One-year prognosis after syncope and the failure of the ROSE decision instrument to predict one-year adverse events.
Reed MJ., Henderson SS., Newby DE., Gray AJ.
STUDY OBJECTIVE: We investigate the incidence of adverse events at 1 year in the cohort of emergency department (ED) syncope patients enrolled in the original Risk Stratification of Syncope in the Emergency Department (ROSE) study, the time to adverse event, and the test performance of the ROSE decision instrument to detect events at 1 year. METHODS: This report details the 1-year follow-up of adult syncope patients presenting to the ED who were enrolled into the ROSE study, a single-center, prospective, observational cohort study. The primary endpoint was the combination of serious outcome and all-cause death at 1 year. Serious outcome encompassed acute myocardial infarction, life-threatening arrhythmia, need for pacemaker/implantable defibrillator, pulmonary embolus, cerebrovascular accident, intracranial or subarachnoid hemorrhage, and interventional procedure or hemorrhage requiring blood transfusion. Secondary endpoints were all-cause death and cardiovascular serious outcome at 1 year. RESULTS: One thousand forty-three patients were available for analysis; 162 patients (15.5%) had a primary outcome. Twenty-eight (17%) of these were within 24 hours, 56 (35%) were within 1 week, and 78 (48%) were within 1 month. The remaining 84 (52%) outcomes occurred between months 2 and 12. At 1 year, 71 (6.8%) patients had died and 76 (7.3%) had a cardiovascular serious outcome. The sensitivity and specificity of the ROSE decision instrument for 1-year serious outcome and all-cause death were, respectively, 71.6% (95% confidence interval [CI] 63.9% to 78.3%) and 71.1% (95% CI 67.9% to 74.0%); for 1-year all-cause death, 76.1% (95% CI 64.2% to 85.1%) and 67.4% (95% CI 64.3% to 70.3%); and for 1-year cardiovascular serious outcome, 75.0% (95% CI 63.5% to 83.9%) and 67.5% (95% CI 64.5% to 70.5%). CONCLUSION: The proportion of patients with serious outcome and all-cause death 1 year after syncope is similar to that reported in recent international syncope studies. We have defined the proportion of patients with cardiovascular serious outcome at 1 year and have shown that most events occurred in the first month, with decreased frequency of events observed after that time especially marked for cardiovascular serious outcome. More than 50% of the outcomes observed, however, occurred after the first month. The ROSE decision instrument does not perform well at predicting 1-year outcome of ED syncope patients.