We read with great interest the article by Bendszus et al1 in which successful coil embolization was followed by fatal re-rupture 2 weeks later.
In reviewing the case presentation, we made a number of observations from which we would like to propose an alternative explanation of the eventual outcome. Although there are obviously institutional differences in how patients like these are managed, we noted a number of unique points that may have contributed to an increase in periprocedural risk.
The conventional and CT angiographic images submitted are of high quality and demonstrate an aneurysm of the basilar terminus measuring approximately two thirds of the diameter of the basilar artery. It is not clear how the aneurysm was determined to be 3 mm in diameter, but because the average basilar artery typically measures approximately 3 mm in diameter, one would suspect that the actual dimension of the aneurysm would more likely be in the range of 2–2.5 mm. The neck of the aneurysm is not well demonstrated by angiography but appears rather well defined by CT angiography and would not be considered wide-necked. On the basis of the images submitted, we would suggest that the coils selected may have been too large and that the aneurysm might well have been successfully treated with fewer 2- or 2.5-mm coils, without the need for balloon assistance. We do not routinely proceed directly to the balloon-assist technique until we have first attempted direct unassisted coil embolization. We would also suggest that although balloon assist is useful in the placement of the coils within wide-necked aneurysms, using a balloon might lead to overpacking in small aneurysms. This can result in a relatively greater degree of pressure against the wall of the aneurysm that could, in turn, lead to an increased risk of rupture. We have long since learned, from the experience of endoluminal balloon embolization of aneurysms, that increased and asymmetric stresses on the wall of an aneurysm predispose to rupture.
One principal advantage of the detachable coil technique over balloon embolization is a lower and more symmetric distribution of radial forces within a treated aneurysm and proved association of a lower incidence of aneurysm rupture. We would point out that the dimensions of the coil mass in Fig 2A are significantly larger than those of the untreated aneurysm in Fig 1A. It would, therefore, appear likely that the aneurysm was overdistended, resulting in a tear of the ventral wall, by use of 12-cm oversized coils and the balloon-assist technique. The patient was subsequently diagnosed with vasospasm. Assuming that the patient was being volume expanded and was hypertensive at the time of her re-rupture, it is quite possible that the increased volume and pressure in the face of an overpacked aneurysm could have contributed to the rebleed. The coil mass likely began to be displaced and finally prolapsed across the torn ventral wall of the aneurysm, resulting in a fatal rebleed.
We cannot disagree with the conclusion that coiling of aneurysms cannot protect all patients from rehemorrhage. There are many examples of how undercoiling can lead to rehemorrhage; however, we believe this to be an example of how overcoiling can lead to rehemorrhage. The hypothesis of re-rupture as a consequence of recanalization of a partially thrombosed aneurysm cannot be entirely excluded; however, we would not expect extrusion of coils into the subarachnoid space if this were the etiology.
Although we agree that so-called bioactive coils may improve permanence of coil embolizations, we disagree with your conclusion that bioactive coils would have made any difference in the outcome in this patient. The stated absence of organized thrombus within the aneurysm lumen after a re-rupture should not lead the reader to believe that no fibrotic reaction occurred at all during the 2 weeks that the coils were in place. It is entirely possible that the more mature thrombus surrounding the coil mass may have extravasated into the subarachnoid space along with the coils.
In conclusion, we do not believe that this case represents a failure of detachable coils but rather a consequence of incorrect selection of coil size and overly aggressive delivery technique.
Reference
Reply:
In summary, it is stated in this letter regarding our case presentation1 that we used oversized coils with balloon assistance, which, in combination, resulted in delayed aneurysm rupture due to increased wall stress. This letter relies on several incorrect assumptions that we cannot accept. First, this aneurysm measured 3 mm as determined by intra-arterial 3D angiography (not CT angiography as stated by the authors of the letter). The assumption of a size of 2–2.5 mm is simply incorrect. Second, this aneurysm did indeed have a wide neck. At our institution, we never use balloon assistance as the first line of treatment. Rather, we always attempt deploying a spheric coil first and use a balloon only when this deployment is unsuccessful. Before blaming an incorrect or overly aggressive technique, one should read the manuscript carefully. (“Because of the wide neck of the aneurysm, it was not possible to place a coil in the aneurysm without it prolapsing into the basilar artery.”1).
Third, the assumption that the coil mass in Fig 2A is larger than the aneurysm in Fig 1A is incorrect. Figures 2A and 2B are more magnified than Fig 1A. Looking at Figs 3A and 3B with a magnification similar to that in Fig 1A, one realizes that the coil mass very closely corresponds to the initial aneurysm size. Fourth, the argument that delayed re-hemorrhage occurred as a result of volume expansion and induced hypertension as a cause of overdistention of the aneurysm is wrong. As we stated in the manuscript, re-rupture occurred 14 days after initial rupture, when the patient had stabilized and was scheduled for rehabilitation the next day. Vasospasm had completely subsided and re-rupture occurred 5 days after cessation of hypervolemic/hypertensive therapy. Fifth, extrusion of coils outside the aneurysm sac is a frequent finding in coiled aneurysms undergoing surgery later and must not be mistaken for overpacking.2
Finally, as we stated in the article, there was no histologic evidence for tissue response such as thrombus organization, macrophages, or fibrin formation. We are amazed at why the authors, without providing evidence, stated that the reader should not be led to believe that no fibrotic reaction occurred at all during the 2 weeks that the coils were in place. As we stated in our article,1 this case differed histologically from findings reported for aneurysms at a similar time after coil embolization.3–5 Re-rupture may have occurred for several reasons in this patient, but we cannot accept the allegation that this was most likely caused by incorrect or overly aggressive treatment.
- Copyright © American Society of Neuroradiology