Pulmonary Embolism: Don't Throw Out That EKG!

 

Author: Paul Trinquero, MD (EM Resident Physician, PGY-2, NUEM) // Edited by: Elizabeth Byrne, MD (EM Resident Physician, PGY-4 NUEM) // Expert Commentary: Daniel Schimmel, MD

Citation: [Peer-Reviewed, Web Publication] Trinquero P, Byrne E (2017, January 17). Pulmonary Embolism: Don't Throw Out That EKG [NUEM Blog. Expert Commentary By Schimmel D]. Retrieved from http://www.nuemblog.com/blog/ecg-in-pe


Introduction

Pulmonary embolism (PE) is on the differential for a variety of common emergency department (ED) complaints and it can often be a tricky diagnosis to nail down. That being said, because it is so commonly considered, most of us have developed a system of risk stratifying patients based on history, exam, and decision tools such as the Well’s Criteria and PE rule-out Criteria (PERC). And as you are probably aware, the only factor even remotely related to the EKG that goes into either of those decision tools is the heart rate. Tachycardia gives a patient 1.5 points on the Well’s score and prevents a patient from being excluded by the PERC rule. But you don’t even need an EKG to diagnose tachycardia. It’s much easier to look at the monitor, scan the triage vitals, or even palpate a pulse. The EKG becomes an afterthought; quickly scanned to rule out STEMI and tossed back into the chart. Everyone knows that the EKG is a poorly sensitive test and cannot be used to reliably exclude PE on its own, especially given that somewhere around 20% of patients found to have a PE will have a normal EKG. So why even write about this topic?  

First off, the EKG is easy to obtain and is certainly useful to potentially exclude or uncover other etiologies of chest pain. Furthermore, an EKG can aid in diagnosis or at least increase suspicion or PE.  For example, T wave inversions in both lead III and v1 (the two rightmost leads) occur frequently in PE (88% according to one study), but only 1% of the time in ACS [3]. Finally, it can be useful in prognosis and risk stratification once a diagnosis is made. 


Diagnosis

While poorly sensitive, there are certain EKG findings that are somewhat specific and in the right clinical context can increase suspicion for PE in the undifferentiated patient. 

Physiology 

In order to better understand the EKG changes associated with PE, it is helpful to conceptualize the physiological framework. Because a PE represents an acute obstruction to the normal blood flow to the lungs, it results in an acute increase in pulmonary vascular resistance (PVR). It does so not only via mechanical obstruction but also because hypoxia leads to pulmonary artery vasoconstriction. The sudden increase in right ventricular (RV) afterload is often poorly tolerated since the RV typically cannon generate a mean pressure greater than 40 mmHg [4].  If severe enough, the increased pulmonary resistance leads to dilatation of the RV and a decrease in stroke volume. Dilatation of the RV leads to increased volume, increased pressure, and increased wall stress (i.e. ischemic strain). Furthermore, RV dilatation can cause or worsen tricuspid regurgitation, which in combination with increased RV pressure can lead to RA dilatation. Decreased RV stroke volume leads to decreased LV preload and subsequent hypotension, a feared complication. To summarize, the EKG changes associated with PE are all the result of either dilatation of the right heart due to increased pulmonary vascular resistance, ischemia of the RV due to increased demand, or an increased sympathetic drive due to compensatory chronotropic stimulation, dyspnea, and/or anxiety. A further discussion follows below. 

  • Dilation of the RV and/or RA  
    • RBBB: RV dilatation can lead to right sided conduction delays, which show up on the EKG as a complete or incomplete RBBB [2]. Incidence of this finding is variably reported but is probably close to 25% and may be more common with more severe, central clot burden. These conduction delays can often resolve over time as the clot dissolves and normal right sided hemodynamics return [1].
    • Right Axis Deviation (RAD): A dilated RV may shift the QRS axis and be identified as new right axis deviation [2]. It is more commonly seen in more severe PE [1].
    • Dominant R wave in v1: An acutely dilated and dysfunctional RV may manifest a large r wave. Sensitivity is low (19% according to one study), but this sign is seen more frequently in patients who go on to suffer cardiogenic shock [1].
    • P-pulmonale: Right atrial dilatation manifests as a tall (>2.5mm) p wave in lead II (or III, avF) [2]. It is seen somewhere between 0-19% of the time and does not appear to have any significant prognostic value [1].
  • Ischemia of the RV 
    • Precordial TWI: Repolarization abnormalities such as T wave inversions in the anterior precordial leads (V1-V4) and/or inferior leads are thought to be due to ischemic strain [1]. T wave inversions are more frequently associated with right ventricular dysfunction, elevated troponin, and mortality [1].
    • S1Q3T3 pattern: This classic pattern of an S wave in lead I, q wave in lead III, and a T wave inversion in lead III is thought to be due to acute right ventricular strain. It is both insensitive for the diagnosis of PE as well as non specific, however it is more likely to be seen in patients with massive or submassive PE manifesting acute RV strain [1].  
    • ST elevations in lead aVR: This manifestation of RV strain is thought to be present in 30-40% of patients with acute PE [1]. However, its presence is associated with increased severity, higher mortality, and higher likelihood of cardiogenic shock.  
  • Increased sympathetic drive 
    • Increased adrenergic tone, often compensatory, can lead to sinus tachycardia or can precipitate tachyarrhythmias such as atrial fibrillation or atrial flutter.  
 

A patient with pulmonary embolism. Shown in this EKG is sinus tachycardia with S1Q3T3 pattern, anterior T-wave inversions and mild elevation in aVR. Levis JT. ECG Diagnosis: Pulmonary Embolism. The Permanente Journal. 2011;15(4):75. 

 

Prognosis

In the patient with a known PE, there are certain EKG findings that if present increase the risk for right ventricular failure and subsequent hemodynamic collapse.  

The American Heart Association categorizes PE as massive, sub-massive, or low risk.  

  • Massive PE is defined by SBP <90 for at least 15 minutes or requiring vasopressor support, pulselessness, or bradycardia. That’s easy enough to distinguish and certainly doesn’t require poring over an EKG.  
  • Submassive PE does not involve hypotension but does demonstrate some degree of RV strain marked by an elevated troponin or BNP, dilatation or dysfunction noted on TTE or CT, or by EKG changes.  
  • Low risk PE has none of the above features and, as the name implies, carries the lowest risk of adverse outcome. 

So it turns out that the EKG has some utility in the identification or sub-massive PE. It stands to reason that the EKG changes that are the result of strain and/or failure of the right heart portend a poorer prognosis. Most in hospital deaths from PE occur within 48 hours of diagnosis and circulatory collapse often develops within 24 hours of diagnosis. Hence, it is important to risk stratify patients as early as possible and the EKG can help with that task. In 2001, Daniel et al. developed an EKG scoring system with a goal of identifying high risk PEs. They created a 21-point ECG scoring system using the following abnormalities: sinus tachycardia (2 points), incomplete RBBB (2), complete RBBB (3), TWI in leads V1–V4 (0–12), S wave in lead I (0), Q wave in lead III (1), inverted T in lead III (1), and entire S1Q3T3 complex (2). Using a 10 point cutoff, their scoring system was 23.5% sensitive and 97.7% specific for the recognition of severe pulmonary hypertension [1]. A recent systematic review and meta-analysis of over 3000 patients identified six EKG findings of RV strain that were associated with increased risk of circulatory collapse, which are displayed in the table to the right [6].  

The majority of these findings overlap with those chosen for the Daniels score, with the notable addition of ST elevation in aVR (a relatively recently described finding) and atrial fibrillation. Other studies have confirmed the predictive value of ST elevations in lead aVR and described it as an independent predictor of mortality. One study of 200 patients with a diagnosis of PE, found a 33% in-hospital mortality amongst the subgroup with ST elevations in aVR [5]. Interestingly, several EKG findings (tachycardia, TWI in v2-v3, and STE in aVR) have an odds ratio for circulatory collapse higher than echo findings of RV strain or an elevated troponin. So next time you have a patient with a PE, take a longer look at that EKG before tossing it in the bin. 


Conclusions

  • Although the EKG is poorly sensitive as a stand alone diagnostic tool, certain EKG findings can and should increase suspicion for PE 
  • EKG changes associated with PE are the result of a sudden increase in pulmonary vascular resistance leading to dilatation of the right heart, ischemia of the RV, and/or an increased sympathetic drive. 
  • The prevalence of these typical findings varies significantly in the literature and is, for the most part, beyond the scope of this post. That being said, it is useful to be aware of these common findings because the presence of multiple signs of right heart strain should increase your suspicion for PE. 
  • After a diagnosis is made, the EKG can still be used (in conjunction with cardiac markers and/or TTE) to identify any signs of right heart strain and to help prognosticate. 

Expert Commentary

Dear Paul, 

Thank you for your expert review of the EKG in the setting of pulmonary embolism (PE).  

To some an exquisite technology and to others an intimidating, crude tool with a lack of sensitivity and specificity, the EKG in PE has stood the test of time as an initial triage tool for cardiopulmonary symptoms.  In 1935, Dr. McGinn and White first described the S1Q3T3 phenomena associated with acute cor pulmonale via a limited version of our present day EKG.  The echocardiogram has evolved from M-Mode to 2-dimensional reconstructions and now 3-dimensional constructs with shadow overlays and color-coding to make the picture seem more like tissue.  The CT scan can reconstruct images to look as detailed as invasive angiograms and even measure flow and evaluate blood volumes.  But the 12 lead EKG stands strong as the technology that has not changed since 1942, with the ability to quickly triage a number of diseases with little to no harm to the patient

PE can be a frustrating syndrome with no one symptom or physical sign clearly suggestive of PE before the diagnostic CT is performed.   While shortness of breath is common, it is still only present in 50% of patients with acute PE.  We employ vital signs, history and exam in order to define a probability of PE to determine who may need laboratory testing and/or imaging to diagnose a potentially life-threatening condition.

We can hardly malign the EKG in its ability or lack thereof to diagnose acute PE.  But as Dr. Trinquero astutely conveys, it can be extremely helpful in triage.

Diagnosis

A summarized ventricular depolarization vector that is shifted to the right should be very concerning for significant cardiopulmonary disease.  This can be seen in the QRS axis of lead I, but also in isolated V1 with changes to the P wave as well.  In a stable patient, acute PE is unlikely but chronic thromboembolic pulmonary hypertension (CTEPH) or acute on CTEPH is certainly possible.  In an undifferentiated patient with EKG findings of RV strain and acute shortness of breath (SOB) without bleeding conditions, I would consider administering anticoagulation before I got my CT scan.

RBBB and incomplete RBBB is a more common finding and present in 1.4% and 4.7% of the population with a 3 times greater prevalence in men.  A number of studies have followed patients to correlate the presence with a higher risk of cardiovascular (CV) mortality.  The largest study of prevalence to date is the Copenghagen City heart study which did suggest a higher CV mortality in complete RBBB with a greater effect in a younger population.  I do see this often in patients with PE and signs of heart strain and will use this to predict PE in the patient in the ED.  I find it less useful in the post-operative patient, patients with preexisting cardiac history and following cardiac arrest.

Ischemia of the RV with T wave inversions in anterior precordial leads, ST elevations in aVR, or ST elevations inferiorly are poor prognostic signs.  They have limited value independently to differentiate between acute coronary syndromes (ACS) and acute PE.  However, a history of acute onset SOB with these findings and normal CXR would shift my differential to acute PE over ACS.  Chest discomfort features that are not pleuritic would anchor me more to the diagnosis of an ACS.  The first pulmonary angiogram I did was on a patient who had the cardiac catheterization lab activated for inferior STEMI.  After coronary angiogram excluded severe CAD, we performed a pulmonary angiogram confirming central PE and lysed the patient immediately resolving his hemodynamic instability.

Because sympathetic drive is present in so many acute conditions and tachyarrhythmias in the elderly are so common, I am very hesitant to treat empirically with anticoagulation or systemic lytics in an unstable patient with atrial fibrillation and a RBBB.  While prognostically significant in PE, making a definitive diagnosis is required.

Prognosis

As alluded above, the EKG is extremely valuable in the prognostic capacity for confirmed acute PE.  Prognosis is helpful in deciding triage and treatments.  Unfortunately, we continue to lack a clear triage and treatment pathway.  Who goes to the ICU and who gets thrombolytic therapy?  Do we give systemic or catheter-directed therapies?  The new Chest guidelines for acute PE published in 2016 appropriately state that systemic anticoagulation is clearly the therapy in most patients with acute PE without shock.  But they also make the caveat that there is a bedside evaluation and dynamic changes in the patient that may prompt the physician to appropriately choose a more intensive therapy.

In the Shopp study referenced, my experience is consistent with their findings.  Ischemia of the RV with TW inversions and ST segment deviations is ominous.  Due to the lack of specificity of atrial tachyarrhythmias and RBBB, these impact my decisions for treatment but are less worrisome.  

Signs of RV strain and biomarker elevations should prompt a discussion for observation in the ICU or a step-down unit for the first 24-48 hours after symptom onset.  At Northwestern, our PE response team uses the addition of risk factors, hemodynamic changes and markers of bleeding to identify patients that may benefit from more advanced therapy.  There are many prognostic calculators that have been evaluated to differentiate the submassive risk group into a low-intermediate versus high-intermediate risk subgroup.  I think the European Society Guidelines from 2014 are the best written to describe and identify higher risk individuals who may benefit from advanced therapy.  But I still also believe that most high-intermediate risk patients will benefit from anticoagulation alone with more intensive monitoring rather than thrombolytic therapies.  Another investigative risk score I believe is valuable is the BOVA score.  However, being simple and needing to quickly triage sick patients, I use my own informal “3 plus one rule.”  (1) RV dilation on CT should be present.  The CT may overcall RV dilation compared with echo but its absence is fairly specific for excluding PE causing acute cor pulmonale.  (2) An elevated BNP confirms elevated right sided filling pressures.  However, depending on when symptoms began and when the patient presented, a serial BNP measurement may be required to identify an acute process if initially negative.  (3) An elevated troponin should be positive to confirm the acuity since RV dilation, an elevated BNP and central filling defects on CT can all be present in CTEPH as well.  The“plus one” is more variable but will inform decisions on advanced therapy and can be tachypnea, hypoxia, an elevated shock index (HR>SBP), RV strain on EKG, a highly functional patient that wants to return to baseline function more quickly.  

Summary

Because of the nonspecific nature of the symptoms, exam findings and prognostic tools, including EKG, for a highly morbid condition, diagnosing and treating PE can be extremely difficult but rewarding for the clinician.  It requires a discerning history and may require coordination between multiple specialists in the acute phase.   It may require an informed discussion with the patient regarding the risks and benefits of the treatment options.  Paul did a wonderful job describing the role of EKG as a tool.  Like every other technology we have for diagnosis and prognosis, clinical correlation is advised.

Daniel Schimmel MD MS
Interventional Cardiology, Assistant Professor of Medicine
Acute Pulmonary Embolism Response Team, Executive Committee
Pharmacy and Therapeutics Medication Safety Committee, Chair
Bluhm Cardiovascular Institute
Northwestern Medicine


References

  1. Digby, G., Kukla, P., Zhan, Z., Pastore, C., Piotrowicz, R., & Schapachnik, E. et al. (2015). The Value of Electrocardiographic Abnormalities in the Prognosis of Pulmonary Embolism: A Consensus Paper. Ann Noninvasive Electrocardiol, 20(3), 207-223. http://dx.doi.org/10.1111/anec.12278
  2. Harrigan RA, Jones K. ABC of clinical electrocardiography. Conditions affecting the right side of the heart. BMJ. 2002 May 18;324(7347):1201-4. Review. PMID: 12016190. 
  3. Kosuge M, Kimura K, Ishikawa T, Ebina T, Hibi K, Kusama I, Nakachi T, Endo M,  Komura N, Umemura S. Electrocardiographic differentiation between acute pulmonary embolism and acute coronary syndromes on the basis of negative T waves. Am J Cardiol. 2007 Mar 15;99(6):817-21. Epub 2007 Jan 30. PMID:17350373. 
  4. Mitarai, T. (2015). High Risk Pulmonary Embolism. In J. Arbo, S. Ruoss, G. Lighthall & M. Jones, Decision Making in Emergency Critical Care: An Evidence-Based Handbook (1st ed., pp. 139-156). Philadelphia: Wolters Kluwer. 
  5. Pourafkari, L., Ghaffari, S., Tajlil, A., Akbarzadeh, F., Jamali, F., & Nader, N. (2016). Clinical Significance of ST Elevation in Lead aVR in Acute Pulmonary Embolism. Ann Noninvasive Electrocardiol. http://dx.doi.org/10.1111/anec.12368
  6. Shopp, J., Stewart, L., Emmett, T., & Kline, J. (2015). Findings From 12-lead Electrocardiography That Predict Circulatory Shock From Pulmonary Embolism: Systematic Review and Meta-analysis. Acad Emerg Med, 22(10), 1127-1137. http://dx.doi.org/10.1111/acem.12769
  7. Levis JT. ECG Diagnosis: Pulmonary Embolism. The Permanente Journal. 2011;15(4):75. 

 

Posted on January 16, 2017 and filed under Cardiovascular.