Article Text

Signs and symptoms of acute pulmonary embolism and their predictive value for all-cause hospital death in respect of severity of the disease, age, sex and body mass index: retrospective analysis of the Regional PE Registry (REPER)
  1. Dušan Predrag Ružičić1,
  2. Boris Dzudovic2,3,
  3. Jovan Matijasevic4,5,
  4. Marija Benic4,
  5. Sonja Salinger6,7,
  6. Ljiljana Kos8,
  7. Tamara Kovacevic-Preradovic8,9,
  8. Irena Mitevska10,
  9. Aleksandar Neskovic11,12,
  10. Bjanka Bozovic13,
  11. Nebojsa Bulatovic13,14,
  12. Vladimir Miloradovic15,16,
  13. Ivica Djuric17 and
  14. Slobodan Obradovic3,17
  1. 1Department of Internal Medicine and Invasive Cardiology, General Hospital Valjevo, Valjevo, Serbia
  2. 2Clinic Of Emergency Internal Medicine, Military Medical Academy, Belgrade, Serbia
  3. 3School of Medicine, University of Defense, Belgrade, Serbia
  4. 4Institute of Pulmonary Diseases of Vojvodina, Novi Sad, Serbia
  5. 5School of Medicine, University of Novi Sad, Novi Sad, Serbia
  6. 6Clinic of Cardiology, Clinical Center Nis, Nis, Serbia
  7. 7School of Medicine, University of Nis, Nis, Serbia
  8. 8Clinic of Cardiology, Clinical Center Banja Luka, Banja Luka, Bosnia and Herzegovina
  9. 9School of Medicine, University of Banja Luka, Banja Luka, Bosnia and Herzegovina
  10. 10Clinic of Cardiology, Intensive Care Unit, School of Medicine Skopje, University of Skopje, Skopje, North Republic of Macedonia
  11. 11Clinic of Cardiology, Clinical Center Zemun, Belgrade, Serbia
  12. 12School of Medicine, University of Belgrade, Belgrade, Serbia
  13. 13Clinic of Cardiology, Clinical Center Podgorica, Podgorica, Montenegro
  14. 14School of Medicine, University of Podgorica, Podgorica, Montenegro
  15. 15Clinic of Cardiology, Clinical Center Kragujevac, Kragujevac, Serbia
  16. 16School of Medicine, University of Kragujevac, Kragujevac, Serbia
  17. 17Clinic of Cardiology, Military Medical Academy, Belgrade, Serbia
  1. Correspondence to Professor Slobodan Obradovic; sloba.d.obradovic{at}


Background The incidence of the signs and symptoms of acute pulmonary embolism (PE) according to mortality risk, age and sex has been partly explored.

Patients and methods A total of 1242 patients diagnosed with acute PE and included in the Regional Pulmonary Embolism Registry were enrolled in the study. Patients were classified as low risk, intermediate risk or high risk according to the European Society of Cardiology mortality risk model. The incidence of the signs and symptoms of acute PE at presentation with respect to sex, age, and PE severity was investigated.

Results The incidence of haemoptysis was higher in younger men with intermediate-risk (11.7% vs 7.5% vs 5.9% vs 2.3%; p=0.01) and high-risk PE (13.8% vs 2.5% vs 0.0% vs 3.1%; p=0.031) than in older men and women. The frequency of symptomatic deep vein thrombosis was not significantly different between subgroups. Older women with low-risk PE presented with chest pain less commonly (35.8% vs 55.8% vs 48.8% vs 51.9%, respectively; p=0.023) than men and younger women. However, younger women had a higher incidence of chest pain in the lower-risk PE group than in the intermediate-risk and high-risk PE subgroups (51.9%, 31.4% and 27.8%, respectively; p=0.001). The incidence of dyspnoea (except in older men), syncope and tachycardia increased with the risk of PE in all subgroups (p<0.01). In the low-risk PE group, syncope was present more often in older men and women than in younger patients (15.5% vs 11.3% vs 4.5% vs 4.5%; p=0.009). The incidence of pneumonia was higher in younger men with low-risk PE (31.8% vs<16% in the other subgroups, p<0.001).

Conclusion Haemoptysis and pneumonia are prominent features of acute PE in younger men, whereas older patients more frequently have syncope with low-risk PE. Dyspnoea, syncope and tachycardia are symptoms of high-risk PE irrespective of sex and age.

  • Pulmonary Embolism

Data availability statement

Data are available in a public, open access repository.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

What is already known on this topic

  • Haemoptysis, deep vein thrombosis, chest pain, dyspnoea, syncope, tachycardia and secondary pneumonia are associated with acute pulmonary embolism (PE).

What this study adds

  • The incidence of the signs and symptoms of acute PE varies depending on disease severity, sex, age and body mass index (BMI).

How this study might affect research, practice or policy

  • Awareness that the signs and symptoms of acute PE depend on disease severity, sex, age, and BMI might contribute to its prompt diagnosis and treatment.


Recent advances in the diagnosis and treatment of acute pulmonary embolism (PE) have significantly improved patient prognoses.1 However, its clinical presentation is variable, and proper management depends on prompt diagnosis.2–4 The most common symptoms of PE are dyspnoea, pleuritic chest pain and cough.4–7 The correlation between clinical presentation and PE severity has not been investigated, and this relationship may change over time. Thus, there is some recent evidence that age and the presence of comorbidities could be associated with some clinical presentations of acute PE, such as pulmonary infarction, which has been demonstrated more in younger patients with fewer comorbidities,8 antiphospholipid syndrome in sarcoidosis9 or acute PE without deep vein thrombosis (DVT) in SARS-CoV-2-positive patients.10 Current guidelines11 suggest using pretest probability scores, such as the revised Le Gal et al12 Kline and Wells13 and YEARS,14 together with D-dimer testing to select normotensive patients with suspected PE for CT-pulmonary angiography (CT-PA) for final diagnosis. These tools use demographic data (age), anamnestic data (history of malignancies, recent surgery, immobilisation or previous venous thromboembolism), signs (tachycardia, DVT and haemoptysis) and symptoms (unilateral lower limb pain).15 However, the clinical presentation varies depending on the severity of PE. For instance, acute severe PE may be oligosymptomatic, presented only as dyspnoea and tachycardia, and represented as a low probability using all three pretest probability scores.4 11 High-risk PE may be easily overlooked because of the lack of common elements in pretest probability rules (eg, haemoptysis and signs of DVT) that are elements of the pretest scores, but they might be rarely present in high-risk PE patients. Thus, the suspicion of acute PE is based on pretest probability scores in emergency settings, where inexperienced doctors often try to diagnose acute PE that can be presented with multiple manifestations and is mixed with the symptoms and signs of adjacent comorbidities.

A better understanding of the association of some important patient characteristics, such as age, sex and body mass index (BMI), with the clinical presentation of acute PE of different severity, is needed for better diagnosis of the disease. This study aimed to investigate the relationship between the clinical presentation of acute PE, PE severity and patient age and sex. We examined the incidence of major clinical signs and symptoms of acute PE in younger and older (according to median age) men and women with respect to the European Society of Cardiology (ESC) mortality risk model.11


The study described here is the retrospective, observational cohort study of the regional acute PE registry (REPER). This study included 1242 consecutive patients with acute PE enrolled from 2015 to the end of 2021, and five university clinics in Serbia (Military Medical Academy (MMA) Belgrade, Institute of Pulmonary Diseases Vojvodina, Clinical Center Nis, Clinical Center Kragujevac and Clinical Center Zemun) and one general hospital (Pancevo), together with the Clinical Center of Banja Luka (Bosnia and Herzegovina) and the Clinical Center from Podgorica (Montenegro), were included in the REPER. All patients had signs and symptoms of acute PE in the previous 14 days, and a thrombus was diagnosed in at least one segmental pulmonary artery on CT-PA at admission. The attending physicians obtained the baseline data.


Women and men were assigned to two groups according to their median age (younger or older than the median age). Patients were also divided into three subgroups according to BMI: normal weight 18.5–24.9 kg/m2, overweight 25.0–29.9 kg/m2 and obese with BMI >30 kg/m2. There were only two patients who had BMI <18.5 kg/m2, and they were not included in the analysis of the association between BMI and PE mortality risk score (online supplemental table S1).

Risk stratification

Patient risk stratification was performed using the ESC mortality risk model11 with three strata: low-risk patients, intermediate-risk patients (normotensive patients with right ventricle (RV) dysfunction on CT-PA and/or echocardiography at admission or with a simplified Pulmonary Embolism Severity Index (sPESI)>0) and high-risk patients who were hypotensive or resuscitated at admission. Three-risk strata, instead of four, were used because the mortality risk of intermediate-low-risk and intermediate-high-risk PE patients was similar in the study by Beccattini et al,16 where the ESC mortality risk score was validated for the prediction of PE-related death in consecutive patients from the registry. The second reason for not using a four-risk stratum model was to avoid more complex, multiple relationships with smaller cohorts of patients.

In addition to the ESC mortality risk score, to estimate PE severity, we used the sPESI as a two-level score (0, low risk; >1, high risk).17

Diagnostic methods

ECG was performed on admission. The typical PE ECG signs were recorded (S1Q3T3 sign, right bundle branch block (RBBB), negative T-waves in the precordial leads, S-waves in the aVL and paroxysmal atrial fibrillation).18

To define right ventricular dysfunction, transthoracic echocardiography was performed in most patients at admission, and the systolic pressure in the RV was measured indirectly using tricuspid regurgitation velocity.19 Cardiac transthoracic ultrasound was performed on admission in 1117/1242 patients (89.9%). We have no precise data on compression ultrasound (CUS) examinations during hospitalisation; however, the main approach was to perform CUS in patients with symptomatic lower limb DVT.

High-sensitivity troponin concentration was measured on admission using commercial laboratory kits.

The signs and symptoms of acute PE were assessed according to sex, age and PE risk (low, intermediate and high). The symptoms and signs of PE used in this study are as follows:

Haemoptysis was defined as the presence of bloody sputum according to the patient’s anamnestic claims, or if it became obvious that the patient had haemoptysis during physical examination.

Symptomatic DVT was defined as lower limb oedema with or without pain. In patients with these symptoms, CUS venography was used to confirm the DVT diagnosis, but it was not used routinely in all patients without symptoms of DVT, although we are aware that some patients could have asymptomatic DVT.

Chest pain is defined as a symptom of pleural pain or retrosternal pain in patients. Unfortunately, from the beginning of our registry, we did not classify these two types of chest pain and could not analyse them separately.

Dyspnoea was defined as breathing effort required at rest or ‘lack of air’ reported by the patient or detected during physical examination.

Syncope was defined as a temporary loss of consciousness with spontaneous full recovery. Tachycardia was defined as 100 pm on ECG at admission.

Pneumonia was defined as fever with condensation of pulmonary tissue on the CT-PA.


Patients were informed during hospitalisation that their medical data, without revealing their identity, would be recorded in the registry of acute PE patients, and their informed consent was obtained. The Ethical Committee of the Military Medical Academy Belgrade, the central institution of the registry and institutional review boards from other hospitals involved gave us permission to publish the results from the registry (code numbers for the decisions of EC are 20/2022, 3/2021 and 160/2019).

Statistical analyses

Data are presented as frequencies, means and SD. The significance of the differences in the incidence of the signs and symptoms of acute PE among the subgroups according to age, sex, BMI and PE mortality risk was assessed using the χ2 test. P values for the significance of differences in the frequencies of symptoms across several cohorts were presented as unadjusted and adjusted p values using the Bonferroni method. The predictive value of the signs and symptoms of PE for all-cause in-hospital mortality was estimated using univariate analysis and three regression models. The first model was adjusted for age and sex, the second for BMI, and the third one was adjusted for age, sex and mortality risk. The results are shown in online supplemental table S2.

Statistical analyses were performed using SPSS Statistics V.19.0 (IBM Corporation, Armonk, New York). The level of significance was set at p<0.05.

Patient and public involvement statement

Patients were not actively involved in the research.


The patient characteristics are presented in table 1. The median ages of men and women were 63 years (IQR, 49–71 years) and 69 years (IQR, 66–77.5 years), respectively. Patient characteristics varied across the subgroups. The prevalence of chronic obstructive pulmonary disease (COPD), chronic heart failure (CHF), coronary artery disease (CAD), previous stroke, diabetes and hypertension was higher in older subgroups (table 1). In contrast, the occurrences of DVT and PE were similar in all four cohorts. The presence of malignant disease did not differ across the subgroups. The presence of typical risk factors for PE, such as immobilisation and major surgery, was higher in women than in men (table 1). The incidence of sPESI scores >0 (moderate and high risk) and the need for resuscitation during hospitalisation was higher in older patients.

Table 1

Baseline characteristics of study patients

The signs and symptoms of acute PE across the four cohorts with respect to age, sex and mortality risk are presented in table 2, and those with respect to BMI are presented in online supplemental table S2.

Table 2

Symptoms and signs of acute PE at presentation to hospital in respect of PE European Society of Cardiology mortality risk model, age and sex9


The incidence of haemoptysis in the intermediate-risk and high-risk PE groups differed significantly among the four cohorts. In intermediate-high risk PE, both younger and older men had a higher incidence of haemoptysis compared with younger and older women (11.7% and 7.5% vs 5.9% and 2.3%, respectively, p=0.010); younger men also had the highest frequency of haemoptysis in high-risk PE, compared with the much lower incidence in the other three cohorts (13.8% vs 2.5% vs 0.0% and 3.7%, p=0.031, in younger men vs older men vs younger women and older women, respectively). In the cohort of younger women, the incidence of haemoptysis decreased with an increase in PE risk (10.5% vs 5.9% vs 0.0%, p=0.047 for low-risk vs intermediate-risk vs high-risk PE, respectively).

Symptomatic lower limb DVT

The incidence of lower limb DVT was 40%–45% and 30%–35% in patients with low-risk and high-risk PE, respectively, without significant differences in distribution across the subgroups of patients.

Chest pain

In lower risk PE patients, chest pain was often present in men and younger women, but rarely in older women (55.5% vs 48.8% vs 51.9% vs 35.8%, p=0.023, in younger men vs older men vs younger women vs older women, respectively). Older men (48.8% vs 34.7% vs 30.0%, p=0.048) and younger women (51.9% vs 36.8% vs 22.7%, p=0.001) had a higher incidence of chest pain in the low-risk PE group then in intermediate-risk and high-risk PE groups.


The incidence of dyspnoea did not differ across the four cohorts in either of the PE severity risk strata. However, the incidence of dyspnoea increased significantly in every cohort from low-to-high-risk PE, except in older men, where the incidence of dyspnoea varied non-significantly from low-to-high-risk patients (78.6%–87.5%, respectively).


A considerable proportion of older men and women in the low-risk PE cohort had syncope at presentation, which was rare but still recorded in younger men and women (15.5% and 11.3% in older men and women, respectively, vs 4.5% in both younger subgroups, p=0.009). The incidence of syncope significantly increased with the risk of PE and was lower in younger patients with low-risk PE.


The incidence of tachycardia was higher in younger men and women with intermediate-high PE (57.9% and 52.6% in younger men and women vs 42.8% and 40.15 in older men and women, p=0.004). The frequency of tachycardia significantly increased with PE severity, especially in younger women; 84.1% of patients with high-risk PE had tachycardia at admission.


Pneumonia at presentation was more often presented in younger men with low-risk (31.8% vs 15.3% vs 10.5% vs 14.2%in younger men compared with older men and younger and older women, p<0.001) and intermediate-risk PE (25.5% vs 17.5% vs 18.4% vs 12.9%, p=0.036 in the same order). There were no differences in the incidence of pneumonia with respect to the PE mortality risk.

BMI and clinical presentation of acute PE

The results regarding BMI and the frequencies of different symptoms of acute PE are presented in online supplemental table S1. Patients with normal weight who had intermediate-risk PE had more frequent haemoptysis than overweight and obese patients (10.5% vs 2.7% vs 2.9%, p=0.001, respectively) and pneumonia at presentation (26% vs 13.6% vs 15.4%, p=0.003). Symptomatic DVT was more frequent in overweight patients with intermediate-risk PE (49.5% vs 36.4% vs 39.3% in overweight vs normal-weight vs obese patients, p=0.023). In low-risk patients, the frequency of dyspnoea increased with BMI (84.2% vs 76.6% vs 65.7% in obese vs overweight vs normal-weight patients, p=0.009). The frequency of syncope and tachycardia increased significantly with PE severity (p>0.001) in all three patient cohorts with respect to BMI.

All-cause hospital mortality in different cohorts of patients and association with symptoms

The in-hospital mortality rate in our cohort was 11.1% (138/1242). The in-hospital mortality rates were 2.8% (12/434), 10.3% (66/641) and 35.9% (60/167) in the low-risk, intermediate-risk and high-risk PE cohorts, respectively. Considering subgroups of patients according to age and sex, in-hospital mortality was 7.0% (20/284), 8.2% (27/329), 12.5% (37/297) and 16.3% (54/332) in the younger male, older male, younger female and older female cohorts, respectively. Interestingly, the hospital mortality rate was higher in normal-weight patients (15.0% (62/414)) than in overweight patients (7.0% (25/357)) and obese patients (9.8% (24/244)) (p=0.002).

Symptoms and signs of PE and hospital mortality

In the univariate analysis, lower limb DVT and chest pain were negatively associated with all-cause mortality, whereas dyspnoea, tachycardia and pneumonia were positively associated with all-cause mortality (online supplemental table S2). The results of the regression analysis, adjusted for age and sex, were similar to those of the univariate analysis (online supplemental table S2). In model 2, in which BMI was adjusted for study symptoms, dyspnoea, syncope, tachycardia and pneumonia were associated with higher hospital mortality, and the presence of symptomatic DVT and chest pain was associated with lower hospital mortality (online supplemental table S2). Model 3 was adjusted for age, sex and ESC mortality risk stratification (online supplemental table S2) and showed that the signs and symptoms of lower limb DVT were negatively associated with all-cause hospital mortality, whereas pneumonia was positively associated with this outcome.


To the best of our knowledge, this is the first study to measure the incidence of the signs and symptoms of acute PE according to age, sex, BMI and PE severity. We demonstrated that the signs and symptoms of acute PE depend on the severity of PE and patient characteristics. However, PE diagnosis remains challenging. The suspicion of PE was based on pretest probability scores, and the diagnosis was confirmed using MDCT-PA.11

This study analysed the incidence of symptoms and signs commonly present in acute PE according to severity, age, sex and BMI. We assume that these factors influence the clinical picture of acute PE and that clinicians should be aware of these relationships.

Haemoptysis is a typical symptom of acute PE and is included in the Geneva revised score and the Wells score.12 13 The reported incidence of haemoptysis in patients with acute PE, regardless of PE severity, was 6%–13%.20–23 In our cohort, the incidence of haemoptysis varied depending on age and sex, with younger men having an incidence of haemoptysis between 11.5% and 16.5% in all strata of PE risk, whereas the incidence of haemoptysis in the other three subgroups was <5.0% in high-risk PE. The presence of haemoptysis is a consequence of pulmonary infarction, and its presence was more likely to be detected in younger patients according to the investigation by Ruaro et al.9 However, this study enrolled 493 patients with acute PE, and there were no data about the association of sex with the presence of pulmonary infarction, especially with respect to the mortality risk score. In that study, the incidence of pulmonary infarction decreased significantly with increasing PE severity, similar to our results.

The frequency of symptomatic lower limb DVT was similar across age, sex and PE mortality risk in our cohort, but the incidence of it was about 7%–11% higher in all cohorts in the low-risk group compared with high-risk groups (the difference did not reach significance in each cohort). These results are partly in concordance with the results of the study by Wang et al,24 who found that symptomatic lower extremity DVT is more often present in low-risk PE patients (63.0%) than in intermediate-risk and high-risk PE patients (36.8%). Patients with intermediate-risk and high-risk PE also have a higher incidence of distal DVT than those with low-risk patients. In our study, we found that the presence of symptomatic DVT was associated with lower hospital mortality. The findings of Wang et al could explain the discrepancy between our finding and the meta-analysis by Becattini et al16 about the prediction value of DVT for the prognosis of acute PE, where the presence of DVT was associated with higher 30-day all-cause death with OR of 1.9 (95% CI 1.5 to 2.4). In this meta-analysis, the diagnosis of DVT was established using CUS in all patients, and the incidence of DVT was calculated as the sum of symptomatic and asymptomatic DVT. As we can see from the study by Wang et al, patients with intermediate-risk and high-risk PE more often have silent DVT and distal symptomatic DVT, which might be explained by the higher embolic potential in patients with intermediate-risk and high-risk PE. We showed that symptomatic DVT, which is more characteristic of lower risk PE, had a better prognosis, and the symptomatic nature of DVT with occluded veins might be explained by the lower embolic capacity and consequently, a better outcome.

In our registry, the incidence of chest pain tended to decrease with an increase in PE risk, especially in younger women. However, in older women, the incidence of chest pain was very similar across the strata of PE mortality risk. Chest pain may involve pleuritic pain, which is more common in peripheral PE and is associated with pulmonary infarction, and myocardial ischaemia-derived central chest pain, which is more common in patients with more severe PE. Unfortunately, we did not separate pleuritic pain from central pain, which may have had an opposite distribution in relation to PE severity. In the PIOPED II (Prospective Investigation of Pulmonary Embolism Diagnosis) registry, pleuritic pain was more common in patients younger than 70 years, which is similar to our study, except that younger women had a significantly lower incidence of chest pain in the high-risk PE group.4 In our investigation, chest pain at presentation of PE was similar in younger cohorts with low-risk and intermediate-risk PE. However, older women had a lower incidence of chest pain in the low-risk PE group than the other three cohorts, and younger women had a lower incidence of chest pain in the high-risk group. Several studies did not find a difference in the incidence of chest pain between sexes,4 22 but they did not consider PE severity as a modifying factor.

Acute dyspnoea is a hallmark of acute PE. The incidence of dyspnoea in patients with acute PE is 66%–82%.4 22 23 In our study, there was a strong association between the incidence of dyspnoea and PE severity, which is consistent with the results of other studies. However, even patients with high-risk PE may not present with dyspnoea, especially older men. Additionally, our data showed that between 20% and 30% of patients with low-risk PE and 8% and 12% with intermediate-risk PE did not complain of dyspnoea, which may deceive the diagnosis of acute PE. The incidence of dyspnoea may correlate with BMI. Thus, overweight and obese patients had significantly more dyspnoea than normal-weight low-risk patients with PE in our study. The incidence of dyspnoea is common in obese patients, even in those without any particular disease, and we must consider PE a differential diagnosis in obese patients with worsening dyspnoea or dyspnoea with exertion.25 Furthermore, rapidly progressive dyspnoea is correlated with acute PE and haemodynamic instability.2 9 In the PIOPED II study,4 dyspnoea was less frequently observed in patients older than 70 years; however, the study did not examine sex-related and PE severity-related differences in dyspnoea. In our study, in the high-risk PE group, older women had the highest incidence of dyspnoea (98.1%); however, older men had the lowest incidence (87.5%), which may be an obstacle to the diagnosis of high-risk PE in older men.

The incidence of syncope tended to increase with the risk of PE, and in patients with high-risk PE was 20%–42%. Additionally, younger patients of both sexes with low-risk PE had a lower incidence of syncope (4.5% in both younger men and women). However, in our study, syncope at PE presentation was relatively common in low-risk older men (15.5%) and women (11.3%), which means that PE must be considered in older patients with syncope, even in those without features of severe PE. This finding is in agreement with a study by Prandoni et al, who found PE in 17.3% of 560 consecutive patients with syncope, of whom 37% did not have a detected thrombus in the main and lobar pulmonary arteries.26 The presence of syncope is considered a feature of severe PE, and patients with syncope and acute PE usually present with haemodynamic instability and right ventricle dysfunction and have adverse short-term outcomes.27 28 However, syncope at presentation may only be associated with early death in women.29 Syncope may be due to the passage of a thrombus through the right ventricle, causing transient arrhythmia, or the presence of a large thrombus in the main pulmonary arteries or right ventricle, causing an obstruction of blood flow with a sudden decrease in arterial pressure or activation of the vaso-vagal reflex.30 The first mechanism may be responsible for syncope in patients with low-risk PE.

The incidence of tachycardia increased with PE severity and tended to be lower in older men, probably due to the higher use of beta-blockers. Consistent with our findings, the CURES (China pUlmonary Thromboembolism REgistry Study) study28 found that the frequency of palpitations was higher in patients with syncope and haemodynamic instability, and the RIETE registry reported that the incidence of tachycardia (≥110 bpm) was higher in patients with massive PE.31 Tachycardia is a common symptom of PE, especially in older patients, and is associated with a worse prognosis.32 However, it is important to note that, in our study, almost 35.0% of older men and 28.8% did not have tachycardia >100 bpm in patients with high-risk PE.

In our study, the incidence of pneumonia was higher in younger men then in other cohorts—in all three strata of PE severity. Interestingly, pneumonia was more common in normal-weight patients with intermediate-risk PE than in overweight or obese patients. Using a single-centre registry, our group has shown that pneumonia with pulmonary parenchymal condensation and fever are more common in younger male patients.3 However, the correlation between pneumonia, as a clinical manifestation of PE, and its severity is unknown. In the present study, the incidence of pneumonia did not depend on the severity of PE across the four cohorts. Younger patients, especially men, are prone to developing pulmonary infarction, which is characteristic of peripheral PE, but pneumonia in intermediate-risk and high-risk patients probably has a different pathophysiology, where stasis and severe airflow-to-blood flow mismatch contribute to the bacterial infection with a much more serious prognosis. In the ICOPER (International Cooperative Pulmonary Embolism Registry) registry,22 atelectasis on chest radiography was more common in men than in women, whereas the incidence of fever was similar between the sexes.

We found that the early presence of pneumonia was an independent predictor of hospital death regardless of PE severity, and that early recognition and therapy of this complication might be very important for survival.

Limitations of the study

Since this was a retrospective analysis of the registry data, some records about symptoms and signs of acute PE were missing or incorrect, and bias towards higher missing data in high-risk patients is likely to exist. However, as the registry was raised, all physicians involved tried to record medical data about patients as soon as possible during the hospitalisation phase, and we think that the quality of data was sufficient to write this manuscript. We did not use the four-strata ESC model for risk stratification, that is, we did not divide the intermediate-risk group into intermediate-risk low-risk and intermediate-high-risk groups. There are three main reasons for this finding. First, we want to avoid the complexity of interpreting results by multiplying subgroups; second, we consider that intermediate-high and intermediate-low-risk subgroups are more similar in the prediction of mortality (Becattini et al), compared with low-risk and high-risk subgroups; third, levels of troponin add little to echocardiography parameters for mortality prediction in intermediate-risk PE patients.33 Unfortunately, we did not separate pleural chest pain from central chest pain (probably due to myocardial ischaemia) because these two symptoms may have different distributions of ESC mortality risk across the main cohorts examined in our study. Pleuritic pain was two times as frequent as non-pleuritic pain in the PIOPED II study, which was probably also the case in our study. Only symptomatic DVT was analysed in our study, but this is in course of the main purpose of this study was actually an association between symptoms associated with acute PE and the main characteristics of the patients through the ESC mortality risk score. Additionally, some symptomatic DVT may have been missed, especially in high-risk PE, where attending physicians did not record the presence of symptomatic DVT in critically ill patients. For physicians who attempt to diagnose acute PE, it is important to know that the presence of characteristic symptoms of the disease might depend greatly on the severity of the disease and that the clinical picture is significantly modulated by the patients’ age, sex and BMI. This work is a comprehensive aid; however, as it is registry based and has several limitations, prospective studies on this topic are warranted.


The incidence of haemoptysis was higher in younger men with intermediate and high-risk PE, and on the contrary, haemoptysis is very rare in younger women with high-risk PE. The presence of symptomatic DVT was more often observed in all cohorts with low-risk and intermediate-risk PE than in those with high-risk PE; however, this was not statistically significant. The presence of chest pain was much more common in younger patients of both sexes, and it remained frequent in high-risk younger men, but much less frequent in younger women. The incidence of dyspnoea was higher in all cohorts with respect to PE severity, except in older men, where the incidence of dyspnoea was similar across the PE severity risk strata. The incidence of syncope was higher in both older subgroups in low-risk patients and increased in all cohorts with PE severity, reaching a maximum in older men with high-risk PE, where 42.5% of patients presented with syncope. Tachycardia was more pronounced in younger patients of both sexes with intermediate-risk PE, and its frequency significantly increased with the risk of PE in all cohorts. The presence of pneumonia at presentation was more distinct in younger men with low-risk and intermediate-risk PE than in the other cohorts. Considering BMI, haemoptysis and pneumonia were more marked in normal-weight patients with intermediate-risk PE, while dyspnoea was a prominent symptom in the low-risk obese cohort. Among all the analysed symptoms and signs, only the presence of pneumonia was associated with higher hospital death, whereas symptomatic DVT was associated with a lower mortality risk, independent of age, sex and ESC mortality risk score.

Data availability statement

Data are available in a public, open access repository.

Ethics statements

Patient consent for publication

Ethics approval

The Ethical Committee of the Military Medical Academy Belgrade, the central institution of the registry and institutional review boards from other hospitals involved gave us permission to publish the results from the registry (code numbers for the decisions of EC are 20/2022, 3/2021 and 160/2019). Participants gave informed consent to participate in the study before taking part.


This work is done under the project 'Risk factors for hospital death outcome, bleeding and late complications in acute Pulmonary Embolism patients' code: MFVMA01/22-24, supported by the Ministry of Defense of the Republic of Serbia, but till now without any financial support, only paper support.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 5.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Collaborators none.

  • Contributors DPR and SO had full access to all data and were responsible for the data integrity and accuracy of the data analysis. Concept and design: DPR and SO. Data acquisition: MB, SS, LK, IM and BB. Data analysis or interpretation: DPR, BD and SO. Drafting of the manuscript: DPR and SO. Critical revision of the manuscript for important intellectual content: all authors. Statistical analysis: DPR, BD and SO. Supervision of the study and registry: JM, AN, VM, NB and SO. The guarantor of the manuscript is SO, and accepts full responsibility for the work, conduct of the study, had access to all data, and controlled the decision to publish.

  • Funding This work was not financially supported by any organisation.

  • Competing interests The interpretation of the data and presentation of this investigation were not influenced by personal or financial relationships with other people or organisations.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.