To investigate the correlation between malnutrition, assessed by the geriatric nutritional risk index (GNRI), and the risk of in-hospital mortality in elderly patients with chronic heart failure (CHF).

A retrospective analysis was conducted of clinical data from 4466 elderly CHF patients admitted to the department of cardiology, the First Medical Center, PLA General Hospital, between February 2008 and December 2019. Based on height, weight, and serum albumin levels on admission, GNRI was calculated, and patients were divided into four groups: normal nutrition (GNRI>98), mild malnutrition (92≤GNRI≤98), moderate malnutrition (82≤GNRI<92), and severe malnutrition (GNRI<82). Data were analyzed using EmpowerStats (version 4.1) and R (version 4.4.2) software. Between-group comparisons were performed using One-way ANOVA, Kruskal-Wallis test, or Chi-square test, as appropriate. Logistic regression analyses were used to examine the correlation between GNRI and hospital mortality. Smooth curve fitting models were employed to explore the nonlinear relationship between GNRI and mortality.

The mean GNRI of 4466 elderly CHF patients was 102.29 ± 10.87. Malnutrition was present in 1504 patients (33.68%), including 718 (16.08%) with mild, 633 (14.17%) with moderate, and 153 (3.43%) with severe malnutrition. The in-hospital mortality rates in the normal-nutrition group, mild malnutrition group, moderate malnutrition group and severe malnutrition group were 1.42%, 2.79%, 5.06% and 12.42%, respectively, and the difference was statistically significant (P<0.001). Multivariate logistic regression analysis identified age, New York Heart Association (NYHA) class ≥ Ⅲ, GNRI, C-reactive protein (CRP) ≥ 0.8 mg/dl, and troponin T (TnT) > 0.2 ng/ml as independent predictors of in-hospital mortality (all P<0.05). A generalized additive model revealed a nonlinear relationship between GNRI and mortality risk (P=0.004). Subgroup analyses showed that the association between malnutrition and in-hospital mortality was statistically significant in patients aged 70–80 years, males, those with NYHA class ≥ Ⅲ, those with CRP < 0.8 mg/dl, and those with coexisting hypertension (all P<0.05).

Malnutrition is common in elderly CHF patients. GNRI is an independent predictor of in-hospital mortality and exhibits a nonlinear relationship with mortality risk, with a greater pronounced predictive value in specific subgroups of elderly CHF patients.

Catheter-based pulmonary vein isolation (PVI) is an effective and well-established intervention for symptomatic paroxysmal atrial fibrillation (PAF). Nevertheless, late recurrences of atrial fibrillation (LRAF) occurring during 3 to 12 months are common, and the underlying mechanisms remain elusive. Circular RNAs (circRNAs) in atrial tissue have been linked to the pathophysiological mechanisms and progression of PAF in a few studies. However, their expression patterns in peripheral blood and regulatory function in LRAF are not clear.

In the present study, the expression profile of circulating circRNAs in three paired nonvalvular PAF patients with or without LRAF was investigated by high-throughput sequencing and validated by quantitative real-time polymerase chain reaction (qRT-PCR). Bioinformatics analyses, including Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis, and circRNA/miRNA regulatory network, were performed to predict the functions and potential regulatory roles of differentially expressed (DE) circRNAs.

A total of 12,834 circRNAs, comprising 5,491 down-regulated and 7,343 up-regulated circRNAs, were found to be DE in blood smaples from the two groups in peripheral blood between LRAF and non-recurrence control individuals. The most enriched GO categories in terms of molecular function, biological process, and cellular component features were catalytic activity, cellular metabolic process, and intracellular part, respectively. The KEGG enrichment study revealed that the most important metabolic process controlled by DE circRNAs is endocytosis. In the circRNA/microRNAs interaction network, four up-regulated circRNAs (hsa_circ_0002665, hsa_circ_0001953, hsa_circ_0003831, and hsa_circ_0040533) and one down-regulated circRNA (hsa_circ_0041103) were predicted to play potential regulatory roles in the pathogenesis of LRAF.

This investigation discovered the expression pattern of circulating circRNAs that is indicative of PAF late recurrence, which may serve as risk markers or therapeutic targets for LRAF after PVI.