P6570Performance of the early clinical judgement for the diagnosis of syncope on the emergency department

Abstract

Abstract Background Clinical judgement of the emergency department (ED) physician at time of discharge outperforms prognostic risk scores but the early clinical judgment (ECJ) regarding diagnosis has never been assessed. Methods We evaluated the diagnostic accuracy of the ED physicians' ECJ 90 minutes after admission of patients >40 years presenting with syncope to the ED in a prospective diagnostic multicenter study. Cardiac syncope, as adjudicated by two physicians based on information available including 1-year follow-up, was the diagnostic endpoint. Death and MACE were the prognostic endpoints. Lasso-regression was used to identify variables contributing most to the ECJ or to the diagnosis of cardiac syncope. Syncope-specific diagnostic and prognostic scores, high-sensitivity cardiac troponin I (hs-cTnI) and B-type natriuretic peptide (BNP) were used for comparison. Results Cardiac syncope was adjudicated in 252/1494 patients (15.2%). The diagnostic accuracy of the ECJ for cardiac syncope, as quantified by the Area Under the Curve (AUC), was 0.87 (95% CI 0.84–0.89) and was superior to the one of biomarkers and diagnostic score, constant between all centers, but poorly calibrated. 16 variables available very-early on the ED achieved a comparable performance (AUC 0.84 (95% CI 0.82–0.87), p=0.136). The extrapolated prognostic accuracy of the ECJ was moderate for MACE (AUC 0.73–0.8) but poor for death (0.58–0.63) over two years follow-up. Figure 1 Conclusion The ECJ performs well for the diagnosis of cardiac syncope but a similar accuracy can be obtained using structured variables obtained very-early in the diagnostic process. Acknowledgement/Funding Swiss National Science Foundation, Swiss Heart Foundation, Cardiovascular Research Foundation, Basel (Switzerland), University Basel (Switzerland)