrated in the external validation set. In terms of clinical applicability, the model yields a net benefit when the threshold is set at 55% (\u003cstrong\u003eFigure14\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eBased on the clinical data of MACE after PCI in patients with CHD, this study constructed a risk prediction model by integrating multidimensional variables to provide clinical decision support.\u003c/p\u003e\n\u003cp\u003eIn terms of model prediction efficacy, the prediction model constructed in this study demonstrated excellent performance in the training set, test set, and external validation set. The AUCs of 1-year, 2-year, and 3-year MACE prediction in the test set are 0.754, 0.747, and 0.771, respectively, which are significantly better than the traditional GRACE scores (C-statistic 0.65-0.72)\u003csup\u003e12\u003c/sup\u003e and the SYNTAX scores (C-statistic 0.68-0.75)\u003csup\u003e14\u003c/sup\u003e , demonstrating that the model of the present research is superiority in risk prediction. In addition, the calibration performance of this study was equally robust, with the test set Brier Score of 0.192 (95% CI 0.170-0.214) at 1 year and 0.205 (95% CI 0.184-0.225) at 3 years, reflecting a better calibration, indicating that the predicted probability was highly consistent with the actual risk. Meanwhile, DCA further validated the utility of the model, suggesting that it can effectively avoid over-intervention in low-risk patients and accurately identify high-risk groups requiring close follow-up. Furthermore, the time-dependent ROC curve showed that the predictive efficacy of the model stabilized over the follow-up period (3-year AUC fluctuation range ± 0.05), indicating that it still has a strong ability to capture long-term risks, which is superior to the existing tools focusing on the short-term or a single event only\u003csup\u003e22\u003c/sup\u003e\u003csup\u003e,\u003c/sup\u003e\u003csup\u003e23\u003c/sup\u003e .\u003c/p\u003e\n\u003cp\u003eIn this study, 11 independent risk factors were screened by LASSO-Cox combination, covering four major dimensions: metabolic, inflammatory, anatomical and behavioral, reflecting the complex pathophysiological mechanisms of MACE. Among the metabolic-related indexes, AIP(HR=2.107) was found to be important in predicting MACE. AIP quantifies the ratio of triglycerides to HDL-C, which may reflect the degree of lipid metabolism disorders, and has been confirmed to be closely related to the instability of atherosclerotic plaques\u0026nbsp;\u003csup\u003e24\u003c/sup\u003e\u003csup\u003e-\u003c/sup\u003e\u003csup\u003e26\u003c/sup\u003e. This finding is consistent with the study by Zhao et al\u0026nbsp;\u003csup\u003e8\u003c/sup\u003e on the association between AIP and coronary artery plaque vulnerability, suggesting that AIP can be used as an important monitoring index for post-PCI patients. Inflammation indicators hsCRP (HR=1.009) is a marker of chronic inflammatory state and endothelial dysfunction\u003csup\u003e27\u003c/sup\u003e , which may promote thrombosis by activating the monocyte-endothelial cell adhesion pathway\u003csup\u003e28\u003c/sup\u003e . The present study demonstrated that each unit of its elevation increased the risk of MACE by 0.9%, which further supports the important role of inflammation in the development of cardiovascular events after PCI. The anatomical parameter GS (HR=1.011) portrays coronary complexity more finely than the conventional SYNTAX score by systematically quantifying the number of lesion branches, degree of stenosis, and calcification characteristics\u003csup\u003e29\u003c/sup\u003e , and the optimization of the weighting of the left main stem and bifurcation lesions in particular improves the predictive specificity\u003csup\u003e30\u003c/sup\u003e . The results of this study support the importance of GS scores in MACE risk prediction, suggesting that clinicians should individualize the assessment by incorporating anatomical features of the coronary artery. Notably, the behavioral factor history of alcohol consumption (HR=1.699) was included in the prediction system, which may be due to the ethanol metabolite acetaldehyde inducing oxidative stress and sympathetic excitation, thus exacerbating myocardial ischemic injury\u003csup\u003e31\u003c/sup\u003e , however, the effect of alcohol consumption on cardiovascular health remains controversial, and its dose effect and racial differences need to be further verified\u003csup\u003e32\u003c/sup\u003e . In addition, prolonged QTC interval (HR=1.006) was also included in the final model. Prolonged QTC interval may indicate autonomic dysfunction and ventricular repolarization abnormality, which may increase the risk of malignant arrhythmia and sudden death\u003csup\u003e33\u003c/sup\u003e , which is echoed by Verevkin et al\u003csup\u003e9\u003c/sup\u003e on the mechanism of arrhythmia complication after PCI.\u003c/p\u003e\n\u003cp\u003eThe main strength of this study lies in multidimensional data integration and methodological optimization. Compared with traditional models that focus on a single data type, this study systematically integrates demographic characteristics (eg, gender, BMI), laboratory indicators (eg, hs-CRP, Homocysteine), imaging parameters (eg, GS score), and postoperative dynamics (eg, hospitalization length). It comprehensively captures the risk signals at different stages of \"preoperative- intraoperative-postoperative\", which improves the prediction accuracy and clinical applicability of the model. In terms of methodology, LASSO regression effectively screens key variables and reduces the covariance problem of high-dimensional data, while Bootstrap resampling and external validation further confirm the robustness and generalization ability of the model, making its applicability in different populations better than the traditional tools relying on single-center retrospective data. In addition, the model adopts a column-line graph visualization tool, which facilitates clinicians to visually assess the MACE risk of individual patients, and assists in the formulation of precise risk stratification and individualized intervention strategies.\u003c/p\u003e\n\u003cp\u003eHowever, this study still has some limitations. First, the data in this study were obtained from a single medical center, which may have sample selection bias and affect the generalization ability of the model. The stability and applicability of the model needs to be further verified by multicenter prospective studies in the future. Second, the present study is still deficient in the coverage of dynamic indicators. Despite the inclusion of length of hospitalization, a postoperative variable, antiplatelet drug responsiveness, inflammatory factor fluctuation trends, or wearable device monitoring data (eg, heart rate variability) were not integrated, whereas previous studies have demonstrated that postoperative DAPT compliance and CYP2C19 gene polymorphisms are strongly associated with MACE risk\u0026nbsp;\u003csup\u003e34\u003c/sup\u003e\u003csup\u003e,\u003c/sup\u003e\u003csup\u003e35\u003c/sup\u003e. In addition, emerging biomarkers such as high-sensitivity troponin, lipoprotein a, and imaging histological features (eg, plaque vulnerability texture analysis) were not included in this study, which may limit the model's ability to parse complex lesions to some extent\u0026nbsp;\u003csup\u003e36\u003c/sup\u003e. Future studies may construct more accurate cross-omics prediction models through multimodal data fusion (integrating information from genomics, proteomics, and imaging genomics) and develop a dynamic monitoring platform to collect real-time postoperative physiological parameters and medication data, which can be combined with machine learning algorithms (eg, XGBoost) to achieve real-time early warning and intervention optimization of individualized MACE risk.\u003c/p\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eThe post-PCI MACE risk prediction model constructed in this study has significantly improved its predictive efficacy and clinical utility through multidimensional data integration and methodological optimization, and its line graph tool provides a quantitative basis for individualized risk management. In the future, the model should be further improved by expanding the sample size and incorporating dynamic and emerging biomarkers to promote the practice of precision medicine in the cardiovascular field, and ultimately reduce the burden of disease and improve the long-term prognosis of patients.\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors sincerely thank all the individuals who contributed to this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding Sources\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study were supported by Hangzhou Biomedical and Health Industry Development Support Technology Special Project (2021WJCY002), Clinical Medical Research Project of Zhejiang Medical Association (2018ZYC-A39), the Zhejiang Administration Bureau of Traditional Chinese Medicine (2023ZR040), the Zhejiang Medical Association Fund (2023ZYC-A13), the Hangzhou TCM Hospital Affiliated to Zhejiang Chinese Medical University Hospital-Level Fund (YJ202305), and the Construction Fund of Medical Key Disciplines of Hangzhou (2020SJZDXK06).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDisclosures\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors report no conflicts of interest in this work.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Sharing Statement \u0026nbsp;\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analyzed during this study are available from the corresponding authors on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthical Approval and Consent to Participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll procedures in studies involving human participants were conducted in accordance with the ethical standards of the institutional and national research committee and with the Helsinki Declaration of 1964 and its subsequent amendments or comparable ethical standards.Due to the retrospective nature of the study, the Ethics Committee of Hangzhou Hospital of Traditional Chinese Medicine (Approval No. 2024KL1216) waived the need of obtaining informed consent.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClinical trial number\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for Publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis was a retrospective cohort study, and an exemption from informed consent was approved by the Ethics Committee of Hangzhou Traditional Chinese Medicine Hospital (Ethical Application Ref: 2024KL1216).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization, X.G. and T.C.; Data curation, Y.L. and S.M.; Formal analysis, X.G.; Funding acquisition, T.C.; Methodology, X.G., Y.L. and S.M.; Project administration, T.C.; Software, X.G., Y.L. and S.M.; Supervision, X.G., Y.L., S.M. and T.C.; Validation, T.C.; Writing—original draft, X.G., Y.L. and S.M.; Writing—review and editing, T.C. All authors have read and agreed to the published version of the manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eCardiovascular diseases (CVDs). Accessed March 20. 2025. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.who.int/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds\u003c/span\u003e\u003cspan address=\"https://www.who.int/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHenderson A. Coronary heart disease: Overview[J]. Lancet. 1996;348(Suppl 1):s1\u0026ndash;4.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhu KF, Wang YM, Zhu JZ, et al. National prevalence of coronary heart disease and its relationship with human development index: a systematic review[J]. Eur J Prev Cardiol. 2016;23(5):530\u0026ndash;43.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBhatt DL. Percutaneous coronary intervention in 2018[J]. JAMA. 2018;319(20):2127\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDoenst T, Haverich A, Serruys P, et al. Pci and cabg for treating stable coronary artery disease: jacc review topic of the week[J]. J Am Coll Cardiol. 2019;73(8):964\u0026ndash;76.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFeng S, Li M, Fei J, et al. Ten-year outcomes after percutaneous coronary intervention versus coronary artery bypass grafting for multivessel or left main coronary artery disease: a systematic review and meta-analysis[J]. J Cardiothorac Surg. 2023;18(1):54.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRinfret S, Baron SJ, Cohen DJ. Percutaneous coronary intervention: finally mature enough[J]. J Am Coll Cardiol. 2015;65(23):2508\u0026ndash;10.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhao B, Zhang J, Li Y, et al. Prevalence, predictors, and clinical presentation of acute pericardial effusion following percutaneous coronary intervention[J]. Front Cardiovasc Med. 2021;8:759164.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVerevkin A, Von Aspern K, Leontyev S, et al. Early and long-term outcomes in patients undergoing cardiac surgery following iatrogenic injuries during percutaneous coronary intervention[J]. J Am Heart Assoc. 2019;8(1):e010940.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiu Y, Wang LF, Yang XC, et al. In-hospital outcome of primary pci for patients with acute myocardial infarction and prior coronary artery bypass grafting [J]. J Thorac Dis. 2021;13(3):1737\u0026ndash;45.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCheng Y, Fang Z, Zhang X, et al. Association between triglyceride glucose-body mass index and cardiovascular outcomes in patients undergoing percutaneous coronary intervention: a retrospective study[J]. Cardiovasc Diabetol. 2023;22(1):75.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCollet JP, Thiele H, Barbato E, et al. 2020 esc guidelines for the management of acute coronary syndromes in patients presenting without persistent st- segment elevation[J]. Eur Heart J. 2021;42(14):1289\u0026ndash;367.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSianos G, Morel MA, Kappetein AP, et al. The syntax score: an angiographic tool grading the complexity of coronary artery disease[J]. EuroIntervention. 2005;1(2):219\u0026ndash;27.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang C, Li M, Liu L, et al. Systemic immune-inflammation index as a novel predictor of major adverse cardiovascular events in patients undergoing percutaneous coronary intervention: a meta-analysis of cohort studies[J]. BMC Cardiovasc Disord. 2024;24(1):189.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVatcheva KP, Lee M, McCormick JB, Rahbar MH. Multicollinearity in Regression Analyses Conducted in Epidemiologic Studies. Epidemiol (Sunnyvale). 2016;6(2):227.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e2023 AHA/ACC/ACCP/ASPC/NLA/PCNA Guideline for the Management of Patients With Chronic Coronary Disease: A Report of the American Heart Association/American College of Cardiology Joint Committee on Clinical Practice Guidelines-All Databases. Accessed March 17, 2025. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://wvpn.zcmu.edu.cn/https/77726476706e69737468656265737421e7f2439321236b597b068aa9d6562f34899051d9fc85a85327\u003c/span\u003e\u003cspan address=\"https://wvpn.zcmu.edu.cn/https/77726476706e69737468656265737421e7f2439321236b597b068aa9d6562f34899051d9fc85a85327\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e/wos/alldb/full-record/WOS:001057951100001\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e2025 ACC/AHA/ACEP/NAEMSP/SCAI Guideline for the Management of Patients With Acute Coronary Syndromes: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines.-All Databases. Accessed March 17. 2025. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://wvpn.zcmu.edu.cn/https/77726476706e69737468656265737421e7f2439321236b597b068aa9d6562f34899051d9fc85a85327\u003c/span\u003e\u003cspan address=\"https://wvpn.zcmu.edu.cn/https/77726476706e69737468656265737421e7f2439321236b597b068aa9d6562f34899051d9fc85a85327\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e/wos/alldb/full-record/MEDLINE:40014670\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003evon Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370(9596):1453\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBeesley LJ, Bondarenko I, Elliot MR, et al. Multiple imputation with missing data indicators[J]. Stat Methods Med Res. 2021;30(12):2685\u0026ndash;700.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFern\u0026aacute;ndez-Mac\u0026iacute;as JC, Ochoa-Mart\u0026iacute;nez AC, Varela-Silva JA, P\u0026eacute;rez-Maldonado IN. Atherogenic Index of Plasma: Novel Predictive Biomarker for Cardiovascular Illnesses. Arch Med Res. 2019;50(5):285\u0026ndash;94.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRampidis GP, Benetos G, Benz DC, et al. A guide for gensini score calculation[J]. Atherosclerosis. 2019;287:181\u0026ndash;3.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSong J, Liu Y, Wang W, et al. A nomogram predicting 30-day mortality in patients undergoing percutaneous coronary intervention[J]. Front Cardiovasc Med. 2022;9:897020.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSong JJ, Liu YP, Wang WY, et al. Development and validation of a nomogram predicting one-year mortality in patients undergoing percutaneous coronary intervention[J]. J Geriatr Cardiol. 2022;19(12):960\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLioy B, Webb RJ, Amirabdollahian F. The association between the atherogenic index of plasma and cardiometabolic risk factors: a review[J]. Healthc (Basel), 2023, 11(7).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRabiee Rad M, Ghasempour Dabaghi G, Darouei B, et al. The association of atherogenic index of plasma with cardiovascular outcomes in patients with coronary artery disease: a systematic review and meta-analysis[J]. Cardiovasc Diabetol. 2024;23(1):119.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAl Shawaf E, Al-Ozairi E, Al-Asfar F et al. Atherogenic index of plasma (aip) a tool to assess changes in cardiovascular disease risk post laparoscopic sleeve gastrectomy[J]. J Diabetes Res, 2020, 2020: 2091341.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFrom C-R. Protein to Interleukin-6 to Interleukin-1 Moving Upstream To Identify Novel Targets for Atheroprotection-All Databases. Accessed March 20, 2025. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://wvpn.zcmu.edu.cn/https/77726476706e69737468656265737421e7f2439321236b597b068aa9d6562f34899051d9fc85a85327/wos/alldb/full-record/WOS:000368455200018\u003c/span\u003e\u003cspan address=\"https://wvpn.zcmu.edu.cn/https/77726476706e69737468656265737421e7f2439321236b597b068aa9d6562f34899051d9fc85a85327/wos/alldb/full-record/WOS:000368455200018\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRizo-T\u0026eacute;llez SA, Sekheri M, Filep JG. C-reactive protein: a target for therapy to reduce inflammation[J]. Front Immunol. 2023;14:1237729.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBoyraz B, Peker T. Comparison of syntax and gensini scores in the decision of surgery or percutaneous revascularization in patients with multivessel coronary artery disease[J]. Cureus. 2022;14(2):e22482.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang N, Liang C. Relationship of gensini score with retinal vessel diameter and arteriovenous ratio in senile chd[J]. Open Life Sci. 2021;16(1):737\u0026ndash;45.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRosoff DB, Davey Smith G, Mehta N, et al. Evaluating the relationship between alcohol consumption, tobacco use, and cardiovascular disease: a multivariable mendelian randomization study[J]. PLoS Med. 2020;17(12):e1003410.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLarsson SC, Burgess S, Mason AM, et al. Alcohol consumption and cardiovascular disease: a mendelian randomization study[J]. Circ Genom Precis Med. 2020;13(3):e002814.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEl Amrawy AM, Abd El Salam S, Ayad SW, et al. Qtc interval prolongation impact on in-hospital mortality in acute coronary syndromes patients using artificial intelligence and machine learning[J]. Egypt Heart J. 2024;76(1):149.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCzarny MJ, Nathan AS, Yeh RW, et al. Adherence to dual antiplatelet therapy after coronary stenting: a systematic review[J]. Clin Cardiol. 2014;37(8):505\u0026ndash;13.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKheiri B, Osman M, Abdalla A, et al. Cyp2c19 pharmacogenetics versus standard of care dosing for selecting antiplatelet therapy in patients with coronary artery disease: a meta-analysis of randomized clinical trials[J]. Catheter Cardiovasc Interv. 2019;93(7):1246\u0026ndash;52.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhou Z, Gao Y, Zhang W, et al. Deep learning-based prediction of percutaneous recanalization in chronic total occlusion using coronary ct angiography[J J]. Radiology. 2023;309(2):e231149.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-cardiovascular-disorders","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bcar","sideBox":"Learn more about [BMC Cardiovascular Disorders](http://bmccardiovascdisord.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bcar/default.aspx","title":"BMC Cardiovascular Disorders","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"risk prediction model, major adverse cardiovascular events (MACE), percutaneous coronary intervention (PCI), coronary heart disease (CHD), LASSO regression","lastPublishedDoi":"10.21203/rs.3.rs-6994881/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6994881/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eOBJECTIVE: \u003c/strong\u003eTo identify predictors of major adverse cardiovascular events (MACE) after percutaneous coronary intervention (PCI) in coronary heart disease (CHD) patients and construct a risk prediction model to identify high-risk patients and optimize postoperative management.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMETHODS:\u003c/strong\u003e A single-center retrospective study enrolled 500 CHD patients who underwent PCI at Hangzhou TCM Hospital between April 2021 and October 2022. Data on demographics, laboratory results, imaging parameters, and postoperative outcomes were collected. Variables were selected using LASSO regression, and a predictive model was built with the Cox proportional hazard model. Model performance was assessed with AUC, Brier score, sensitivity (TP), specificity (TN), positive predictive value (PPV), and negative predictive value (NPV), and visualized using a column-line plot.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRESULTS:\u003c/strong\u003e The 1-, 2-, and 3-year MACE rates were 32.8%, 37.0%, and 37.4%, respectively. Eleven independent predictors were identified, and the AUCs for 1-, 2-, and 3-year MACE predictions in the test set were 0.754 (95% CI: 0.661-0.847), 0.747 (0.630-0.864), and 0.771 (0.546-0.996), outperforming traditional scores. The model effectively stratified risk (log-rank P\u0026lt;0.05). Calibration curves showed high agreement between predicted and actual risks (Brier score\u0026lt;0.25), and decision curve analysis (DCA) indicated significant clinical benefit.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCONCLUSION:\u003c/strong\u003e This study provides robust evidence for the accurate management of post-PCI patients, enhancing predictive efficacy, risk stratification, and clinical applicability through multidimensional data integration, advanced variable selection, and visualization tools.\u003c/p\u003e","manuscriptTitle":"Development of a Risk Prediction Model for Major Adverse Cardiovascular Events After PCI in Patients with Coronary Heart Disease","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-07 10:22:54","doi":"10.21203/rs.3.rs-6994881/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"55443121895392763167988234448451023192","date":"2025-08-04T10:07:04+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"135760519133966177543052430349489984765","date":"2025-08-04T06:43:55+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-04T06:37:23+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-07-08T04:25:07+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-07-04T14:25:04+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-04T14:25:00+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Cardiovascular Disorders","date":"2025-06-28T01:34:08+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-cardiovascular-disorders","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bcar","sideBox":"Learn more about [BMC Cardiovascular Disorders](http://bmccardiovascdisord.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bcar/default.aspx","title":"BMC Cardiovascular Disorders","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"1bb1c9b2-503b-41ce-a111-f4ea97b42fba","owner":[],"postedDate":"August 7th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-08-07T10:22:54+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-07 10:22:54","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6994881","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6994881","identity":"rs-6994881","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}