When a patient comes to the office or emergency department complaining of shortness of breath, acute pulmonary embolism is a diagnosis that must be considered.

The signs and symptoms of pulmonary embolism (PE) – which include tachycardia, shortness of breath, and chest pain – are nonspecific. So, it is important to have a well-thought-out approach to make the diagnosis in patients who have PE and avoid unnecessary tests and risks in patients with a low likelihood of PE.

New guidelines from the American College of Physicians suggest a graded approach to diagnostic testing based on a patient’s estimated pre-test probability of having PE.

When deciding whether to test for PE and which test to order, it is essential to determine the likelihood that a patient’s symptoms are due to PE. Validated decision-support tests include the Well’s Criteria, Pulmonary Embolism Rule-Out Criteria (PERC), and the revised Geneva score. Although they have been recommended for years, these tests often are not used. The guidelines note that the accuracy of an experienced clinician’s gut sense appears to be similar to that of the validated decision tools.

The first decision is whether to do any testing at all. PERC was developed in response to the growing and inappropriate use of D-dimer testing to rule out PE in situations with very low clinical suspicion. The ACP guidelines discuss a meta-analysis of 12 studies, which determined that the overall proportion of missed PEs in patients who had a negative PERC score was 0.3%.

PERC criteria are age younger than 50 years, heart rate less than 100 beats per minute, oxygen saturation greater than 94% on room air, no unilateral leg swelling, no hemoptysis, no surgery or trauma, no history of venous thromboembolism, and no estrogen use. The PERC test is negative when an individual meets all of the criteria above.

It is recommended that, in patients felt to be at low risk and whose PERC test is negative, then no other workup should be done. In this circumstance, “the risk of PE is lower than the risk of testing.”

If the PERC test is positive, then low-risk patients should undergo a highly sensitive D-dimer test. When used this way, the PERC tool decreases the use of D-dimer testing in patients who otherwise would have been tested inappropriately.

For intermediate-risk patients, the first test to obtain is a highly sensitive D-dimer test. The ACP guidelines summarized two recent studies – one using the Well’s Criteria and the other a revised Geneva score – which examined D-dimer testing specifically in intermediate-risk groups.

The studies evaluated 1,679 and 330 patients, respectively, at intermediate risk and found that a normal D-dimer level in these patients was 99.5% and 100% sensitive, respectively, for excluding PE on CT. Therefore, patients at both low- and intermediate-level risk should be screened first with a D-dimer rather than going directly to imaging.

When evaluating D-dimer results, a threshold of greater than 500 ng/mL usually indicates a positive test. However, in patients older than 50 years, the guidelines note that it may be more accurate to use a D-dimer level equal to a patient’s age multiplied by 10 ng/mL. Because it is a very sensitive test, a negative D-dimer test indicates no further testing is needed, and there is no reason to obtain a CT scan.

In patients believed to be at high risk of having PE, either through the use of a validated clinical decision tool or by clinical gestalt, a negative D-dimer test is not sensitive enough to rule out PE. Patients who are deemed to be at high risk of having PE should go directly to evaluation by CT pulmonary angiography (CTPA). Pulmonary ventilation/perfusion scanning should be used when CTPA is unavailable or contraindicated.

The bottom line

• The first step when evaluating a patient is to determine his or her pretest probability of PE using either a clinical tool or clinical judgment. The Well’s Criteria and Geneva score have been validated and are considered equally accurate, and neither has been shown to be superior to risk stratification using clinical gestalt.

Low pretest probability of PE: First, use the PERC criteria. In those patients who meet all eight rule-out criteria, DO NOT order a D-dimer; the risk of PE is lower than the risk of testing. Those who do not meet all eight criteria should undergo a D-dimer. A normal D-dimer level is sufficient to rule out PE, and imaging studies are not needed. An elevated plasma D-dimer, ideally adjusted for age, should prompt evaluation by CTPA.

Intermediate pretest probability of PE: D-dimer testing is the first step. A negative D-dimer has sufficient negative predictive value to eliminate the need for further testing. An elevated D-dimer, ideally adjusted for age, should prompt evaluation by CTPA.

High pretest probability of PE: In patients with a high pretest probability secondary to either clinical gestalt or a clinical prediction tool, evaluation by CTPA is warranted.


• “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 Nov 3;163(9):701-11 .

Dr. Skolnik is associate director of the family medicine residency program at Abington (Pa.) Memorial Hospital. Dr. Vandergrift is a first-year resident in the program.