Abstract
Background
The objective was to explore the value of serum carbohydrate antigen 125 (CA125) combined with N-terminal pro B-type natriuretic peptide (NT-proBNP) in predicting the clinical prognosis of patients with acute heart failure (AHF).
Methods
We prospectively observed 213 patients with AHF. CA125 (U/ml) and NT-proBNP (pg/ml) were dichotomised based on ROC curve analysised prognostic cutpoints, and a variable with four groups was formed (CA125 and NT-proBNP): C1 = CA125 < 47.6 and NT-proBNP <3790 (n = 100); C2 = CA125 < 47.6 and NT-proBNP ≥3790 (n = 29); C3 = CA125 ≥ 47.6 and NT-proBNP < 3790 (n = 26); C4 = CA125 ≥ 47.6 and NT-proBNP ≥3790 (n = 58). Kaplan–Meier curve was drawn and multivariate COX regression analysis was performed to analyse the prognostic efficacy of CA125 combined with NT-ProBNP in patients with AHF.
Results
The levels of CA125 and NT-proBNP in death group were obviously higher than those in non-death group [56.20 (45.70, 78.00) vs 31.10 (19.48, 47.68), p < 0.001; 5619.00 (2924.00, 10066.00) vs 2203.00 (1460.50, 5070.25), p < 0.001]. The ROC curve showed that the best cut-off values of CA125 and NT-proBNP for predicting the prognosis of AHF were 47.6 and 3790, respectively. Multivariate COX regression analysis showed that CA125 ≥ 47.6 and NT-proBNP ≥ 3790 were independent predictors of 1-year all-cause death in patients with AHF (HR = 3.05, 95%CI: 1.50–6.20, p = 0.002) and (HR = 2.34, 95%CI: 1.19–4.61, p = 0.014). At 12 months, 55 deaths (25.8%) were identified. The cumulative rate of mortality was highest for patients in C4 (56.9%), intermediate for C2 and C3 (24.1% and 34.6%, respectively), and lowest for C1 (6.0%), and p-value for trend <0.05. After adjusting for established clinical risk factors, compared with C1: C2 (HR = 4.58, 95%CI: 1.53–13.77, p = 0.007), C3 (HR = 5.24, 95%CI: 1.85–14.82, p = 0.002), C4 (HR = 7.75, 95%CI: 3.09–19.45, p < 0.001).
Conclusion
Elevated CA125 is an independent predictor of poor prognosis in patients with AHF, and combined with NT-proBNP can improve the efficiency of risk identification.
Acknowledgments
The authors thank the team members and clinical laboratory technicians for their hard work. The authors are also grateful to all participants who voluntarily took part in our studies.
Disclosure statement
No potential conflict of interest for all authors.