Recurrence and retreatment of anterior communicating artery aneurysm after endovascular treatment CURRENT STATUS: UNDER REVIEW

Background To describe our of anterior (Acom) aneurysms (AcomAs), and to evaluate the incidence of recurrence and risk factors for recurrence and retreatment Methods The study comprised 260 patients who were treated at a single center between January 2010 and December 2018. Patients who had EVT, including stent-assisted coiling of AcomA, were included. All medical records were retrospectively reviewed. The incidence of recurrence and risk factors for recurrence and retreatment were evaluated. Univariate analysis and multivariate logistic regression analysis were conducted. Results Recurrence of AcomA occurred in 38 (14.6%) patients. Mean follow-up duration was 27 months. Multivariate logistic regression analysis indicated that ruptured aneurysm (odds ratio OR 3.55, P=0.001), dome direction (anterior) (OR 3.86, P=0.002), maximal diameter (OR 1.19, P=0.02), and mean age (OR 0.96, P=0.02) were independent risk factors for aneurysm recurrence. Among cases of recurrence, 10 (3.8%) patients had major recurrence. Ruptured aneurysm (OR 14.7, P=0.004), maximal diameter (OR 1.56, P=0.02), inflow angle (OR 1.04, P=0.03), and Raymond-Roy classes II and III (incompletely occluded status of coiling) (OR 6.19, P=0.03) showed significant relation to retreatment in multivariate logistic regression analysis.


Introduction
Anterior communicating artery aneurysm (AcomA) has a higher risk of rupture than other aneurysms. 1 Therefore, treatment of AcomA is essential and critical for prevention of aneurysmal subarachnoid hemorrhage (SAH). However, surgical treatment of AcomA remains a challenge as it can cause morbidity to patients because it may require rectus gyrus suction and may even cause olfactory nerve injury. AcomA shows a variety of anomalies and complex anatomy like fenestrated, triplicated, and azygous anterior cerebral artery (ACA). It also has many perforator and associated vessels, like the recurrent artery of Heubner and bilateral A1 and A2. [2][3][4] Since endovascular treatment has rapidly replaced surgical clipping, Alshekhlee reported that hospital mortality was higher in patients who had surgical clipping than in those who had endovascular coiling (1.6% and 0.57% respectively). 5 However, retreatment after aneurysmal treatment was performed in 17.4% of patients in the endovascular treatment (EVT) group and in 3.8% of patients in the surgical clipping group in an International Subarachnoid Aneurysm Trial (ISAT) study. 6 The study also found that young age, large lumen size, and incomplete occlusion were risk factors for retreatment after EVT. O'Neil et al reported in their systemic analysis that endovascular coiling resulted in significantly lower treatment-related morbidity compared with clipping, but clipping resulted in significantly lower angiographic recurrence and retreatment. 7 The purpose of this study was to evaluate recurrence and retreatment rates and their risk factors after endovascular treatment of AcomA.

Materials And Methods Patients' Population
This study was approved by our Institutional Review Board, and the requirement for patients' informed consent was waived due to its retrospective design.
Between January 2010 and December 2018, a total of 514 AcomAs were consecutively treated by either surgical or endovascular method in our institute. Patients who had vasculitis or infectious, fusiform, and blood blister-like aneurysms were excluded from this study. Patients whose aneurysms were treated with flow diverter were also excluded, and finally, 260 AcomA patients who underwent coiling with or without stents were included. All clinical and radiological data were obtained from electronic medical records and a prospectively registered aneurysm database, and they were retrospectively reviewed.

Radiologic Evaluation
All radiologic images were retrospectively reviewed by two investigators.
At the initial coiling, all aneurysms were assessed with 3D angiogram (Allura Xper FD20/20 and Allura Clarity, Philips Medical Systems, Best, The Netherlands). Based on 3D angiogram, aneurysm factors, including maximal size, size ratio, inflow angle, hypoplastic A1, and dome direction were evaluated.
Size ratio was calculated by dividing the aneurysm maximal size (mm) by the average size of both A1 segments (mm). 8 Inflow angle was defined as the angle between the maximal height of the aneurysm and the parent vessel. 9 Hypoplastic A1 was defined as A1 with diameter less than 50% of the diameter of contralateral A1. 10 Additionally, we dichotomized aneurysm direction into anterior or posterior by drawing an imaginary perpendicular line to the anterior cranial fossa using sagittal view of computed tomography angiography (CTA). 11 For the correction of different values between the two investigators, mean values were obtained.
After the initial coiling, routine imaging follow-up was conducted with 3.0-T magnetic resonance angiography (MRA) at 6, 18, 30, and 60 months. Post-treatment and follow-up angiographic occlusion grade were assessed according to Raymond-Roy classification, wherein class I is defined as complete occlusion, class II as neck remnant, and class III as sac remnant. 12 Recurrence was defined as any progression of Raymond-Roy class. All class III recurrent aneurysms were further evaluated with 3D angiogram.

Retreatment
Retreatment was considered for aneurysms if volume of recurred sac was ≥ 20.0 mm 3 on 3D angiogram. Retreatment was planned based on consensus by a multidisciplinary team.

Statistical Analysis
Univariate analysis was performed to determine the association of procedure-related complications with other factors. The statistical significance of recurrence was analyzed using the chi-square test for categorical, nominal variables, or the logistic regression test for continuous, numerical variables.
Multivariate logistic regression analysis was performed for variables with an unadjusted effect and a P-value < 0.20 in univariate analysis to determine independent associations of recurrence and retreatment with other factors. Results of binary logistic regression were reported as odd ratios (ORs) with P-value < 0.05 for a 95% confidence interval (CI), which was considered statistically significant.
Correlation analysis and chi-square test were performed to identify the strength of relationships between age and statistically significant factors such as maximal diameter, rupture status, and anterior dome direction. All statistical analyses were performed with Statistical Package for the Social Sciences (SPSS) version 19.0 (IBM Corp., Armonk, NY, USA).
All aneurysms were successfully treated without (n = 208, 80.0%) or with (n = 52, 20.0%) stents, and  These results are shown in Table 1. Correlation analysis and chi-square test performed for evaluation of relation of age and statistically significant factors. As a result, p value of correlation analysis between age and maximal diameter was 0.068 which was not statistically significant at the significance level of 5%, but the p-value was close to being statistically significant. The other factors; ruptured status and dome direction was also not statistically significant correlation with age.  (Fig. 1), and the other six patients with EVT using stent assisted coiling (SAC) (Fig. 2). The detail of factors is shown in

Discussion
This study evaluated the incidence of recurrence and risk factors for recurrence and retreatment of anterior communicating artery aneurysm after endovascular treatment. We found that the recurrence rate of AcomA after EVT was 14.6%, and its associated factors were younger age, rupture status, bigger aneurysm size, and anterior dome direction. The rate of retreatment, which signifies major recurrence, was 3.8%. Patients with AcomA with a large inflow angle, ruptured status, large aneurysm size, and incomplete aneurysm occlusion may be at a high risk of retreatment.
Many studies have shown that surgical procedure for AcomA can lead to post-operative deficits such as memory impairment and personality changes. 13, 14 Ramos et al also reported that gyrus rectus resection can cause cognitive and psychiatric dysfunction caused by orbital prefrontal cortex lesion or a disconnection in the ventromedial circuits. 15 Another key problem of this surgery is olfactory nerve injury. In some cases of clipping of AcomA, the frontal lobe inevitably is retracted to some degree in order to access the aneurysm. During frontal lobe retraction, the olfactory nerve is also retracted from the cribriform plate. Park et al reported objective olfactory dysfunction rate of 10.8% in their study. 16 Because of the risks of these complications, minimal invasive surgery like keyhole approach with eyebrow or palpebral incision was introduced in AcomA surgery. 17-19 These approaches reduced the amount of brain retraction and rectus gyrectomy. Some authors even advocated for use of orbitotomy approach, 20,21 and recently, endoscopic endonasal approach was reported for AcomA surgery. 22 Vascular anomalies such as multiple, fenestrated, or azygous artery are common in Acom. [23][24][25][26] These anomalies originate during embryologic development. The prevalence of duplication of Acom is 18% and that of fenestration of Acom is 12-21%. 27  Corns et al reported that younger age predisposed to a higher risk of recurrence in ruptured aneurysm. 33 However, their study did not give definite explanation for this observation. In some studies, younger age was a predictor of growth of aneurysm after clipping, 34,35 but this age effect is difficult to understand. Our study also showed that younger patients had more recurrence than older patients. This age effect can be explained in two ways. One, there was a bias toward more frequent surveillance imaging in younger patients. Second, is the correlation with other factors that signify recurrence. We therefore performed a correlation analysis, after which correlation between age and maximal diameter was not statistically significant at the significance level of 5%, although the P-value of 0.068 was close to being statistically significant. Therefore, we cannot completely ignore the association (correlation coefficient = 0.114).
Anterior dome direction is known to reflect aneurysm hemodynamics, including wall shear stress and flow velocity, which play important roles in the growth and rupture of aneurysm. 36 This may explain its significant association with recurrence in this study.

Retreatment
In ISAT study, younger age, large lumen size, and incomplete occlusion were risk factors for late retreatment after EVT. 6 Smoking is known as one of the most important risk factors for formation and rupture of intracranial aneurysm, 37-39 and that association was explained by inhibitory effect of cigarette smoke on alpha 1-antitrypsin (A1AT). 40 Brinjikji however reported that smoking was not an independent risk factor for aneurysm recurrence (OR = 1.04, P = 0.87) and retreatment (OR = 0.82, P = 0.50) for patients receiving EVT for aneurysm. 41 Recently, Futchko found that history of smokingwhether current or former-was associated with a significantly increased risk of aneurysm recurrence.
The odds ratios for aneurysm recurrence in current and former smokers were 2.73 and 2.69, respectively, compared with never smokers. The author accounted for the difference between the Brinjikji's study and their study. The former study exclusively used balloon-assisted coiling, whereas the latter study used stent-assisted coiling. 42 Our study showed no association between smoking and recurrence.
Using computational fluid dynamic (CFD) analysis, Merih et al identified inflow angle as an independent and robust rupture status differentiator in intracranial aneurysm. 9 CFD showed that increasing inflow angle led to deeper migration of flow with higher peak flow velocities and a greater transmission of kinetic energy into the dome. Wenjun et al also revealed that an inflow angle of over 90° and incomplete occlusion were associated with aneurysm recurrence in unruptured aneurysm after EVT. Our study showed that inflow angle was not related with recurrence, but associated with retreatment.
Rupture status and large aneurysm size were not only associated with recurrence, but also with retreatment. On the other hand, incomplete coil packing status was associated with retreatment but not with recurrence.

Limitation
This study had several limitations. First, the decision to retreat relied on the clinician's decision, so it was highly subjective. This could have led to bias. Second, recurrence and retreatment cases were few compared with the other group which showed no recurrence and retreatment. This could be a factor in lowering the statistical reliability of the results. Third, inflow angle was not accurate. Even though we used 3D rotation image, there could be inter-observer difference in measurement. Fourth, inflow angle could change after EVT especially in SAC. However, in this study, only 52 patients (20%) were treated with SAC.

Conclusions
In our study, recurrence rate of AcomA after endovascular treatment was 14.6%. Younger age, rupture status, large aneurysm size, and anterior dome direction were significantly associated with recurrence. Retreatment rate of recurrent AcomA cases after EVT was 3.8%. Patients with AcomA with a large inflow, rupture status, large size, and incomplete occlusion may be at a high risk of retreatment of recurring aneurysm. These patients need to be closely followed up after EVT.

Consent for publication
Not applicable

Availability of data and materials
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no conflict of interest.

Funding
No funding was received for this study.

Authors' contributions
CKJ had contributions to the concept of the study, interpretation of data, drafting the manuscript, and revising the manuscript. JC had contributions to the acquisition of data, analysis of data, and revising the manuscript. JWL and SKH had contributions to the interpretation of data and revising the manuscript. NHS had contributions to statistical significance analysis. KYP contributed to the concept of the study, revising and editing the manuscript, and supervising the whole process of the study. All authors have approved the final version of the manuscript.