江继承
阜外华中心血管病医院 大数据中心
BACKGROUND:The impact of preoperative statin use on postoperative acute kidney injury (AKI) is uncertain. We aimed to examine the association of statin therapy before cardiac surgery with postoperative AKI.METHODS:The retrospective cohort study consisted of 1581 patients undergoing cardiac surgery. Postoperative AKI were identified by the modified KDIGO definition. Propensity-score matching was employed to control for selection bias, and logistic regression was used to control for confounders. Subgroup and interaction analyses were performed to evaluate the robustness of the findings.RESULTS:The overall incidence of postoperative AKI and severe AKI were 42.19% and 12.27%, respectively. Preoperative moderate-dose statin was significantly associated with a reduced incidence of postoperative AKI (28.9% vs 43.0%, OR (95%CI): 0.54 (0.38, 0.77), p < 0.001) and severe AKI (6.9% vs 13.7%, OR (95%CI): 0.46 (0.26, 0.83), p = 0.009). The beneficial effect on postoperative AKI persisted after adjusting for major confounding factors (OR (95%CI): 0.47 (0.34, 0.66)). Decreased risk of postoperative AKI was observed in patients with preoperative statin duration of 7 ∼ 14 days (OR (95%CI): 0.41 (0.25, 0.65)) and over 14 days (OR (95%CI): 0.43 (0.28, 0.65)), but not in those with preoperative statin duration of <7 days. Similar favorable effects were noted in most subgroup patients, except for those with high-risk factors such as diabetes mellitus, previous congestive cardiac failure, arrhythmia, preoperative ACEI/ARB, aortic cross-clamping or IABP.CONCLUSION:Preoperative moderate-dose statin was significantly related to a decreased risk of postoperative AKI, especially in patients who received statins for a longer duration. Further large-scale multicenter randomized controlled trials are needed to ascertain the impact of statin dose, duration, and timing on postoperative AKI in cardiac surgery patients.
Current medical research and opinion 2024
OBJECTIVE:Postoperative acute kidney injury (PO-AKI) is a common complication after cardiac surgery. We aimed to evaluate whether machine learning algorithms could significantly improve the risk prediction of PO-AKI.METHODS:The retrospective cohort study included 2310 adult patients undergoing cardiac surgery in a tertiary teaching hospital, China. Postoperative AKI and severe AKI were identified by the modified KDIGO definition. The sample was randomly divided into a derivation set and a validation set based on a ratio of 4:1. Exploiting conventional logistic regression (LR) and five ML algorithms including decision tree, random forest, gradient boosting classifier (GBC), Gaussian Naive Bayes and multilayer perceptron, we developed and validated the prediction models of PO-AKI. We implemented the interpretation of models using SHapley Additive exPlanation (SHAP) analysis.RESULTS:Postoperative AKI and severe AKI occurred in 1020 (44.2%) and 286 (12.4%) patients, respectively. Compared with the five ML models, LR model for PO-AKI exhibited the largest AUC (0.812, 95%CI: 0.756, 0.860, all P < 0.05), sensitivity (0.774, 95%CI: 0.719, 0.813), accuracy (0.753, 95%CI: 0.719, 0.781) and Youden index (0.513, 95%CI: 0.451, 0.573). Regarding severe AKI, GBC algorithm showed a significantly higher AUC than the other four ML models (all P < 0.05). Although no significant difference (P = 0.173) was observed in AUCs between GBC (0.86, 95%CI: 0.808, 0.902) and conventional logistic regression (0.803, 95%CI: 0.746, 0.852), GBC achieved greater sensitivity, accuracy and Youden index than conventional LR. Notably, SHAP analyses showed that preoperative serum creatinine, hyperlipidemia, lipid-lowering agents and assisted ventilation time were consistently among the top five important predictors for both postoperative AKI and severe AKI.CONCLUSION:Logistic regression and GBC algorithm demonstrated moderate to good discrimination and superior performance in predicting PO-AKI and severe AKI, respectively. Interpretation of the models identified the key contributors to the predictions, which could potentially inform clinical interventions.
BMC nephrology 2023
OBJECTIVE:We aimed to assess hypertension prevalence and management in Central China.METHODS:In this cross-sectional study conducted from February 2019 to February 2020, we applied stratified multistage random sampling to investigate residents aged 35 to 75 years in Dancheng county of Zhoukou city and Xincai county of Zhumadian city, both in Central China.RESULTS:We enrolled 63,940 participants in this survey. A total of 43.6% (95% confidence interval [CI]: 43.2-44.0) of participants had hypertension. Of these, 49.3% (95% CI: 48.7-49.9) were aware of their diagnosis, 36.5% (95% CI: 35.9-37.1) took antihypertensive medication, and 14.3% (95% CI: 13.9-14.7) had their blood pressure under control. Only 31.4% of hypertensive people receiving antihypertensive treatment had achieved control. The hypertension prevalence was lower in urban areas than in rural areas, with higher rates of awareness, treatment, and control. Among subgroups, rural men had the highest prevalence of hypertension. Prevalence, awareness, and treatment rates all increased with age, except for control rates, which declined in the 65- to 75-year-old group.CONCLUSIONS:People in Central China have a high hypertension prevalence but low rates of awareness, treatment, and control. Great effort is needed to improve the prevention and management of hypertension in this region.
The Journal of international medical research 2023
To comprehensively estimate the association of gestational diabetes mellitus (GDM) risk with maternal red blood cell (RBC) folate, plasma/serum folate, dose and duration of folic acid supplement (FAS) intake and vitamin B12 separately. PubMed, Web of science, CNKI, and Wanfang Databases were searched through March 26, 2021. We synthesized data using random-effects model meta-analysis in Stata 12.0. Sensitivity, subgroup and dose-response analyses were also performed. The certainty of evidence was evaluated using the Grading of Recommendations, Assessment, Development and Evaluations (GRADE). Twenty six datasets from thirteen eligible observational studies were included in the study. We found a significant increase of GDM risk with the highest versus lowest category of RBC folate (OR = 1.96, 95% CI: 1.48-2.61, I2 = 0.0%, moderate-certainty evidence) and plasma/serum folate (OR = 1.23, 1.02-1.48, I2 = 57.8%, low-certainty evidence). The dose-response analysis revealed that each 200 ng/ml increase in RBC folate was significantly associated with 8% higher GDM risk. No significant association between dose of FAS intake and GDM risk was found with very low cetainty. Meanwhile, longer duration (≥3 months) of FAS conferred 56% significant higher GDM risk (OR = 1.56, 1.02-2.39, very low certainty evidence). No significant association of GDM risk with highest plasma/serum B12 was observed compared to lowest B12 (OR = 0.77, 0.58-1.02, very low-certainty evidence). Moderate-certainty evidence suggests that higher RBC folate appears to significantly increase GDM risk. Higher plasma/serum folate may increase GDM risk but with low certainty. Further well-designed trials or prospective studies are needed.
European journal of clinical nutrition 2022
BACKGROUND:Machine learning (ML) can include more diverse and more complex variables to construct models. This study aimed to develop models based on ML methods to predict the all-cause mortality in coronary artery disease (CAD) patients with atrial fibrillation (AF).METHODS:A total of 2037 CAD patients with AF were included in this study. Three ML methods were used, including the regularization logistic regression, random forest, and support vector machines. The fivefold cross-validation was used to evaluate model performance. The performance was quantified by calculating the area under the curve (AUC) with 95% confidence intervals (CI), sensitivity, specificity, and accuracy.RESULTS:After univariate analysis, 24 variables with statistical differences were included into the models. The AUC of regularization logistic regression model, random forest model, and support vector machines model was 0.732 (95% CI 0.649-0.816), 0.728 (95% CI 0.642-0.813), and 0.712 (95% CI 0.630-0.794), respectively. The regularization logistic regression model presented the highest AUC value (0.732 vs 0.728 vs 0.712), specificity (0.699 vs 0.663 vs 0.668), and accuracy (0.936 vs 0.935 vs 0.935) among the three models. However, no statistical differences were observed in the receiver operating characteristic (ROC) curve of the three models (all P > 0.05).CONCLUSION:Combining the performance of all aspects of the models, the regularization logistic regression model was recommended to be used in clinical practice.
BMC cardiovascular disorders 2021
