Endovascular treatment for unruptured small wide-necked ophthalmic segment aneurysms: Technique feasibility, efficacy and mid-term follow-up
Purpose: The aim of this study was to evaluate the feasibility of endovascular treatment (EVT) for unruptured small wide-necked ophthalmic segment aneurysms (OSAs) of the internal carotid artery (ICA). Materials and Methods: Between January 2010 and May 2013, 17 patients with 19 unruptured small wide-necked OSAs received EVT. Occlusion rates were classified as total/near-total (95-100%), subtotal (80-95%) and partial (<80%) occlusions. This was determined by immediate and follow-up angiography. Follow-up outcome was assessed by using modified ranking scale (mRS). Results: EVT was successfully performed in all the patients: Coiling only in 2 (10.5%) and stent-assisted coiling in 17 (89.5%) of the aneurysms. The immediate total/near-total occlusion was seen in 1 (5.3%), sub-total occlusion in 8 (42.1%) and partial occlusion in 10 (52.6%). Follow-up angiography performed at 9.4 (±4.7) months revealed total/near-total occlusion in 13 (68.4%), subtotal occlusion in 5 (26.3%) and partial occlusion in 1 (5.3%). At the end of the follow-up period of 17.4 (±6.9) months, no aneurysm rupture was found and 16 (94.1%) of the patients had mRS scores of: Grade 0 in (5.9%) an mRS 1 in the remaining. Conclusions: EVT may be feasible and effective treatment option for unruptured small wide-necked ophthalmic aneurysms of the ICA.
Keywords: Aneurysm, endovascular treatment, ophthalmic segment
Ophthalmic segment aneurysms (OSAs) arise from the internal carotid artery (ICA), reaching from the distal dural ring to the origin of the posterior communicating artery (Pc0mA) and are the C6 segment aneurysms coined by Bouthillier et al. , OSAs are often intradural with risk of lethal subarachnoid hemorrhage (SAH). However, the complex adjacent anatomy, proximity to the optic chiasma and the clenoid process, makes the treatment of OSAs by microsurgery difficult. , With the development of techniques of stent assistance and balloon remodeling techniques, endovascular treatment (EVT) for OSAs has become a popular treatment option. ,, The International Study of Unruptured Intracranial Aneurysms (ISUIA) and the California Unruptured Aneurysm Database have provided evidence that endovascular therapy may be safer than open microsurgical clipping for many patients with unruptured aneurysms. , However, for unruptured small wide-necked OSAs, endovascular coiling remains challenging. This study investigated the feasibility and efficacy of EVT for the treatment of unruptured small wide-necked OSAs.
Between January 2010 and May 2013, 17 patients (7 men, 10 women and mean age 59.0 ± 9.2 years, range 39-73 years) with 19 unruptured small wide-necked OSAs have received EVT. To obtain outcome data: Radiographic studies, details of endovascular procedure and follow-up data were retrospectively reviewed from the case records. Informed written consent was obtained from all the patients prior to inclusion in the study.
The inclusion criteria: (1) Unruptured OSA with or without associated clinical features, confirmed by computed tomography (CT) scan and CT-angiography, magnetic resonance angiography or digital subtraction angiography (DSA); (2) OSAs of small size (≤7 mm) on angiography; (3) a wide-necked OSAs with a neck width ≥4 mm or a fundus-to-neck ratio ≤2; (4) high risk cases for surgical clipping; and (5) availability and willingness for at least one follow-up angiographic assessment (>3 months postop).
Aneurysm size, including the neck and sac, was measured on the angiogram that could best display the configuration, with 3-dimensional (3D)-DSA reconstruction being the reference, for aneurysm categorization and calculation of the mean values.
Treated OSAs were classified into four types based on the angiographic findings and endovascular considerations from previous reports:  Type A: Aneurysms arising from the superior ophthalmic segment of the ICA and related to the ophthalmic artery; Type B: Aneurysms arising from the superior ophthalmic segment of the ICA, without involvement of the ophthalmic artery; Type C aneurysms arising from the ventral ophthalmic segment of the ICA; Type D: Aneurysms arising from the medial or lateral ophthalmic segment of the ICA [Figure 1].
EVT was performed under general anesthesia and systemic heparinization. All patients underwent cerebral angiography and rotational angiography with 3D image reconstruction using a bi-planar angiography unit (Artis zee biplane system; Siemens, Germany). Endovascular strategies included coil embolism alone in 2 (10.5%) and stent-assisted coiling in 17 (89.5%) aneurysms (Neuroform, Boston Scientific, Natick, MA and Solitaire AB, eV3, Plymouth, MN, USA). After steam shaping based on 3D images, the microcatheters (Enchalon-10, eV3, Plymouth, MN, USA) were introduced over a microguide wire (Transcend Platinum; Boston Scientific, Natick, MA, USA) and navigated into the aneurysm sacs. Various types of detached platinum coils were used for aneurysm sac embolism (Boston Scientific, Natick, MA, Microventions, Tustin, CA and Cordis, Miami, FL, USA).
After a guiding catheter (Envoy, Cordis and Miami, FL, USA) was positioned at the C2 segment of the ICA, a bolus of 5,000 U of heparin was injected intravenously and an additional 1000 U was added every hour to maintain heparinization during the procedure. Non-enhanced brain CT scans were obtained in all patients before and after the procedure to exclude any intracranial hemorrhage. CT scans were also performed when new neurological symptoms occurred. Detailed neurological examinations were conducted by neurointerventional radiologists and neurosurgeons before the procedure, immediately after the procedure and thereafter daily until hospital discharge. If a stent was used, all patients received double anti-platelet therapy, 300 mg clopidogrel and 300 mg aspirin as a loading dose. Heparin, 125,000 U diluted in 500 ml saline/day was infused at the rate of 18 ml/h for 3 days after the procedure and fallowed by 75 mg/day clopidogrel and 100 mg/day aspirin for at least another 6 months.
Immediate and follow-up angiographic data were reviewed by two independent experienced neurosurgeons and classified into three categories: Total/near-total occlusion (95-100%), subtotal occlusion (80-95%) or partial occlusion (<80%. Recurrent aneurysms after coiling were categorized as "reopened" and as "residual sac enlargement".
Patient status on admission and clinical outcomes were evaluated using the modified ranking scale (mRS). Procedure-related morbidity was defined as EVT-related neurological deficit 24 h after the procedure. The neurological deficit, or events including cerebral hemorrhage or infarction, was considered EVT-related if it was caused by aneurysm rupture, coil migration, stent migration, acute in-stent thrombosis, thrombo-embolic events or vasospasm.
Immediate angiographic results and complications
All treated aneurysms were saccular, with mean aneurysm neck and sac diameters of 4.1 ± 1.2 mm and 2.9 ± 0.8 mm, respectively. On the basis of our The OSAs classification was: 3 (15.8%) as Type A, 3 (15.8%) as Type B, 4 (21.1%) as Type C and 9 (47.4%) as Type D [Table 1].
The success rate of either procedure was 100%. Microcatheter reshaping was performed according to the aneurysm type, in which an anti-'S'-shaped microcatheter head was usually built for Type A and Type B OSAs, a 2D 'pigtail' for Type C and a 3D 'pigtail' for Type D. In all 17 OSAs that required stent assistance, the stent-jailing technique was mostly frequently applied (15, 88.2%), especially when a closed-cell stent such as Solitaire AB was used for superior aneurysm neck remodeling. The stent jacket technique was used in the remaining two aneurysms due to failure of jailing. Immediate angiography data revealed successful total/near-total occlusion in 1 (5.3%) of the aneurysms, sub-total occlusion in 8 (42.1%) and partial occlusion in 10 (52.6%) [Figure 2] and [Figure 3].
EVT-related complications included coil escape into the parent artery after detachment in one patient with OSA Type D. The coil was then compressed to the arterial wall by a Neuroform stent. CT-confirmed brain infarction in one patient with OSA Type D, who presented with transient left lower limb weakness. It was considered an intra-procedure thrombo-embolic event.
Angiographic follow-up results
The mean angiographic mid-term follow-up was 9.4 ± 4.7 months (range, 3-20 months). Follow-up total/near-total occlusion rate was 68.4% (n = 13), subtotal occlusion rate was 26.3% (n = 5) and the partial occlusion rate was 5.3% (n = 1). The recurrence rate was 10.5% (n = 2): Including neck reopening and residual sac enlargement in 1 patient each, both of which required no further treatment [Table 1]. Of the different OSA types, both Type A and B aneurysms achieved total occlusion rates of 100%, 50% in Type C and 55.6% in Type D aneurysms. Angiographic results also confirmed higher recurrence rates in OSA Types C (25%) and D (11.1%) aneurysms than in Types A (0%) and B (0%) aneurysms [Table 2].
Clinical baseline and follow-up results
On admission, SAH was not present in any of the patients as confirmed by CT. The mean clinical follow-up period was 17.4 ± 6.9 months (range, 9 to 33 months). In 16 (94.1%) patients mRS Score was 0 and 1 in 1 (5.9%), scores similar to the pre-procedures period. The two patients with coil escaping or thrombo-embolism showed no residual or deteriorated neurological symptoms. No cases of bleeding or death were observed [Table 2].
In this retrospective study of treating unruptured small wide-necked OSAs with coiling only and stent-assisted coiling techniques, the success rates were each 100%. Most patients achieved good immediate and follow-up angiographic outcomes, with a recurrence rate of 10.5% and good follow-up clinical outcomes.
The site of origin, projection and relationship of aneurysms arising from the ophthalmic segment of ICA to adjacent structures are complex and make their surgical management often difficult.  The key features of successful surgical treatment of these lesions include establishing control of the proximal artery, adequate exposure of the aneurysm neck and successful obliteration of the aneurysm with minimal manipulation of the optic nerve. , However, procedure-related permanent morbidities associated with surgical clipping of OSAs could be up to 8.7% and complete visual loss can occur in about 3.5% of patients. , For OSAs with ventral or medial projection, at branching sites or wide-necked, both surgical clipping and endovascular embolism are considered challenging. Advances in neurointerventional techniques have facilitated more effective and safer treatment of OSAs. In recent, studies complete and near-complete occlusion of OSAs approached 62%, with a complication rate less than 5%. , EVT is suitable for OSAs and rarely causes visual impairment as it avoids damage to the optic tract when compared with surgical clipping. 
Unruptured cerebral aneurysms are diagnosed with greater frequency with improved imaging techniques, however their treatment remains controversial. This is because of incomplete and conflicting data about the natural history of aneurysm and the risks ng associated with treatment. ISUIA study recommends treatment for unruptured aneurysm either by surgical repair or by endovascular embolism considering the factors such as aneurysm size, location and age.  The California unruptured aneurysm database analysis indicated that EVT was superior to surgical clipping, with a mortality rate of 0.5% versus 3.5%.  However, compared with aneurysms in other ICA locations or posterior circulation aneurysms, OSAs have a lower rupture risk ,, OSAs with smaller bodies or wide necks, are likely to have a higher coiling risk. Until now, only a few studies have focused on EVT for these challenging OSAs. Our results even though in a small number of patients with unruptured small, wide-necked OSAs suggest that EVT for these OSAs may achieve a high technical success rate with good follow-up angiographic total/near-total occlusion rates. The rate of total/near-total occlusion in our study was comparable to the observation in the previous similar studies. The complete occlusion rate was 87.8% in 73 patients with paraclinoid aneurysms.  and 70.9% in 67 patients with paraclinoid aneurysms.  The reported rates of major or permanent complications ranged between 3.8% and 8.3%. ,, However, we have not experienced any major EVT-related complications, only two minor complications.
In our institute stent-assisted techniques are more frequently used for the treatment of small, wide-necked and unruptured aneurysms because in our experience it can facilitate a higher coil packing density and a more stable aneurysm neck sealing and reduce the risk of coil herniation into the parent artery.  Ogilvy et al. in their study have reported that 60% of the paraclinoid aneurysms treated with stent assistance had a post-treatment occlusion rate of ≥95%, a result comparable with the rate of 54.2% after coiling alone.  Moreover, stent placement in patients with unruptured aneurysms may have a lower risk of in-stent thrombosis rate than ruptured aneurysms. However, the immediate post procedure total/near total occlusion rate in our study was only 5.3%, which was much lower than that of coiling-treated aneurysms in other study.  This result is probably due to a poorer microcatheter stability in the aneurysm sac at this segment and the lower coil density in small aneurysm sacs. With regard to stent selection in the presence of a tortuous parent artery, we had selected a neuroform stent because of its good compliance with the arterial wall. Otherwise, the retrievable Solitaire stent may be selected because repeated placement adjustments can be made before stent detachment.
Classification of OSAs is not only useful for surgical clipping but also gives directions for endovascular coiling. In our experience, ventral OSAs (Type C) are most difficult for clipping, whereas medial OSAs (Type D) are the most difficult for coiling. Difficult manipulation is often likely to result in technique-related complications, as in our study, where both coil escaping and a thrombo-embolic event occurred in OSAs Type D. Microcatheter shaping is one of the most important technological components of EVT for OSAs. Generally, an anti-'S' figure is effective for Type-A and -B OSAs, a simple 'cobro' or 2D "pigtail" shaping for Type C and a 3D "pig-tail" for Type D. , Different types of OSAs may also have different rates of total occlusion, as suggested by our mid-term follow-up angiographic results. In our study, Types A and B had higher total/near total occlusion rates than Type C and Type D. The suggested reason is that aneurysms arising from the superior ophthalmic segment of the ICA may be subjected to lower wall shear stress or favorable blood flow pattern than aneurysms at other sites, associated with a reduced coil compression rate and facilitated thrombus formation in sacs. ,
In recent years, clinical application of flow-diverting devices has become an alternative effective technique for treating aneurysms, especially for unruptured complicated aneurysms arising from the ICA. ,, However, whether the high metal surface area coverage of the flow-diverting device (approximately 30-35%) may cause ischemia events, influences long-term patency of the ophthalmic artery or causes aneurysm rupture due to aneurysm wall inflammation induced by hemodynamic alternations has to be further studied before routine application can be considered.
Our results suggest that EVT may be a safe and effective treatment option for unruptured small wide-necked OSAs. However, the major limitation is the small patient sample size and the relatively short follow-up period. Longer-term follow-up angiographic and clinical data from a larger study population are needed to determine firmly the safety and efficacy of EVT for unruptured complex OSAs. Moreover, considering a relative low risk of rupture and more difficulty handling of the ophthalmic aneurysms, the decision whether to perform active treatment of such kind of aneurysm or not probably mostly depends on experiences of different centers.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2]