https://doi.org/10.1007/s00392-025-02625-4
1Universitätsklinikum Heidelberg Klinik für Innere Med. III, Kardiologie, Angiologie u. Pneumologie Heidelberg, Deutschland; 2Katholisches Marienkrankenhaus gGmbH Kardiovaskuläre MRT Hamburg, Deutschland; 3Universität Leipzig Medizinische Fakultät Institut für Medizinische Informatik, Statistik und Epidemiologie (IMISE) Leipzig, Deutschland
Background:
Reduced left ventricular ejection fraction (LV-EF) and left ventricular hypertrophy (LVH) are associated with high mortality and morbidity. Early and accurate diagnosis is essential, but non-invasive, cost-effective diagnostic tools are still needed for clinical application. This study evaluated the diagnostic performance of a modified vectorcardiography algorithm, using Cardisiography (CSG), a technique which records cardiac electrical activity in three dimensions, for detecting reduced LV-EF and LVH.
Methods:
This prospective, single-center, case-control study included patients with reduced LV-EF (< 40%), and LVH, defined as indexed LV mass > 55 g/m², compared with controls without cardiac pathology. Cardiac magnetic resonance imaging was performed as part of routine clinical care and used as the reference standard for measuring LV-EF and LV mass. Patients were enrolled consecutively. The CSG analyzed 583 vectorcardiography parameters per heartbeat to classify cardiac status, consisting of 533 regular parameters and 50 Frenet-Serret parameters. Parameters were selected based on a low p-value from the Mann-Whitney-Wilcoxon test, indicating statistical significance. Bayes' theorem was used to update prior probabilities and to derive posterior probabilities for reduced LV-EF and LVH when compared with controls.
Results:
A total of 280 patients were included in the analysis. The group with reduced LV-EF (n=40) had a mean age of 56 ± 16 years and was predominantly male (78%), with a median LV-EF of 31.5% (IQR: 23.4-36.3%). The LVH group (n=209) had a mean age of 60 ± 16 years (87% male) with an indexed LVM of 67g/m² (IQR: 61-77 g/m²). Controls (n=31) had a mean age of 50 ± 16 years (61% male), normal LV-EF of 62 ± 5.6% and indexed LV mass of 40g/m² (IQR: 35-47g/m²).
A CSG parameter with p-value <0.1% after conservative Bonferroni adjustment and AUC of 87.5% was identified among CSG parameters and demonstrated a sensitivity of 80.0% and specificity of 85.7% for detecting reduced LV-EF (accuracy 82.9%). For detecting LVH, it achieved a sensitivity of 74.5% and specificity of 68.6% (accuracy 73.6%) using 3 CSG parameters. In cases with patients revealing both reduced LV-EF and LVH (n=64), the CSG demonstrated a sensitivity of 79.3% and a specificity of 85.7% (accuracy 82.8%).
Conclusion:
The modified vectorcardiography algorithms showed diagnostic value for detecting reduced LV-EF and LVH. CSG could serve as a fast, non-invasive, and cost-effective method to assist clinicians in identifying significant cardiac conditions and to guide further diagnostic steps.