BACKGROUND: Syncope is a very common complaint in primary care and is often very difficult to diagnose. Most previous studies have focused only on high-risk patients and on selected diagnostic tests.

POPULATION STUDIED: All patients were eligible for inclusion who were 18 years or older and who presented to the emergency department (ED) of a large primary and tertiary care teaching hospital with a chief complaint of syncope. Syncope was defined as a sudden transient loss of consciousness with an inability to maintain postural tone, with spontaneous recovery. Patients with a seizure disorder, vertigo, dizziness, coma, or shock were excluded. Of 788 eligible patients, 115 did not complete the standardized evaluation, and 23 refused to participate. The remaining 650 patients ranged in age from 18 to 93 years (mean age = 60 years) and represented both men and women equally.

STUDY DESIGN AND VALIDITY: Patients underwent a standard evaluation including a complete history and physical examination, laboratory evaluation (hematocrit, serum creatine kinase and glucose), electrocardiogram (EKG), testing for orthostatic hypotension, and bilateral carotid massage unless contraindicated. If this approach did not lead to a diagnosis, a second series of tests was conducted: 24-hour Holter monitoring, ambulatory loop monitoring or electrophysiologic studies as guided by an abnormal EKG, or a tilt-table test to identify neurocardiogenic or orthostatic syncope. A committee of 2 internists and a cardiologist reviewed the findings of each case, and explicit and reproducible criteria were used to verify the etiology of the syncope. Some of the diagnoses relied on clinical judgment, since no gold standard reference was available.

OUTCOMES MEASURED: The cause of syncope for each patient based on the sequential evaluation. Follow-up information about mortality and recurrent syncope was obtained at 3 6-month intervals from primary physicians, patients, or their families.

RESULTS: A diagnosis was made in 69% of patients following the initial round of examination. In this group, vasovagal disorders accounted for 53% of the diagnoses, along with orthostatic hypotension (35%), arrhythmia (5%), and other causes (5%). Targeted testing was performed in 67 patients, and the suspected diagnosis was confirmed in an additional 49 patients (8%). Extensive cardiovascular testing of 122 of the remaining 155 patients established a diagnosis in 30 of them through the use of Holter or ambulatory loop monitoring, tilt-table, or electrophysiologic testing. No etiology was found in 92 patients (14%). Overall mortality (9% over 18 months) and sudden death were more common among patients with cardiac causes of syncope compared with other causes of syncope.

Sequential evaluation of patients with syncope is useful in identifying causes for most cases in an unselected patient population. The initial work-up includes a complete history and physical examination, laboratory evaluation, EKG, testing for orthostatic hypotension, and bilateral carotid massage unless contraindicated. These diagnostic maneuvers will lead to diagnosis in 69% of patients and suggests a cause that can be confirmed by selective diagnostic testing in an additional 8%. Undiagnosed patients require further cardiovascular evaluation. In the absence of abnormal EKG findings, other extensive cardiovascular testing has little yield. No diagnosis may be uncovered for 14% of patients. Finally, patients evaluated in the ED most likely represent a different subsample of patients suffering from syncope than those seen in the office; therefore, the diagnostic yields may be different.