• 
• 

Three Part Question

In a [patient with chest pain] what is the [sensitivity of a normal chest radiograph] in ruling out [aortic dissection]?

Clinical Scenario

A 52 year old man attends the emergency department with central chest tightness and left arm heaviness. ECG shows anterior ST elevation of 3mm in three consecutive leads. He has a normal mediastinum on chest Xray, but as you administer the thrombolytic agent, you wonder just how sensitive this investigation is in ruling out an aortic dissection.

Search Strategy

Medline 1966-11/03 using the OVID interface.
[exp Aneurysm, Dissecting OR dissecti$.mp OR aneurysm.mp] AND [exp AORTA OR exp AORTA, THORACIC OR aort$.mp.] AND [X-ray.mp OR exp X-Rays OR exp Radiography, Thoracic OR radiograph$.mp] LIMIT to human AND English language.

Search Outcome

Altogether 557 papers were found. One recent literature review included the relevant papers with the exception of 3 additional papers.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Hartnell GG et al, 1993, UK	Chest radiographs from 18 patient with aortic dissection, and 25 patients with acute myocardial infarction Chest radiographs reviewed by conxultant radiologists	Case control study	Sensitivity of chest radiographs interpretation of 1st observer	88.9%	Unclear how MI chest radiographs were obtained Only two radiologists interpreted the radiographs. There was a large interobserver variability. It would have been more informative if a large number of radiologists was used. Radiologists knew that around half the patients had a dissection and half an MI This study does not use A&E doctors
			Sensitivity observer 2	72.2%
Vu KH et al, 1994, Australia	42 patients with diagnosis of aortic dissection	Retrospective cohort	The following results are calculated using 19 patients with dissection (16 confirmed by surgery/autopsy, but also includes 3 patients who died without autopsy)		Gold standard not applied to all patients – only 16 had confirmation of dissection at surgery or on PM. No information given regarding the diagnostic imaging for the other 26. Retrospective review open to bias Chest radiographs reported by a senior radiologist – not A&E doctors Unclear how patients found retrospectively
			Sensitivity wide mediastinum	52.6%
			Sensitivity dilated aortic arch	47%
			Sensitivity displacement of calcified aortic plaques	17%
			Tracheal deviation	5%
			Pleural effusion	15%
			Cardiomegally	21%
Hennessy TG et al, 1996, Ireland	55 consecutive patients referred with suspected thoracic dissection to a cardiology department 35/55 had a dissecting thoracic aneurysm	Retrospective cohort	Sensitivity wide mediastinum	65.7%	Retrospective review using case notes. Unclear who reported CXR Only includes those referred to cardiology department. It is unclear which investigations each patient had undergone prior to referral Not clear whether gold standard applied to all patients
			Normal CXR	14/35 normal
			Other abnormality	7/35 showed another abnormality
Klompas M, 2002, USA	Patients with clinically suspected aortic dissection or confirmed dissection	Literature review including prospective and retrospective cohorts in 21 studies	Abnormal aortic contour	Sensitivity 61% (CI 56-84)	Search terms not clearly stated. It would not be possible to repeat this search from the information given Vast majority of papers are retrospective reviews of patients with known dissections, unblinded and heavily open to bias Not all studies looked at the same CXR findings
			Pleural effusion	Sensitivity 16% (CI 12-21)
			Displaced intimal calcification	Sensitivity 9% (CI 6-13)
			Wide mediastinum	Sensitivity 64% (CI 44-80)
			Abnormal CXR	Sensitivity 90% (CI 87-92)

Comment(s)

All these studies are of poor quality. There is an enormous lack of quality prospective studies recruiting consecutive patients presenting to the emergency department with chest pain.

Clinical Bottom Line

The classical chest radiological findings of a wide mediastinum or abnormal aortic contour do not appear sufficiently sensitive to rule out aortic dissection in a patient with chest pain.

References

1. Hartnell GG, Wakeley CJ, Tottle A, et al. Limitations of chest radiography in discriminating between aortic dissection and myocardial infarction: implications for thrombolysis. J Thorac Imaging 1993;8(2):152-5.
2. Vu KH, Young N, Soo YS. Imaging of thoracic aortic dissection. Australas Radiol 1994;38(3):170-175.
3. Hennessey TG, Smith D, McCann HA, et al. Thoracic aortic dissection or aneurysm: clinical presentation, diagnostic imaging and initial management in a tertiary referral centre. Ir J Med Sci 1996;165(4):259-263.
4. Klompas M. Does this patient have an acute thoracic aortic dissection? JAMA 2002;287(17):2262-72.