| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Correspondence |
The Editor—Sir,
Ching et al [1] describe an adult patient, who presented with bilateral haemothoraces, in whom they diagnosed a "congenital arteriovenous malformation of the posterior mediastinal aorta" on the basis of the demonstration of a cluster of abnormally dilated vessels in the posterior mediastinum on CT and angiography. They state that this diagnosis was confirmed by surgery and subsequent histology. There are a number of comments that should be made about this case report.
First, it is difficult to understand how the authors reached a diagnosis of "congenital arteriovenous malformation of the posterior mediastinal aorta". The figures in their case report show at least two tortuous and aneurysmal vessels within the mediastinum, and it is clear from the arteriograms that the vessel that was selectively catheterized supplies an area within the left lower lobe. The second vessel, demonstrated on the CT images, obviously lies at a much higher level. The fact that this vessel was not visible on angiography (the authors mention that only descending thoracic aortograms were performed) suggests that it arose from a site other than the descending thoracic aorta. Possibilities include the inferior aspect of the aortic arch or a branch of the left subclavian artery. This patient, therefore, is likely to have an abnormality within the left lower lobe that has a systemic arterial supply. The authors state that he had "no past medical or traumatic history", by which we assume that he had no prior respiratory symptoms, which makes pulmonary sequestration the most likely diagnosis. This has been complicated by bilateral haemothoraces due to rupture into the mediastinum of one (or more) of the aneurysms involving the hypertrophied arteries arising from the aorta, a complication that has been described several times previously [2–5]. Multiple feeding arteries to an area of lung sequestration are described [6] and may arise from sites other than the descending thoracic aorta [7]. The authors describe venous drainage into the "azygos–hemiazygos system", which would suggest that their patient had a sequestration of extralobar type. However, the angiogram shown in their Figure 1e appears to clearly demonstrate pulmonary venous drainage, which would favour an intralobar lesion, which would be much more likely in view of the patient's age.
The differential diagnosis would include an acquired abnormality within the left lower lobe, with subsequent bronchial artery hypertrophy. Bronchial artery enlargement and spontaneous rupture have been described at least 14 times previously [8], even in the absence of pulmonary disease, and may mimic the symptoms of an aortic dissection, as was the case in the patient presented in this report. This diagnosis is less likely for two reasons: the vessel that has been selectively catheterized in this report does not have the appearance of a left bronchial artery because of its relationship to the left main stem bronchus; and it is common to see systemic-to-pulmonary artery shunting in the lung when chronic pulmonary inflammation is present, which was not seen in this case.
Second, we are unconvinced by the demonstration of contrast medium extravasation in Figure 1e. The arrows do not appear to be pointing at anything. What this arteriogram does show is a parenchymal blush, presumably within the sequestrated segment that is compressed by the haemothorax, with a pulmonary vein draining contrast medium superiorly behind the enlarged cardiac silhouette towards the left atrium.
Third, the fact that a "pathological examination confirmed arteriovenous malformations" is not surprising, as the histologist will have been presented with a specimen containing multiple large tortuous arteries, quite likely accompanied by a request form suggesting that this was an arteriovenous malformation! In this case, however, these vessels were those supplying an abnormality within the lung and that had subsequently become hypertrophied and aneurysmal. What would be interesting to know is whether these vessels were elastic or muscular in type; the former would be expected in the case of a pulmonary sequestration whilst the latter would be more likely if the vessels were bronchial arteries supplying an acquired pulmonary abnormality [9].
Fourth, in the discussion, the authors state that "DSA demonstrated the exact site and extent of the vascular abnormality", which is obviously not the case in view of the comments made above. There was at least one further artery higher in the mediastinum at the level of the aortic arch and carina that was not seen angiographically. This is clearly an important point, as an alternative therapeutic approach in this patient would have been embolisation of the abnormal mediastinal vessels from which the haemorrhage was likely to have arisen. The fact that the second vessel was not identified in this case means that such treatment would have been incomplete and the patient would have remained at high risk of recurrent bleeding.
Finally, at the beginning of the discussion the authors describe arteriovenous malformations as being congenital or acquired, which is incorrect. It is important that the term arteriovenous malformation is used only to describe congenital vascular abnormalities, i.e. malformed blood vessels. An arteriovenous communication resulting from iatrogenic or accidental trauma is termed an arteriovenous fistula.
Yours etc.,
Department of Imaging, Hammersmith Hospital, Du Cane Road, London W12 0NN, UK
Footnotes
See the reply on page 566
Received for publication January 30, 2001. Accepted for publication February 16, 2001.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
