A 44-year-old woman sought care at the emergency department (ED) because she was having difficulty breathing and felt faint. She had been fine until that morning. Three days earlier the patient, who had a history of high blood pressure and elevated cholesterol levels, had driven from Connecticut to New York and back, spending a total of 4 hours in her car. The patient indicated that she’d been taking oral contraceptives (OCPs) for several years, but she did not smoke. There was no history of hemoptysis, recent surgery, or trauma. Neither blood clots nor cancer were part of her or her family’s history.
In the ED, the patient did not have any signs or symptoms of a deep venous thrombosis (DVT). She was obese, with a body mass index of 40.3 kg/m2; other vitals were: blood pressure (BP), 134/88 mm Hg; heart rate (HR), 64 beats per minute (bpm); respiratory rate (RR), 12; and O2 saturation, 99% with ambulation.
The ED physician strongly suspected a pulmonary embolism (PE), but the patient’s score on a clinical probability algorithm (using the Wells criteria) was a 3, indicating only “moderate probability“ of a PE (TABLE 1). (She scored a 3 because an “alternative diagnosis [was] less likely than PE.”) In addition, her D-dimer level was 160 ng/mL using the Triage D-Dimer Test by Biosite, Inc (normal <400 ng/mL), which ruled out a PE. (Many ED physicians at our institution are more cautious when using this D-dimer assay and use a lower cutoff value.)
Given these results, the ED physician did not order imaging studies because the expense and radiation exposure outweighed the probability of the patient having a PE. A subsequent coronary work-up was also negative. The patient was discharged to home and advised to follow up with her primary care physician a few days later.
Two days later we saw the patient at our office. Not only had her dyspnea gotten worse while the presyncope remained, but she now had left-sided pleuritic chest pain. She also reported mild pain in her right calf. On examination, the patient’s BP was 126/86 mm Hg, HR was 82 bpm, RR was 12, and O2 saturation was 96% with ambulation. Her Wells score was now 6, still a moderate probability for PE. (She received another 3 points for the new DVT symptoms—“clinically suspected DVT.”)
Although the patient did not also have signs of a DVT, her additional symptoms along with the original symptoms’ persistence and the existence of other risk factors (OCP use and obesity) led us to reconsider a PE diagnosis. These suspicions prompted us to send the patient back to the ED, where a Doppler ultrasound of the right lower extremity was negative, but the D-dimer was positive at 565 ng/mL.
A pulmonary computed tomography angiogram (CTA) showed 2 small pulmonary emboli within the distal left upper lobe pulmonary arteries.
The patient was treated with heparin and warfarin and discharged without complications.
TABLE 1
Calculating and interpreting the Wells score4,5,7,9,10
| Clinical parameter | Points |
|---|---|
| Clinically suspected DVT | 3.0 |
| Alternative diagnosis less likely than PE | 3.0 |
| Tachycardia | 1.5 |
| Immobilization/surgery (within 4 weeks) | 1.5 |
| History of DVT or PE | 1.5 |
| Hemoptysis | 1.0 |
| Malignancy (treatment within 6 months, palliative) | 1.0 |
| TOTAL | |
| Score | Traditional interpretation |
| <2.0 | Low probability of PE |
| 2.0-6.0 | Moderate probability of PE |
| >6.0 | High probability of PE |
| Score | Alternative classification scheme |
| ≤4.0 | PE unlikely |
| >4.0 | PE likely |
| DVT, deep venous thrombosis; PE, pulmonary embolism. | |
Discussion
The incidence of PE in the United States varies significantly: Individuals younger than 40 have a risk of 1 in 10,000 compared with 1 in 100 for those older than 80.1 Mortality associated with undiagnosed PE varies widely, from 9.2% to 51%.2 This percentage is significant given that half of all PEs go undiagnosed.3 In addition, when left untreated, PE will recur in 30% to 50% of patients, with a fatality rate of 10% to 45%.1 Further, up to 4% of patients with acute PE develop chronic PE and subsequent pulmonary hypertension.4,5 Given the consequences of failing to diagnose a PE, clinicians must consider this condition in patients who present with unexplained hypotension, dyspnea, or chest pain.6
Not an easy diagnosis
This case report demonstrates the inherent difficulty in diagnosing a PE. Still, certain clinical symptoms/signs can aid in the decision-making process. Fever, crackles, and wheezes decrease the probability of PE, whereas syncope, hemodynamic shock, leg edema, and hemoptysis increase its likelihood.7 Despite the many commonly reported risk factors for PE, only malignancy, recent surgery, or a history of DVT/PE significantly increase the risk of developing a clot.8
The Wells criteria. This scoring system groups patients according to the probability of having a PE: low (score: <2), moderate (score: 2-6), and high (score: >6).6 An alternative classification scheme divides patients into 2 groups: likely to have a PE (score: >4) or unlikely to have a PE (score: ≤4).8