The specific goals of this study were to determine the ability of multidetector CTA to rule out or detect pulmonary embolism, and to evaluate whether the addition of computed tomographic venography (CTV) improves the diagnostic accuracy of CTA.
Methods
Using a technique similar to PIOPED I, the investigators performed a prospective, multi-center trial using a composite reference standard to confirm the diagnosis of pulmonary embolism. Once again, for ethical reasons, the use of pulmonary artery digital-subtraction angiography was limited to patients whose diagnosis could neither be confirmed nor ruled out by less invasive tests. In contrast to PIOPED I, a clinical scoring system was used to assess the clinical probability of pulmonary embolism. Central readings were performed on all imaging studies except for venous ultrasonography.
Results
Of the 7,284 patients screened for the study, 3,262 were eligible, and 1,090 were enrolled. Of those, 824 patients received a completed CTA study and a reference standard for analysis. In 51 patients, the quality of the CTA was not suitable for interpretation, and these patients were excluded from the subsequent analysis. Pulmonary embolism was diagnosed in192 patients.
CTA was found to have a sensitivity of 83% and a specificity of 96%, yielding a likelihood ratio for a positive multidetector CTA test of 19.6 (95% confidence interval, 13.3 to 29.0), while the likelihood ratio for a negative test was 0.18 (95% confidence interval, 0.13 to 0.24). The quality of results on CTA-CTV was not adequate for interpretation in 87 patients; when these patients were excluded from analysis, the sensitivity was 90% with a specificity of 95%, yielding likelihood ratios of 16.5 (95% confidence interval, 11.6 to 23.5) for a positive test and 0.11 (95% confidence interval, 0.07 to 0.16) for a negative test.
Conclusions
Multidetector CTA and CTA-CTV perform well when the results of these tests are concordant with pre-test clinical probabilities of pulmonary embolism. CTA-CTV offers slightly increased sensitivity compared with CTA alone, with no significant difference in specificity. If the results of CTA or CTA-CTV are inconsistent with the clinical probability of pulmonary embolism, additional diagnostic testing is indicated.
Commentary
CTA has been used widely, and in many centers has largely replaced other diagnostic tests for pulmonary embolism. This well-done study incorporated recent advances in technology with multidetector CTA-CTV, along with a clinical prediction rule to better estimate pre-test probabilities of pulmonary embolism.2 It is important to recognize that 266 of the 1,090 patients enrolled were not included in the calculations of sensitivity and specificity for CTA-CTV because they did not have interpretable test results.
Although the specificity of both CTA and the CTA-CTV combination were high, the sensitivity was not sufficient to identify all cases of pulmonary embolism. This result contrasts to the recent outcomes studies of CTA, in which low rates of venous thromboembolism were seen in follow-up of patients with negative multidetector CTA.3,4 Although multidetector CTA has a higher sensitivity than single-slice technology, this test may still miss small subsegmental thrombi that might be detected using other diagnostic tests (ventilation-perfusion scintigraphy and/or pulmonary digital-subtraction angiography).
An important take-home message from this study is to recognize once again the importance of utilizing established clinical prediction rules for venous thrombosis and pulmonary embolism (such as the Wells clinical model).2 As with the majority of diagnostic tests at our disposal, when our clinical judgment is in contrast with test results, as in the case of a high likelihood of a potentially fatal disease like pulmonary embolism with a normal CTA result, additional diagnostic testing is necessary.
References
- The PIOPED Investigators. Value of the ventilation/perfusion scan in acute pulmonary embolism: results of the prospective investigation of pulmonary embolism diagnosis (PIOPED). JAMA. 1990;263:2753-2759.
- Wells PS, Anderson DR, Rodger M, et al. Excluding pulmonary embolism at the bedside without diagnostic imaging: management of patients with suspected pulmonary embolism presenting to the emergency department by using a simple clinical model and d-dimer. Ann Intern Med. 2001 Jul 17;135(2):98-107.
- Perrier A, Roy PM, Sanchez O, et al. Multidetector-row computed tomography in suspected pulmonary embolism. N Engl J Med. 2005 Apr;352(17):1760-1768.
- van Belle A, Buller HR, Huisman MV, et al. Effectiveness of managing suspected pulmonary embolism using an algorithm combining clinical probability, D-dimer testing, and computed tomography. JAMA. 2006 Jan 11;295(2):172-179.