This was a single-center prospective cohort study conducted over a one-month follow-up period at the Cardiovascular Center, Hue Central Hospital, from 01/10/2024 to 01/12/2024. The study was approved by Duy Tan University (approval number: 4792/QĐ-ĐHDT) and adhered fully to the principles of the Declaration of Helsinki (2013 version). The reporting followed the STROBE guidelines to ensure the quality of observational research. As this was a pilot study, no formal sample size calculation was required (16, 17).

Convenience sampling was employed, and an initial pool of 240 patients was recruited. Informed consent was obtained from all participants prior to their enrollment in the study. After the exclusion process, 80 participants were included in the study. These participants were then categorized into three groups: the AHF group, the chronic heart failure (CHF) group, and the control group. Details of the sampling process are illustrated in Figure 1.

The AHF and CHF groups were diagnosed on the basis of the 2021 European Society of Cardiology guidelines for heart failure (2), whereas the control group consisted of hospitalized individuals without heart failure. The exclusion criteria included patients with malignant diseases (e.g., ovarian, endometrial, gastric, liver, lung, or breast cancers), benign gynecological conditions (e.g., uterine fibroids, endometriosis, or pelvic inflammatory disease), pregnant women, and patients with fluid overload due to noncardiac conditions (e.g., tuberculous peritonitis, cirrhosis with ascites, or pleural effusion).

Patients and/or the public were not involved in the design, conduct, reporting, or dissemination plans of this study.

Data were collected comprehensively via structured research forms. Details of the questionnaire are provided in Supplementary Document S1. The primary outcome of the study was all-cause in-hospital mortality, with a follow-up period of up to one month from admission.

Serum CA125 levels were measured via an electrochemiluminescence immunoassay based on the sandwich principle. The Elecsys CA 125 II was performed on the Cobas Pro 2 system (Roche, Germany). Because CA125 has a relatively long half-life of several days, it is recommended to measure it not only at admission but also at least seven days later to obtain information about the response to therapy (13). Therefore, blood samples for CA125 quantification were collected at the time of admission before any therapeutic intervention and after 7 days of treatment. The normal reference range for CA125 in this system is 0–35 U/ml.

All the statistical tests were two-sided, with the significance level set at p < 0.05. The statistical analysis followed the SAMPL guidelines to ensure clarity and accuracy in reporting (18).

Statistical analyses were performed via R (http://www.R-project.org, The R Foundation), SPSS Version 26 (IBM, New York, United States), and GraphPad Prism Version 10 (GraphPad Software, Boston, United States).

Data normality was assessed via the Kolmogorov‒Smirnov test. Normally distributed continuous variables are presented as the means ± standard deviations, whereas nonnormally distributed variables are expressed as medians with interquartile ranges. Categorical variables are summarized as frequencies and percentages.

Fisher's exact test was used to assess differences in categorical variables between groups. For continuous variables, either the unpaired t test or the Mann‒Whitney U test was used, depending on the data distribution. One-way ANOVA with multiple comparisons was used to compare normally distributed variables across groups, whereas the Kruskal‒Wallis test was used for nonparametric comparisons of three or more groups. Missing data were excluded from all analyses.

Spearman's correlation coefficient was applied for nonnormally distributed variables, and Pearson's correlation coefficient was used for normally distributed variables to determine correlations between continuous variables.

In-hospital mortality rates were estimated using Kaplan–Meier analysis and compared between groups using the log-rank test. Simple hazard ratios (HRs) were calculated using the Mantel–Haenszel method and were displayed alongside the Kaplan–Meier survival curves. To evaluate the association between CA125 levels and in-hospital mortality, a Cox proportional hazards regression model was employed, treating CA125 levels as a continuous variable. HRs with corresponding 95% confidence intervals (CIs) were estimated. To account for potential nonlinear associations, cubic spline functions and smoothing curve fitting were applied. The results were visualized as HR curves with shaded 95% CIs based on model predictions.

The demographic data revealed no significant difference in age between the AHF and CHF groups (P = 0.245). However, BMI was significantly lower in the AHF group than in the CHF group (P = 0.015). With respect to medical history, no significant differences were observed between the two groups in terms of smoking status, sedentary lifestyle, hypertension, diabetes, dyslipidemia, atrial fibrillation, or coronary artery disease (P > 0.05).

In terms of current medications, the use of beta-blockers was significantly lower in the AHF group than in the CHF group (P = 0.035). Conversely, no significant differences were noted for the use of angiotensin receptor-neprilysin inhibitors/angiotensin-converting enzyme inhibitors/angiotensin receptor blockers, mineralocorticoid receptor antagonists, or sodium‒glucose cotransporter-2 inhibitors (P > 0.05).

When leading symptoms were evaluated, peripheral edema, jugular vein distention, dyspnea, and hepatomegaly were significantly more common in the AHF group than in the CHF group (P < 0.001). Moreover, vital signs at admission revealed a significantly greater heart rate in the AHF group than in the CHF group (P = 0.033), whereas no significant differences were observed in systolic or diastolic blood pressure (P > 0.05).

Echocardiographic findings revealed a trend toward larger left atrial diameters in the AHF group than in the CHF group; however, this difference did not reach statistical significance (P = 0.051). Similarly, no significant differences were observed in the left ventricular ejection fraction (P = 0.241).

Laboratory assessments revealed significantly higher N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels in the AHF group than in the CHF group (P = 0.003). In contrast, other parameters, including creatinine, urea, cholesterol, triglycerides, sodium, and potassium, did not differ significantly between the groups (P > 0.05). Further details are presented in Table 1 and Supplementary Table S1.

Figure 2A shows that CA125 levels were markedly elevated in the AHF group (127.5 U/ml) compared with both the CHF group (15.8 U/ml, P < 0.001) and the control group (10.4 U/ml, P < 0.001). Additionally, CA125 levels in the CHF group were significantly higher than those in the control group (P = 0.047). Figure 2B reveals no significant difference in CA125 levels between patients with EFs ≤ 40% and those with EFs > 40% (P = 0.68). Furthermore, the dynamic changes in CA125 and NT-proBNP levels after 7 days of treatment in the AHF group are illustrated in Supplementary Figure S1.

When analyzing the correlations between CA125 levels and other laboratory parameters in heart failure patients, including both AHF and CHF, CA125 demonstrated a moderate positive correlation with NT-proBNP (r = 0.59), whereas its correlation with creatinine was weak (r = 0.22), indicating a minimal association. In contrast, NT-proBNP showed a stronger correlation with creatinine (r = 0.48), highlighting a more pronounced relationship than CA125. Further details are presented in Figure 3.

Figure 4 shows a significant difference in survival probability between patients with increased and decreased CA125 levels after 7 days of treatment (HR: 37.50, P = 0.022). Patients with increased CA125 levels have a lower probability of survival than those with decreased CA125 levels. Conversely, for NT-proBNP, no significant difference in survival probability was observed between patients with increased and decreased levels after 7 days of treatment (HR: 4.27, P = 0.342). Furthermore, Figure 5 illustrates the relationship between CA125 levels and in-hospital mortality risk. The HR increases exponentially with increasing CA125 levels.

Our study revealed that patients with AHF had significantly higher CA125 levels at admission than those with CHF and the control group. This finding aligns with the study by Nassiba Menghoum et al., which reported elevated CA125 levels in patients with heart failure with preserved ejection fraction compared with controls (22). Similarly, Wussler et al. reported that patients who experienced acute dyspnea caused by AHF had higher CA125 levels than did those with dyspnea from other causes (23). However, studies that directly compare CA125 levels between AHF patients and CHF patients are lacking. Despite being a small-scale study, the pronounced difference in our findings underscore the potential role of CA125 in AHF and provide a strong rationale for conducting larger studies in the future.

Our study also demonstrated promising results regarding the prognostic value of CA125. Patients with increased CA125 levels after 7 days of treatment had a significantly greater risk of in-hospital mortality than those with decreased CA125 levels (p = 0.022). Additionally, a positive correlation was observed between admission CA125 levels and in-hospital mortality rates. Remarkably, despite the small sample size, the findings were encouraging.

In comparison, while increased NT-proBNP levels after 7 days of treatment were associated with higher in-hospital mortality rates, this difference was not statistically significant. These results highlight the potential of CA125 as a prognostic marker and suggest the need for larger studies to strengthen the evidence.

Other studies have also supported the role of CA125 in predicting outcomes in AHF patients. For example, a retrospective analysis from the BIOSTAT-CHF trial revealed a positive correlation between CA125 levels and 1-year mortality and heart failure-related rehospitalization rates (24). Similar findings have been reported in other studies (25–28). Notably, a study by Gonzalo et al. demonstrated that CA125, rather than NT-proBNP, was a useful marker for identifying AHF patients with congestive intrarenal venous flow patterns (26). In addition, the study by Meritxell Soler et al. in patients with AHF and severe tricuspid regurgitation showed that CA125 outperformed NT-proBNP in long-term prognostication (29). These findings emphasize the role of CA125 in long-term prognostication and rehospitalization risk assessment.

In contrast, our study focused on short-term in-hospital outcomes to help clinicians develop appropriate treatment strategies aimed at reducing CA125 levels. However, long-term studies are needed to further validate and expand upon the findings of our research.

The promising results of our study suggest that larger, multicenter studies should be considered to draw more robust and generalizable conclusions. As a pilot study, this research provides a foundation for focusing time, resources, and funding on more extensive, long-term investigations to further explore the role of CA125 in heart failure prognosis.