Machine learning-based risk models for procedural complications of radiofrequency ablation for atrial fibrillation.

BACKGROUND: Radiofrequency ablation (RFA) for atrial fibrillation (AF) is associated with a risk of complications. This study aimed to develop and validate risk models for predicting complications after radiofrequency ablation of atrial fibrillation patients. METHODS: This retrospective cohort study included 3365 procedures on 3187 patients with atrial fibrillation at a single medical center from 2018 to 2021. The outcome was the occurrence of postoperative procedural complications during hospitalization. Logistic regression, decision tree, random forest, gradient boosting machine, and extreme gradient boosting were used to develop risk models for any postoperative complications, cardiac effusion/tamponade, and hemorrhage, respectively. Patients' demographic characteristics, medical history, signs, symptoms at presentation, electrocardiographic features, procedural characteristics, laboratory values, and postoperative complications were collected from the medical record. The prediction results were evaluated by performance metrics (i.e., the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, F score, and Brier score) with repeated fivefold cross-validation. RESULTS: Of the 3365 RFA procedures, there were 62 procedural complications with a rate of 1.84% in the entire cohort. The most common complications were cardiac effusion/tamponade (28 cases, 0.83%), and hemorrhage (21 cases, 0.80%). There was no procedure-related mortality. The machine learning algorithms of random forest (RF) outperformed other models for any complication (AUC 0.721 vs 0.627 to 0.707), and hemorrhage (AUC 0.839 vs 0.649 to 0.794). The extreme gradient boosting (XGBoost) model outperformed other models for cardiac effusion/tamponade (AUC 0.696 vs 0.606 to 0.662). CONCLUSIONS: The developed risk models using machine learning algorithms showed good performance in predicting complications after RFA of AF patients. These models help identify patients at high risk of complications and guiding clinical decision-making.

Development of interpretable machine learning models for prediction of acute kidney injury after noncardiac surgery: a retrospective cohort study.

BACKGROUND: Early identification of patients at high-risk of postoperative acute kidney injury (AKI) can facilitate the development of preventive approaches. This study aimed to develop prediction models for postoperative AKI in noncardiac surgery using machine learning algorithms. The authors also evaluated the predictive performance of models that included only preoperative variables or only important predictors. MATERIALS AND METHODS: Adult patients undergoing noncardiac surgery were retrospectively included in the study (76 457 patients in the discovery cohort and 11 910 patients in the validation cohort). AKI was determined using the KDIGO criteria. The prediction model was developed using 87 variables (56 preoperative variables and 31 intraoperative variables). A variety of machine learning algorithms were employed to develop the model, including logistic regression, random forest, extreme gradient boosting, and gradient boosting decision trees. The performance of different models was compared using the area under the receiver operating characteristic curve (AUROC). Shapley Additive Explanations (SHAP) analysis was employed for model interpretation. RESULTS: The patients in the discovery cohort had a median age of 52 years (IQR: 42-61 years), and 1179 patients (1.5%) developed AKI after surgery. The gradient boosting decision trees algorithm showed the best predictive performance using all available variables, or only preoperative variables. The AUROCs were 0.849 (95% CI: 0.835-0.863) and 0.828 (95% CI: 0.813-0.843), respectively. The SHAP analysis showed that age, surgical duration, preoperative serum creatinine, and gamma-glutamyltransferase, as well as American Society of Anesthesiologists physical status III were the most important five features. When gradually reducing the features, the AUROCs decreased from 0.852 (including the top 40 features) to 0.839 (including the top 10 features). In the validation cohort, the authors observed a similar pattern regarding the models' predictive performance. CONCLUSIONS: The machine learning models the authors developed had satisfactory predictive performance for identifying high-risk postoperative AKI patients. Furthermore, the authors found that model performance was only slightly affected when only preoperative variables or only the most important predictive features were included.

Blood pressure response index and clinical outcomes in patients with septic shock: a multicenter cohort study.

BACKGROUND: Sepsis is a leading cause of mortality in intensive care units and vasoactive drugs are widely used in septic patients. The cardiovascular response of septic shock patients during resuscitation therapies and the relationship of the cardiovascular response and clinical outcome has not been clearly described. METHODS: We included adult patients admitted to the ICU with sepsis from Peking Union Medical College Hospital (internal), Medical Information Mart for Intensive Care IV (MIMIC-IV) and eICU Collaborative Research Database (eICU-CRD). The Blood Pressure Response Index (BPRI) was defined as the ratio between the mean arterial pressure and the vasoactive-inotropic score. BRRI was compared with existing risk scores on predicting in-hospital death. The relationship between BPRI and in-hospital mortality was calculated. A XGBoost's machine learning model identified the features that influence short-term changes in BPRI. FINDINGS: There were 2139, 9455, and 4202 patients in the internal, MIMIC-IV and eICU-CRD cohorts, respectively. BPRI had a better AUROC for predicting in-hospital mortality than SOFA (0.78 vs. 0.73, p = 0.01) and APS (0.78 vs. 0.74, p = 0.03) in the internal cohort. The estimated odds ratio for death per unit decrease in BPRI was 1.32 (95% CI 1.20-1.45) when BPRI was below 7.1 vs. 0.99 (95% CI 0.97-1.01) when BPRI was above 7.1 in the internal cohort; similar relationships were found in MIMIC-IV and eICU-CRD. Respiratory support and latest cumulative 12-h fluid balance were intervention-related features influencing BPRI. INTERPRETATION: BPRI is an easy, rapid, precise indicator of the response of patients with septic shock to vasoactive drugs. It is a comparable and even better predictor of prognosis than SOFA and APS in sepsis and it is simpler and more convenient in use. The application of BPRI could help clinicians identify potentially at-risk patients and provide clues for treatment. FUNDING: Fundings for the Beijing Municipal Natural Science Foundation; the National High Level Hospital Clinical Research Funding; the CAMS Innovation Fund for Medical Sciences (CIFMS) from Chinese Academy of Medical Sciences and the National Key R&D Program of China, Ministry of Science and Technology of the People's Republic of China.

Multidimensional dynamic prediction model for hospitalized patients with the omicron variant in China.

Purpose: To establish dynamic prediction models by machine learning using daily multidimensional data for coronavirus disease 2019 (COVID-19) patients. Methods: Hospitalized COVID-19 patients at Peking Union Medical College Hospital from Nov 2nd, 2022, to Jan 13th, 2023, were enrolled in this study. The outcome was defined as deterioration or recovery of the patient's condition. Demographics, comorbidities, laboratory test results, vital signs, and treatments were used to train the model. To predict the following days, a separate XGBoost model was trained and validated. The Shapley additive explanations method was used to analyze feature importance. Results: A total of 995 patients were enrolled, generating 7228 and 3170 observations for each prediction model. In the deterioration prediction model, the minimum area under the receiver operating characteristic curve (AUROC) for the following 7 days was 0.786 (95% CI 0.721-0.851), while the AUROC on the next day was 0.872 (0.831-0.913). In the recovery prediction model, the minimum AUROC for the following 3 days was 0.675 (0.583-0.767), while the AUROC on the next day was 0.823 (0.770-0.876). The top 5 features for deterioration prediction on the 7th day were disease course, length of hospital stay, hypertension, and diastolic blood pressure. Those for recovery prediction on the 3rd day were age, D-dimer levels, disease course, creatinine levels and corticosteroid therapy. Conclusion: The models could accurately predict the dynamics of Omicron patients' conditions using daily multidimensional variables, revealing important features including comorbidities (e.g., hyperlipidemia), age, disease course, vital signs, D-dimer levels, corticosteroid therapy and oxygen therapy.

Differential mechanistic investigation of protective effects from imperatorin and sec-O-glucosylhamaudol against arsenic trioxide-induced cytotoxicity in vitro.

BACKGROUND AND PURPOSE: The clinical use of arsenic trioxide (As2O3) for treating acute promyelocytic leukemia (APL) is limited due to its severe cardiotoxicity. The possible mechanisms of As2O3-induced cardiotoxicity include DNA fragmentation, reactive oxygen species (ROS) generation, cardiac ion channel changes and apoptosis. The present study is designed to investigate the protective effects of imperatorin and sec-O-glucosylhamaudol and to explore their mechanistic involvement in As2O3-induced cytotoxicity. EXPERIMENTAL METHODS: Cell viability assay, Lactate dehydrogenase (LDH) release, Acridine orange/ethidium bromide (AO/EB) double staining, Caspase-3 activity assay, ROS generation, cellular calcium levels, mRNA expression levels by qRT-PCR and protein expression levels by Western blotting were measured in H9c2 cells in combination with As2O3 and imperatorin or sec-O-glucosylhamaudol. KEY RESULTS: We observed that H9c2 cells treated with imperatorin or sec-O-glucosylhamaudol were more resistant to As2O3-induced cell death. Both imperatorin and sec-O-glucosylhamaudol reduced H9c2 cell apoptosis, but both imperatorin and sec-O-glucosylhamaudol had no effects on Caspase-3 activity and intracellular calcium accumulation. Furthermore, imperatorin was capable of suppressing ROS generation, while sec-O-glucosylhamaudol did not show this effect. Moreover, imperatorin and sec-O-glucosylhamaudol triggered Nrf2 activation, which resulted in upregulation of downstream phase II metabolic enzymes and antioxidant protein/enzyme, probably offering cellular protection to As2O3-induced cardiotoxicity via the Nrf2 signal pathway. CONCLUSIONS AND IMPLICATIONS: Imperatorin and sec-O-glucosylhamaudol can ameliorate As2O3-induced cytotoxicity and apoptosis in H9c2 cells, the mechanisms probably related to antioxidation. As2O3 in combination with imperatorin or sec-O-glucosylhamaudol could be considered as a novel strategy to expand the clinical application of As2O3.

Chromatographic fingerprints of industrial toluic acids established for their quality control.

The chromatographic fingerprints of industrial o-toluic acid, m-toluic acid and p-toluic acid have been established by HPLC-UV detection according to their impurity groups. HPLC separation of all relative substances involved in the groups was developed on a Kromasil C(18) column by using methanol-water-NH(4)Ac-HAc buffer (100mM, pH 4.70) 15/65/20 (v/v/v) as the mobile phase at a flow rate of 1.5mL/min, and detection was operated by UV adsorption at a wavelength of 254nm. The ultraviolet spectra corresponding to each chromatographic peak were also recorded for further identification of all components. Whether the limits of relative impurities residues in a toluic acid product are qualified or not can be intuitively estimated by analyzing its chromatogram with comparison to the fingerprint. This protocol has successfully provided some Chinese manufacturers with a simple and feasible method for quality control of toluic acids for industrial use.