Optimal treatment strategies for traumatic ICA pseudoaneurysms are controversial. In this study, we reviewed 12 patients with 12 traumatic ICA pseudoaneurysms treated with LVIS stent-assisted coiling. With the exception of two patients who discontinued treatment and one patient who died of airway damage, which was unrelated to the pseudoaneurysm, most patients (8/9, 88.9%) recovered well during follow-up.
In the past, various surgical methods have been performed to treat pseudoaneurysms, and were proven to have different shortcomings. Direct clipping is not always feasible and durable because pseudoaneurysms lack a true neck. Wrapping-clipping, microsurgical suturing of the vessel defect, and trapping with or without bypass are complex and challenging, especially in the petrous, cavernous, and paraclinoid segments. Ligation of the ICA may result in severe ischemic complications, and collateral retrograde blood flow may result in treatment failure. Nowadays, surgical treatment is only applied for pseudoaneurysms accompanied by a large hematoma exerting a significant mass effect, and minimal invasive endovascular techniques are more preferred in clinical practice.
Endovascular techniques, including coiling, stent-assisted coiling, flow diversion device implantation with or without coils, covered stent implantation, and occlusion of the ICA, have been used for the treatment of traumatic ICA pseudoaneurysms. Packing of the aneurysmal cavity with coils has been reported in several case reports [1, 2, 4, 9]. Because of absence of a true aneurysm neck, aneurysm recurrence is a major complication [2, 9]. Occlusion of the ICA is not recommended as first-line therapy because patients may develop ischemic complications [7].
Cohen et al. [6] have reported on the treatment of three ICA pseudoaneurysms by balloon-expandable bare stent-assisted coiling. In this report, two aneurysms were completely occluded and one aneurysm reoccurred. Ogilvy et al. [13] reported on a successful treatment of a supraclinoid ICA pseudoaneurysm using single Enterprise stent-assisted coiling. Lim et al. [10] described a supraclinoid ICA pseudoaneurysm that was successfully cured by two overlapping Enterprise stents-assisted coiling.
Pipeline device implantation was also reported for the treatment of ICA pseudoaneurysms [5]. The Pipeline device has a high metal coverage rate of 30–35%, and therefore has a better flow diversion effect than bare stents. Nevertheless, it is stiffer than traditional stents, which leads to more difficult navigation in the tortuous ICA, and thereby may worsen the vascular damage during its deployment.
Two case series have reported the successful treatment of traumatic ICA pseudoaneurysms with the Willis covered stent [18, 19]. Covered stent implantation can immediately exclude the aneurysm from the parent artery and restore the normal blood flow. The major limitation is that the covered stents may occlude important branch vessels adjacent to the aneurysms. There is a high requirement for antiplatelet therapy with both flow diversion devices and covered stent implantation, which is accompanied with additional bleeding risk for patients with acute cerebral hemorrhage or multiple injury.
There are two reasons why the LVIS stent was chosen here. Firstly, compared with traditional stents, it has a higher metal coverage rate and a smaller cell size. The LVIS stent has a mean metal coverage rate (23%) between conventional laser cut stents (6–11%) and flow diversion devices (30–35%). A higher metal coverage rate results in improved blood flow diverting effects, such as reduced blood flow velocity and wall shear stress at the aneurysmal wall. Theoretically, the LVIS stent promotes aneurysm thrombosis better than conventional stents. Wang et al. [17] demonstrated that a single LVIS stent resulted in more blood flow decrease than 2 overlapping Enterprise stents. The cell size (1.0 mm × 0.3 mm) of LVIS stents is smaller than that of non-braided stents, which provides protection for small coil prolapse. Secondly, the damaged parent artery wall is incomplete and vulnerable. The use of stents with strong radial force may cause further damage to the parent artery. Also, the push-pull technique used in the release of flow diversion devices may result in further damage. In contrast, the release of the LVIS stent is simple, especially the 3.5 mm diameter version. It is important to note that attempts to increase the metal coverage rate at the pseudoaneurysmal neck by compressing the LIVS stent is dangerous, as this can lead to further damage to the artery wall.
Because of the vulnerability of the pseudoaneurysmal wall, a tight coil embolization of the pseudoaneurysmal cavity is more dangerous than in the real saccular aneurysmal cavity. In case 2 in our study, the coils had separated clearly from the parent artery during follow-up (Fig. 1f). This demonstrates that the pseudoaneurysmal wall was consisted of a vulnerable hematoma. With the absorption of the extravascular hematoma and repairment of the damaged artery, the coils may separate from the parent artery.
We have gained some experience on coil insertion that is shared here. First, the microcatheter should only be navigated to the aneurysmal neck, which in fact is the defect of the parent artery. Second, two-dimensional coil or soft coil should be selected instead of three-dimensional coil. The diameter of the coil should be 1–2 mm smaller than that of the pseudoaneurysm. Finally, if tight embolization of the aneurysmal cavity cannot be completed, the maximum number of coils should be inserted at the aneurysmal neck. Coils that protrude into the parent artery can be applied when necessary to form a “cap effect” between the stent and the parent artery.
Of the ten patients with angiographic follow-up, two (20%) patients received additional coiling because of pseudoaneurysm recurrence. Considering the high recurrent rate of pseudoaneurysm, close angiographic follow-up is recommended. The first follow-up visit should be scheduled within 1 month.
There are some limitations in this study. First, our study was retrospective with a small sample size. Second, most patients’ follow-up time was short. Finally, considering the heterogeneity of pseudoaneurysms with regard to etiologies, locations, sizes, and comorbidities of patients between our study and previous studies, no direct comparison of efficacy and safety of different treatment methods could be made.