Diagnosing pulmonary embolism

A pulmonary embolism can present in a variety of clinical scenarios; however, there are several features that are typical.

Case presentation

A 59-year-old woman presented to the ED with pleuritic chest pain, dyspnea, and syncope. She reported worsening dyspnea over the past month that had been evaluated by several clinicians and treated as an asthma exacerbation. The day prior to admission she had an episode of syncope at home with associated dyspnea and dizziness. The patient reported no recent travel, immobility, history of cancer, or hormonal replacement. Her medical history was significant for asthma. She had no history of venous thromboembolism and no family history of thrombophilia. The patient did not smoke tobacco or drink alcohol. Outpatient medications included prednisone, 40 mg/d (started recently for asthma exacerbation); cetirizine, 10 mg/d; fluticasone/salmeterol, 250 μg/50 μg two puffs daily; nebulized albuterol as needed; and pantoprazole, 40 mg/d.

The patient was not in acute distress and was afebrile, with a blood pressure of 109/76 mm Hg, a heart rate of 174 beats/min, a respiratory rate of 20 breaths/min, and oxygen saturation of 91% on 15 L/min via non-rebreather mask. Physical exam was remarkable for an irregularly irregular heart rhythm. Lung exam revealed clear lung fields without wheezing; lower extremities were without edema. The remainder of the exam was normal. An electrocardiogram demonstrated atrial fibrillation with a rate of 160 beats/min and normal axis without ST-T wave abnormalities. Admission labs revealed a hemoglobin level of 11.7 g/dL, a white blood cell count of 11.6 cells/mm3, a platelet count of 98,000 cells/mm3, a creatinine level of 1.5 mg/dL, a troponin level of 0.10 ng/mL, and a B-type natriuretic peptide level of 124 pg/mL; other labs were normal.

Clinical presentation

A pulmonary embolism can present in a variety of clinical scenarios; however, there are several features that are typical. Dyspnea or pleuritic chest pain is the presenting symptom in most pulmonary embolism cases and may be the primary reason for the patient to seek medical care (1). These symptoms can occur suddenly, likely at the time of acute embolization, or can develop over several days to weeks (1). Other clinical features of pulmonary embolism include tachypnea, tachycardia, and hemoptysis (2). Indicators of a deep venous thrombosis, including unilateral leg swelling, redness, or pain, have been reported in approximately 30% of patients (2).

Syncope has been investigated recently as a presenting symptom of pulmonary embolism. The Pulmonary Embolism in Syncope Italian Trial (PESIT) reported that of the patients who presented to the ED with syncope and required hospitalization, 17.3% were found to have a pulmonary embolism (3). Other studies have reported syncope in 5% to 14% of pulmonary embolism cases (4, 5, 6). The discrepancy in this data may be due to several of these studies being observational cohorts; further studies are needed to conclude whether syncope is a common presentation for pulmonary embolism. In patients who present with syncope, it is reasonable not to evaluate for pulmonary embolism if there is an alternative diagnosis.


Once a pulmonary embolism is suspected, several algorithms can be used to determine its probability. Elements of the clinical presentation, physical exam, and laboratory results are combined by decision tools, such as the Wells criteria or Geneva score (7, 8). In the Wells criteria, points are assigned to various elements of the patient's presentation: suspicion of deep venous thrombosis, no alternative diagnosis that better explains the illness, heart rate greater than 100 beats/min, immobilization for greater than three days or surgery in the previous four weeks, history of deep venous thrombosis or pulmonary embolism, presence of hemoptysis, or presence of malignancy (9, 10). These criteria will separate patients into low, intermediate, or high-risk probability of a pulmonary embolism.

Patients found to have a high probability of pulmonary embolism should have immediate imaging. The preferred imaging modality, in the absence of contraindications to contrast dye, is CT angiography (8).

In patients with an intermediate probability of a pulmonary embolism, a D-dimer level should be ordered. If negative, this has sufficient negative predictive value to rule out a pulmonary embolism and no further imaging should be ordered (7, 8). A positive D-dimer warrants further evaluation with imaging studies.

Previously, it was recommended that patients with a low probability have a D-dimer to rule out a pulmonary embolism. However, factors including pregnancy, chronic illness, malignancy, and elderly age can cause an elevated D-dimer level, making it ineffective as a rule-out test in many populations (11). More recently, an additional scoring system called the PERC rule was developed to rule out pulmonary embolism in low-probability patients without a D-dimer, avoiding any additional testing or radiation exposure (8). The PERC rule has a set of eight criteria that must be met to rule out a pulmonary embolism: age younger than 50 years, initial heart rate less than 100 beats/min, initial oxygen saturation greater than 94% on room air, no unilateral swelling, no hemoptysis, no surgery or trauma within four weeks, no history of venous thromboembolism, and no estrogen use (12). If any of these criteria are positive, a D-dimer level should be checked. A normal D-dimer level has sufficient negative predictive value, and imaging should not be ordered (8). If the D-dimer is positive, imaging should be completed.

Very recently, the YEARS clinical decision rule has been demonstrated as a new predictive model for pulmonary embolism that utilizes only three items of the original Wells criteria (clinical signs of deep venous thrombosis, hemoptysis, and pulmonary embolism as the most likely diagnosis) in conjunction with D-dimer (13). Using this rule, CT angiography is not recommended in the following scenarios: no YEARS items and D-dimer level less than 1,000 ng/mL, or one or more YEARS items and D-dimer level less than 500 ng/mL (13). This algorithm led to a 14% decrease in the need for CT angiography (13). Further validation of this method is needed before it is incorporated into practice.

Determining severity

After a pulmonary embolism has been diagnosed, its severity should be determined to identify patients at high risk for mortality. Determining severity is also useful to identify patients who are at the lowest risk for death and can be safely treated as outpatients. Several scoring tools have been developed to assist in severity classification: the Pulmonary Embolism Severity Index (PESI), the simplified PESI (sPESI), and the Hestia criteria (14, 15, 16).

The PESI incorporates several clinical factors to estimate a patient's 30-day mortality risk from pulmonary embolism. The index includes the following variables: older age, male sex, cancer, heart failure, chronic lung disease, pulse greater than 110 beats/min, systolic blood pressure less than 100 mm Hg, respiratory rate greater than 30 breaths/min, temperature less than 36 °C, altered mental status, and arterial oxygen saturation less than 90% (14). Based on these variables, patients are separated into five risk categories: class I (very low risk), class II (low risk), class III (intermediate risk), class IV (high risk), and class V (very high risk). Class I has a 30-day mortality rate of 1.6% or less and class II has a 30-day mortality rate of 3.5% or less, while class V patients have a 30-day mortality rate up to 24.5%. Based on data from a clinical trial comparing outpatient and inpatient treatment for pulmonary embolism, patients in classes I and II may be safely treated as outpatients (14, 17).

The main benefit of the PESI score is to identify patients who can be safely treated as outpatients. The simplified PESI score is streamlined for settings like a busy ED and eliminates factors that were not found to be significantly associated with mortality (15). The variables included are age (>80 years), history of cancer, history of chronic cardiopulmonary disease, heart rate greater than 110 beats/min, systolic blood pressure less than 100 mm Hg, and oxygen saturation less than 90%. Based on the score, patients are separated into low- and high-risk groups (15).

As with the PESI and simplified PESI, the Hestia criteria were developed to identify patients who could be safely treated as outpatients (16). In addition to hemodynamic instability and oxygen saturation less than 90%, the criteria also take into account other factors, such as need for thrombolysis or embolectomy, active bleeding or high risk for bleeding, pulmonary embolism diagnosed during anticoagulant treatment, administration of IV pain medications for greater than 24 hours, medical or social reasons for treatment in the hospital, creatinine clearance less than 30 mL/min, severe liver impairment, pregnancy, or history of heparin-induced thrombocytopenia (16, 18). If one or more of the criteria are met, the patient should be hospitalized for pulmonary embolism treatment.

Although these scoring systems are useful for helping clinicians determine the severity of a patient's pulmonary embolism, they should be used in conjunction with the overall clinical picture.


In conclusion, several predictive models, such as the Wells criteria and the YEARS clinical decision rule, in conjunction with the patient's clinical presentation, can guide a physician to the diagnosis of pulmonary embolism. Once the diagnosis is made, clinical tools such as the PESI or Hestia criteria can help determine the severity of pulmonary embolism and perhaps guide the ideal setting for treatment.

Back to the case

Our patient would have a Wells score of 1.5, placing her in the low-risk group. The patient did not meet the PERC rule criteria (age, heart rate, oxygen saturation). In accordance with the algorithm, a D-dimer level was obtained and was greater than 20 μg/mL. Since the D-dimer level was elevated, CT angiography was performed and demonstrated a large pulmonary embolism with large clot burden distending the right main pulmonary artery and extending into the upper and lower branches and a saddle embolism extending across the lower lobe pulmonary artery branches on the left.

The patient's clinical presentation alone would warrant admission, and her PESI score was 110, placing her into the high-risk category (class IV). While in the ED, the patient became hypotensive and was given systemic tissue plasminogen activator, 100 mg, over two hours. The patient was also started on IV unfractionated heparin. She improved clinically and was subsequently bridged to warfarin prior to discharge home.