Autonomic dysfunction in patients with chronic obstructive pulmonary disease (COPD) may increase the risks of arrhythmia and sudden death. We studied cardiac autonomic function in patients with acute exacerbation of COPD (AECOPD). Patients with AECOPD were classified into ventricular tachycardia (VT) and non-VT groups according to the presence or absence of VT. The following parameters derived from 24-h Holter monitoring were compared between groups: average heart rate, heart rate deceleration capacity (DC), heart rate acceleration capacity (AC), standard deviation of normal RR intervals (SDNN), standard deviation of average RR interval in 5-min segments (SDANN), root mean square of standard deviations of differences between adjacent normal RR intervals (rMSSD), low-frequency power (LF), high-frequency power (HF) and LF/HF ratio. Seventy patients were included, 22 in the VT group and 48 in the non-VT group. The groups had similar clinical characteristics (except for more common amiodarone use in the VT group, P < 0.05) and general ECG characteristics. DC, SDNN, SDANN and rMSSD were lower and AC higher in the VT group (P < 0.05). In the VT group, DC was correlated positively with SDNN (r = 0.716), SDANN (r = 0.595), rMSSD (r = 0.571) and HF (r = 0.486), and negatively with LF (r = -0.518) and LF/HF (r = -0.458) (P < 0.05). AC was correlated negatively with SDNN (r = -0.682), SDANN (r = -0.567) and rMSSD (r = -0.548) (P < 0.05). DC decreased and AC increased in patients with AECOPD and VT, reflecting an imbalance in autonomic regulation of the heart that might increase the risk of sudden death.
Background: The pathophysiologic heterogeneity of heart failure (HF) necessitates a more detailed identification of diagnostic biomarkers that can reflect its diverse pathogenic pathways. Methods: We conducted weighted gene and multiscale embedded gene co-expression network analysis on differentially expressed genes obtained from HF and non-HF specimens. We employed a machine learning integration framework and protein–protein interaction network to identify diagnostic biomarkers. Additionally, we integrated gene set variation analysis, gene set enrichment analysis (GSEA), and transcription factor (TF)-target analysis to unravel the biomarker-dominant pathways. Leveraging single-sample GSEA and molecular docking, we predicted immune cells and therapeutic drugs related to biomarkers. Quantitative polymerase chain reaction validated the expressions of biomarkers in the plasma of HF patients. A two-sample Mendelian randomization analysis was implemented to investigate the causal impact of biomarkers on HF. Results: We first identified COL14A1, OGN, MFAP4, and SFRP4 as candidate biomarkers with robust diagnostic performance. We revealed that regulating biomarkers in HF pathogenesis involves TFs (BNC2, MEOX2) and pathways (cell adhesion molecules, chemokine signaling pathway, cytokine–cytokine receptor interaction, oxidative phosphorylation). Moreover, we observed the elevated infiltration of effector memory CD4+ T cells in HF, which was highly related to biomarkers and could impact immune pathways. Captopril, aldosterone antagonist, cyclopenthiazide, estradiol, tolazoline, and genistein were predicted as therapeutic drugs alleviating HF via interactions with biomarkers. In vitro study confirmed the up-regulation of OGN as a plasma biomarker of HF. Mendelian randomization analysis suggested that genetic predisposition toward higher plasma OGN promoted the risk of HF. Conclusions: We propose OGN as a diagnostic biomarker for HF, which may advance our understanding of the diagnosis and pathogenesis of HF.
Heart failure (HF) is the main manifestation of cardiovascular disease. Recent studies have shown that various RNA molecules and their complex connections play an essential role in HF’s pathogenesis and pathological progression. This paper aims to mine key RNA molecules associated with HF. We proposed a Prior-knowledge Driven Joint Deep Semi-Negative Matrix Factorization (PD-JDSNMF) model that uses a hierarchical nonlinear feature extraction method that integrates three types of data: mRNA, lncRNA, and miRNA. The PPI information is added to the model as prior knowledge, and the Laplacian constraint is used to help the model resist the noise in the genetic data. We used the PD-JDSNMF algorithm to identify significant co-expression modules. The elements in the module are then subjected to bioinformatics analysis and algorithm performance analysis. The results show that the PD-JDSNMF algorithm can robustly select biomarkers associated with HF. Finally, we built a heart failure diagnostic model based on multiple classifiers and using the Top 13 genes in the significant module, the AUC of the internal test set was up to 0.8714, and the AUC of the external validation set was up to 0.8329, which further confirmed the effectiveness of the PD-JDSNMF algorithm.