Electrocardiograms in Pulmonary Embolus

An acute central pulmonary embolus (PE) causes pulmonary artery obstruction and elevates pulmonary artery and right ventricular pressures, resulting in right ventricular and atrial strain and dilation. Acute massive PE may produce obstructive shock and subsequent death, and thus an accurate and timely diagnosis is important for the implementation of potentially lifesaving treatments including systemic fibrinolytic therapy or catheter-based treatments. Clinical conditions at the time of presentation that impact patient outcomes include amount of obstruction in the pulmonary artery, preexisting cardiopulmonary disease, and the presence of deep vein thrombus with risk for a second thromboembolic event.1 Imaging with echocardiography or computed topography (CT), along with cardiac troponins, are often used to guide treatment decisions. In a study by Kasper et al1 of patients with suspected PE, the presence of right ventricular afterload stress identified during echocardiography was associated with a poor prognosis, whereas absence of right ventricular afterload stress was generally associated with a benign clinical course, even in patients with confirmed PE.Patients with PE present with a wide range of symptoms that complicate diagnostic decision-making. Less than 50% of patients with PE are diagnosed at the time of presentation,2 and between 10% and 30% are not diagnosed until autopsy.3 Elderly patients have a higher risk for not being identified at the time of presentation.2 The electrocardiogram (ECG) cannot be used clinically to confirm or exclude PE because of its limited sensitivity and specificity.4 Many patients with confirmed PE have a normal ECG or one with only nonspecific or temporary ECG changes.2,3 Sreeram et al5 reported a normal ECG on admission in 25% of patients with PE, with 18% maintaining a normal ECG throughout admission. Although limited in diagnostic utility, the ECG is readily available and is obtained routinely in patients presenting with many of the characteristic symptoms of PE. Despite limitations, the ECG may be a helpful diagnostic tool when used in conjunction with clinical presentation,2 particularly a physical exam finding of hepatojugular reflux.6 In patients with PE, clinical evidence of right-sided heart failure is associated with a poor prognosis, with mortality as high as 32%.1 Any ECG evidence of possible PE in the presence of exam findings of right-sided heart failure warrant immediate definitive diagnostic testing for PE. The ECG may also be helpful in determining the severity and prognosis of patients presenting with known PE.3The ECG in patients with PE may show abnormalities including arrhythmias, conduction disturbances, axis deviation, nonspecific ST-T wave abnormalities, morphology changes, and right precordial lead changes.2,3,7Table 1 outlines ECG changes seen in PE. Many ECG changes, including right bundle branch block (RBBB) and T-wave inversions, will be transient, resolving with improvement of right ventricle (RV) hemodynamics.2,8 ST-T wave changes are the most common ECG findings in patients with PE,5 with nonspecific ST depression found in up to 50% of patients.9 T-wave inversion may be seen in lead III (T3), aVF, and/or in leads V1, V2, V3, or V4.5,9An RBBB or incomplete RBBB may be present in as many as 67% of patients.3 An RBBB in PE may be associated with an upright T wave and with ST-segment elevation in V1; thus it may mimic an anteroseptal myocardial infarction.5 Although right-axis deviation (RAD) is frequently associated with PE, left-axis deviation greater than 30° and inde-terminant axis may occur as frequently. The cardiac axis in PE is determined by the patient’s preexisting cardiopulmonary disease.3,5QRS morphology changes are also common. Delayed R-wave progression in the precordial leads is a potential finding in PE, with the amplitude of the R wave and S wave being most equal in lead V5 rather than in lead V4.5 S waves may be seen in lead I (S1) or aVL.5,6 Q waves may frequently be seen in lead III (Q3) or aVF but are not seen in lead II in patients with PE.5,6 Additional QRS findings include a late R wave in lead aVR and an S slur in leads V1 and/or V2.9 Limb lead low voltage (QRS 2.5 mV) due to right atrial hypertrophy may be present, however it is more commonly seen in patients with preexisting lung disease.5 PR-segment deviations may also be present in PE.9The presence of S1Q3T3 together, although often a transient finding, was one of the earliest identified ECG findings in PE; the reported incidence ranges from 10% to 50%. S1Q3T3 in one study was found equally in those with PE and those without.10 S1Q3T3 do not always coexist together and can occur at different frequencies.5 The Figure shows an ECG in a patient with PE with T-wave inversions in leads II, III, and aVF, V1 through V4, and the S1Q3T3 pattern.Pulmonary embolus most commonly impacts the ECG by producing hemodynamic and ischemic changes.3 However, not all ECG changes during PE can be ascribed to ischemia or hemodynamic alteration. For example, ECG changes may be seen before there is evidence of right-sided heart strain on the echocardiogram, and ECG changes have been shown to persist after pulmonary artery pressure normalizes.3 Autonomic and metabolic changes also affect the ECG. Less common ischemic ECG changes in PE are caused by paradoxical coronary embolism through an atrial septal defect or a patent foramen ovale.11Several ECG patterns are associated with evidence of right-sided heart strain, including S1Q3; S1Q3T3; notched S wave in lead V1; RBBB; or Qr pattern in lead V1 (Table 2).11–13 Right-sided heart strain can also produce an enlarged P wave in lead II.3 Right bundle branch block represents RV strain and dilation along with ischemia of the right bundle branch.11,14 Qr in lead V1 signifies RV dilation and may represent more disease severity than does the presence of RBBB. Qr in V1 is an independent predictor of poor outcomes in patients with PE.11,13 A notched S wave in lead V1 may represent acute RV strain and may be an early finding in the development of RBBB.11Right ventricular strain in PE accompanied by hypotension, hypoxemia, and neurohormonal activation can lead to both transmural ischemia of the RV and to subendocardial ischemia of the left ventricle (LV).11 ST elevation in lead aVR may represent ischemia of the RV outflow tract or right paraseptal area, and/or it may represent a reciprocal change to subendocardial ischemia of the LV.11 Three ischemic patterns have been identified in PE on the basis of the presence of RV transmural ischemia, LV subendocardial ischemia, or a combination of both (Table 3). In acute PE, ST elevation representing RV transmural ischemia is often seen during hemodynamic instability, whereas anterior T-wave inversion is proposed to represent postischemic evolutionary changes. ST elevation is associated with more RV strain and dysfunction than T-wave inversion alone.12The ECG may be useful in the diagnosis of PE, although the ECG alone cannot be the sole diagnostic tool.11 Sreeram et al5 found the presence of PE probable in patients without preexisting lung disease if 3 or more of the following were present on the ECG: RBBB with T-wave inversion or ST-segment elevation in lead V1; prominent S wave in leads I and aVL; R-wave transition shift to equiphasic R and S waves in V5; Q waves in leads III and aVF; RAD or indeterminate axis; low voltage in limb leads; or T-wave inversion in leads III, aVF, or V1 through V4. However, in one study the above guide of 3 or more findings had only 26.7% sensitivity and 57.1% positive predictive value in patients with clinical suspicion for PE (although the same guide of findings had 94.2% specificity and 81.7% negative predictive value).10 In patients with chronic obstructive pulmonary disease, the above rule results in a high false-positive rate because preexisting lung disease can also cause many of the same ECG features seen in PE.3,5In one study, only 2 ECG changes, S1 and Q3, were found significantly more often in patients with confirmed PE6; yet another study found only sinus tachycardia and incomplete RBBB significantly more common in confirmed PE.10 A study by Kucher et al13 confirmed the results of previous studies that the ECG alone cannot be used to exclude PE; however, in this study the specificity and positive predictive value for Qr in V1 was each 100%. Although Qr in V1 is highly specific for PE, its prevalence is low.13Some ECG findings in the setting of acute PE have prognostic significance. ST-segment depression and T-wave inversion in leads V1 and V2 have been associated with an increased number of underperfused lung segments on imaging.3,9 Geibel et al15 identified several ECG characteristics (atrial arrhythmias, RBBB, limb lead low voltage, Q waves in leads III and aVF, and ST elevation or depression in the left pre-cordial leads) associated with worse survival in patients presenting with acute central PE. One or more of these ECG findings was found to be an independent predictor of poor outcome. In this prospective study, 29% of patients with 1 or more of these characteristics did not survive to hospital discharge compared with 11% who did not have any of the above ECG findings. T-wave inversion in leads V1 through V4 did not show prognostic significance.15Sreeram et al5 found inconsistent association between ECG changes and acute RV volume and pressure overload identified on echocardiograms, suggesting that echocardiography may be more sensitive than the ECG to acute changes in RV size and pressure.5 However, Kucher et al13 showed Qr in lead V1 in patients with PE to be the strongest predictor of RV strain and an independent predictor of subsequent adverse outcome, including increased early mortality. Qr in V1 and a negative T wave in V2 or V3 were closely related to the presence of moderate to severe RV dysfunction and were predictors of a complex hospitalization.13 Qr and ST elevation in V1 were also associated with troponin leak.13 Zhan et al11 reported ST elevation in lead aVR and ST depression in leads I and V4 through V6 during hemodynamic instability in patients with PE. In a study of 386 patients with known PE and normal blood pressure, ECG signs of right-sided heart strain (complete or incomplete RBBB; S1Q3T3 pattern; and T-wave inversion V1 through V4) were associated with adverse outcomes independent of echocardiogram findings.14Electrocardiogram changes are best seen in patients with a massive or submassive PE. Daniel et al16 proposed a 21-point scoring system based on ECG findings to predict a massive PE. A score of 10 or greater was specific to severe pulmonary hypertension caused by PE. The ECG scoring system weighted the following: sinus tachycardia; incomplete RBBB; complete RBBB; any T-wave inversion in V1, V2, or V3; T-wave inversion in all leads V1 through V4; S wave in lead I; q wave in lead III; inverted T wave in lead III; and the entire S1Q3T3 complex. This scoring system has been proposed to identify patients with massive PE from those without. In a study by Ferrari et al8 in which 74% of patients had massive PE, 91% had ECG changes, with negative T waves in the precordial leads being the most common finding; this finding was present in 85% of patients with massive PE and in only 19% of patients with nonmassive PE. Negative T waves in the precordial leads was accompanied by negative T waves in the inferior leads in 50% of patients. There was not a significant difference in the presence of S1Q3T3 and RBBB or incomplete RBBB between massive and nonmassive PE.8 However, the emergence of a new RBBB during ECG monitoring in patients with confirmed PE has been associated with a massive main pulmonary trunk PE.17The finding of T-wave inversion has been shown to resolve with resolution of PE after treatment with fibrinolytic therapy. The reversibility of the anterior T-wave inversion within 6 days of massive PE is associated with a lower mean pulmonary artery pressure.8 Yoshinaga et al18 found an increase in the peak negative amplitude of T waves after treatment with fibrinolytic therapy for massive PE to also be associated with lower pulmonary artery pressures.There is no ECG sign confirming certainty of diagnosis of PE; ECG findings are limited in the evaluation of possible PE in patients with preexisting lung disease. The ECG in a patient with signs and symptoms suspicious for PE may reveal the presence of other clinical conditions such as acute coronary syndrome that require urgent intervention. The information obtained from the ECG in patients with clinical suspicion for PE and in those with confirmed PE should guide frequency of assessment to identify signs of right-sided heart strain. Nursing assessment may identify signs such as hepatojugular reflux, and bedside monitoring can identify signs such as the development of a new RBBB. The ECG in conjunction with the clinical picture may raise suspicion of PE and indicate a need for confirmatory diagnostic testing. Additionally, the ECG may aid in the evaluation of the severity of the PE and the patient’s prognosis.