RESUMO
Aims: The diagnostic application of artificial intelligence (AI)-based models to detect cardiovascular diseases from electrocardiograms (ECGs) evolves, and promising results were reported. However, external validation is not available for all published algorithms. The aim of this study was to validate an existing algorithm for the detection of left ventricular systolic dysfunction (LVSD) from 12-lead ECGs. Methods and results: Patients with digitalized data pairs of 12-lead ECGs and echocardiography (at intervals of ≤7 days) were retrospectively selected from the Heart Center Leipzig ECG and electronic medical records databases. A previously developed AI-based model was applied to ECGs and calculated probabilities for LVSD. The area under the receiver operating characteristic curve (AUROC) was computed overall and in cohorts stratified for baseline and ECG characteristics. Repeated echocardiography studies recorded ≥3 months after index diagnostics were used for follow-up (FU) analysis. At baseline, 42 291 ECG-echocardiography pairs were analysed, and AUROC for LVSD detection was 0.88. Sensitivity and specificity were 82% and 77% for the optimal LVSD probability cut-off based on Youden's J. AUROCs were lower in ECG subgroups with tachycardia, atrial fibrillation, and wide QRS complexes. In patients without LVSD at baseline and available FU, model-generated high probability for LVSD was associated with a four-fold increased risk of developing LVSD during FU. Conclusion: We provide the external validation of an existing AI-based ECG-analysing model for the detection of LVSD with robust performance metrics. The association of false positive LVSD screenings at baseline with a deterioration of ventricular function during FU deserves a further evaluation in prospective trials.
RESUMO
Introduction: Type 2 diabetes mellitus (T2DM) is a leading cause of chronic kidney disease (CKD) globally. Both conditions substantially worsen patients' prognosis. Current data on German in-hospital CKD cohorts are scarce. The multinational CaReMe study was initiated to evaluate the current epidemiology and healthcare burden of cardiovascular, renal and metabolic diseases. In this substudy, we share real-world data on CKD inpatients stratified for coexisting T2DM derived from a large German hospital network. Methods: This study used administrative data of inpatient cases from 89 Helios hospitals from 01/01/2016 to 28/02/2022. Data were extracted from ICD-10-encoded discharge diagnoses and OPS-encoded procedures. The first case meeting a previously developed CKD definition (defined by ICD-10- and OPS-codes) was considered the index case for a particular patient. Subsequent hospitalizations were analysed for readmission statistics. Patient characteristics and pre-defined endpoints were stratified for T2DM at index case. Results: In total, 48,011 patients with CKD were included in the present analysis (mean age ± standard deviation, 73.8 ± 13.1 years; female, 44%) of whom 47.9% had co-existing T2DM. Patients with T2DM were older (75 ± 10.6 vs 72.7 ± 14.9 years, p < 0.001), but gender distribution was similar to patients without T2DM. The burden of cardiovascular disease was increased in patients with T2DM, and index and follow-up in-hospital mortality rates were higher. Non-T2DM patients were characterised by more advanced CKD at baseline. Patients with T2DM had consistently higher readmission numbers for all events of interest, except for readmissions due to kidney failure/dialysis, which were more common in non-T2DM patients. Conclusion: In this study, we present recent data on hospitalized patients with CKD in Germany. In this CKD cohort, nearly half had T2DM, which substantially affected cardiovascular disease burden, rehospitalization frequency and mortality. Interestingly, non-diabetic patients had more advanced underlying renal disease, which affected renal outcomes.