The diagnosis of pericardial effusion and cardiac tamponade by 12-lead ECG. A technology assessment.

OBJECTIVE This study was designed to determine the diagnostic value of 12-lead ECG for pericardial effusion and cardiac tamponade. DESIGN Cross-sectional study. SETTING University hospital. PATIENTS Hospitalized patients with and without pericardial effusion and cardiac tamponade. MEASUREMENTS AND RESULTS In a blinded manner, we reviewed 12-lead ECGs from 136 patients with echocardiographically diagnosed pericardial effusions (12 of whom had cardiac tamponade) and from 19 control subjects without effusions. We examined the diagnostic value of three ECG signs: low voltage, PR segment depression, and electrical alternans. We found that all three ECG signs were specific but not sensitive for pericardial effusion (specificity, 89 to 100%; sensitivity, 1 to 17%) and cardiac tamponade (specificity, 86 to 99%; sensitivity, 0 to 42%). None of the ECG signs were associated with pericardial effusions of all sizes, but low voltage was associated with large and moderate pericardial effusions (odds ratio = 2.5; 95% confidence interval [CI] = 0.9 to 6.5; p = 0.06) and with cardiac tamponade (odds ratio = 4.7; 95% CI = 1.1 to 21.0; p = 0.004). In contrast, PR segment depression was associated only with cardiac tamponade (odds ratio = 2.0; 95% CI = 1.0 to 4.0; p = 0.05), while electrical alternans was not associated with either pericardial effusion or cardiac tamponade. CONCLUSIONS Low voltage and PR segment depression are ECG signs that are suggestive, but not diagnostic, of pericardial effusion and cardiac tamponade. Because these ECG findings cannot reliably identify these conditions, we conclude that 12-lead ECG is poorly diagnostic of pericardial effusion and cardiac tamponade.