Diagnosing Pulmonary Embolism During Pregnancy: A Noteworthy Approach

The evaluation of potential pulmonary embolism (PE) is one of the great challenges in emergency medicine. PE is notoriously overevaluated (with D-dimer testing and imaging) and also underevaluated, resulting in adverse outcomes. Medical providers have for many years faced a difficult time deciding when to initiate a workup with lab testing or imaging for possible PE.

Is clinical gestalt enough to decide on when to work up the patient? If so, how much clinical experience and knowledge of the literature is necessary to rely on that gestalt? When should D-dimer testing be performed? Which decision instruments (Pulmonary Embolism Rule-Out Criteria, or PERC; Wells criteria; revised Geneva score) are reliable and when should they be used? Concerns are continually raised about the financial costs associated with unnecessary workups as well as the dangers of contrast and radiation associated with pulmonary imaging.

In 2015, Raja and colleagues^[1] published a wonderful "best practice advice" paper on behalf of the Clinical Guidelines Committee of the American College of Physicians that reviewed the literature and provided an evidence-based approach to the workup of the patient with possible PE. The suggested algorithm incorporated decision instruments and, in some cases, D-dimer testing to decide when the workup could be terminated versus when it should proceed to pulmonary imaging.

However, the algorithm didn't make any special provision for pregnant patients with suspected PE, which left many of us still perplexed about the most rational workup. Is there a sensible approach for these patients as well?

Diagnosis of Pulmonary Embolism During Pregnancy

Righini and colleagues^[2] conducted a multicenter, multinational, prospective diagnostic management outcome study. Patients included in the study were pregnant women over 18 years of age presenting to one of the study centers with signs or symptoms concerning for PE, such as acute onset of dyspnea or chest pain without another obvious cause. A total of 395 patients were included in the study.

Patients underwent a workup that included pretest probability (PTP) assessment using the revised Geneva score. Patients with a high PTP underwent bilateral compression ultrasonography (CUS) of the legs. Patients with a low or intermediate PTP underwent D-dimer testing. If this test was positive, they moved on to CUS. If the D-dimer was negative, the workup was discontinued.

Patients with a positive CUS were assumed to have PE and were treated with anticoagulation. Patient with a negative CUS underwent CT pulmonary angiography (CTPA). If the CTPA was negative, the workup was discontinued. If the CTPA was positive, PE was diagnosed and the patient was treated with anticoagulation. If the CTPA was inconclusive, a ventilation/perfusion (V/Q) scan was performed to decide on whether to discontinue the workup versus initiate treatment for PE.

In total, 28 (7.1%) patients were diagnosed with PE (positive CUS in seven patients, positive CTPA in 19 patients, and high probability V/Q scan in two patients). PE was excluded in 367 patients utilizing this algorithm. During 3-month follow-up of the patients in whom PE was excluded, none developed evidence of symptomatic venous thromboembolic events.

Final Thoughts on the Algorithm

PE is one of the leading causes of maternal death in first-world countries.^[3] That fact that two patients are at risk for death—mother and baby—is an even greater cause for angst amongst providers. Thus far, unfortunately, clear recommendations for the workup of these patients have been lacking.

Righini and colleagues have provided an outstanding algorithm for the workup of pregnant patients with possible PE. If this algorithm is validated in a larger study, we will finally have a reasonable, evidence-based approach to managing this challenging patient presentation.

Follow Medscape on Twitter for more emergency medicine news: @Medscape