Background: Chest Pain and dyspnea are two of the most common causes of presentation to emergency departments. Since particular cardiac diseases like myocardial infarction or valvular heart disease are associated with a high mortality, patients must be assessed quickly. To optimize this approach, chest pain units (CPUs) have been certified by the German Society of Cardiology since 2008. Echocardiography can safely rule out many cardiac causes of chest pain/dyspnea such as structural heart disease, heart failure or severe cardiac hypertrophy. However, this requires an experienced examination and sufficient resources, which may be limited, especially when the number of patients in the CPU is high.

Objectives: With the present study we aimed to define cut-offs for high-sensitive cardiac troponin I (hs-cTNI) at 0/2hours and brain natriuretic peptide (BNP) to identify patients without severe or moderate structural heart diseases, heart failure or severe cardiac hypertrophy.

Methodology and Results: We performed a retrospective analysis of 400 patients presenting from October to November 2021 in our CPU with chest pain and/or dyspnea as a main symptom. Every patient received an echocardiography by an experienced sonographer. Subsequently we correlated cardiac biomarkers with the echocardiographic evaluation. In our calculation, a BNP <= 184pg/ml was determined to be an "optimal" cut-off value to exclude severe valvular heart disease (defined as any severe-grade stenosis or insufficiency ≥3 ) with a probability of 99.3%. In comparison to other cardiac biomarkers, receiver-operating characteristics curve (ROC curve) analysis for association of severe valvular heart disease for BNP was higher (AUC 0,89, 95% CI 0.83-0.95) than for hs-cTNI at 0 hours (AUC 0,73, 95% CI 0.67- 0.80) or after 2 hours (AUC 0.69, 95% CI 0.62-0.77) (Figure 1). Moderate and severe valvular heart disease can be excluded by a BNP <=22pg/ml with a probability of 95,1% (AUC 0.80, 95% CI 0.81-0.91) . A reduced ejection fraction below 45% was not found in patients presenting with a BNP <=22pg/ml with a probability of 98,4% and an AUC of 0.86 (95% CI 0.80-0.90). Cardiac hypertrophy defined as interventricular septum –thickness (IVSD) > 16mm did not correlate well with either BNP (AUC 0.65, 95% CI 0.55-0.75) or hsTnI (AUC 0.70, 95% CI 0.60-0.79).

Conclusion: The numbers of patients presenting with acute chest pain or dyspnea in emergency rooms are increasing and in such patients guidelines recommend to perform echocardiography as an early assessment. However, experienced personnel and time is necessary to perform high-quality echocardiography in all patients. With the present study we demonstrate that the biomarker BNP is more suitable opposed to hs-cTNI to exclude severe valvular heart disease or reduced ejection fraction in patients presenting at our CPU. The perspective of this study is to securely identify patients in the emergency room at very low risk for structural heart disease or heart failure, in which echocardiographic assessment is dispensable. This approach can help to reduce the length of in-hospital stay, amount of personnel work and costs without affecting patient’s outcomes.