Derivation of the San Francisco Syncope Rule to predict patients with short-term serious outcomes

The causes of syncope are usually benign but are occasionally associated with significant morbidity and mortality. We derive a decision rule that would predict patients at risk for short-term serious outcomes and help guide admission decisions. This prospective cohort study was conducted at a university teaching hospital and used emergency department (ED) patients presenting with syncope or near syncope. Physicians prospectively completed a structured data form when evaluating patients with syncope. Serious outcomes (death, myocardial infarction, arrhythmia, pulmonary embolism, stroke, subarachnoid hemorrhage, significant hemorrhage, or any condition causing a return ED visit and hospitalization for a related event) were defined at the start of the study. All patients were followed up to determine whether they had experienced a serious outcome within 7 days of their ED visit. Univariate analysis was performed with χ 2 and nonparametric techniques on all predictor variables. κ Analysis was performed on variables requiring interpretation. Variables with κ more than 0.5 and a P value less than .1 were analyzed with recursive partitioning techniques to develop a rule that would maximize the determination of serious outcomes. There were 684 visits for syncope, and 79 of these visits resulted in patients' experiencing serious outcomes. Of the 50 predictor variables considered, 26 were associated with a serious outcome on univariate analysis. A rule that considers patients with an abnormal ECG, a complaint of shortness of breath, hematocrit less than 30%, systolic blood pressure less than 90 mm Hg, or a history of congestive heart failure has 96% (95% confidence interval [CI] 92% to 100%) sensitivity and 62% (95% CI 58% to 66%) specificity. If applied to this cohort, the rule has the potential to decrease the admission rate by 10%. The San Francisco Syncope Rule derived in this cohort of patients appears to be sensitive for identifying patients at risk for short-term serious outcomes. If prospectively validated, it may offer a tool to aid physician decisionmaking.