Pulmonary Embolism: Causes and How Doctors Diagnose It
Updated
A pulmonary embolism (PE) is a blockage in an artery of the lungs, and in most people it is caused by a blood clot that formed in a deep vein of the leg or pelvis, broke loose and travelled to the lungs. Surgery, injury, immobility, cancer, hormones, pregnancy and a previous clot all make that more likely, yet in about 40% of people with PE no predisposing factor is found. Because the symptoms overlap with many other conditions, doctors don't diagnose it from symptoms alone: they follow a pathway of scores, a blood test and a scan. This guide covers the causes, then each step of that pathway, from the Wells score to the risk class that guides treatment.
- Usual causeA deep vein clot from the leg or pelvis (NHLBI)
- No trigger foundIn about 40% of people with PE (ESC 2019)
- Ruling it outWells or PERC, then D-dimer, age-adjusted after 50
- Confirming itCT pulmonary angiography, or a V/Q scan
Each step of that pathway has its own cut-off, and the reason for a doctor's choice is rarely explained at the bedside. AskMed is a medical question app made for doctors, nurses, pharmacists, medical students and researchers. Instead of answering from memory, it searches clinical guidelines, journals and drug labels first, then writes an answer with a link after every claim. Ask it why a doctor might skip the D-dimer test and go straight to a CT scan, and it explains the role of clinical probability, citing the ESC guideline and the studies behind it.
Ask AskMed about a step in the pulmonary embolism workup:
What causes a pulmonary embolism?
A pulmonary embolism is caused by material that travels through the veins and lodges in a lung artery, and in the great majority of cases that material is a blood clot. The US National Heart, Lung, and Blood Institute (NHLBI) describes pulmonary embolism as an embolus from a deep vein blood clot, or from a blood vessel in the lungs, that breaks loose and blocks an artery in the lung. The NHLBI explains three ways such a clot forms: blood flow in the veins slows down; veins are damaged by surgery or trauma; and changes in the body make the blood more likely to clot. Because the leg clot and the lung clot are one disease, doctors group them as venous thromboembolism (VTE). The 2019 European Society of Cardiology (ESC) guideline puts the yearly incidence of pulmonary embolism at 39 to 115 per 100,000 people, and VTE is almost eight times more common at age 80 or over than in people in their forties.
Less often, the blockage is not a blood clot at all. A 2009 review in the European Respiratory Journal by Jorens and colleagues defines nonthrombotic pulmonary embolism as the arrival in the lung circulation of cells, such as fat cells, amniotic fluid cells, placental cells or tumour cells, or of bacteria, fungi, foreign material or gas. These forms often cause unusual signs that are easily overlooked, can set off strong inflammation in the lungs, and are sometimes confirmed only by examining tissue. They are rare and follow their own diagnostic route, so the rest of this guide is about the common, clot-based pulmonary embolism.
Which risk factors make a pulmonary embolism more likely?
The risk factors for pulmonary embolism are those for a deep vein clot, and the ESC 2019 guideline ranks them by how strongly they raise the odds. Strong risk factors, each with an odds ratio above 10, are a broken leg, a hip or knee replacement, major trauma, a spinal cord injury, a previous clot, and, within the previous 3 months, a hospital stay for heart failure or atrial fibrillation or a heart attack. Moderate factors, with odds ratios of 2 to 9, include cancer, with the highest risk in metastatic disease, chemotherapy, oral contraceptives, hormone replacement therapy, in vitro fertilisation, the weeks after childbirth, infections such as pneumonia, inflammatory bowel disease, central venous lines and inherited thrombophilia. Weak factors, with odds ratios below 2, include bed rest for more than 3 days, long car or air travel, older age, obesity, diabetes and pregnancy. The NHLBI adds that the chance of VTE nearly doubles every 10 years after age 40.
Family history counts too: the NHLBI says the brothers and sisters of someone who has had VTE have more than double the risk of the general population. Still, a cause is not always found, and the ESC guideline reports that no predisposing factor is found in 40% of patients with PE. Whether a temporary trigger such as an operation came before the clot matters later, when the length of treatment is decided. The same facts also feed the first step of diagnosis. Recent surgery or immobilisation, a previous clot, cancer, hormone use and age are items in the Wells score or the PERC rule, so the questions you are asked about your history are part of the test, not small talk.
Step 1: How likely is it? The Wells score and PERC
The first step in diagnosing a pulmonary embolism is estimating how likely it is before any test, because that pretest probability decides which test is safe to use next. The most used tool is the Wells score, published by Wells and colleagues in Thrombosis and Haemostasis in 2000 from a regression on 40 clinical variables. It gives points for clinical signs of a deep vein clot (3), no alternative diagnosis (3), a heart rate over 100 (1.5), immobilisation or surgery in the previous 4 weeks (1.5), a previous DVT or PE (1.5), coughing up blood (1) and cancer (1). A score under 2 means low probability, 2 to 6 moderate and over 6 high; in the two-level version, 4 or less means PE is unlikely. The ESC 2019 guideline reports that PE is confirmed in about 10% of low, 30% of moderate and 65% of high-probability patients, and in about 12% of PE-unlikely and 30% of PE-likely patients.
- Clinical probability: the Wells score (or the revised Geneva score) before any test.
- Very low risk: if all eight PERC criteria are met, the workup can stop.
- Not high, or PE unlikely: D-dimer, with an age-adjusted cut-off over 50; a negative result excludes PE.
- High, likely, or D-dimer raised: CT pulmonary angiography, or a V/Q scan.
- PE confirmed: a risk class from PESI or sPESI, right ventricular imaging and troponin.
For people a doctor already judges to be at very low risk, the pulmonary embolism rule-out criteria (PERC) can end the workup without a blood test. Kline and colleagues derived PERC in 2004 from 3,148 emergency department patients at 10 US hospitals: age under 50, pulse under 100, oxygen saturation above 94%, no swelling of one leg, no coughing up blood, no recent trauma or surgery, no previous clot and no hormone use. All eight must be met. The PROPER trial (Freund and colleagues, 2018) tested this in 1,916 low-risk patients in 14 French emergency departments: with PERC, one PE (0.1%) was found during 3 months of follow-up, against none with usual care, CT pulmonary angiography fell from 23% to 13% of patients, and the emergency stay was about 36 minutes shorter. The ESC guideline cautions that PE was uncommon in the PERC studies, which limits how widely the results apply. You can add up the points with the Wells score for PE calculator and check the eight items with the PERC rule calculator.
Step 2: The D-dimer test and the age-adjusted cut-off
The D-dimer test measures a breakdown product of blood clots, and in suspected pulmonary embolism its value lies in ruling the diagnosis out, not in confirming it. The ESC 2019 guideline says quantitative laboratory D-dimer tests have a sensitivity of 95% or more, so a normal result in a patient without a high clinical probability makes PE very unlikely. A raised result proves little, because D-dimer is also more often high in cancer, in hospitalised patients, in severe infection or inflammation and during pregnancy. Age matters most: the guideline notes that specificity falls steadily with age, to about 10% in people over 80, so the standard cut-off of 500 µg/L flags most older patients. For that reason the ESC allows an age-adjusted cut-off of age × 10 µg/L in people over 50, which gives a threshold of 750 µg/L at age 75. In the pathway tested in the ADJUST-PE study, only patients without a high or likely clinical probability had D-dimer as their next test.
ADJUST-PE (Righini and colleagues, JAMA 2014) followed 3,346 emergency outpatients with suspected PE in 19 centres in Belgium, France, the Netherlands and Switzerland. Patients whose D-dimer lay between 500 µg/L and their age-adjusted cut-off had no CT scan and no anticoagulant; over 3 months, 1 of 331 (0.3%) had a clot. In patients aged 75 or over without a high clinical probability, the age-adjusted cut-off raised the share in whom PE could be excluded by D-dimer from 6.4% to 29.7%, with no additional false-negative results.
The YEARS study (van der Hulle and colleagues, Lancet 2017) took another route in 3,465 patients in 12 Dutch hospitals. It checks three items, clinical signs of a deep vein clot, coughing up blood and whether PE is the most likely diagnosis, and excludes PE when none is present and D-dimer is below 1,000 ng/mL, or when one or more is present and D-dimer is below 500 ng/mL (ng/mL and µg/L are the same unit). Of the 2,946 patients in whom PE was ruled out and left untreated, 18 (0.61%) had a clot within 3 months, and CT was not needed in 48% of patients, against 34% had the Wells rule with a fixed cut-off been used.
The two shortcuts don't always agree. For a 72-year-old with none of the YEARS items and a D-dimer of 850 µg/L, YEARS would exclude PE, while the age-adjusted cut-off of 720 µg/L would not. That is the kind of question AskMed handles well: describe the situation, ask which rule applies and why, and the answer sets out what ADJUST-PE, the YEARS study and the ESC guideline each say, with a link to each source, so you can check it against your hospital's pathway.
Compare the D-dimer rules for a real case in AskMed:
Step 3: Confirming the clot with a CT scan or a V/Q scan
CT pulmonary angiography (CTPA) is the main test that confirms a pulmonary embolism: a CT scan taken while contrast dye fills the lung arteries, so that a clot shows up as a gap in the dye. The ESC 2019 guideline calls multidetector CTPA the method of choice for imaging the lung arteries in suspected PE. Its accuracy was measured in the PIOPED II study (Stein and colleagues, New England Journal of Medicine 2006): among 824 patients, CTPA was inconclusive in 51 because of poor image quality, and in the rest it had a sensitivity of 83% and a specificity of 96%, with sensitivity rising to 90% when the leg veins were scanned as well. The scan has to be read together with the clinical probability. The ESC accepts a negative CTPA as enough to exclude PE when the probability is low or intermediate, but notes that its negative predictive value was only 60% when the pretest probability was high.
A ventilation-perfusion (V/Q) scan is the main alternative. It compares where air and blood reach in the lungs, needs no iodine contrast and, according to the ESC, gives an effective radiation dose of about 2 mSv, against 3 to 10 mSv for CTPA. The ESC lists the groups in whom it may be preferred: outpatients with a low clinical probability and a normal chest X-ray, young patients, particularly women, pregnant women, people who have had a severe allergic reaction to contrast and people with severe kidney failure. The trade-off is certainty: a planar V/Q scan is inconclusive in about half of cases, while the SPECT technique is non-diagnostic in under 3%. In pregnancy, the ESC says a normal perfusion scan and a negative CTPA appear equally safe for ruling out PE, and that D-dimer levels rise steadily as pregnancy goes on.
Step 4: How severe is it? PESI, heart strain and troponin
Once a pulmonary embolism is confirmed, the ESC 2019 guideline sorts it by the risk of early death, because that risk decides where and how it is treated. The first question is whether the circulation is failing: cardiac arrest, obstructive shock, or a systolic blood pressure below 90 mmHg, or a fall of 40 mmHg or more, lasting longer than 15 minutes, makes it high-risk PE. For everyone else, the ESC combines a clinical score with tests of the heart. The Pulmonary Embolism Severity Index (PESI), derived by Aujesky and colleagues in 2005 from 15,531 hospital discharges in 186 Pennsylvania hospitals, uses 11 characteristics and gives five classes, with 30-day mortality of 0% to 1.6% in class I rising to 10.0% to 24.5% in class V. The simplified PESI (sPESI) by Jiménez and colleagues keeps six items: age over 80, cancer, chronic heart or lung disease, a heart rate of 110 or more, a systolic pressure below 100 mmHg and an oxygen saturation below 90%.
In the sPESI derivation cohort, the 30.7% of patients who scored 0 had a 30-day mortality of 1.0%, against 10.9% in those who scored 1 or more. The heart tests look for strain on the right ventricle, the chamber that pumps against the blocked arteries. The ESC names a right-to-left ventricular diameter ratio of 1.0 or more, on echocardiography or CT, and a tricuspid annular movement (TAPSE) under 16 mm on echocardiography among the findings most linked to a poor outcome. Troponin, a marker of heart muscle injury, is raised in 30% to 60% of patients, depending on the assay. In a 2007 meta-analysis of 20 studies by Becattini and colleagues, 19.7% of patients with a raised troponin died in the short term, against 3.7% with a normal level, and the link held in patients with stable blood pressure.
- High risk: cardiac arrest, obstructive shock or persistent low blood pressure.
- Intermediate-high: stable, PESI class III to V or sPESI 1 or more, with right ventricular strain on imaging and a raised troponin.
- Intermediate-low: stable, PESI class III to V or sPESI 1 or more, with one of those findings or neither.
- Low risk: PESI class I to II or sPESI 0, with no right ventricular strain or raised troponin.
You can score both versions with the PESI score calculator. Real results often fall between the boxes, such as an sPESI of 0 with a right-to-left ventricular ratio of 1.1 on CT. Ask AskMed how the ESC classes that patient and what the class means for monitoring and early discharge, and the answer cites the guideline table and the studies behind each cut-off, so a student can follow the logic and a clinician can check it quickly.
Work through a PE risk class in AskMed:
How is a pulmonary embolism treated, and when is it an emergency?
A pulmonary embolism is treated mainly with anticoagulants, the medicines often called blood thinners, and the risk class decides what is added. The ESC 2019 guideline advises starting anticoagulation while test results are awaited when the clinical probability is high or intermediate. The anticoagulant classes it covers are low-molecular-weight heparin, fondaparinux, unfractionated heparin, the non-vitamin K oral anticoagulants (NOACs, also called DOACs) and vitamin K antagonists, and it reports about a 40% lower risk of major bleeding with NOACs than with vitamin K antagonists. For high-risk PE, the ESC calls primary reperfusion, in most cases systemic thrombolysis with a clot-dissolving drug, the treatment of choice, with surgical embolectomy or catheter-directed treatment as alternatives. In intermediate- and low-risk PE, routine thrombolysis is not recommended and is kept as rescue treatment for patients whose circulation deteriorates. Vena cava filters are mainly for people who cannot take anticoagulants because of active bleeding.
For low-risk PE, the ESC says early discharge, with anticoagulant treatment continued at home, should be considered when the person has no serious other illness and outpatient care is feasible. The choice of medicine and the length of treatment depend on the cause of the clot, kidney function, pregnancy, cancer and bleeding risk, and are set by the treating team. For the symptoms, the hospital stay and recovery, see the guide to lung embolism; for clots that form in the lung arteries themselves and for late complications, see pulmonary thrombosis.
A pulmonary embolism can be life-threatening, so symptoms come before any score. The NHS advises calling 999 (911 in the US) or going to an emergency department for difficulty breathing that comes on suddenly, pain in the chest or upper back, a very fast heartbeat, or if someone has passed out. Coughing up blood needs urgent advice from a doctor or NHS 111.
Frequently asked questions
What is the most common cause of a pulmonary embolism?
A blood clot from a deep vein, usually in the leg or pelvis, that breaks loose and travels to the lungs. Surgery, injury, immobility, cancer and hormones raise the risk, but the ESC 2019 guideline reports that no predisposing factor is found in 40% of patients.
Can a normal D-dimer rule out a pulmonary embolism?
Yes, when the clinical probability is not high or PE is unlikely on a score such as Wells. Laboratory D-dimer tests have a sensitivity of 95% or more, but with a high probability doctors go straight to imaging.
What is the age-adjusted D-dimer cut-off?
From age 50, the cut-off is age × 10 µg/L instead of the standard 500 µg/L, so 700 µg/L at age 70. In the ADJUST-PE study it raised the share of patients aged 75 or over in whom PE could be excluded from 6.4% to 29.7%.
Is a CT scan always needed to diagnose a pulmonary embolism?
No. PERC or a negative D-dimer can exclude PE without a scan in suitable patients. When a scan is needed, CT pulmonary angiography is the usual test, and a V/Q scan is an alternative, for example in pregnancy or severe kidney failure.
What does an sPESI score of 0 mean?
It means a low risk of early death: in the study that created the score, patients with an sPESI of 0 had a 30-day mortality of 1.0%. If imaging and troponin show no heart strain, the ESC classes the PE as low risk, and early discharge may be considered.
Sources
- Konstantinides SV, Meyer G, Becattini C, et al. 2019 ESC Guidelines for the diagnosis and management of acute pulmonary embolism developed in collaboration with the European Respiratory Society (ERS). Eur Heart J. 2020;41(4):543-603.
- Wells PS, Anderson DR, Rodger M, et al. Derivation of a simple clinical model to categorize patients probability of pulmonary embolism: increasing the models utility with the SimpliRED D-dimer. Thromb Haemost. 2000;83(3):416-420.
- Kline JA, Mitchell AM, Kabrhel C, Richman PB, Courtney DM. Clinical criteria to prevent unnecessary diagnostic testing in emergency department patients with suspected pulmonary embolism. J Thromb Haemost. 2004;2(8):1247-1255.
- Freund Y, Cachanado M, Aubry A, et al. Effect of the Pulmonary Embolism Rule-Out Criteria on Subsequent Thromboembolic Events Among Low-Risk Emergency Department Patients: The PROPER Randomized Clinical Trial. JAMA. 2018;319(6):559-566.
- Righini M, Van Es J, Den Exter PL, et al. Age-adjusted D-dimer cutoff levels to rule out pulmonary embolism: the ADJUST-PE study. JAMA. 2014;311(11):1117-1124.
- van der Hulle T, Cheung WY, Kooij S, et al. Simplified diagnostic management of suspected pulmonary embolism (the YEARS study): a prospective, multicentre, cohort study. Lancet. 2017;390(10091):289-297.
- Stein PD, Fowler SE, Goodman LR, et al. Multidetector computed tomography for acute pulmonary embolism. N Engl J Med. 2006;354(22):2317-2327.
- Aujesky D, Obrosky DS, Stone RA, et al. Derivation and validation of a prognostic model for pulmonary embolism. Am J Respir Crit Care Med. 2005;172(8):1041-1046.
- Jiménez D, Aujesky D, Moores L, et al. Simplification of the pulmonary embolism severity index for prognostication in patients with acute symptomatic pulmonary embolism. Arch Intern Med. 2010;170(15):1383-1389.
- Becattini C, Vedovati MC, Agnelli G. Prognostic value of troponins in acute pulmonary embolism: a meta-analysis. Circulation. 2007;116(4):427-433.
- Jorens PG, Van Marck E, Snoeckx A, Parizel PM. Nonthrombotic pulmonary embolism. Eur Respir J. 2009;34(2):452-474.
- National Heart, Lung, and Blood Institute. Venous Thromboembolism: Causes and Risk Factors. NIH.
- NHS. Pulmonary embolism.
