NT-proBNP in chest pain: elevated values in patients with atrial fibrillation

https://doi.org/10.1007/s00392-025-02625-4

Fabian Lang (Regensburg)1, J. Hupf (Regensburg)1, A. Luchner (Regensburg)2, L. S. Maier (Regensburg)1, C. G. Jungbauer (Regensburg)1

1Universitätsklinikum Regensburg Klinik und Poliklinik für Innere Med. II, Kardiologie Regensburg, Deutschland; 2Krankenhaus Barmherzige Brüder Regensburg Klinik für Kardiologie Regensburg, Deutschland

 

Aims
Acute decompensated heart failure is a prevalent condition leading to hospital admission. A common comorbidity among these patients is atrial fibrillation (AF). This study aimed to identify effects of AF and congestive heart failure (CHF) on NT-proBNP in a real world chest pain cohort.

Methods
Blood concentrations of NT-proBNP were measured in a cohort of 785 patients presenting with acute chest pain in the emergency department of the University Hospital Regensburg (mean age 60.5 y). Physical examination (orthopnea, peripheral edema and neck vein distention) and the routinely performed ECG were used to detect CHF and AF. A total of 32 patients were identified suffering from AF and CHF. Only CHF  was present in 94 patients contrary to 47 with AF, but no CHF.

Results
Patients suffering from combined AF and CHF were characterized by significantly elevated NT-proBNP concentrations compared to AF as well as CHF only (each p < 0,05). Patients without AF or CHF had significantly lower concentrations compared to AF and/or CHF (each p < 0.05). However, in patients with only AF NT-proBNP was not significantly different to only CHF, with a trend to increased levels for AF (p = n. s.).
Promising AUC values were seen in a ROC analysis detecting patients with AF and CHF  in  the study cohort     (AUC = 0.900). A cutpoint of 940   pg/ml revealed a sensitivity of 97% and a specificity of 81%. Similarly, satisfactory results were seen in a ROC analysis for detection of combined AF and CHF opposite to only one of both (AUC = 0.696). At a cut-off of 1180 pg/ml a sensitivity of 88% and specificity of 45% is given.
In a binary regression analysis, NT-proBNP and preexisting AF were significant independent predictors for patients with AF and CHF (p < 0.05), opposite to age, gender, diabetes, arterial hypertension and creatinine (p = n. s.).

Conclusion
Overall, significantly increased NT-proBNP levels in CHF patients with AF were detected. In conclusion, NT-proBNP might be potentially beneficial for detection of AF in decompensated patients. In addition, this study provides further evidence to adjust the currently established NT-proBNP cut-off values for the diagnosis of heart failure according to the heart rhythm.
Diese Seite teilen