Wei Zhao

Objective To develop and validate prediction models for peak oxygen uptake (VO₂peak) in patients with coronary heart disease (CHD) using submaximal cardiopulmonary exercise testing (CPET) indicators and deep learning methods. Design Retrospective model development and validation study. Setting Cardiac Rehabilitation Centre, Peking University Third Hospital, China. Participants A total of 10 538 patients with CHD who underwent CPET between January 2014 and December 2019. Methods Clinical data and CPET indicators were collected. Multiple machine learning and deep learning models were developed and compared. Model performance was assessed using R², mean absolute error (MAE), bias, Bland–Altman analysis and SHapley Additive exPlanations (SHAP) feature importance ranking. Results The neural network model achieved the best performance (R² = 0.82, MAE=1.55 mL/kg/min, bias=0.08). XGBoost was the best-performing traditional machine learning model (R² = 0.74). SHAP analysis identified eight top-ranked features, including VO₂@AT, OUES, weight, VE/VCO₂ slope, VE/VCO₂@AT, age, gender and HR@AT. Conclusion The CPET deep learning model shows potential for predicting VO₂peak in CHD patients, but further external validation and prospective studies are required before clinical application.

Background: There has been insufficient research on the assessment of exercise capacity in patients with coronary heart disease (CHD) following percutaneous coronary intervention (PCI) who exhibit inspiratory muscle weakness (IMW). Methods: A retrospective cohort study involving CHD patients who underwent PCI at Peking University Third Hospital Heart Rehabilitation Center between January 2019 and December 2021 was conducted. Patients who had undergone inspiratory muscle testing and cardiopulmonary exercise testing (CPET) were included, and their clinical data were collected and analyzed. Results: A total of 571 post-PCI CHD patients were included in the study. The average age was 60.8 ± 4.3 years, and 479 male patients (83.9%) were included. The average maximal inspiratory pressure (MIP) of the enrolled patients was 90.7 ± 26.1 cm H₂O, with 56 patients (9.8%) presenting with IMW. The IMW group had lower peak oxygen uptake (VO₂peak) (17.4 ± 4.2 vs. 19.3 ± 5.1 ml/min/kg, P < 0.001) and oxygen uptake efficiency slopes (OUES) (1464.7 ± 368.5 vs. 1619.2 ± 400.4, P=0.004). MIP correlated with VO₂peak (r = 0.719, P < 0.001) and OUES (r = 0.622, P < 0.001). Multivariate regression analysis revealed that VO₂peak (OR = 0.917, 95% CI = 0.858 ~ 0.980) and history of chronic obstructive pulmonary disease (COPD) (OR = 1.705, 95% CI = 0.934 ~ 3.112) were independent risk factors for IMW. Conclusions: After PCI, CHD patients exhibiting IMW, especially those with comorbid COPD, demonstrated reduced exercise tolerance and oxygen uptake efficiency. Keywords: coronary heart disease inspiratory muscle weakness exercise tolerance cardiopulmonary exercise testing chronic obstructive pulmonary disease

Accurately quantifying cardiorespiratory fitness (CRF) through cardiopulmonary exercise testing (CPET) is increasingly important for improving risk assessment and guiding clinical decisions. However, research on V̇O₂peak reference values and predictive equations for the Chinese population remains limited. This study aimed to establish a V̇O₂peak predictive equation for Chinese adults. This study analyzed healthy participants who underwent CPET at Peking University Third Hospital (PUTH) from September 1, 2017, to September 1, 2023. Data from September 1, 2017, and August 31, 2021 were used as the derivation cohort, and September 1, 2021, to September 1, 2023 were utilized as an external validation cohort for temporal validation. The derivation cohort underwent backward multivariate regression analysis to generate the V̇O₂peak prediction equation, which was compared with the widely-used Wasserman, FRIEND and Xiangya equations. The PUTH derivation cohort (N = 4531, mean age: 50.7 years, 18–88 years) and validation cohort (N = 4624, mean age: 46.1 years, 18–89 years) included 48.8 and 48.5% men, respectively. With increasing age, both men and women V̇O₂peak exhibited a general decline. The predictive equation for V̇O₂peak was established based on the derivation cohort: V̇O₂peak (mL·min⁻¹) = −24364.9 − 621.3 × Sex (Women = 1, Men = 2) −10.7 × Age + 0.2 × Height² (cm) + 6464.7 × Log(BMI) −24997.2 × Log(BSA) +12388.6 × Log(LBM) (adjusted R² = 0.624, p < 0.001). It demonstrated higher consistency between measured and predicted results compared to Xiangya, Wasserman, and FRIEND equations. This study presents the PUTH equation, a new V̇O₂peak prediction equation for Chinese adults. Compared to existing equations, the PUTH equation shows reduced bias and improved accuracy, providing a more reliable tool for assessing CRF and guiding clinical interventions in the Chinese population.

Objective: To investigate the cut-off value of the breathing reserve for predicting a decline in cardiorespiratory fitness (CRF) among healthy middle-aged Chinese individuals. Methods: Healthy middle-aged individuals who underwent cardiopulmonary exercise testing (CPET) at the Peking University Third Hospital from May to October 2021 were selected. The study included 321 participants, with an average age of 48.8 ± 5.7 years. They were divided into two groups based on the peak oxygen uptake (VO₂peak): the adequate CRF group and the CRF decline group. Multivariate logistic regression analysis was used to explore the factors influencing CRF. Results: In the male CRF decline group, heart rate, alanine aminotransferase, end-tidal partial pressure of carbon dioxide (PETCO₂), and breathing reserve (BR%) were significantly higher, while the oxygen uptake at the anaerobic threshold (VO₂@AT) was lower. An elevated BR% was independently associated with CRF decline (OR = 1.111, 95% CI: 1.068–1.156). The female CRF decline group had significantly higher FEV1/FVC and BR% and significantly lower age, fasting glucose, hemoglobin, and VO₂@AT compared to the adequate CRF group. Elevated BR% was independently associated with CRF decline (OR = 1.086, 95% CI: 1.038–1.137). The receiver operating characteristic (ROC) curve for the males showed an area under the curve (AUC) of 0.769 (95% CI: 0.703–0.827) with an appropriate BR% cut-off value of 49.9%, sensitivity of 59.9%, and specificity of 77.8%. For the females, the ROC curve displayed an AUC of 0.694 (95% CI: 0.607–0.773) with an appropriate BR% cut-off value of 57.0%, sensitivity of 58.7%, and specificity of 86.0%. Conclusions: The breathing reserve was independently associated with CRF. The appropriate cut-off values for BR% to predict CRF decline were 49.9% for the males and 57.0% for the females.

This study investigated the influencing factors of exercise systolic blood pressure response (ESBPR) by cardiopulmonary exercise test (CPX) in nonalcoholic fatty liver disease (NAFLD) in people aged 40-60 years. A total of 603 adults were enrolled in this study. The inclusion criteria of this cross-sectional study were adults who underwent health checks and CPX. There were significant differences in body mass Index (BMI) (26.80 ± 2.64 VS 23.31 ± 2.41, p < 0.001) kg/m2, fasting blood glucose (FPG) (5.56 ± 0.94 VS 5.13 ± 0.55, p < 0.001) mmol/L, alanine aminotransferase (ALT), aspartate transaminase (AST), triglycerides (TG), high-density lipoprotein-cholesterol (HDL-C) (1.13 ± 0.22 VS 1.43 ± 0.33, p < 0.001) mmol/L, low-density lipoprotein-cholesterol (LDL-C) (3.21 ± 0.79 VS 2.99 ± 0.68, p = 0.001) mmol/L, resting systolic blood pressure (SBP) (123.53 ± 14.73 VS 118.79 ± 14.79, p < 0.001) mmHg, resting diastolic blood pressure (DBP) (80.29 ± 9.62 VS 75.27 ± 10.41, p < 0.001) mmHg, peak SBP (184.01 ± 23.50 VS 172.81 ± 24.95, p < 0.001) mmHg, peak DBP (87.47 ± 10.50 VS 84.01 ± 11.46, p = 0.001) mmHg, oxygen pulse (VO2/HR) (0.88 ± 0.15 VS 0.92 ± 0.16, p = 0.004) ml/beat, exercise maximum heart rate, carbon dioxide Ventilation equivalent (VE/VCO2) (25.84 ± 4.43 VS 25.12 ± 3.58, p = 0.038), peak oxygen uptake (VO2 peak) (1.78 ± 0.45 VS 1.56 ± 0.46, p < 0.001) ml/min between the NAFLD and control groups. VE/VCO2 (OR = 0.822, p = 0.036) and oxygen uptake/work rate (VO2/WR) (OR = 0.517, p = 0.021) ml/min/watt were associated with a lower risk of ESBPR in NAFLD subjects. Resting SBP was associated with a higher risk of ESBPR in NAFLD patients (OR = 1.059, p = 0.003) and overweight NAFLD subjects (OR = 1.075, p = 0.002). ESBPR (OR = 1.268, p = 0.045), skeletal-muscle mass (OR = 1.305, p < 0.001), and SMI (OR = 1.315, p < 0.001) were linked to an elevated risk of NAFLD in individuals. Our findings indicate that ESBPR is associated with an increased risk of NAFLD in individuals aged 40-60 years. Furthermore, in NAFLD subjects, VE/VCO2 and VO2/WR were found to be correlated with a decreased risk of ESBPR, whereas resting SBP was linked to an elevated risk of ESBPR. This will provide a research basis for the NAFLD subjects who have ESBPR at risk of adverse events during exercise.

This study aimed to determine whether the waist-to-thigh ratio (WTTR) is associated with the incidence of metabolic-associated fatty liver disease (MAFLD) in health care workers. There were 4517 health care workers with baseline data and results from 2 follow-up examinations. We divided the subjects into 3 groups according to baseline WTTR and used the Cox hazard regression model to estimate MAFLD risk. The WTTRs were categorized by tertiles at baseline using the values 1.58 and 1.66. Patients with higher WTTR tended to have significantly greater values for the following factors, body mass index (BMI), fasting blood glucose (FPG), systolic blood pressure, diastolic blood pressure, total cholesterol (TC), triglycerides (TG), low-density lipoprotein-cholesterol (LDL-C) and neck circumference. The incidence of MAFLD significantly increased with increasing WTTR tertiles (5.74%, 12.75% and 22.25% for the first, second and third tertiles, respectively, P < 0.05 for trend). Kaplan-Meier(K-M) survival analysis revealed a significant tendency towards increased MAFLD risk with increasing WTTR tertile. In the fully adjusted model, the hazard ratios (95% CIs) for MAFLD in the second, third WTTR tertiles compared with the first quartile were 2.17(1.58,2.98), 3.63(2.70,4.89), respectively, third neck circumference tertiles compared with the first quartile were 2.84(1.89,4.25), 8.95(6.00,13.35), respectively. Compared with those of individuals with a BMI > 23 kg/m2, the associations between WTTR and MAFLD incidence were more pronounced in subjects with a BMI < 23 kg/m2. Similarly, the difference in neck circumference was more pronounced in these patients with a BMI < 23 kg/m2. Our results revealed that the WTTR is an independent risk factor for MAFLD, and there was a dose‒response relationship between the WTTR and MAFLD risk. The neck circumference was significantly different in subjects with a BMI < 23 kg/m2. This approach provides a new way to predict the incidence rate of MAFLD.

Prescribing appropriate exercise is an important means to improve the safety and efficacy of cardiac rehabilitation. Improper exercise may induce an increased cardiovascular risk in older persons with coronary heart disease. Cardiopulmonary exercise testing (CPET)-guided cardiac rehabilitation could be helpful for providing clinical evidence for cardiac rehabilitation therapy in older persons after percutaneous coronary intervention (PCI). We retrospectively included older persons who underwent PCI and cardiac rehabilitation based on CPET at the Cardiac Rehabilitation Center of Peking University Third Hospital from January 2014 to December 2019. Patients’ baseline and follow-up clinical data were collected. A total of 403 older persons after PCI were included in the study. The mean age was 80.5 ± 4.3. The mean follow-up time was 12 ± 2 months. During the follow-up period, no significant exercise-related adverse events occurred, and the peak oxygen uptake (VO₂peak) increased compared with baseline (15.5 ± 3.8 ml/min/kg vs. 17.3 ± 4.1 ml/min/kg). Among the 90 patients (22.2%) without exercise habits at baseline who started regular exercise during follow-up, the improvement in VO₂peak was most significant, at 3.2 ± 0.4 ml/min/kg. Cardiac rehabilitation based on CPET improved exercise habits and exercise tolerance in older persons with coronary heart disease after PCI.

Background: Prescribing appropriate exercise is an important means to improve the safety and efficacy of cardiac rehabilitation. Improper exercise may induce an increased cardiovascular risk in older persons with coronary heart disease. Cardiopulmonary exercise testing (CPET)-guided cardiac rehabilitation could be helpful for providing clinical evidence for cardiac rehabilitation therapy in older persons after percutaneous coronary intervention (PCI). Methods: We retrospectively included older persons who underwent PCI and cardiac rehabilitation based on CPET at the Cardiac Rehabilitation Center of XXX Hospital from January 2014 to December 2019. Patients' baseline and follow-up clinical data were collected. Results: A total of 403 older persons after PCI were included in the study. The mean age was 80.5±4.3. The mean follow-up time was 12±2 months. During the follow-up period, no significant exercise-related adverse events occurred, and the peak oxygen uptake (VO2peak) increased compared with baseline (15.5±3.8 ml/min/kg vs. 17.3±4.1 ml/min/kg). Among the 90 patients (22.2%) without exercise habits at baseline who started regular exercise during follow-up, the improvement in VO2peak was most significant, at 3.2±0.4 ml/min/kg. Conclusions: Cardiac rehabilitation based on CPET improved exercise habits and exercise tolerance in older persons with coronary heart disease after PCI.

Purpose of Review: Heart failure is a serious global health problem, and coronary artery disease is one of the main causes. At present, the treatment options for ischemic heart failure (IHF) are limited. This article mainly aims to explore the evidence of enhanced external counterpulsation (EECP) as a non-invasive cardiac rehabilitation method in patients with IHF and to make a preliminary exploration of its mechanisms. Recent Findings: According to the existing evidence, the standard course of EECP is safe in patients with IHF and can significantly improve the quality of life of these patients. The effect of EECP on systolic function is still unclear, while EECP has a significant improvement effect on cardiac diastolic function. At the same time, this treatment can reduce the re-hospitalization rate and emergency visit rate of patients within 6 months. In terms of mechanisms, in addition to the immediate hemodynamic effect, existing evidence mostly suggests that its improvement of cardiac function may come from its upregulation of shear stress to improve myocardial perfusion. Summary: EECP is safe to use in patients with stable ischemic heart failure, and it can improve the performance status of patients and may be beneficial to cardiac function and reduce the short-term re-hospitalization rate.

Background: The triglyceride-glucose (TyG) index has been recognized as being an alternative cardiometabolic biomarker for insulin resistance associated with the development and prognosis of cardiovascular disease (CVD). However, the prospective relationship between baseline and long-term trajectories of the TyG index and carotid atherosclerosis (CAS) progression has yet to be investigated. Methods: This longitudinal prospective cohort study included 10,380 adults with multiple general health checks at Peking University Third Hospital from January 2011 to December 2020. The TyG index was calculated as ln (fasting triglyceride [mg/dL] × fasting glucose [mg/dL]/2). The latent class trajectory modeling method was used to analyze the TyG index trajectories over the follow-up. Based on univariate and multivariate Cox proportional hazards analyses, hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated for the baseline and trajectory of the TyG index. Results: During a median follow-up period of 757 days, 1813 participants developed CAS progression. Each 1-standard deviation (SD) increase in the TyG index was associated with a 7% higher risk of CAS progression after adjusting for traditional CVD risk factors (HR = 1.067, 95% CI 1.006–1.132). Similar results were observed when the TyG index was expressed as quartiles. According to different trajectory patterns, participants were categorized into low-stable, moderate-stable, and high-increasing groups. After multivariate adjustment, the moderate-stable group had a 1.139-fold (95% CI 1.021–1.272) risk of CAS progression. The high-increasing trajectory of the TyG index tended to be associated with CAS progression (HR = 1.206, 95% CI 0.961–1.513). Conclusions: Participants with higher baseline and moderate-stable trajectory of the TyG index were associated with CAS progression. Long-term trajectories of the TyG index can help to identify individuals at a higher risk of CAS progression who deserve specific preventive and therapeutic approaches.