Previously Coiled Aneurysms

Aneurysm Residual, Aneurysm Recurrence, and the Risk of Rebleeding

The major concerns after occlusion of an intracranial aneurysm is whether or not the risk of bleeding (or rebleeding in the case of ruptured aneurysms) has been removed and whether not there will be regrowth of the treated lesion. Coil embolization has been shown to be an effective initial treatment for intracerebral aneurysms, with complete or near complete occlusion rates as high as 91%. However, the fate of coiled aneurysms (completely occluded or not) after initial treatment is less clear.

Early studies looking at long-term follow-up of coiled aneurysms suggested that there was a significant incidence of aneurysm recurrence, be it from compaction of coils or growth of areas of residual filling. Cognard et al. reported recanalization in 20 of 148 (14%) “totally occluded” aneurysms and growth of the remnant in 6 of 18 (33%) of “subtotally occluded” aneurysms during follow-up out to 38 months. Other early series also suggested long-term angiographic recurrence rates of 28% to 49%. However, in almost every series there was a low incidence of aneurysm hemorrhage. ^{, , ,}

The degree of occlusion of coiled aneurysms appears to have some prognostic significance. The Modified Raymond and Roy Occlusion Classification (MRROC) system has been widely used to assess the degree of occlusion in coiled aneurysms. The system divides lesions into class I (complete occlusion), class II (residual filling of the aneurysm neck), class IIIa (residual filling of the aneurysm within the coil pack), and class IIIb (residual filling of the aneurysm between the coil pack and the aneurysm wall). Several recent studies suggest that aneurysms with an “adequate” degree of occlusion (complete or near complete occlusion equating to class I or II lesions) at 6 months have very low rates of long-term recurrence. ^,

Mortimer et al. found an overall rate of reopening of 6% for “adequately occluded” aneurysms (MRROC class I or II), with only 2% showing “significant” recanalization. Gallas et al. reported that 96% of ruptured aneurysms that were completely occluded at 1 year remained occluded at follow-up out to 36 months. Mascitelli et al. found low rates of recanalization in both short- and long-term follow-up (out to >48 months) with class I and II aneurysms. There were no ruptures in either group. They also found that class IIIa aneurysms were much less likely to show regrowth or hemorrhage as compared with class IIIb aneurysms. Stapleton et al., in a retrospective study of 167 class III aneurysms, found similar differences between class IIIa and IIIb lesions.

The actual risk of hemorrhage of coiled aneurysms is low in almost all long-term studies. The Cerebral Aneurysm Re-rupture after Treatment (CARAT) study found that the risk of rerupture after coil embolization was related to the initial degree of aneurysm occlusion, with the risk of rerupture being as low as 1.1% for aneurysms with complete occlusion (class I), 2.9% for aneurysms more than 90% occluded (class II), 5.9% for incompletely coiled aneurysms (70% to 90% or class IIIa), and as high as 17.6% for aneurysms with less than 70% occlusion (class IIIb). In a multicenter retrospective analysis of 626 aneurysms with varying degrees of residual filling, Munich et al. found low rates of bleeding after detachable microcoil treatment in both unruptured (0.6%) and ruptured aneurysms (3.4%).

Other factors may contribute to recanalization or bleeding in previously coiled aneurysms. Anatomic, technical, technological, and perhaps even social factors likely play a role. Aneurysm size, location, and wall parameters may affect the stability of even well-treated aneurysms. Ferns et al. concluded that aneurysm size greater than 10 mm was associated with increased rates of recurrence and retreatment. Munich et al. found that aneurysms greater than 25 mm were associated with increased rates of bleeding (as were aneurysms less than 10 mm; however, this second finding may be a statistical aberration). Both studies suggest that location in the posterior fossa is a risk factor for recurrence/bleeding. Mortimer found that size greater than 10 mm and neck size greater than 4 mm were significant risk factors; however, location did not seem to be a significant variable in their study. Type of coil does not appear to have an effect on the incidence of aneurysm regrowth or hemorrhage. The presence of a stent may be helpful, but no statistically significant advantage has been shown. Though less measurable, factors such as genetics, underlying medical conditions, and lifestyle likely have an effect. It is logical (though not proven) that conditions such as connective tissue disease, hypertension, and inflammatory conditions could adversely affect the growth or rupture of coiled or virgin cerebral aneurysms. Certainly high-risk behaviors such as tobacco, drug, or alcohol abuse can add to the risk of aneurysm rupture.

Stapleton et al. found that the predictors of recanalization across their entire cohort were male gender, ruptured aneurysm, intraluminal thrombus, and MRROC IIIb status at closure. They found factors such as cavernous internal carotid artery location, larger aneurysm size, increased aneurysm neck size, presence of blebs or dome irregularities, presence of intraluminal thrombus, ruptured aneurysm, and decreased coil packing density were all associated with initial MRROC IIIb closure status. As these factors have been found to be of some consequence in predicting recanalization/bleeding in other studies, a number of interrelated findings may help to predict aneurysms that will recanalize or bleed.

Surveillance and Imaging

Initial management of aneurysms treated with endovascular coiling involves surveillance with periodic imaging of the treated lesion. Protocols on when and how often to perform imaging and what modalities are to be used for imaging vary greatly. Protocols are influenced by the training and experience of the surgeon. In general, the authors obtain follow-up digital subtraction angiography (DSA) or magnetic resonance angiography (MRA) in patients with ruptured aneurysms treated with detachable coil therapy at 6 months and at 1 year posttreatment. A similar schedule is used for unruptured coiled aneurysms. Given the significance of the degree of occlusion at 6 months as a predictor of recanalization in a number of studies, this may be a prudent starting point for follow-up studies. However, in our practice, the 6-month study is not always performed if the initial occlusion was complete or almost complete and the patient remains asymptomatic. Although long-term data (>10 years) are scarce, asymptomatic patients who have studies demonstrating stable complete occlusions or aneurysms with minimal neck residuals for more than 5 years are allowed to decrease follow-up imaging to every 2 years or longer if they prefer.

As to the choice of imaging procedure, DSA is considered the gold standard for the evaluation of aneurysms treated by either endovascular techniques or surgery. ^{, ,} A great many surgeons always perform DSA to follow coiled aneurysms, citing a superior ability to detect small neck remnants or changes in the coil pack’s configuration. However, DSA is an invasive procedure and does have the inherent problems of ionizing radiation and nephrotoxicity. These concerns are magnified somewhat by the need for repeat examinations over time. Noninvasive methods such as magnetic resonance angiography (MRA) and in some cases computed tomography angiography (CTA) have improved to the point where they are often used for following previously coiled aneurysms. In general, because of the significant streak artifact caused by the metallic coil pack and the fact that CT still requires ionizing radiation, CTA is not widely employed to follow coiled aneurysms (although some centers use CTA in aneurysms with stent coiling and for clipped aneurysms).

A number of studies have demonstrated the efficacy of MRA as an imaging tool for following coiled intracranial aneurysms. The authors tend to prefer obtaining both contrast-enhanced MRA (CE-MRA) and time-of-flight MRA (TOF-MRA—a protocol utilizing the principal of flow-related enhancement and therefore not requiring contrast). We also tend to prefer 3.0 T magnets versus 1.5 T magnets, though imaging at both magnet strengths has been shown to be adequate. , , In a recent meta-analysis by Ahmed et al. of MRA versus DSA for follow-up imaging, a subgroup analysis found a sensitivity of 88% for any recanalization with both TOF-MRA and CE-MRA. Specificity was 94% and 96%, respectively. Results for the detection of small neck residuals were also good, with a sensitivity of 80% using both techniques. Interestingly, the study suggests that MRA techniques could provide a high degree of sensitivity and specificity for residual filling even in patients who had undergone stent-assisted coiling. Unfortunately, the small number of studies and comingling of the data with aneurysms treated with flow diverters makes these conclusions less definitive. In a prospective trial of CE-MRA and TOF-MRA at both 1.5 and 3 T, Kauffman et al. found that the sensitivity for the detection of any aneurysm remnant was good for all techniques and all magnet strengths. However, CE-MRA had greater sensitivity for larger (class IIIa or IIIb) aneurysm remnants than TOF-MRA at 1.5 T. As these remnants are more concerning clinically, the CE-MRA sequences may be the more valuable in characterizing treated lesions. A preference for 3 T over 1.5 T was seen owing to theoretical physical advantages and demonstrated the superiority of 3 T MRA over 1.5 T MRA for uncoiled aneurysms. One caveat to follow-up with MRA is that equipment varies greatly by institution, and MRA pulse sequence parameters continue to evolve. Timing of imaging after bolus injection of contrast is also variable per patient, and all of these factors can greatly affect results. Still, for many surgeons, the use of MRA for the follow-up imaging of coiled aneurysms has become routine, with DSA reserved for cases in which there are problems obtaining MRA, suboptimal images or questions about the MRA appearance, or other reasons for a higher level of suspicion.