p.15 抽出テキスト

August 1, 2017 Circulation. 2017;136:e60–e122. DOI: 10.1161/CIR.0000000000000499 e74

3.2.5. Electrophysiological Study: Recommendations The EPS can identify a substrate for clinical bradyarrhythmia or tachyarrhythmia as a potential cause of syncope after a nondiagnostic initial evaluation. Despite these purported benefits, EPS has a limited role in the evaluation of syncope, especially in patients without known heart disease or with low suspicion of an arrhythmic etiology.117,187,188 The sensitivity and specificity of EPS to assess sinus node dysfunction and AV conduction disease in patients with syncope are variable, depending on patient selection and pretest probability of a bradycardia substrate.189–191 Inducible ventricular tachycardia (VT) in patients with syncope, ischemic heart disease, and a prior history of myocardial infarction is predictive of spontaneous VT and prognosis. The causal relationship between the inducible VT during EPS and syncope requires clinical correlation. The lack of an inducible sustained monomorphic VT predicts lower risk of spontaneous VT and better prognosis.192 The overall role of EPS in the evaluation of ventricular arrhythmias (VA) in patients with syncope has diminished in the past 2 decades. This is primarily due to the use of ICD as a Class I indication for the primary prevention of SCD in patients with ischemic or nonischemic cardiomyopathy and significant LV dysfunction (ejection fraction ≤35%). An EPS is no longer required in patients with syncope before consideration of ICD therapy. However, although ICDs may reduce risk of death, they may not prevent syncope. The role of EPS in patients with syncope suspected to be due to VA and acquired nonischemic heart disease is unproven.193–198

Figure 2. Patient Disposition After Initial Evaluation for Syncope. Colors correspond to Class of Recommendation in Table 1. ED indicates emergency department; and pts, patients.

Recommendations for EPS

COR LOE Recommendations IIa B-NR EPS can be useful for evaluation of selected patients with syncope of suspected arrhythmic etiology.91,151,199–205

See Online Data Supplement 14. Diagnostic results detected during EPS occur predominantly in patients who have cardiac disease (eg, conduction system delay, coronary artery disease, cardiomyopathy, and valvular heart disease). Most of the literature evaluating EPS as a means to diagnose syncope is relatively old, and the data were obtained in referral centers where there was a high pretest probability of an arrhythmia. Eight of these small retrospective studies91,199–205 (total n=625) found that, of the 406 patients with cardiac disease or an abnormal ECG, 41% had a positive result (of these, 21% had VT and 34% had a bradycardia).151 Of 219 patients without evidence of heart disease, only 5% had a positive result (1% with VT and 10% with evidence of substrate for symptomatic bradycardia). Overall, the diagnostic yield of EPS was approximately 50% and 10% in patients with and without structural heart disease, respectively. II: No Benefit B-NR EPS is not recommended for syncope evaluation in patients with a normal ECG and normal cardiac structure and function, unless an arrhythmic etiology is suspected.205–207

See Online Data Supplement 14. One prospective evaluation of 247 patients with syncope of undetermined etiology who underwent EPS found that the diagnostic yield was significantly higher in patients with an abnormal ECG than in those with a normal ECG (22% versus 3.7%) and that the diagnostic yield was low in patients with a normal ECG and without cardiac disease (2.6%).206 In another small series of 34 patients with unexplained syncope who had normal ECGs and normal testing otherwise and who underwent EPS,205 the results were diagnostic in only 4 patients; the results were abnormal but not diagnostic in 2 patients and were normal in the remaining 28 patients. In another evaluation of 421 patients with undiagnosed syncope who underwent noninvasive testing as a means of predicting abnormal EPS findings, a normal ECG and ambulatory monitor were associated with a lower risk of EPS abnormalities than were an abnormal ECG and ambulatory monitor (9% versus 82%).207

Downloaded from http://ahajournals.org by on October 7, 2026