Relative GSR mRNA expression was significantly lower and miR-144 expression was significantly higher in patients with atrial fibrillation compared to controls, with both markers remaining associated with AF in multivariable logistic regression, though neither yet supports routine clinical use as diagnostic or risk-stratification biomarkers.
Key Findings
Results
GSR mRNA expression was significantly lower in patients with atrial fibrillation compared to controls.
Median relative GSR mRNA expression was 0.1539 in the AF group versus 0.2058 in controls (p < 0.001).
The study included 94 patients with AF and 95 without AF, all hospitalized adults.
GSR mRNA was assessed by quantitative real-time PCR in whole blood samples.
In multivariable logistic regression adjusted for clinical confounders, lower GSR expression remained independently associated with AF (OR = 0.343, p < 0.001).
Bootstrap internal validation supported statistical stability of the GSR association (optimism-corrected AUC = 0.709, bootstrapped 95% CI: 0.622–0.785).
Results
Circulating miR-144 expression was significantly higher in patients with atrial fibrillation compared to controls.
Median relative miR-144 expression was 0.3636 in the AF group versus 0.2550 in controls (p = 0.0183).
miR-144 expression was evaluated in serum samples.
In multivariable logistic regression adjusted for clinical confounders, higher miR-144 expression remained independently associated with AF (OR = 1.336, p = 0.048).
Bootstrap internal validation showed a more modest and less stable association for miR-144 (optimism-corrected AUC = 0.598, bootstrapped 95% CI: 0.519–0.684).
The miR-144 signal is described as exploratory and does not yet support routine clinical use as a diagnostic or risk-stratification biomarker in AF.
Results
Both GSR mRNA and miR-144 showed exploratory correlations with selected clinical and biochemical variables.
Correlations were evaluated using Spearman's rank correlation coefficient.
Clinical, anthropometric, biochemical, and echocardiographic data were collected for all participants.
The paper describes these correlations as hypothesis-generating and states they warrant further investigation.
The cross-sectional design means all reported relationships should be interpreted as statistical associations rather than evidence of causality.
Methods
The study design was a case-control study of 189 hospitalized adults with model stability assessed using bootstrap internal validation.
The study included 189 hospitalized adults: 95 without AF and 94 with AF.
Group differences were assessed using nonparametric tests.
Multivariable logistic regression was used to identify variables independently associated with AF.
Bootstrap internal validation was used to assess model stability.
The cross-sectional design limits causal inference; all associations are described as statistical rather than causal.
Background
Atrial fibrillation is associated with oxidative stress and inflammation, providing the rationale for examining glutathione reductase and miR-144 as redox-related markers.
Glutathione reductase (GR) and microRNA-144 (miR-144) were proposed to participate in redox-related pathways.
The study aimed to evaluate relative GSR mRNA and miR-144 expression in peripheral blood samples from patients with and without AF.
The study also aimed to explore associations of these markers with selected clinical characteristics.
Neither marker currently supports routine clinical use as a diagnostic or risk-stratification biomarker in AF based on these findings.
What This Means
This research suggests that two molecules measurable in blood — an enzyme called glutathione reductase (GR) and a small RNA molecule called miR-144 — have different levels in people with atrial fibrillation (an irregular heart rhythm) compared to people without it. Specifically, people with atrial fibrillation had lower levels of the gene that makes glutathione reductase and higher levels of miR-144 in their blood. Both differences remained statistically meaningful even after accounting for other health factors, suggesting they are not simply explained by other characteristics of the patients.
The study involved 189 hospitalized adults (94 with atrial fibrillation and 95 without) and used standard laboratory and statistical techniques, including a bootstrap validation method to check how reliable the findings are. The glutathione reductase finding was more statistically stable than the miR-144 finding, but both are described by the authors as exploratory. Both molecules are involved in how the body manages oxidative stress — a type of cellular damage linked to many heart conditions — which provides a biological reason why they might be connected to atrial fibrillation.
This research suggests that oxidative stress pathways involving these two molecules may play a role in atrial fibrillation, but the authors are clear that this study cannot prove cause and effect because it only measured patients at one point in time. The findings are intended to generate hypotheses for future research rather than to immediately change clinical practice, and the authors explicitly state that neither marker is ready for use as a routine diagnostic or risk-assessment tool for atrial fibrillation.
Różycka-Kosmalska M, Grabarczyk M, Śliwińska A, Kozłowska M, Kosmalski M, Wranicz J, et al.. (2026). Peripheral Expression Profiles of Glutathione Reductase and miR-144 in Patients with Atrial Fibrillation.. Medical sciences (Basel, Switzerland). https://doi.org/10.3390/medsci14040456