Florian A Wenzl, Victor Schweiger, Maria A Smolle, Daniel Zech, Alexander Akhmedov, Peizhi Wang, Tetiana Lapikova-Bryhinska, Yujing Wang, Luca Liberale, Karine Santos, Yulia Ilina, Simon Kraler, Andreas Bergmann, Paul Kaufmann, Florian Kahles, Arnold von Eckardstein, Marco Roffi, Lorenz Räber, Victoria L Cammann, Jelena R Ghadri, Davide Di Vece, Thomas F Lüscher, Christian Templin, InterTAK and SPUM-ACS investigators
TTS is characterized by a distinct biomarker profile reflecting pathophysiological differences from ACS. The non-subjective biomarker-based BioTAK score can support the early identification of TTS, help streamline diagnostic pathways, and might reduce potentially avoidable invasive procedures in patients presenting with suspected ACS.
BACKGROUND AND AIMS: Takotsubo syndrome (TTS) closely mimics acute coronary syndrome (ACS). Early differentiation remains challenging because diagnosis currently relies on medical, psychiatric, and neurological history, repeated cardiac imaging, and exclusion of a culprit epicardial coronary artery lesion, generally with invasive coronary angiography (ICA). The aim of this study was to develop and externally validate a biomarker-based diagnostic score supporting the early differentiation of TTS from ACS before ICA.
METHODS: In this study, 3615 patients with ACS or TTS from overlapping registries were investigated. Circulating levels of established and emerging cardiovascular biomarkers were measured before ICA. A biomarker-based diagnostic score for TTS (BioTAK) was developed in a cohort of 1823 patients (ACS n = 1754; TTS n = 69) using machine learning-guided feature selection and logistic regression and externally validated in an independent cohort of 1792 patients (ACS n = 1715; TTS n = 77).
RESULTS: The final five-item score incorporated biomarkers reflecting myocardial stress (N-terminal pro-B-type natriuretic peptide), vasoconstriction and anxiety regulation (peptidylglycine α-amidating monooxygenase), atherosclerotic plaque instability (soluble lectin-like oxidized low-density lipoprotein receptor-1), and lipid metabolism (low-density lipoprotein cholesterol), along with sex. BioTAK demonstrated excellent discrimination between TTS and ACS in the development cohort [area under the receiver operating characteristic curve (AUC) .97, 95% confidence interval (CI) .95-.98] and maintained high performance on external validation (AUC .93, 95% CI .90-.96), with good calibration. Using predefined rule-in and rule-out thresholds, the score stratified almost 90% of patients to a likely diagnosis prior to ICA.
CONCLUSIONS: TTS is characterized by a distinct biomarker profile reflecting pathophysiological differences from ACS. The non-subjective biomarker-based BioTAK score can support the early identification of TTS, help streamline diagnostic pathways, and might reduce potentially avoidable invasive procedures in patients presenting with suspected ACS.
CLINICAL TRIAL REGISTRATION: NCT01000701, NCT01947621.