Header

UZH-Logo

Maintenance Infos

Prospective validation of prognostic and diagnostic syncope scores in the emergency department


Abstract

BACKGROUND Various scores have been derived for the assessment of syncope patients in the emergency department (ED) but stay inconsistently validated. We aim to compare their performance to the one of a common, easy-to-use CHADS score.
METHODS We prospectively enrolled patients ≥ 40 years old presenting with syncope to the ED in a multicenter study. Early clinical judgment (ECJ) of the treating ED-physician regarding the probability of cardiac syncope was quantified. Two independent physicians adjudicated the final diagnosis after 1-year follow-up. Major cardiovascular events (MACE) and death were recorded during 2 years of follow-up. Nine scores were compared by their area under the receiver-operator characteristics curve (AUC) for death, MACE or the diagnosis of cardiac syncope.
RESULTS 1490 patients were available for score validation. The CHADS-score presented a higher or equally high accuracy for death in the long- and short-term follow-up than other syncope-specific risk scores. This score also performed well for the prediction of MACE in the long- and short-term evaluation and stratified patients with accuracy comparative to OESIL, one of the best performing syncope-specific risk score. All scores performed poorly for diagnosing cardiac syncope when compared to the ECJ.
CONCLUSIONS The CHADS-score performed comparably to more complicated syncope-specific risk scores in the prediction of death and MACE in ED syncope patients. While better tools incorporating biochemical and electrocardiographic markers are needed, this study suggests that the CHADS-score is currently a good option to stratify risk in syncope patients in the ED.
TRIAL REGISTRATION NCT01548352.

Abstract

BACKGROUND Various scores have been derived for the assessment of syncope patients in the emergency department (ED) but stay inconsistently validated. We aim to compare their performance to the one of a common, easy-to-use CHADS score.
METHODS We prospectively enrolled patients ≥ 40 years old presenting with syncope to the ED in a multicenter study. Early clinical judgment (ECJ) of the treating ED-physician regarding the probability of cardiac syncope was quantified. Two independent physicians adjudicated the final diagnosis after 1-year follow-up. Major cardiovascular events (MACE) and death were recorded during 2 years of follow-up. Nine scores were compared by their area under the receiver-operator characteristics curve (AUC) for death, MACE or the diagnosis of cardiac syncope.
RESULTS 1490 patients were available for score validation. The CHADS-score presented a higher or equally high accuracy for death in the long- and short-term follow-up than other syncope-specific risk scores. This score also performed well for the prediction of MACE in the long- and short-term evaluation and stratified patients with accuracy comparative to OESIL, one of the best performing syncope-specific risk score. All scores performed poorly for diagnosing cardiac syncope when compared to the ECJ.
CONCLUSIONS The CHADS-score performed comparably to more complicated syncope-specific risk scores in the prediction of death and MACE in ED syncope patients. While better tools incorporating biochemical and electrocardiographic markers are needed, this study suggests that the CHADS-score is currently a good option to stratify risk in syncope patients in the ED.
TRIAL REGISTRATION NCT01548352.

Statistics

Citations

Dimensions.ai Metrics

Altmetrics

Additional indexing

Item Type:Journal Article, refereed, original work
Communities & Collections:04 Faculty of Medicine > University Hospital Zurich > Institute of Clinical Chemistry
Dewey Decimal Classification:610 Medicine & health
540 Chemistry
Language:English
Date:15 October 2018
Deposited On:28 Feb 2019 11:51
Last Modified:28 Feb 2019 12:03
Publisher:Elsevier
ISSN:0167-5273
OA Status:Closed
Publisher DOI:https://doi.org/10.1016/j.ijcard.2018.06.088
PubMed ID:30224031

Download

Full text not available from this repository.
View at publisher

Get full-text in a library