Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED II)

BACKGROUND:

Approximately 600,000 Americans sustain pulmonary embolism each year; one-third of these episodes are fatal. Unfortunately, pulmonary embolism is underdiagnosed and, therefore, under-treated. A substantial body of evidence suggests that the diagnosis of pulmonary embolism is not made in the majority of patients in whom it causes or contributes to death.

In the main, there are two explanations for the failure to diagnose pulmonary embolism: pulmonary embolism may be clinically silent, and there is no definitive, noninvasive diagnostic test. Indeed, ventilation perfusion lung scans are nondiagnostic in the majority of patients with suspected acute pulmonary embolism. Pulmonary angiography may be used to establish a diagnosis in such patients, but it is underutilized because of a mortality risk around 1 percent.

Recently, relatively small studies have suggested that contrast enhanced spiral computed tomography (CT) scanning is a useful diagnostic test for pulmonary embolism, with sensitivity as high as 80 percent and specificity as high as 95 percent. Spiral CT is widely available and much less invasive than pulmonary angiography. If spiral CT could be established as a useful diagnostic test, pulmonary embolism would be diagnosed more effectively and more patients would receive proper treatment.

DESIGN NARRATIVE:

The study evaluates the role of spiral CT scan in the diagnosis of PE by comparison with a composite reference test, including pulmonary angiography, V/Q lung scan in patients without prior PE and compression ultrasound of the lower extremities in patients with no prior deep venous thrombosis (DVT).