Abstract Body (Do not enter title and authors here): Introduction
Cardiac Time Intervals (CTIs), derived from mechanical events of the cardiac cycle, offer valuable insights into myocardial performance and hemodynamics. With advancements in non-invasive technologies such as HEMOTAG, it is now possible to longitudinally monitor CTIs in real-world settings within minutes. However, the clinical implications of CTI variability in high-risk populations, such as diabetic patients, remain underexplored.
Research Questions
To identify distinct CTI patterns using unsupervised clustering and evaluate their associations with biomarkers (NT-proBNP, N-terminal pro B-type natriuretic peptide), comorbidities, and medication profiles in a diabetic population. Our hypothesis was that diabetes patients with abnormal CTI patterns would demonstrate different clinical profiles and medication regimens compared to those with normal patterns, potentially informing targeted therapeutic approaches.
Methods
Daily multi-channel recordings from the HEMOTAG device and electronic health records were analyzed for 56 diabetic patients. CTI features such as mitral valve closing (MC), aortic valve opening (AO), aortic valve closing (AC), mitral valve opening (MO), isovolumic contraction time (IVCT), and derived variability metrics, were extracted. K-means clustering identified patient groups based on CTI patterns. Associations with NT-proBNP, comorbidities, and medication use/dosages were assessed using appropriate statistical tests.
Results
Two CTI-based clusters were identified: Stable and Abnormal. Patients in the Abnormal group showed higher NT-proBNP levels (p = 0.0139), also higher mean and variability in CTIs, during 30-day daily home management, suggesting increased cardiac stress. Medication analysis revealed that drugs such as lisinopril (p = 0.0012), losartan (p = 0.0268), and atorvastatin (p = 0.0061) differed both in frequency and dosage between groups, implicating potential pharmacologic influences on CTI dynamics. Notably, a higher proportion of patients in the Abnormal CTI group had undergone angioplasty. This trend may suggest a potential association between abnormal CTI patterns and underlying coronary pathology, warranting further investigation in larger cohorts.
Conclusions
Abnormal CTI patterns are associated with elevated cardiac biomarkers and distinct medication profiles. CTI analysis may offer a non-invasive, physiologically grounded tool for risk stratification and personalized cardiovascular therapy in diabetic patients.