Diagnosing Pulmonary Embolism

April 2011

Diagnosing Pulmonary Embolism

Search Strategy: You first conduct a PUBMED Clinical Query (broad/sensitive) for “pulmonary embolism” and combine the results with a search for “Annals of Emergency Medicine” (http://tinyurl.com/4rovoug) which identifies two of the manuscripts selected below from amongst 64 citations. Next you combine the Clinical Query results with first “diagnostic accuracy” and then “computed tomography” (http://tinyurl.com/48sgmxe) or “ventilation scan” (http://tinyurl.com/4qoszow) to obtain the remaining citations.

Chest pain. Dyspnea. Palpitations. Everywhere you look while working today’s shift, all chief complaints seem to return to the thoracic cavity. You wistfully recall watching “ER” on television in high school amazed at the variety of conditions and effortless miracles portrayed each week. But now you live in the real world of 21st Century emergency medicine and your next six patients have some variant of the above complaints. Amongst a differential diagnoses including acute coronary syndrome and chronic obstructive pulmonary disease exacerbations, you contemplate the diagnosis of pulmonary embolism for each patient.

Three patients have low-risk as computed using the Well’s score and are “PERC-negative” so you exclude the diagnosis of PE without a D-dimer and document your rationale in the medical records moving on to other diagnostic possibilities. Three more patients are low-risk using Well’s criteria, but not PERC-negative so you order D-dimer tests. Two are negative and one positive. You do not pursue the diagnosis of PE in the negative D-dimer cases, but feel compelled to evaluate further in the positive case.

Ms. Z is a 27 year old patient with 18-hours of pleuritic left-sided chest discomfort in the mid-clavicular line at ribs 4-6 with associated dyspnea, but no cough, fever, edema, or visible rash/trauma. The pain is not reproducible and her ECG is normal sinus rhythm without any old ones for comparison. She takes birth control pills (hence her non-low risk PERC score), is not currently pregnant, and has no family history of venous thromboembolism. You are about to reflexively order a PE protocol CT, but your attending challenges you to describe the state of the literature for the diagnosis of PE in the ED.

PICO Question

Population: Adult emergency department patients with acute chest pain clinically concerning for PE

Intervention: Diagnostic tests (history, physical exam, V/Q, CT, MRI?) for ovarian torsion

Comparison: N/A

Outcome: Diagnostic accuracy (sensitivity, specificity, likelihood ration) for bedside physical exam, CT, V/Q, MRI..


First years: Clinical Policy: critical issues in the evaluation and management of adult patients presenting with suspected pulmonary embolism, Ann Emerg Med 2003; 41: 257-270. (http://pmid.us/12548278)

Second years: Clinical features from the history and physical examination that predict the presence or absence of pulmonary embolism in symptomatic emergency department patients: results of a prospective, multicenter study, Ann Emerg Med 2010; 55: 307-315. (http://pmid.us/20045580)

Third years: Multidetector computed tomography for acute pulmonary embolism, NEJM 2006; 354: 2317-2327 (http://pmid.us/16738268)

Fourth years: Computed tomographic pulmonary angiography vs ventilation-perfusion lung scanning in patients with suspected pulmonary embolism: a randomized controlled trial, JAMA 2007; 298: 2743-2753. (http://pmid.us/18165667)


Article 1: Clinical Policy Critical Issues in the Evaluation and management of Adult Patients Presenting with Suspected Pulmonary Embolism, Ann Emerg Med 2003; 41257-270

Article 2: Clinical Features from the History and Physical Examinations that Predict The Presence or Evidence of PE in Symptomatic ED Patients Results of a Prospective, Multicenter Study, Ann Emerg Med 2010; 55307 ANSWER KEY

Article 3: Multidetector Computed Tomography for Acute Pulmonary Embolism (PIOPED II), NEJM 2006; 3542317-2327

Article 4: Computed Tomographic Pulmonary Angiography vs Ventilation-Perfusion Lung Scanning in Patients with Suspected Pulmonary Embolism A Randomized Controlled Trial, JAMA 2007; 2982743-2753 ANSWER KEY

Bottom Line

Pulmonary embolism (PE) is an elusive diagnosis that occurs in 1-2 individuals per 1000 persons in the United States each year. Patients diagnosed with PE in the emergency department have an overall low mortality (1% attributed to PE) and high functional status. PE can present with a variety of symptoms (dyspnea, chest pain, palpitations, syncope) and most patients with these symptoms do not have a PE. Unfortunately, the diagnosis of PE using clinical findings alone is inaccurate (Hoellerich 1986, Moser 1994, Hampson 1995). Although there is a role for clinical gestalt in experienced physicians, clinical prediction rules (such as the Well’s criteria described below) should be used based upon overall accuracy and the potential to reduce test ordering variability particularly amongst less experienced clinicians. However, many physicians fail to use these decision aids such as the Well’s Criteria or PERC rule while assessing for PE.

In contemplating the diagnosis of PE, clinicians are balancing the risks of missing the often fatal PE against the radiation risks (and sometimes logistical risks of transferring patients to tertiary centers where the technology is available) of V/Q or CT pulmonary angiography. Here are two epidemiological facts to consider. One large database suggests that only 1.4% of patients evaluated for PE will ultimately be diagnosed with ACS. And lower extremity Doppler are not a reasonable first-line screen for PE since emergency physicians do not suspect DVT in 90% of PE patients.

ACEP Guidelines (circa 2003) recommend that Well’s criteria (below) or similarly validated decision aids like the Wicki criteria or Kline criteria should be used prior to test-ordering to risk stratify ED patients with suspected PE. Once PE probability stratification is deduced, use of quantitative D-dimer (ELISA or turbidimetric) can exclude patients with low pre-test probability. In low to moderate pretest probability patients, a normal V/Q scan excludes clinically significant PE. The 3-month risk of DVT or PE in patients with a negative spiral CT is 0.5%. Updated ACEP Guidelines have been written but await ACEP Board of Director approval and have not yet been published. The new guidelines will hopefully provide evidence-based recommendations for the use of emergency physician-performed DVT ultrasound, the role of the PERC score to avoid preventable false-positive D-dimer testing, as well as the indications for repeat imaging in patients with known PE who frequently present to the ED with potentially related complaints.

Well’s Criteria

Risk FactorPoints
Suspected DVT3
Alternative Diagnosis Less Likely than PE3
Heart Rate >1001.5
Immobilization or surgery previous 4 weeks1.5
Previous DVT or PE1.5
Malignancy (last 6 months or palliative)1
ScoreMean Probability PE (%)% with this ScoreRisk Interpretation
0 – 2 points3.640Low
3 – 6 points20.553Moderate
> 6 points66.77High

Several risk factors that are not currently part of validated clinical decision rules can increase (thrombophilia, pleuritic pain, family history of VTE) or decrease (substernal pain, current smoker) the probability of PE when contemplating test-treat thresholds. V/Q scanning should still play a role in the diagnostic evaluation of PE. Although based upon 3-month outcomes (death or new PE or DVT) suggests no differences between a normal V/Q or nondiagnostic V/Q + negative LE ultrasound with Wells score <4.5, clinicians are less accepting of V/Q than CTPA. In addition to the radiation-risk and contrast-dye adverse side effects, clinicians should be aware that a portion of PE’s diagnosed by CTPA may be clinically inconsequential as evidenced by the lower overall PE detection rate in the randomized sampling of the PGY IV article with no increase in mortality or morbidity.