Recurrence of a large intracranial fusiform aneurysm treated with overlapping Pipeline embolization devices: illustrative case

Abstract

BACKGROUND

Flow diversion, specifically with the Pipeline embolization device (PED), represents a paradigm shift in the treatment of intracranial aneurysms. Several studies have demonstrated its efficacy and at times superiority to conventional treatment modalities for aneurysms with a fusiform morphology, giant size, or wide neck. However, there may be a nonsignificant risk of recurrence after flow diversion of these historically difficult-to-treat aneurysms, relative to aneurysms with a more favorable morphology and size (i.e., saccular, narrow necked). To date, only three papers in the literature have demonstrated the recurrence of a completely occluded aneurysm on follow-up.

OBSERVATIONS

The authors describe a patient with a giant middle cerebral artery fusiform aneurysm treated with multiple telescoping PEDs. On the 3-month follow-up angiogram, there was complete occlusion of the aneurysm. The patient was lost to follow-up and presented 4 years later with a recurrence of the aneurysm between PED segments, requiring retreatment. The patient represented 3 years posttreatment with the need for repeat treatment of the fusiform aneurysm due to separation of the existing PEDs along with stent reconstruction. At the 20-month follow-up after the third treatment, the initial aneurysm target was found to be occluded.

LESSONS

This case illustrates the need for long-term follow-up, specifically for patients with giant wide-necked or fusiform aneurysms treated with overlapping PEDs.

The introduction of flow-diverting stents, such as the Pipeline embolization device (PED), has revolutionized the endovascular management of intracranial aneurysms.¹ By inducing the disruption of blood flow near the aneurysm-parent vessel interface, flow diverters function by altering hemodynamics at the aneurysm site, resulting in aneurysm thrombosis and neointimal growth over the stent.²

Flow-diverting stents have been shown to afford better neck reconstruction than conventional endovascular treatment options. Thus, flow diversion offers a significant improvement in treatment strategies for fusiform or giant aneurysms.³ However, this treatment option may require the use of telescoping PEDs to cover the length of the aneurysm.⁴ Long-term occlusion of fusiform aneurysms treated with multiple overlapping flow diverters can be impacted by device migration and foreshortening as well as by limited control of pore density, which can disrupt the overall crucial process of endoluminal reconstruction and contribute to delayed recurrence.⁵ These disturbances may partially explain the lower complete occlusion rate of fusiform aneurysms relative to those with a saccular morphology.³ However, only a few studies in the literature have described the recurrence of a once completely occluded fusiform aneurysm on follow-up.^6–8

Illustrative Case

A 38-year-old male with a past medical history of type 2 diabetes, hyperthyroidism, and hypertension presented with a 4-month history of headaches. He was a nonsmoker and had no family history of cerebral aneurysms.

Investigations

Initial computed tomography angiography (CTA) demonstrated a 3.5 × 2.7 × 2.5–cm fusiform aneurysm arising from the proximal and mid M1 segments. There was subtle dilation of the distal M1 and proximal M2 branches. These findings were further confirmed with cerebral angiography (Fig. 1A–C).

FIG. 1.

A: Initial computed tomography scan demonstrating a well-circumscribed, round lesion involving the MCA. B and C: Anteroposterior (AP) and lateral digital subtraction angiography (DSA) demonstrating a 3.5 × 2.7 × 2.5 cm fusiform aneurysm involving the proximal 38 mm of the M1 segment of the left MCA. There is a 4.5-mm diameter fusiform dilation of the distal M1 and subtle dilation of the proximal M2 branches. D and E: Unsubtracted lateral and oblique angiograms obtained following embolization, showing stasis of flow in the aneurysm well into the venous phase.

Considerations

Both microsurgical and endovascular strategies were considered. Microsurgical treatment mandated an extracranial-to-intracranial bypass with proximal ligation of the aneurysm. However, because of the proximal location of the aneurysm on M1, an intermediate- or high-flow bypass or a double-barrel superficial temporal artery-middle cerebral artery (MCA) bypass would likely have been required. Additionally, the location of the aneurysm on the dominant side increased the risk of operative morbidity. Therefore, an endovascular strategy was pursued. Because of the wide neck and fusiform nature of the aneurysm, coil embolization with or without stent or balloon assistance was not feasible. However, flow diversion was believed to represent a reasonable treatment option. Prior to treatment, aspirin (650 mg) and clopidogrel (300 mg) were administered at therapeutic levels.

Treatment

The distal superior M2 division was accessed with a Phenom 27 microcatheter (Medtronic). A 5 × 30 mm PED (Medtronic) was initially placed 16 mm distal to the aneurysm neck. The proximal landing zone terminated within the aneurysm sac. The construct was extended with a 5 × 25 mm PED with approximately 30% overlap with the initial device. A third 5 × 25 mm PED was placed from the second PED to the internal carotid artery (ICA) just distal to the origin of the ophthalmic artery. Despite a similar 30% overlap between the second and third devices, these two devices became disengaged, and an additional (fourth) 5 × 20 mm PED was used to buttress the stent construct. A final angiogram demonstrated good wall apposition of all PEDs with stasis of flow in the aneurysm sac and improved flow to the distal M2 branches (Fig. 1D and E).

Outcome and Follow-Up

The patient tolerated the procedure well and was discharged home on postprocedure day 1 without changes in his neurological examination. The next day, after the patient presented with symptomatic subcortical infarcts in the setting of subtherapeutic aspirin levels, his dose was doubled without complications. A follow-up angiogram at 3 months demonstrated a patent stent construct, without evidence of stenosis and no residual filling of the left MCA aneurysm (Fig. 2A and B).

FIG. 2.

AP (A) and lateral (B) DSA demonstrating no filling of the left MCA aneurysm at the 3-month follow-up.

The patient was subsequently lost to follow-up and presented 4 years later with headaches. CTA demonstrated a recurrence of the prior aneurysm with separation of the PED between the most distal and middle devices, as well as between the middle and second most proximal devices (Fig. 3A and B). The aneurysmal dilations measured 9.9 × 17.7 × 10.7 mm and 11.0 × 8.0 × 16 mm, respectively. Given that CTA performed immediately after the symptomatic subcortical infarcts had demonstrated good flow in all devices, we believed that this complication had not contributed to the recurrence; thus, we did not alter the retreatment plan. This recurrent aneurysm was treated with flow diversion utilizing additional PEDs. A 5 × 25 mm PED was advanced and deployed spanning from the midpoint of the most distal PED (first) to the midpoint of the middle PED (second). A 5 × 30 mm PED was then deployed from the midpoint of the second PED to the midpoint of the adjacent (third) PED (Fig. 3C). A total of six PEDs had now been deployed across the aneurysm. The patient tolerated the procedure well and was discharged home the following day.

FIG. 3.

A: Three-dimensional (3D) reconstruction CTA demonstrating separation of the PED and recurrence of the aneurysm between PED segments at the 4-year follow-up. B: DSA performed prior to retreatment, demonstrating a recurrent aneurysm between the PEDs. C: DSA performed after the placement of two additional PEDs between separated segments, demonstrating a decreased size of recurrent aneurysms. D: 3D reconstruction of follow-up CTA at 3 months after retreatment showing complete occlusion of the aneurysm. E: 3D reconstruction of CTA showing recurrence of aneurysm between flow diverters at the 3-year follow-up (red arrow). F: DSA at 20 months after retreatment, demonstrating occlusion of the initial aneurysm and dilation of MCA past the stent construction.

A 3-month follow-up with CTA demonstrated occlusion of the aneurysms (Fig. 3D). The patient was again lost to follow-up. However, he presented 3 years later with a recurrence of the aneurysm. Cerebral angiography demonstrated separation of the third and fourth flow-diverting stents, resulting in a recurrence of the aneurysm measuring 18 × 15.7 × 13.5 mm (Fig. 3E), which was treated with one Surpass Evolve 5 × 30 stent placed between the distal PED and the proximal PED with a 15-mm overlap. He was discharged without complications and placed on 6 months of dual antiplatelet therapy.

At 20 months after retreatment, repeat angiography demonstrated aneurysmal dilation of the MCA continuing past the stent construct involving the bifurcation and both M2 vessels, although the initially treated fusiform aneurysm was occluded (Fig. 3F).

Patient Informed Consent

The necessary patient informed consent was obtained in this study.

Discussion

Observations

Here we report a patient with a fusiform aneurysm initially treated with multiple overlapping PEDs. Although angiography showed complete occlusion of the aneurysm at 3 months, aneurysm recurrence and PED migration was found 4 years posttreatment, requiring deployment of additional flow diverters. Despite a short period of angiographic occlusion, separation of existing flow diverters mandated retreatment of the aneurysm with additional telescoping PEDs at 3 years following the second treatment. At the 20-month follow-up, angiography showed occlusion of the initial fusiform aneurysm with dilation distal to the stent reconstruction. However, cases of fusiform aneurysm recurrence after short-term radiographic observance of complete occlusion are rare. Our case presentation is unique because of the fusiform morphology of the aneurysm, long-term radiographic follow-up period, and the persistent use of flow diverters to achieve angiographic occlusion.

Lessons

Flow diversion has quickly become a valuable treatment option in patients with a variety of intracranial aneurysms, with satisfactory occlusion and complication rates compared to those following other surgical and endovascular treatment strategies.^9,10 Specifically, for large unruptured aneurysms with a saccular morphology, PED treatment has been associated with a higher rate of complete obliteration (86%) compared to coiling (41%) with lower rates of retreatment (2.8% vs 37%, respectively).⁹

By providing a scaffold for neo-endothelization and vessel remodeling, flow diversion potentially offers a significant improvement in prior treatment strategies for fusiform aneurysms, which historically have been challenging to treat either surgically or endovascularly.³ Notably, rates of complete occlusion at follow-up are lower for these aneurysms than for those with a saccular morphology. In a study of 29 patients with 30 fusiform aneurysms treated with the PED at a single institution, 60% of aneurysms were completely occluded and 76% were favorably occluded at the 17.4-month follow-up.³

Telescoping of flow diverters may be required to accommodate the entire length of fusiform aneurysms.⁴ However, the use of multiple PEDs has been correlated with device migration/foreshortening and limits control of pore density and porosity of the PED with unknown effects on flow diversion, all of which may interfere with long-term aneurysm occlusion.⁵ Bohara et al.⁶ described the recurrence of a giant ICA aneurysm between PED segments at the 4-year follow-up despite angiographic occlusion at the 3-year follow-up. Coxon et al.⁷ reported on a recurrent MCA fusiform aneurysm with areas of dilation, which had previously been surgically treated. Flow diversion with two PEDs treated the recurrence, resulting in favorable remodeling of the proximal end and mild persistent dilation of the distal segment at 8 months. However, the aneurysm recurred with dilations between the two previously placed PEDs at the 5.5-year follow-up. Chen et al.⁸ reported the regrowth of a large ICA aneurysm treated with two overlapping PEDs, which demonstrated a small remnant on a 4-month angiogram and subsequent regrowth to its original size at 10 months.

Although aneurysm recurrence after a period of complete occlusion has been hypothesized to occur as a result of chronic inflammation and neovascularization in the aneurysm wall, in our case, delayed separation of telescoped PEDs due to shortening of the device or continued aneurysm growth most likely caused the recurrence, resulting in inadequate neck coverage.¹¹ This may have been avoided by placing a larger stent within the smaller one. However, at the time of both treatments, the largest PED available was 5 mm, limiting our ability to do so. The introduction of new devices, such as Surpass (Stryker), with larger diameters and constant pore density may be a more effective solution for these types of aneurysms.

Additionally, fusiform or giant wide-necked aneurysms may be more susceptible to device migration, because they may require a longer period for endothelialization to occur. In a human histopathologic study of giant fusiform aneurysms, endothelial cell coverage of the aneurysmal segment was not observed until at least 13 months despite angiographic occlusion.¹² Alternatively, giant, fusiform aneurysms may face increased disruptions in the endothelialization process, leading to subsequent recurrence. Darsaut et al.¹³ created an experimental giant fusiform aneurysm model using canines, who were then treated with multiple overlapping flow-diverting stents. At 12 weeks after treatment, histopathology demonstrated the presence of small open pores in the neointimal coverage over the surface of the device. Persistence of these leaks may be responsible for continued aneurysmal recurrence in our patient. Therefore, in all cases, longer imaging follow-ups with angiography and magnetic resonance imaging should be pursued for patients with giant fusiform aneurysms to better characterize changes in the dome.

Author Contributions

Conception and design: Park, Kumar, Capek, Crowley. Acquisition of data: Kumar, Dabhi, Capek, Crowley, Kalani. Analysis and interpretation of data: Dabhi, Capek. Drafting the article: Kumar, Dabhi. Critically revising the article: Park, Kumar, Dabhi, Raper, Capek, Kellogg. Reviewed submitted version of manuscript: Park, Dabhi, Raper, Capek, Kellogg. Approved the final version of the manuscript on behalf of all authors: Park. Administrative/technical/material support: Capek. Study supervision: Park. Operating surgeon for several of the surgeries: Crowley.