Purpose: To review the literature on diagnostic testing in syncope and provide recommendations for a comprehensive, cost-effective approach to establishing its cause.
Data sources: Studies were identified through a MEDLINE search (1980 to present) and a manual review of bibliographies of identified articles.
Study selection: Papers were eligible if they addressed diagnostic testing in syncope or near syncope and reported results for at least 10 patients.
Data extraction: The usefulness of tests was assessed by calculating diagnostic yield: the number of patients with diagnostically positive test results divided by the number of patients tested or, in the case of monitoring studies, the sum of true-positive and true-negative test results divided by the number of patients tested.
Data synthesis: Despite the absence of a diagnostic gold standard and the paucity of data from randomized trials, several points emerge. First, history, physical examination, and electrocardiography are the core of the syncope workup (combined diagnostic yield, 50%). Second, neurologic testing is rarely helpful unless additional neurologic signs or symptoms are present (diagnostic yield of electroencephalography, computed tomography, and Doppler ultrasonography, 2% to 6%). Third, patients in whom heart disease is known or suspected or those with exertional syncope are at higher risk for adverse outcomes and should have cardiac testing, including echocardiography, stress testing. Holter monitoring, or intracardiac electrophysiologic studies, alone or in combination (diagnostic yields, 5% to 35%). Fourth, syncope in the elderly often results from polypharmacy and abnormal physiologic responses to daily events. Fifth, long-term loop electrocardiography (diagnostic yield, 25% to 35%) and tilt testing (diagnostic yield < or = 60%) are most useful in patients with recurrent syncope in whom heart disease is not suspected. Sixth, psychiatric evaluation can detect mental disorders associated with syncope in up to 25% of cases. Seventh, hospitalization may be indicated for patients at high risk for cardiac syncope (those with an abnormal electrocardiogram, organic heart disease, chest pain, history of arrhythmia, age > 70 years) or with acute neurologic signs.
Conclusions: Many tests for syncope have a low diagnostic yield. A careful history, physical examination, and electrocardiography will provide a diagnosis or determine whether diagnostic testing is necessary in most patients.