Objective To evaluate the significance of lactate dehydrogenase (LDH) as a predictor of in-hospital mortality in patients with acute aortic dissection(AAD). Methods We conducted a retrospective analysis of the clinical data of 445 AAD patients who were admitted to the Second Xiangya Hospital of Central South University and the Changsha Central Hospital from January 2014 to December 2017 within a time interval of ≤14 days from the onset of symptoms to hospital admission, including 353 males and 92 females with the age of 45-61 years. LDH levels were measured on admission and the endpoint was the all-cause mortality during hospitalization. Results During hospitalization, 86 patients died and 359 patients survived. Increased level of LDH was found in non-survivors compared with that in the survived [269.50 (220.57, 362.58) U/L vs. 238.00 (191.25, 289.15) U/L, P<0.001]. A nonlinear relationship between LDH levels and in-hospital mortality was observed. Using multivariable logistic analysis, we found that LDH was an independent predictor of in-hospital mortality in the patients with AAD [OR=1.002, 95% CI (1.001 to 1.014), P=0.006]. Furthermore, using receiver operating characteristic (ROC) analysis, we observed that the best threshold of LDH level was 280.70 U/L, and the area under the curve was 0.624 (95% CI 0.556 to 0.689). Conclusion LDH level on admission is an independent predictor of in-hospital mortality in patients with AAD.
Objective To investigate predictors for mortality among patients with Stanford type A acute aortic dissection (AAD) and to establish a predictive model to estimate risk of in-hospital mortality. Methods A total of 999 patients with Stanford type A AAD enrolled between 2010 and 2015 in our hospital were included for analysis. There were 745 males and 254 females with a mean age of 49.8±12.0 years. There were 837 patients with acute dissection and 182 patients (18.22%) were preoperatively treated or waiting for surgery in the emergency department and 817 (81.78%) were surgically treated. Multivariable logistic regression analysis was used to investigate predictors of in-hospital mortality. Significant risk factors for in-hospital death were used to develop a prediction model. Results The overall in-hospital mortality was 25.93%. In the multivariable analysis, the following variables were associated with increased in-hospital mortality: increased age (OR=1.04, 95% CI 1.02 to 1.05, P<0.000 1), acute aortic dissection (OR=2.49, 95% CI 1.30 to 4.77, P=0.006 1), syncope (OR=2.76, 95% CI 1.15 to 6.60, P=0.022 8), lower limbs numbness/pain (OR=7.99, 95% CI 2.71 to 23.52, P=0.000 2), type Ⅰ DeBakey dissection (OR=1.72, 95% CI 1.05 to 2.80, P=0.030 5), brachiocephalic vessels involvement (OR=2.25, 95% CI 1.20 to 4.24, P=0.011 7), acute liver insufficiency (OR=2.60, 95% CI 1.46 to 4.64, P=0.001 2), white blood cell count (WBC)>15×109 cells/L (OR=1.87, 95% CI 1.21 to 2.89, P=0.004 9) and massive pericardial effusion (OR=4.34, 95% CI 2.45 to 7.69, P<0.000 1). Based on these multivariable results, a reliable and simple bedside risk prediction tool was developed. Conclusion Different clinical manifestations and imaging features of patients with Stanford type A AAD predict the risk of in-hospital mortality. This model can be used to assist physicians to quickly identify high risk patients and to make reasonable treatment decisions.
ObjectiveTo systematically evaluate prediction models for in-hospital mortality risk in patients with acute myocardial infarction (AMI). MethodsA comprehensive search was conducted in PubMed, Embase, Web of Science, Cochrane Library, and CNKI databases from inception to May 30, 2025, to identify studies related to AMI in-hospital mortality prediction models. Risk of bias and applicability were assessed using the Prediction Model Risk of Bias Assessment Tool (PROBAST). Relevant data were extracted for model quality assessment. ResultsA total of 29 studies involving 75 AMI in-hospital mortality prediction models were included. Key predictive factors identified included Killip classification, neutrophil count, renal insufficiency, age, systolic blood pressure, and left ventricular ejection fraction. The area under the receiver operating characteristic curve (AUC) ranged from 0.580 to 0.998. Internal validation was reported in 21 studies, external validation in 4, and both in 4 studies. Model calibration was evaluated in 23 studies. Most models were presented as nomograms. All studies demonstrated good applicability, though 25 were rated as high risk of bias overall. ConclusionCurrent AMI in-hospital mortality prediction models show generally good predictive performance, with some variables exhibiting stable predictive effects. However, the lack of external validation and high risk of bias remain prevalent issues. Future studies should focus on prospective, multicenter, high-quality designs to enhance the practical and clinical value of these models.
ObjectiveThis study aimed to develop early mortality risk prediction models for patients with severe acute pancreatitis (SAP) based on eight machine learning algorithms, and to identify the major risk factors. MethodsClinical data of SAP patients diagnosed at West China Hospital of Sichuan University between January 2020 and August 2023, were retrospectively collected and randomly divided into a training set (n=878) and a validation set (n=376) in a 7∶3 ratio. Eight machine learning algorithms, including random forest, logistic regression, support vector machine, multilayer perceptron, XGBoost, Gaussian naive Bayes, CatBoost, and AdaBoost, were applied to construct early mortality prediction models for SAP. The models were evaluated using the area under curve (AUC), decision curve analysis (DCA), and Shapley additive explanations (SHAP). ResultsA total of 1 254 SAP patients were finally included in this study, with an early mortality rate of 15.79% (198/1 254). The random forest algorithm demonstrated the best predictive performance in both the training and validation sets, with AUCs of 0.913 and 0.844, respectively. In the DCA, random forest also yielded the greatest net benefit. SHAP analysis ranked seven key predictors of early mortality in SAP by importance: age, body mass index, heart rate, need for assisted ventilation, hemoglobin, interleukin-6, and lactate dehydrogenase, with the need for assisted ventilation being the most critical predictor. ConclusionThe random forest model developed in this study can assist clinicians in more accurately identifying high-risk SAP patients at an early stage, thereby enabling timely interventions to reduce early mortality.