Helical CT in the diagnosis of pulmonary embolism

Journal Club

CJEM 2000;2(2):110-111

Clinical question

What is the utility of contrast-enhanced helical CT of the chest in the diagnosis of pulmonary embolism?

Article chosen

Drucker EA, Rivitis SM, Shepard JAO, Boiselle PM, Trotman-Dickenson B, Welch TJ, et al. Acute pulmonary embolism: assessment of helical CT for diagnosis. Radiology 1998;209:235-41.

The search

MEDLINE: 1990 to present
MeSH headings: pulmonary embolism/di [diagnosis] AND tomography, x-ray computed Yield: 106 citations. Exclusion of reviews, editorials, comments and letters and including a hand search of the references of the remaining articles yielded 4 citations prospectively comparing helical CT to pulmonary angiography.

Clinical bottom line

Some previous studies have suggested that contrast-enhanced helical CT of the chest is a sensitive and specific test for acute pulmonary embolism (PE).1,2This study by Drucker and colleagues is consistent with a previous study3 concluding that helical CT accuracy is interpreter-dependent, that helical CT is specific but not sensitive (especially for subsegmental emboli), and that it cannot be recommended as a first-line screening test for PE.

The Evidence

Design: Prospective survey.

Population: Forty-seven adult patients with suspected PE who were referred for pulmonary angiography.

Intervention: All patients underwent helical CT within 24 hours of pulmonary angiography.

Outcome measured: Helical CT interpretations from two groups of radiologists (experienced vs. inexperienced) were compared to pulmonary angiogram results. Sensitivity, specificity and accuracy were calculated using angiography as the reference standard.

Results: During the study period, 149 adult patients underwent angiography for suspected PE. Of these, 47 had helical CT. The most common reasons for exclusion were: patient unable to consent, patient or physician refused study, investigators or CT unavailable, and patient required mechanical ventilation.

Of 47 patients who underwent helical CT, 15 had angiographically proven PE and 32 did not. Radiologists who were experienced in the CT diagnosis of PE correctly identified 8 of 15 confirmed PE and correctly ruled out 31 of 32 “non-PE” cases. Radiologists who were inexperienced correctly identified 9 of 15 confirmed PE and correctly ruled out 26 of 32 “non-PE” cases. Diagnostic parameters for experienced and inexperienced radiologists are presented in Tables 1 and 2.

Comments

The high “miss rate” and poor sensitivities demonstrated in this study suggest that helical CT cannot be used to rule out acute PE. These conclusions agree with those drawn by Goodman and coworkers3 but conflict with those of Remy-Jardin and collaborators,1,2 who reported helical CT sensitivities of 86%–100% for PE. It is important to note, however, that in the latter studies, the higher sensitivities were only achieved by retrospectively excluding patients with subsegmental emboli.

If helical CT is from 81% to 97% specific, as reported in the Drucker study, this compares favourably with the 88% specificity of a high-probability ventilation–perfusion (V/Q) scan,4 making it reasonable to treat for PE based upon a positive helical CT (assuming the patient has at least intermediate pretest probability of disease). Conversely, helical CT lacks adequate sensitivity to rule out PE; therefore, patients with negative studies will often require pulmonary angiography.^1,2

Helical CT is non-invasive and, in many cases, may identify other causative intra-thoracic pathology; however, it requires similar contrast doses to angiography. The combined contrast load for both tests is within acceptable limits for patients with normal renal function, but patients at risk of contrast-mediated toxicity may be best served by undergoing only angiography, which remains the more definitive test.

Drucker and colleagues note that with further advances in CT scanner technology and interpreter experience, helical CT diagnostic accuracy will likely increase. Data from the European Multicenter trial (ESTIPEP), comparing V/Q scan, CT scan and angiography, are pending and will help clarify the future role of helical CT in the diagnosis of PE. Currently, however, emergency physicians should not feel reassured by a negative helical CT in a patient with suspected PE.

References

1. The PRISM-PLUS Study Investigators. Inhibition of the platelet glycoprotein IIb/IIIa receptor with tirofiban in unstable angina and non-Q-wave myocardial infarction. N Engl J Med 1998;338: 1488-96.
2. The PARAGON Investigators. International randomized control-led trial of lamifoban (a platelet glycoprotein IIb/IIIa inhibitor), heparin, or both in unstable angina. Circulation 1994;97:2386-95.

Date studied: November 1999