Table 2.
Overview of the information of the included prediction models.
| Author year | Variable selection | Model development method | Calibration method | Clinical application | Validation method | Final predictors | Model performance (AUC/C-statistic) | Model presentation |
|---|---|---|---|---|---|---|---|---|
| Stefano (26) | Univariate analysis and multivariate logistic regression analysis | Multivariable logistic regression model | Hosmer-Lemeshow test | NI | External validation | Killip class (II, III), Killip class (IV), Diabetes, Anterior STEMI, Age > 75 years, eGFR<60 ml/min/1.73m2 | D:0.8379(0.802-0.8738); V:0.84 |
Risk score |
| Yinghua (19) | Univariate and multivariate regression analyses | Multivariate regression analysis | Hosmer-Lemeshow test, Calibration plot | DCA | Internal(self-sampling method) and external validation | Age, eGFR, TyG index, PNI | D:0.785(0.729-0.841); V:0.802(0.699-0.905) |
Nomogram model |
| Benjamin (23) | Backward selection | Multivariable logistic regression model | Hosmer-Lemeshow test | NI | Internal cross-validation |
Age, history of CKD, eGFR, LVEF, LVEDP, whether the patient was hypotensive, whether the patient received an IABP | D:0.77(0.70-0.83); V:0.76(0.70-0.82) |
A web-based tool |
| Yuhei (27) | Stepwise backward | Multivariable logistic regression model | Hosmer-Lemeshow test | NI | Internal validation | Blood sugar(BS) ≥200 mg/dL, high-sensitivity troponin I(hsTnI) >1.6 ng/dL (normal upper limit×50), Albumin ≤3.5 mg/dL, eGFR <45 mL/min/1.73 m2 | D:0.754(0.733-0.846); V:0.754(0.644-0.839) |
Risk score |
| Akaphol (24) | Backward elimination | Multivariable logistic regression model | Hosmer-Lemeshow test, Calibration plot | NI | Internal validation | Age, baseline creatinine, LVEF < 40%, multi-vessel pPCI, treated with thrombus aspiration, inserted IABP, pre and intra-procedural cardiogenic shock, congestive heart failure | D:0.78(0.75-0.82); V:0.75(0.72-0.79) |
An online web |
| Pei-Chun (30) | Clinically relevant variables | Multivariable logistic model | NI | NI | Internal validation | Age, DM, ventilator use, prior AKI, number of intervened vessels, CKD, IABP use, cardiogenic shock |
D:0.874(0.868-0.881); V:0.8624(0.8515-0.8733) |
ADVANCIS Score |
| Amir (33) | Lasso, SHAP | Machine learning(NB, LR, CB, LMP, RF) | NI | NI | Internal validation (five-fold cross-validation) | LVEF, FPG, creatinine, mean creatinine, eGFR | AUC=0.775 | Random Forest model |
| Hang (20) | Univariate analysis, LASSO and multivariate logistic regression analysis | Multivariate logistic regression analysis | Hosmer-Lemeshow test, Calibration curves | DCA | Internal validation (Bootstrap) | DM, LVEF, SII,NT-proBNP, hsCRP | D:0.84(0.790-0.890); V:0.844(0.762-0.926) |
Nomogram model |
| Faysal (28) | Univariate analysis,Lasso and multivariate logistic regression analysis | Multivariable logistic regression analysis | Calibration plot | NI | Internal validation (A bootstrap of 200 replicates) | Age, Hypertension, Hemoglobin, eGFR, Albumin, SIIRI, LVEF, Lesion length, Pain-to-balloon time | AUC=0.97 | Nomogram model |
| Kai (21) | LASSO regression and multivariable logistic regression analysis | Multivariate logistic regression analysis | Hosmer–Lemeshow test, Calibration plot | DCA | Internal validation (Bootstrap internal verification method) | Age >75, LVEF, DM, FAR, hsCRP, lymphocyte count | D:0.835(0.800-0.871); V:0.767(0.711-0.824) |
Nomogram model |
| Yue (22) | LASSO regression and multivariate analyses | Multivariate logistic regression analysis | Calibration curves | DCA | Internal validation (Bootstrap self-sampling method) | Subtypes of ACS, age>75, multivessel coronary artery disease, hyperuricemia, LDL-C, TyG index, eGFR | D:0.811(0.766-0.844); V:0.773(0.712-0.829) |
Nomogram model |
| Sukrisd (25) | Univariate analysis and multivariate logistic regression analysis | Multivariable logistic regression analysis | Hosmer-Lemeshow, Calibration plot | NI | Internal validation(1,000 replicates bootstrapped sampling) | Ejection fraction < 40%, Triple-vessel disease, Use of IABP | D:0.83(0.76-0.90); V:0.77(0.68-0.85) |
Risk Stratification Score |
| Kai-Yang (32) | Univariate analysis and multivariate logistic regression analysis | Multivariate logistic regression analysis | Hosmer-Lemeshow test | NI | Internal validation (bootstrap method) |
Age>75, baseline SCr>1.5 mg/dl, hypotension, the use of IABP | V:0.828(0.737-0.920) | Risk score |
| Hui (29) | Univariate logistic regression analysis and multivariate logistic regression analysis |
Multivariate logistic regression analysis | Hosmer-Lemeshow test, Calibration chart | NI | External validation | Age > 72, ejection fraction of no more than 40%, baseline SCr > 102.7 mmol/L, RDW > 13.15, MDCLs | D:0.721(0.652-0.790); V:0.731(0.624-0.838) |
Risk score |
| Ling (31) | Boruta algorithm | Machine learning(DT, SVW, RF, KNN, NB, GBM) | NI | NI | Internal validation Ten-fold cross-validation | Neutrophil percentage, age, Free triiodothyronine, Preoperation hypotension, SCr, Hemoglobin, LDL-C, Total triglycerides, Brain natriuretic peptide, WBC, HDL-C, Heart rate, BMI, Cardiac troponin I, SBP | D:1.000(1.000-1.000); V:0.82(0.76-0.87) |
Random Forest model |
| Xuejun (18) | Univariate logistic regression analysis and multivariate logistic regression analysis |
Multivariate logistic regression analysis | Hosmer-Lemeshow test, Calibration plot | NI | Internal and external validation | Hemoglobin, contrast volume >100ml, hypotension before the procedure, eGFR, logBNP, age | D:0.775(0.732-0.819); V:0.715(0.631-0.799) |
Nomogram model |
NI, no information; D, derivation; V, validation; LVEF, left ventricular ejection fraction; LVEDP, left ventricular end diastolic pressure; IABP, intra-aortic balloon pump; DCA, decision curve analysis; eGFR, estimated glomerular filtration rate; DM, diabetes mellitus; SII, immune-inflammatory index; NT-proBNP, N-terminal pro-brain natriuretic peptide; hsCRP, highly sensitive C-reactive protein; SIIRI, Systemic immune-inflflammation response index; FAR, fibrinogen-to-albumin ratio; LDL-C, low-density lipoprotein cholesterol; TyG, triglyceride-glucose; SCr, serum creatinine; RDW, red blood cell distribution width; MDCLs, middistal segment culprit lesions; HDL-C, High-density lipoprotein cholesterol; BMI, Body mass index; PNI, prognostic nutritional index; SBP, systolic blood pressure; WBC, white blood cell; BNP, B-type natriuretic peptide.