Discussion

Endovascular treatment of ruptured intracranial aneurysms is increasingly performed with satisfactory clinical and angiographic results.1 Nevertheless, aneurysm or neck reopening occurs in approximately 20% of patients, necessitating retreatment in about half of them.2 There is no consensus or recommendations regarding retreatment after aneurysm coiling.3 Indications for retreatment have to take into account the risks of retreatment and the risk of bleeding/rebleeding. Recent long-term analysis of the International Subarachnoid Aneurysm Trial (ISAT) shows that the risk of rebleeding of coiled aneurysms is low.4 The Cerebral Aneurysm Rerupture After Treatment (CARAT) study showed that the risk of early rebleeding after treatment of ruptured aneurysms was strongly related to aneurysm occlusion status.5 In their recent literature review, Tsurumi et al6 confirmed that delayed aneurysm rupture after coiling occurs in the majority of cases when there is an aneurysm remnant, but is also possible in cases of neck remnant. The risk is also higher in patients with large and giant aneurysms, as in our case.

Evolving remnants and aneurysm remnants are usually retreated because they are considered to have a relatively high risk of rerupture.4 In our case, the remnant after the second treatment was classified as a neck remnant and was not evolving. However, it should be noted that the differentiation between neck and aneurysm remnants is not always easy, as illustrated by the moderate interobserver reproducibility of the Montreal scale. To make a decision regarding retreatment, other factors must be taken into account such as age and risk factors (eg, smoking and elevated blood pressure). The modalities of retreatment are also relatively unclear. Coiling is probably the first choice for retreatment, but adjunctive devices are also useful for avoiding further growth of the aneurysm (eg, singular flow diverters).7 Clipping is also considered a viable option. In our case, initial retreatment would now probably consist of coiling with a flow diverter, but this option was not available in 2007.

This case also illustrates the fact that aneurysm growth is not a regular process. Until recently, aneurysms and remnants were considered to have a relatively linear growth. However, experience shows that aneurysms and remnants can remain stable for a long time and then have a rapid or sudden growth. Statistical models have also confirmed that the aneurysm growth process does not have a constant time-independent rate but is irregular and discontinuous, which results in periods with and without aneurysm growth and with high and low risks of rupture.8 This phenomenon was perfectly illustrated in our case with a long-term stable neck remnant followed by the rupture of a probably rapidly growing aneurysm. However, the voluminous aneurysm pouch observed at the time of rebleeding was probably not exclusively the result of aneurysm regrowth, but also partially a pseudoaneurysm created by the rupture itself.

The unpredictable evolution of aneurysms makes it difficult to plan effective follow-up strategies. Because of the low rate of late rebleeding, it is debatable whether long-term follow-up imaging should be performed. In addition, there is no consensus regarding which modality of imaging follow-up should be used. DSA is the gold standard for follow-up but has some disadvantages including rare neurological complications and radiation exposure. MRA (three-dimensional time-of-flight) is an appropriate technique for the follow-up of coiled aneurysms with 3 T, but has some limitations for aneurysms treated with intravascular devices.9 ,10 CT and CT angiography have not been well evaluated for this purpose. Whatever the technique used, it is noteworthy that it gives mostly morphological information including whether the aneurysm is completely occluded and whether there is a neck or aneurysm remnant. There is actually no imaging tool that evaluates the scarring process at the level of the neck, and further work is needed in this field. Regarding the timing, early imaging follow-up is usually performed 3–6 months after treatment, with further follow-up at 12 or 18 months. Later follow-up varies from one center to another and must be adapted to the aneurysm occlusion status at 12/18 months. The case presented here shows that follow-up does not afford complete protection against rebleeding.