Plasma BNP and Diuretics As Predictors of Cardiovascular Events in Patients with MI and IGT

Abstract

Background: Impaired glucose tolerance (IGT) promotes cardiovascular events in myocardial infarction (MI) patients. However, our randomized trial (the ABC study) showed that alpha-glucosidase inhibitors do not prevent cardiovascular events in patients with MI and IGT. The aim of the present study is to identify potential clinical factors for cardiovascular and coronary events in patients with a MI and IGT. Methods: Using the limitless-arity multiple testing procedure, an artificial intelligence (AI)-based data mining method, we analyzed 385,391 combinations of fewer than four clinical parameters. Findings: We identified 380 combinations predicting the occurrence of (1) all-cause hospitalisation, (2) hospitalisation due to worsening of heart failure (HF), (3) hospitalisation due to non-fatal MI, and (4) hospitalisation due to percutaneous coronary intervention (PCI) and coronary artery bypass grafting (CABG) for stable angina among 385,391 combinations in 853 patients. Among these, either plasma BNP levels >200 pg/dl or diuretic use exclusively predicted (1) all-cause hospitalisation, (2) hospitalisation due to worsening of HF, and (3) hospitalisation due to a non-fatal MI, with plasma BNP levels >200 pg/dl being the sole predictor of hospitalisation due to PCI and CABG. Importantly, each finding was verified by independently drawn Kaplan-Meier curves, revealing the unexpected role of plasma BNP levels in the progression of coronary stenosis determined as the necessity of PCI and CABG for stable angina. Interpretation: In patients with MI and IGT, plasma BNP levels >200 pg/dl predicted the occurrence of coronary stenosis, recurrent MI and worsening of HF, whereas diuretic use did not predict the progression of coronary stenosis but non-fatal MI and worsening of HF We also demonstrated the use of AI in cardiovascular medicine. Funding: Grants-in-Aid from the Ministry of Health, Labour and Welfare of Japan; and Grants-in-Aid from the Japan Agency for Medical Research and Development. Declaration of Interest: Relationships to industry do not exist for K.S., H.F., T.H., S.I., A.I., M.A., J.K., H.A., O.T. and T.W. M.K. reports grants and personal fees from Takeda, during the conduct of the study; grants from Japanese government, grants from Japan Heart Foundation, grants from Japan Cardiovascular Research Foundation, grants and personal fees from Asteras, grants and personal fees from Sanofi, personal fees from Daiichi-sankyo, grants and personal fees from Pfizer, grants and personal fees from Ono, personal fees from Bayer, grants and personal fees from Novartis, personal fees from Bheringer, grants and personal fees from Tanabe-mitubishi, personal fees from Kowa, grants and personal fees from Kyowa-hakko-kirin, personal fees from Dainihon-sumitomo, personal fees from Sawai, personal fees from MSD, grants and personal fees from Abott, grants and personal fees from Otsuka, grants from Calpis, grants from Nihon Kohden, personal fees from Shionogi, personal fees from Astrazeneca, personal fees from Asahikasei Med., personal fees from Novo nordisk, personal fees from Fuji-film RI, personal fees from Japan Medical Data, outside the submitted work. Ethical Approval: Approval had been obtained from the institutional review boards and ethics committees of all involved hospitals. All patients provided written informed consent. The study was performed in accordance with the principles of the Declaration of Helsinki and the Japanese ethical guidelines for clinical research.