Abstract
Background
To study trends in the clinical presentation, electrocardiograms, and diagnostic imaging in patients with pulmonary embolism presenting as ST segment elevation.
Methods
We performed a systematic literature search for all reported cases of pulmonary embolism mimicking ST-elevation myocardial infarction. Pre-specified data such as clinical presentation, electrocardiogram changes, transthoracic echocardiographic findings, cardiac biomarkers, diagnostic imaging, therapy, and outcomes were collected.
Results
We identified a total of 34 case reports. There were 23 males. Mean age of the population was 56.5 ± 15.5 years. Patients presented with dyspnea (76.4%), chest pain (63.6%), and tachycardia (71.4%). All patients presented with ST-elevations, with the most common location being in the anterior-septal distribution, lead V3 (74%), V2 (71%), V1 (62%) and V4 (47%). ST-segment elevations in the inferior distribution were present in lead II (12%), III (18%), and aVF (21%). Presentation was least likely in the lateral distribution. Troponin was elevated in 78.9% of cases. Right ventricular strain was the most common echocardiographic finding. Over 80% of patients had findings consistent with elevated right ventricular pressure, with 50% reported RV dilatation and 20% RV hypokinesis. The most commonly used imaging modality was contrast-enhanced pulmonary angiography. There was a greater incidence of bilateral compared to unilateral pulmonary emboli (72.4% vs. 10%). About 65% patients received anticoagulation and 36.3% were treated with thrombolytics. Forty-six percent of patients required intensive care and 18.7% intubation. Overall mortality was 25.8%.
Conclusions
A review of the literature reveals that in patients presenting with pulmonary embolism, electrocardiogram findings of ST-segment elevations will occur predominantly in the anterior-septal distribution.
Introduction
Pulmonary embolism (PE) is a sudden blockage of an artery in the lung caused by a venous thrombosis (DVT) clot from a deep vein. There are about 300,000–600,000 venous thromboembolism cases in the United States annually, with an estimate of 10%–30% mortality within the first 30 days, with most deaths occurring in patients presenting with pulmonary emboli. 1
PE mimicking an ST-elevation myocardial infarction (STEMI) is a rare entity described only in case reports. We reported an interesting case and carefully analyzed the available literature on PE mimicking ST-elevation myocardial infarction to describe the clinical presentation, electrocardiograms (ECGs), diagnostic trends, and prognosis.
A 60-year-old man who underwent percutaneous coronary intervention with drug-eluting stent seven months previously for worsening angina, presented for planned surgical procedure. Twelve days prior to presentation the patient underwent computed tomography (CT)-guided drainage of an abdominal wall abscess. The medical history is notable for diabetes mellitus type 2, hypertension, and portal vein thrombosis with splenorenal shunt four years prior to admission. The patient’s medications included aspirin, carvedilol, verapamil, pravastatin, metformin, and sitagliptin. On post-operative day 1, cardiology was consulted for intermittent left-sided chest pain. Blood pressure was 122/80 mmHg, heart rate 93 beats per minute, and respiratory rate 18 breaths per minute. Physical exam was unremarkable. ECG revealed q-waves in leads III and aVF. Laboratory analysis returned a troponin T of < 0.01 ng/mL. Four hours later, the rapid response team was called—the patient was found diaphoretic and disoriented with a systolic blood pressure of 73 mmHg, heart rate of 136 beats per minute, and an increased respiratory rate. ECG now revealed sinus tachycardia, an SI/QIII/TIII pattern, and ST-segment elevations in leads III, aVF, aVR, and V1, with reciprocal changes in the lateral and precordial leads consistent with an STEMI (Figure 1). The patient was taken for emergent coronary angiography that revealed no obstructive coronary lesions. Right heart catheterization showed mildly elevated right-sided pressures (mean pulmonary artery pressure 29 mmHg, right ventricular pressure 29/7 mmHg, right atrial pressure 7 mmHg, and capillary wedge pressure of 13 mmHg). Laboratory analysis sent off returned a troponin T of 0.42 ng/mL and creatinine kinase (CPK) of 185 U/L. This result was not consistent with ECG findings of the inferior wall myocardial infarction.

Electrocardiogram on POD1 with ST segment changes; ST-segment elevations in Leads III, aVF. Avr, and V1, with an SI/QIII/TIII pattern.
Therefore, a massive PE was suspected and an immediate CT PE protocol revealed a massive saddle PE with right heart strain (Figure 2). Echocardiogram revealed a left ventricular ejection fraction of 70%, severe right ventricular dilation and hypokinesis, and a large mobile thrombus across the tricuspid valve (Figure 3). The patient was treated with catheter-directed localized tissue plasminogen activator for 24 h. An inferior vena cava filter was also placed due to planned second intervention. The clinical condition improved progressively and repeat echocardiogram prior to discharge showed normalized right ventricular function. The ECG normalized the day before discharge with disappearance of ST changes. He was discharged on hospital day 23 with intact cardiovascular and neurological function.

Computed tomography pulmonary angiography revealing massive, saddle pulmonary embolism.

Echocardiogram parasternal view of the right ventricular inflow tract revealing a large mobile thrombus across the tricuspid valve in the right ventricle.
Materials and methods
A protocol for this systematic review was created, posted online, and registered in PROSPERO (CRD42017059163). We followed the guidelines outlined by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). A computerized literature search, of Pubmed, EMBASE CENTRAL, and Google search engine was conducted. Search term keywords included “pulmonary embolism,” “ST-Elevation,” and “myocardial infarction” as well as combinations of these terms. No language restrictions were enforced. Articles not written in English were translated. Inclusion criteria for publications in this systemic review were: (1) Articles including patients with at least radiological or autopsy confirmation of PE; (2) Articles including patients with electrocardiographic evidence suggesting STEMI; (3) Articles providing data on at least one of the following: clinical presentation, ECG description or original ECG, serum cardiac markers, any radiologic images. Exclusion criteria were (1) Review articles; (2) Articles including patients with PE presenting with electrocardiographic evidence not consistent with STEMI. (3) PE presenting as STEMI with patent foramen ovale or atrial septal defect. (4) Articles including other conditions that mimic myocardial infarction.
The following data were extracted: age, sex, presenting symptoms (dyspnea, chest pain, lower extremity edema, syncope), fever > 37.5°C, white blood cell count, serum cardiac markers (CK and troponin), Pro-BNP, D-dimer, Wells’ criteria for PE, ECG characteristics, echocardiographic findings, CT findings, V/Q scan findings, angiography findings, hemodynamics (blood pressure and heart rate), oxygen saturation, risk factors (history of smoking, active cancer, recent surgery, history of PE or DVT), need for mechanical ventilation, inotropes, and intensive care unit (ICU) admission, treatment strategy, and outcomes (dead or alive). The mean age of patients and standard deviations were calculated. All available ECG descriptions for each case were obtained and original data were analyzed and grouped into ST-segment elevation and ST-segment depression. Serum cardiac markers, Pro-BNP, and D-dimer were classified as normal or elevated. Fever was defined as a temperature > 37.5°C, leukocytosis as WBC count > 10,000/mm3, hypotension as systolic BP < 90 mmHg and diastolic BP <60 mmHg, tachycardia > 100/min, oxygen desaturation < 90%, location of PE, defined as unilateral, bilateral, or main pulmonary artery. Treatment method, described as by thrombolytics, localized thrombolysis, or anticoagulation. Imagining findings were obtained from the report. The prevalence of the different measured variables was calculated from the extracted data. The “not available” data cases were not considered in the calculation. Statistical analyses were performed using SPPS version 20 (IBM SPSS, Chicago, Illinois).
Results
The literature search yielded a total of 2732 publications. Following the exclusion criteria, 2699 citations were excluded, leaving 91 articles reviewed, 33 articles included,2–34 with 34 adult cases for analysis.
Demographic and risk factors
There were 34 adult cases. The mean age at presentation was 56.5 ± 15.5 years (range: 25–85 years), with 23 males (67.6%) in the population. Patients with active cancer and recent surgery made up 6.5% and 12.5% of the population, respectively. Of note, no history of PE was reported in the patient population prior to presentation (Table 1).
Clinical, electrocardiographic, laboratory, and imaging findings of pulmonary embolism mimicking an ST-elevation myocardial infarction.
BNP: brain natriuretic peptide; BP: blood pressure; CK: creatine kinase; DVT: deep venous thromboembolism; HR: heart rate; ICU: intensive care unit; PE: pulmonary embolism; TTE: transthoracic echocardiogram; WBC: white blood cell.
aBased on reported data.
Clinical presentation, electrocardiographic, and laboratory findings
Dyspnea, pleuritic chest pain, syncope, and lower extremity swelling were the most common reported symptoms. Hypotension was present in 25% and over 70% of cases presented with tachycardia and 38% with oxygen desaturation. Seven patients were hemodynamically unstable upon presentation. Variability in choice of biomarkers was seen in the cases reviewed. Prevalence of reported elevated troponin assay was high (79%, n = 15). Pro-BNP although routinely used for prognostication for risk stratification was checked in only four patients. Interestingly, 44% of patients had low risk for PE by Wells criteria. The most common ECG finding was the presence of ST-segment elevation in the anterior-septal distribution, lead V3 (74%), V2 (71%), V1 (62%), and V4 (47%). ST-segment elevations in the inferior distribution were present in lead II (12%), III (18%), and aVF (21%). Presentation was least likely in the lateral distribution, lead I (3%) and lead aVL (0%). Figure 4 shows the distribution of ST-segment elevations. Moreover, ST-segment depressions were more common in the lateral distribution, lead V5 (15%), V6 (15%), and I (15%).

Distribution of ST-segment elevations.
Imaging
Over 80% of patients had echocardiographic findings consistent with elevated right ventricular pressure, with 50% reported RV dilatation and 20% RV hypokinesis. Surprisingly, 17 cases diagnosed PE by catheter-based pulmonary angiography; it is unclear if pulmonary artery angiography or V/Q scan was non-diagnostic in these cases. Autopsy was diagnostic in eight (24.2%) cases (Table 1).
Outcomes
There was a greater incidence of bilateral compared to unilateral pulmonary emboli (72.4% vs. 10%), with 28% of bilateral PEs also found to have a concomitant embolus within the main pulmonary artery. In patients presenting with ST-segment elevations of all leads in the anterior-septal distribution (n = 8 cases), dyspnea was the most common reported symptom (88%), with an average Wells score of 1.87. In addition, diagnosis of RV strain (100%), RV dysfunction (100%), and incidence of bilateral pulmonary emboli (75%) were highly likely. Furthermore, mortality rate was 50% in patients presenting with ST-segment elevations in the anterior-septal distribution. Conversely, when examining patients who presented with ST depressions in the lateral distribution (n = 5 cases), these patients were more likely to present with dyspnea (80%), chest pain (60%), and syncope (40%), with an average Wells score of 4.7. Three patients were found to have bilateral pulmonary emboli and there was one mortality.
Treatment and mortality
Overall, 64% (20 cases) of patients received anticoagulation and 36% (11 cases) were treated with anticoagulation and thrombolytics, of which only two patients received catheter-directed thrombolytics. Forty-six percent of patients required intensive care and 18.7% intubation. Ten patients had cardiac arrest and the overall mortality was 26%.
Discussion
Venous thromboembolism affects an estimated 900,000 people in the U.S. each year. 1 The clinical presentation of PE is variable. Small pulmonary emboli are usually without cardiopulmonary symptoms and asymptomatic since the lungs do not have pain fibers. Typically, clinical manifestations in PE patients range from new dyspnea (79%) at rest or with exertion, tachypnea (57%), tachycardia (26%), and orthopnea (38%). Evidence of pulmonary hypertension or elevated right atrial or ventricular pressure overload occurred in 22%. Pleuritic pain, transmitted from pain fibers in the parietal pleura in the setting of a pulmonary infarction, occurred in 47% of patients.35,36 Similarly, in our review and pooled analysis, the major presenting symptoms in patients’ presenting with ST-elevations were dyspnea (76.4%) and chest pain (63.6%); however, this review showed greater evidence of pulmonary hypertension (40%).
ECG findings are inconsistent and too non-specific to diagnosis a pulmonary embolus. Studies have reported the most common finding to be sinus tachycardia and an S1Q3T3 pattern. Typical ECG findings include an S1Q3 or S1Q3T3 pattern, a rightward axis shift, signs of complete or incomplete right bundle branch block, T-wave inversions in the right precordial leads, and sinus tachycardia. 6 Evidence of right heart strain on ECG is also indicative of PE, including T-wave inversion in precordial leads, a transient right bundle branch block, and new right or left axis deviation. 37 ST-segment elevations are not typically associated with a pulmonary embolus. Sreeram et al. 38 looked at 49 hospitalized PE patients and found PE to be considered probable with the presence ≥ 3 of the following ECG abnormalities: incomplete or complete right bundle branch block (RBBB) associated with ST-segment elevation and positive T wave in lead V1, S waves in leads I and aVL of > 1.5 mm, a shift in the transition zone in the precordial leads to V5, Q waves in leads III and aVF (but not in lead II), right-axis deviation, a low voltage QRS axis of < 5 mm in the limb leads, and T-wave inversions in leads III and aVF or leads V1 to V4. In our case, the initial ECG showed sinus tachycardia with an SI/QIII/TIII pattern, and ST-segment elevations in leads III, aVF, aVR, and V1 revealing potentially left main and proximal LAD involvement and RV-infarction. Our systematic review of the literature revealed ECG findings in 34 patients presenting with a PE to have ST-segment elevation generally in the anterior-septal distribution. ST-segment elevations in the inferior distribution were present in a lower proportion of the cases. Presentation was least likely in the lateral distribution (<3%). Suggested mechanisms for ST-segment elevation in the literature include the acute elevation in right ventricular pressures and afterload, resulting in RV dilatation and myocardial ischemia from stretch vs. ischemia induced from catecholamine surge, hypoxia, or hypotension. As seen in the case report by Zhan et al., 8 we postulate the inferior ST elevations in our case resulted from RV ischemia secondary to severely increased RV afterload and myocardial stretch from a massive saddle embolism.
The presentation of PE requires rapid risk stratification based on hemodynamic status to ensure the highest quality of care. A massive PE is characterized by systemic hypotension or evidence of shock by end-organ hypoperfusion, hypoxia, altered level of consciousness, oliguria, or cool extremities. Conversely, a submassive PE will present with hemodynamic stability, but with cardiac strain, evident from right ventricular dysfunction or myocardial ischemia. 39 In our analysis, PE patients presenting with ST-elevations will have frequent poor prognostic factors indicating right heart strain, such as troponin elevation (78.9%) and RV dysfunction (85%), with evidence of RV strain (87.5%), dilation (50%) and hypokinesis (20%), but only around 25% presented with hypotension. These findings may reflect the importance of bedside echocardiography for the use of early stratification of patients suspected of acute coronary syndromes, including patients presenting with ST-elevations and would be useful in the setting of discordant EKG to wall motion findings. The findings of regional wall motion abnormalities sparing the right ventricular apex (McConnell’s sign) and RV dilation/hypokinesis would have been useful in the setting of discordant EKG to wall motion findings. In our clinical context, there was a high suspicion for ACS and the case was expedited for coronary angiography.
In a recent study, poor prognostic factors include systolic blood pressure < 90 mmHg, Shock index > 1, pulmonary embolism severity index (PESI) greater then class III, ECG with T-wave inversions in precordial leads, RV dysfunction on echo, proximal clot, and elevated troponins or pro-BNP. 40 Hospital mortality will range from 4.6% to 12.9% in normotensive patients with RV dysfunction to 24.5% in cardiogenic shock.41–43
PE management centers on the initial hemodynamic stabilization of the patient. Resuscitative therapies range from supplemental oxygen to ventilator support, hemodynamic support, or empiric anticoagulation. 44 Thrombolytic therapy is recommended as standard first-line treatment in patients with massive PE. Our analysis found most patients were treated with heparin (64.5%), which may be due to initial concerns for MI. Thrombolytics were administered in 36.6% of patients. The use of mechanical ventilation (18.7%) and inotropes (16.6%) was low, signaling stable hemodynamic status on presentation. However, these patients still experienced a high overall mortality – 26%. In our analysis, 46% of patients required ICU admission.
Although ST-segment elevation with hemodynamic instability in setting of coronary artery disease initially focused the medical team to acute coronary syndrome, acute PE should have remained high on the differential considering recent surgery and reduced mobility and ECG findings compatible with PE, including the most common, sinus tachycardia. Whereas it is difficult to assume in our systematic review, that a delay in diagnosis of PE led to higher mortality; previous observational studies have shown, a delay in PE diagnosis is associated with worse outcomes and is more likely to occur in the elderly and female patients, in addition to patients without pleuritic chest pain.45–47
Therapeutic decision-making has become increasingly complex and dependent on provider experience and expertise. Having a formal process to activate a rapid response team for PE may increase improved resource utilization and identifications of individuals with expertise in particular procedures such as catheter-directed thrombolysis and mechanical thrombectomy. 48 A multidisciplinary PE response team offers a unique coordinated approach to patient care and can be implemented successfully with improved outcomes for patients with PE.49,50
Limitations
First, the number of studies and included patients was modest, hence limiting the validity of our findings. Second, included cases were heterogeneous in terms of PE image modality for diagnosis, treatment, biomarkers, outcome definition, rendering a comparison across cases difficult. Third, publication bias may have affected the findings of our analysis of published cases. Fourth, incomplete data reporting were frequent across studies, particularly for baseline, imaging and patient characteristics. Fifth, outcome data are predominantly limited to the index admission; Sixth, the prolonged period of time between the publication of the first and the last case; therefore, results should always be interpreted with caution when they are included in reviews and pooled analyses.
Conclusion
The results of this systematic review and pooled-analysis of PE cases presenting with ST-segment elevations found ECG findings predominantly in the anterior-septal distribution. Our review describes the wide presentation in patients presenting with PE and ECG ST changes and highlights the importance of recognizing PE as part of the differential in patients presenting with ST-segment elevation.
Patient consent
Informed consent was obtained from the patient to use the clinical information for research purposes and for submitting for publication.
Footnotes
Authors’ contribution
PAV and PPV have contributed equally to this paper.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
