Does a Negative CT Scan Rule Out Pulmonary Embolism?

Abstract & Commentary

By David J. Pierson, MD, Editor, Pulmonary and Critical Care Medicine, Harborview Medical Center, University of Washington, is Editor for Critical Care Alert

Professor of Pulmonary and Critical Care Medicine, Harborview Medical Center, University of Washington

Synopsis: Meta-analysis of studies reporting follow-up of patients who were not anticoagulated after a negative CT angiogram showed that this test excludes pulmonary embolism as effectively as does conventional pulmonary angiography.

Source: Quiroz R, et al. Clinical validity of a negative computed tomography scan in patients with suspected pulmonary embolism: a systematic review. JAMA. 2005;293:2012-2017.

In this meta-analysis, Quiroz and colleagues attempted to identify all English-language reports published since 1990 that included at least 3 months of follow-up in patients clinically suspected of having pulmonary thromboembolism (PTE), whose CT angiograms were negative and who did not receive anticoagulant therapy. The purpose was to determine the incidence of subsequent episodes of PTE and thus the negative likelihood ratio (NLR) and negative predictive value (NPV) of a negative CT scan in such patients.

Of 22 studies identified through PubMed, MEDLINE, EMBASE, the Cochrane Database, and other sources, 15 studies (3500 patients) met Quiroz et al’s criteria for inclusion in the meta-analysis. Twelve of these studies used single-slice CT, 2 used multidetector-row CT, and 1 used electron-beam CT. Among the 3500 patients in the included studies, there were 36 episodes of PTE and 6 additional instances of deep venous thrombosis during the 3 to 12 months of follow-up. Fifteen deaths were attributed to PTE.

The overall NLR of VTE after a negative CT angiogram was 0.07 (95% confidence interval [CI], 0.05-0.11), and the NPV was 99.1% (95% CI, 98.7-99.5%). There was no difference between single-slice CT (NLR, 0.08; 95% CI, 0.05-0.13) and multidetector-row CT (NLR, 0.15; 95% CI, 0.05-0.43) with respect to ruling out PTE in the included studies. Quiroz et al conclude that the clinical validity of using CT to rule out PTE is similar to that of conventional pulmonary angiography.


The results of this study suggest that clinical outcome is not adversely affected if anticoagulant therapy is withheld based on a negative CT scan. The overall NPV for this test compares favorably with the NPVs previously reported for conventional pulmonary angiography, and is superior to those reported for negative- or low-probability ventilation-perfusion scans.

The principal limitation of CT scanning, which has delayed acceptance of this test as a gold standard for excluding PTE, is its questionable ability to detect isolated peripheral emboli. The clinical importance of this limitation is uncertain, however, particularly in the evaluation of patients who are ill enough to be admitted to the ICU. In the absence of severe co-morbidities such as obstructive lung disease or advanced congestive heart failure, it is doubtful whether sub-segmental PTE in the absence of more central clots would explain acute life-threatening illness. Although the present study found no difference in the performance of single-slice vs multiple-detector CT scanners, it is likely that the latest machines are better able to detect small peripheral emboli than their predecessors of 10 or 15 years ago.

I think we can now regard CT scanning as a second (and more readily available) gold standard in ruling out all but the tiniest pulmonary emboli—provided that equipment of relatively recent manufacture has been used, and the study is judged technically adequate by the interpreting radiologist. For patients in the ICU, there remains a role for ventilation-perfusion scanning in those few patients suspected of having PTE who cannot be given intravenous contrast material and who do not have serious concomitant cardiopulmonary disease. However, in most instances, a CT angiogram is the procedure of choice. If it is negative, pulmonary embolism can be excluded confidently enough to obviate the need for anticoagulation.