陈歌
中国医学科学院阜外医院 26病区
BACKGROUND:FAVOR III China (F3C) is a large-scale randomized trial comparing QFR-guided and angiography-guided percutaneous coronary intervention (PCI) strategies. The aim of current study was to assess the feasibility of predicting the 1-year outcomes of the F3C trial using simulation of retrospectively assessed quantitative flow ratio (QFR) data obtained from the all-comers PANDA III trial.METHODS:Among 2348 subjects from the PANDA III trial, angiography from 1391 patients was able to be analyzed with QFR. Each subject from the F3C was matched to a PANDA III patient according to the five baseline characteristics (age, sex, diabetes, multivessel disease, and existence of any vessel with diameter stenosis % >90% and thrombolysis in myocardial infarction flow <3) through a bootstrapping sampling process. Outcome predictions were based on these blinded baseline data. The primary endpoint was a composite of death, myocardial infarction, or revascularization at 1 year.RESULTS:Among the patients with analyzable QFR, 814 patients were able to be matched to F3C patients undergoing a QFR-guided treatment strategy. After 10,000 simulations, the patients in the QFR-guided group were simulated to have a 1.9% (95% predictive intervals: -3.5% to -0.3%) absolute reduction of the occurrence of the primary study endpoint compared with the angiography-guided group. In total, 72.7% (7266/10,000) simulated point estimates fell within the actual 95% CI of F3C (-4.7% to -1.4%).CONCLUSIONS:Using a simulation process based on a comparison to an existing trial cohort, the primary results of a prospectively conducted randomized controlled trial could be predicted with reasonable precision.
Journal of evidence-based medicine 2023
PURPOSE:We sought to investigate the short- and long-term outcomes in patients with right ventricular infarction in China.METHODS:Data from China Acute Myocardial Infarction (CAMI) Registry for patients with right ventricular infarction between January 2013 and September 2014 were analyzed.RESULTS:Of the 1,988 patients with right ventricular infarction, 733 patients did not receive reperfusion therapy, 281 patients received thrombolysis therapy, and 974 patients underwent primary PCI. Primary PCI and thrombolysis were all associated with lower risks of in-hospital (3.1 vs. 12.6%; adjusted OR: 0.48; 95% CI: 0.27-0.87; P = 0.0151 and 5.7 vs. 12.6%; adjusted OR: 0.43; 95% CI: 0.22-0.85; P = 0.0155, respectively), and 2-year all-cause mortality (6.3 vs. 20.9%; adjusted HR: 0.50; 95% CI: 0.34-0.73; P = 0.0003 and 11.0 vs. 20.9%; adjusted HR: 0.59; 95% CI: 0.38-0.92; P = 0.0189, respectively), compared with no reperfusion therapy. Meanwhile, primary PCI was superior to thrombolysis in reducing the risks of in-hospital atrial-ventricular block (4.2 vs. 8.9%; adjusted OR: 0.46; 95% CI: 0.23-0.91; P = 0.0257), cardiogenic shock (5.3 vs. 13.9%; adjusted OR: 0.43; 95% CI: 0.23-0.83; P = 0.0115), and heart failure (8.5 vs. 23.5%; adjusted OR: 0.35; 95% CI: 0.22-0.56; P < 0.0001). Primary PCI could reduce the risk of 2-year major adverse cardiac and cerebrovascular event (19.1 vs. 33.3%; adjusted HR: 0.72; 95% CI: 0.56-0.92; P = 0.0092) relative to no reperfusion therapy, whereas thrombolysis may increase the risk of 2-year revascularization (15.5 vs. 8.7%; adjusted HR: 1.90; 95% CI: 1.15-3.16; P = 0.0124) compared with no reperfusion therapy.CONCLUSIONS:Timely reperfusion therapy is essential for patients with right ventricular infarction. Primary PCI may be considered as the default treatment strategy for patients with right ventricular infarction in the contemporary primary PCI era.
Frontiers in cardiovascular medicine 2022
Background: A single measurement of grip strength (GS) could predict the incidence of cardiovascular disease (CVD). However, the long-term pattern of GS and its association with incident CVD are rarely studied. We aimed to characterize the GS trajectory and determine its association with the incidence of CVD (myocardial infarction, angina, stroke, and heart failure). Methods: This study included 5,300 individuals without CVD from a British community-based cohort in 2012 (the baseline). GS was repeatedly measured in 2004, 2008, and 2012. Long-term GS patterns were identified by the group-based trajectory model. Cox proportional hazard models were used to examine the associations between GS trajectories and incident CVD. We identified three GS trajectories separately for men and women based on the 2012 GS measurement and change patterns during 2004-2012. Results: After a median follow-up of 6.1 years (during 2012-2019), 392 participants developed major CVD, including 114 myocardial infarction, 119 angina, 169 stroke, and 44 heart failure. Compared with the high stable group, participants with low stable GS was associated with a higher incidence of CVD incidence [hazards ratio (HR): 2.17; 95% confidence interval (CI): 1.52-3.09; P <0.001], myocardial infarction (HR: 2.01; 95% CI: 1.05-3.83; P = 0.035), stroke (HR: 1.96; 95% CI: 1.11-3.46; P = 0.020), and heart failure (HR: 6.91; 95% CI: 2.01-23.79; P = 0.002) in the fully adjusted models. Conclusions: The low GS trajectory pattern was associated with a higher risk of CVD. Continuous monitoring of GS values could help identify people at risk of CVD.
Frontiers in cardiovascular medicine 2021
Background: Mechanically ventilated patients in the intensive care unit (ICU) have high mortality rates. There are multiple prediction scores, such as the Simplified Acute Physiology Score II (SAPS II), Oxford Acute Severity of Illness Score (OASIS), and Sequential Organ Failure Assessment (SOFA), widely used in the general ICU population. We aimed to establish prediction scores on mechanically ventilated patients with the combination of these disease severity scores and other features available on the first day of admission. Methods: A retrospective administrative database study from the Medical Information Mart for Intensive Care (MIMIC-III) database was conducted. The exposures of interest consisted of the demographics, pre-ICU comorbidity, ICU diagnosis, disease severity scores, vital signs, and laboratory test results on the first day of ICU admission. Hospital mortality was used as the outcome. We used the machine learning methods of k-nearest neighbors (KNN), logistic regression, bagging, decision tree, random forest, Extreme Gradient Boosting (XGBoost), and neural network for model establishment. A sample of 70% of the cohort was used for the training set; the remaining 30% was applied for testing. Areas under the receiver operating characteristic curves (AUCs) and calibration plots would be constructed for the evaluation and comparison of the models' performance. The significance of the risk factors was identified through models and the top factors were reported. Results: A total of 28,530 subjects were enrolled through the screening of the MIMIC-III database. After data preprocessing, 25,659 adult patients with 66 predictors were included in the model analyses. With the training set, the models of KNN, logistic regression, decision tree, random forest, neural network, bagging, and XGBoost were established and the testing set obtained AUCs of 0.806, 0.818, 0.743, 0.819, 0.780, 0.803, and 0.821, respectively. The calibration curves of all the models, except for the neural network, performed well. The XGBoost model performed best among the seven models. The top five predictors were age, respiratory dysfunction, SAPS II score, maximum hemoglobin, and minimum lactate. Conclusion: The current study indicates that models with the risk of factors on the first day could be successfully established for predicting mortality in ventilated patients. The XGBoost model performs best among the seven machine learning models.
Frontiers in medicine 2021
