Syncope has been reported to occur as the initial presentation of pulmonary embolism (PE) in approximately 6% of patients. It has proven to be a difficult clinical correlation to make, especially in patients without obvious risk factors for venous thrombosis.1,2 PE considers a differential diagnosis, particularly in the case of syncope with dyspnea or without chest pain. Previous studies showed a prevalence of PE, amounting to around 10%, among patients with syncope. However, in a new 2018 meta-analysis, the pooled estimate of PE prevalence in the emergency department (ED) syncope patients was only 0.8%.3–5 A cross-sectional study conducted by Prandoni et al6 reported a significantly higher prevalence of PE in patients presenting with syncope to the ED (3.8%) or hospitalized because of syncope (17.3%).
The prevalence of PE in the clinical setting of syncope differs in different countries, depending on the accuracy of diagnosis and the new medical diagnostic approach. These discordant and controverted findings challenged us to conduct this study to determine the prevalence and clinical characteristics of PE in patients who presented with a first episode of syncope to the ED and in those hospitalized for syncope in a tertiary care hospital.
MATERIALS AND METHODS
This study was approved by the Ethics Committee of the Emergency County Hospital Baia Mare, Romania. Oral informed consent was obtained from all enrolled patients. Patients' records/information were anonymized and deidentified before the analysis.
We performed a prospective observational study that was aimed at determining the prevalence and the clinical characteristics of PE among adult patients (>18 years old), who presented themselves consecutively to the ED or were admitted for syncope at the Emergency County Hospital Baia Mare. The hospital provides medical services for 800,000 inhabitants from across 2 departments in the north west of Romania, and, usually, the patients with syncope from 7 general hospitals of the region are referred to our hospital's ED for further investigation.
Syncope was defined by the last European Society of Cardiology and American Heart Association guidelines as a transient loss of consciousness with a rapid-onset, short duration, and spontaneous recovery.1,7 The patients who had obvious causes of loss of consciousness, such as epileptic seizure, stroke, or head trauma, were ruled out. The exclusion criteria were as follows: history of previous syncope, undergoing anticoagulation therapy (for atrial fibrillation, atrial flutter, prosthetic valves, venous thrombosis, or other reasons), and pregnancy.
Collection of clinical data
Each patient's medical history was obtained, including the description of the symptoms experienced and the presence of the following risk factors for syncope and/or PE: history of venous thromboembolism, presence of cardiac disease, recent bleeding or volume depletion, administration of hypotensive and antiarrhythmic drugs, immobilization longer than 1 week, active cancer, hormonal treatment, recent surgery, trauma, or infectious disease within the previous month.
Every patient with a first syncopal episode in the ED was systematically evaluated for possible PE. We performed a clinical assessment of all the patients in the ED by evaluating the presence of hypotension (defined by systolic blood pressure <100 mm Hg), tachypnea (defined by respiratory rate >20/min), tachycardia (>100 heart beats/min), hypoxemia (oxygen saturation <90%), and signs of deep-vein thrombosis. The paraclinical examinations of all patients comprised routine blood tests with D-dimer assessment, electrocardiography, and arterial blood gas test, in case of hypoxemia. To determine the etiology of syncope, if applicable, we performed head computed tomography (CT) scan, echocardiography to diagnose a structural heart disease, a 24-hour Holter monitoring to record possible arrhythmias or conduction disorders, a tilt test to confirm a vasovagal cause, and an evaluation of orthostatic hypotension.
Diagnosis of PE
The presence or absence of PE was confirmed by an identified filling defect in the pulmonary artery system with computed tomography pulmonary angiography (CTPA). Thoracic CT scans were performed using a 16-detector multisectional CT scanner (Siemens SOMATOM Sensation 16 CT scanner) with an intravenously injected contrast agent. CTPA results were categorized as positive for PE when an intraluminal filling defect was observed within a pulmonary arterial vessel. In addition, we used a validated algorithm that was based on pretest clinical probability and the result of the D-dimer assay.8 The quantitative assay used was PATHFAST D-Dimer, with a cutoff of 0.5 μg/mL fibrinogen-equivalent units. The pretest clinical probability of PE was defined for all patients enrolled in the study, according to the simplified Wells score.9 PE was ruled out in patients who had a low (“unlikely”) pretest clinical probability, according to the Wells score and a negative D-dimer assay. For patients with a high (“likely”) clinical probability pretest, a positive D-dimer result, or both, CTPA was performed.
To stratify and assess the early mortality risk in patients diagnosed with PE, according to the criteria of 2014 European Society of Cardiology Guidelines, we evaluated the markers of right ventricular dysfunction by NT-pro BNP, the signs of right ventricular dysfunction by echocardiography, and the markers of myocardial injury by high sensitive troponin. We calculated the PESI class (PE severity index), a PESI class >III, indicating a moderate to very high 30-day mortality risk.10
Categorical variables are given as numbers or percentages, and data for continuous variables are given as mean ± SD. Participants were categorized as with and without PE. For comparing the categorical data, the χ2 test or the Fisher exact test was performed. The 95% confidence intervals and P values were calculated according to the normal approximation at the 2-sided 0.05 significance level. Statistical analysis was performed using SPSS Version 21.0 (SPSS, Inc, Chicago, IL).
From January 1, 2012 to December 30, 2017, a total of 1782 patients, 849 women and 933 men, with a median age of 71.3 ± 13.24, min = 20 years, max = 97 years), visited the ED of our hospital for syncope (Figure 1).
As many as 945 patients were discharged because they had vasovagal syncope, situational syncope, drug-induced hypotension, volume depletion, or declined hospitalization.
Eight hundred thirty-seven patients (46.97%) were admitted, from whom 445 were excluded because they had previous episodes of syncope (297 patients) and were receiving ongoing anticoagulation therapy (148 patients) for atrial fibrillation, atrial flutter, prosthetic valves, and venous thrombosis.
Three hundred ninety-two patients with a first episode of syncope were enrolled in the study, of whom 350 were admitted to cardiology, 19 to neurology, 15 to internal medicine, 2 to diabetes and metabolic disorders, and 2 to hematology.
Of these, 185 were women and 207 were men, with a median age of 71.3 years [95% confidence interval (CI) 69.9 to 72.6, SD ± 13.24, maximum age = 90 years, minimum age = 22 years]; 230 patients of the 392 were >70 years old.
The baseline characteristics of the patients are provided in Table 1.
The potential explanations for the syncope are listed in Table 2.
CTPA was performed in the patients who had positive D-Dimer assay and a high pretest clinical probability; 45 patients were diagnosed with PE. In the entire cohort, the prevalence of PE in patients hospitalized with syncope was 11.47%, which is 45 patients of 392 (CI 95% 8.48–15.04). The prevalence of PE in patients presenting with syncope to the ED was 2.52%, which is 45 patients of 1782 (CI 95% 1.8–3.3).
There were significant differences between the patients with PE versus the patients without PE. In the group suffering from PE, there was a statistically higher prevalence of tachypnea (44.4% vs. 6.62%), tachycardia (62.22% vs. 16.1%), hypotension (40% vs. 25.64%), hypoxemia (48.8% vs. 13.25%), clinical signs of deep vein thrombosis (35.5% vs. 3.74%), and risk factors of vein thrombosis such as prolonged immobilization, trauma, or recent surgery (Table 1).
Among the PE patients presenting with syncope, the most frequent location of the embolus was proximal bilateral in 24 patients (53.33%), in a main pulmonary artery in 10 patients (22.22%), in a lobar artery in 10 patients (22.22%), and in a segmental artery in 1 patient (2.22%).
Thirteen patients were classified in PESI class <III, identifying those with a low risk of 30-day mortality, and 32 patients in PESI class ≥III were associated with a moderate to very high 30-day mortality risk (15 patients in PESI class III, 4 patients in PESI class IV, and 13 patients in PESI class V). Combining the clinical presentation with the paraclinical findings, 4 patients were found to be at low risk, 26 patients at intermediate risk (15 with intermediate-low risk, 11 with intermediate-high risk), and 15 patients at high risk. The thrombolytic therapy was used in 10 PE patients (22.2%) in the syncope group. The hospital mortality rate of PE patients presenting a first episode of syncope was 4.4%: 2 patients died of cardiogenic shock in the high-risk group. At 30-day follow-up, the rates of adverse events and mortality remained unchanged.
Our study included a large number of patients who presented themselves to the ED, or were hospitalized for a first episode of syncope, and revealed a high prevalence of PE among them.
PE was confirmed in approximately 1 of every 9 patients (11.47%) admitted to the hospital for syncope and in 1 of every 40 patients who visited the ED for syncope (2.52%). The high prevalence of PE among patients, presenting themselves to the ED or admitted to the hospital, found in our study is in concordance with the results listed by Prandoni et al.6 On the other hand, our results are in contrast with the data reported in other studies, in which the prevalence is very low, varying from 0.2% to 2.5% in patients presenting themselves to the ED and ranging from 0.6% to 1.3% in those hospitalized.5,11,12
The main mechanisms of PE causing syncope include acute right ventricular failure, especially when the occlusion of the pulmonary vascular tree is greater than 50%, Bezold–Jarisch type vasovagal reflex that leads to neurogenic syncope, and associated dysrhythmias. However, hypoxemia secondary to ventilation or perfusion abnormalities may also play an important role in syncope development. The location of the embolus is extremely important for clinical presentation and prognosis, because only the obstruction of a main pulmonary artery or a lobar artery, which leads to a transient depression in cardiac output, is associated with syncope.3,13,14
In our study, almost all the patients had a proximal obstruction on CTPA: more than half of the patients (53.3%) had a bilateral obstruction, 22.2% had an obstruction in a main pulmonary artery, 22.2% in a lobar artery, and only 1 patient had a segmental artery obstructed.
The prevalence of PE in the clinical setting of syncope differs in different studies and countries, depending on the accuracy of diagnosis. A well-documented study shows that the use of CTPA has led to false-positive results in one quarter of CT pulmonary angiograms, which were initially interpreted as positive. In that study, a majority of false-positive readings involved subsegmental locations, due in part to various scanning artifacts.15 Furthermore, PE diagnostics is questionable in the presence of small emboli that were subclinical. This is seen in the case of a meta–analysis by Oqab et al,5 in which the authors explicitly excluded from their analysis the syncope-prevalence study by Prandoni et al.6 Then, Oqab et al principally discussed the study by Prandoni et al and focused their interest on one-third of its diagnosed PEs, which were possibly irrelevant because of segmental or subsegmental thrombi, but did not discuss the two-thirds of its diagnosed PEs that were large. However, if the segmental and subsegmental PEs from the study by Prandoni were excluded, the demonstrated syncope prevalence is still about 2.5% in ED-PE patients and 11.5% among PE-syncope hospitalized patients, both much higher than what previous prevalence studies have indicated.16 The prevalence of PE presenting as syncope in our study, 2.52% in ED-PE patients and 11.47% in hospitalized patients, is very similar to the prevalence found in the study by Prandoni's for PE patients with large thrombi. From this point of view, our results are very interesting because thrombi were segmental in only 1 patient, so maybe irrelevant as far as triggering a syncope is concerned. Thames et al3 , among 132 cases of angiographically documented PEs, found syncope in 13% of the patients and described significant differences in terms of hemodynamic changes and pulmonary angiographic obstructions between PE with and without syncope. Similarly, Stein et al observed syncope in 4% of submassive and 13% of massive PEs (P < 0.01), which was classified on the basis of Walsh's angiographic score. This figure is almost identical to that reported in the International Cooperative Pulmonary Embolism registry, grouping over 2000 individuals with PE, and in a multicenter survey, including nearly 400 patients, performed in France, which found a 14% prevalence of syncope.4,17,18 Also, the incidence of syncope was remarkably higher (35%) in the Management strategies and determinants of outcome in acute major pulmonary embolism registry, in which 1001 patients with major PE were enrolled.19 From this point of view, our study is concordant with the previous observations that the prevalence of syncope in patients with PE in the various populations is higher with massive PE and an average around 15%.20 At the time of writing this article, a new meta-analysis found a low prevalence of PE (0.06%–0.55%) in patients presenting to the ED with syncope and of 0.15%–2.1% in those who were hospitalized for syncope. Investigators conducted a retrospective study of 5 databases in 4 countries (Canada, Denmark, Italy, and the United States) from 2000 to 2016.21 Again, the analysis was performed retrospectively, and for syncope patients, who lacked specific symptoms or signs of PE, testing for PE has never been a standard practice. In the absence of such an approach for syncope evaluation, a number of patients with small pulmonary emboli may have had missed PE. In our study, as well in the study by Prandoni, every patient with a first syncopal episode in the ED was systematically evaluated for possible PE.
As in other studies, we found that tachypnea, tachycardia, hypotension, hypoxemia, and risk factors of vein thrombosis, such as prolonged immobilization, trauma, or recent surgery, are very frequent in patients with PE who have had a syncopal episode (Table 1). Because the differential diagnosis between syncope caused by PE and that due to other pathological conditions may be difficult, particularly in individuals aged >70 years, the association of the above-mentioned clinical symptoms with the electrocardiogram criteria of acute cor pulmonale should lead to further workup.
The thrombolytic therapy was used in 10 patients with PE (22.2%) with persistent hypotension lasting for ≥24 hours and who met the criteria of intermediate-high risk or high risk. The hospital PE mortality rate of 4.4% is not high, and we agree with the observation of Seyyedi et al22 that this paradox in outcome may be due to the earlier diagnosis of PE in patients presenting syncope.
PE is included in the differential diagnosis in the syncope guidelines, but those for the diagnosis of acute PE pay little attention to establishing a diagnostic workup in these patients, considering that there is no evidence to support additional pharmacological or invasive treatments other than those indicated by the current guidelines.1,7,10,23
Finally, our study indicated that patients with syncope more commonly present a centrally located thrombus and highlighted the fact that syncope can be a sign of a life-threatening PE, which should be considered in the differential diagnosis of every syncopal event in the ED and care units.
This study has some limitations, beginning with the fact that it is not multicentric. To date, no such studies have focused on PE among Romanian patients. A national registry is in progress, which is sustained by the Romanian Society of Cardiology. The small number of patients with PE is a limiting factor in the interpretation of the results. We tried to adopt a standard approach to the evaluation of syncope, but a number of patients with PE may have had syncope that was associated with another condition and was missed. We can say nothing of those who may have died of it because the rate of autopsies drastically decreased over the previous years.
PE was confirmed in 11.47% of patients who were hospitalized for a first episode of syncope, who were not receiving anticoagulation therapy, and in 2.52% of patients who visited the ED for syncope. Almost all the patients had a proximal obstruction of the pulmonary artery, half of them bilateral, which clearly explains the clinical presentation of syncope. The occurrence of syncope, if not explained otherwise, should alert one to consider PE. A prompt diagnostic approach to exclude PE should be performed when syncope is associated with specific symptoms or signs of PE such as dyspnea, persistent noncardiac chest pain, prolonged immobilization, leg swelling, unexplained tachycardia, or unexplained hypoxia.
1. Brignole M, Moya A, De Lange FJ, et al. 2018 ESC Guidelines for the diagnosis and management of syncope
: the task force for the diagnosis and management of Syncope
of the European Society of Cardiology. Eur Heart J. 2018;00:1–69.
2. Pollack CV, Schreiber D, Goldhaber SZ, et al. Clinical characteristics, management, and outcomes of patients diagnosed with acute pulmonary embolism
in the emergency department: initial report of EMPEROR (Multicenter Emergency Medicine Pulmonary Embolism
in the Real-World Registry). J Am Coll Cardiol. 2011;57:700–706.
3. Thames MD, Alpert JS, Dalen JE. Syncope
in patients with pulmonary embolism
. JAMA. 1977;238:2509–2511.
4. Ferrari E, Baudouy M, Cerboni P, et al. Clinical epidemiology of venous thromboembolic disease: results of a French multicentre registry. Eur Heart J. 1997;18:685–691.
5. Oqab Z, Ganshorn H, Sheldon R. Prevalence of pulmonary embolism
in patients presenting with syncope
: a systematic review and meta-analysis. Am J Emerg Med. 2018;36:551–555.
6. Prandoni P, Lensing AWA, Prins MH, et al. Prevalence of pulmonary embolism
among patients hospitalized for syncope
. N Engl J Med. 2016;375:1524–1531.
7. Shen WK, Sheldon RS, Benditt DG, et al. 2017 ACC/AHA/HRS guideline for the evaluation and management of patients with syncope
: a report of the American College of Cardiology/American Heart Association Task force on clinical practice guidelines and the Heart Rhythm Society. Circulation. 2017;136:e25–e59.
8. van Belle A, Buller HR, Huisman MV, et al. Effectiveness of managing suspected pulmonary embolism
using an algorithm combining clinical probability, d-dimer testing, and computed tomography. JAMA. 2006;295:172–179.
9. Wells PS, Anderson DR, Rodger M, et al. Derivation of a simple clinical model to categorize patient's probability of pulmonary embolism
: increasing the models utility with the SimpliRED D-dimer. Thromb Haemost. 2000;83:416–420.
10. Konstantinides S, Torbicki A, Agnelli G, et al. Guidelines on the diagnosis and management of acute pulmonary embolism
: the task force for the Diagnosis and Management of Acute Pulmonary Embolism
of the European Society of Cardiology (ESC). Eur Heart J. 2014;35:3033–3069.
11. Frizell A, Fogel N, Steenblik J, et al. Prevalence of pulmonary embolism
in patients presenting to the emergency department with syncope
. Am J Emerg Med. 2018;36:253–256.
12. Thiruganasambandamoorthy V, Sivilotti M, McRae A, et al. LO01: prevalence of pulmonary embolism
among Canadian emergency department patients with syncope
: a multicenter prospective cohort study. Can J Emerg Med. 2017;19:S27.
13. Castelli R, Tarsia P, Tantardini C, et al. Syncope
in patients with pulmonary embolism
: comparison between patients with syncope
as the presenting symptom of pulmonary embolism
and patients with pulmonary embolism
. Vasc Med. 2003;8:257–261.
14. Jenab Y, Lotfi-Tokaldany M, Alemza-deh-Ansari MJ, et al. Correlates of syncope
in patients with acute pulmonary thromboembolism. Clin App Thromb Hemost. 2015;21:772–776.
15. Wiener RS, Schwartz LM, Woloshin S. Time trends in pulmonary embolism
in the United States: evidence of overdiagnosis. Arch Intern Med. 2011;171:831–837.
16. Pallin DJ, Dressler DD. Pulmonary embolus in syncope
patients: maybe not so common, after all. NEJM Journal Watch. December 4, 2017.
17. Stein PD, Willis PW III, De Mets DL. History and physical examination in acute pulmonary embolism
in patients without preexisting cardiac or pulmonary disease. Am J Cardiol. 1981;47:218–223.
18. Goldhaber SZ, Visani L, De Rosa M. Acute pulmonary embolism
: clinical outcomes in the International Cooperative Pulmonary Embolism
Registry (ICOPER). Lancet. 1999;353:1386–1389.
19. Kasper W, Konstantinides S, Geibel A, et al. Management strategies and determinants of outcome in acute major pulmonary embolism
: results of a multicenter registry. J Am Coll Cardiol. 1997;30:1165–1171.
20. Morpurgo M, Zonzin P. Syncope
in acute pulmonary embolism
. Ital Heart J. 2004;5:3–5.
21. Costantino G, Martin H, Quinn J, et al. Prevalence of pulmonary embolism
in patients with syncope
. JAMA Intern Med. 2018;178:356–362.
22. Seyyedi SR, Jenab Y, Tokaldany ML, et al. Syncope
paradox in the outcome of patients with pulmonary thromboembolism: short-term and midterm outcome. Clin Respir J. 2016;10:90–97.
23. Raja AS, Greenberg JO, Qaseem A, et al. Evaluation of patients with suspected acute pulmonary embolism
: best practice advice from the clinical Guidelines Committee of the American College of Physicians. Ann Intern Med. 2015;163:701–711.