We retrospectively reviewed the records of nine (9) patients (5 males- 4 females) who underwent additional microsurgical clipping after initial endovascular treatment in our department over a 12-year period (2010-2022). Patients’ medical records and radiographic studies were retrospectively reviewed in order to obtain clinical and radiographic data.

All patients underwent follow up with digital subtraction angiography for post embolization evaluation at 6 and 18 months. In some cases where extra information was needed (e.g. preparation for microsurgical clipping), magnetic resonance angiography or computed tomography angiography were done.

The assessment of the angiographic results was conducted based on the Raymond-Roy Occlusion Classification (RROC) [44]. Embolization was considered to be complete (Raymond Roy class 1) if there was no contrast filling of the dome, body, or neck of the aneurysm. Neck remnant (Raymond Roy class 2) was defined as residual filling of the neck without opacification of the aneurysmal sac. Residual aneurysm (Raymond Roy class 3) was indicated by the circulation of contrast agent in the dome of the aneurysm. Neurological assessment has been performed preoperatively and at final follow-up by using the modified Rankin scale score.

Meanwhile, modified Rankin Scale (mRS) and Glasgow Outcome Scale (GOS) were used to categorize the literature’s patient population into three (3) major categories: Good outcome (mRS 0-2, GOS 5-6), Bad outcome (mRS 3-5, GOS 3-2) and Death (mRS 6, GOS 1).

A medical council composed of an endovascular neurosurgeon and an interventional neuroradiologist was responsible for the evaluation of the retreatment options.

The study was performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. Due to the study’s retrospective nature, ethics committee approval was not required.

The search for studies eligible for inclusion in our review was conducted using PubMed and Google Scholar, incorporating papers whose publication dates range from 2000 to 2021. For our search strategy, we combined, using both ‘AND’ and ‘OR’ combinations, the following keywords: endovascular treatment, microsurgical treatment, recurrence, initial treatment, retreatment, embolization, coil, intracranial aneurysms. Studies were only reviewed if the published manuscript was in English. We included studies involving microsurgically retreated cases after initial endovascular treatment with GDCs and accepted studies that are comprised of at least one case. Publications using the same patient population were excluded. There were no anatomical exclusion criteria. Studies with incomplete data regarding patients’ sex and age, median latency and post-surgical obliteration rate were still included, while studies with incomplete surgical retreatment and surgical indication data were excluded.

A total of 9 patients (5 males- 4 females) were included in our study with a mean age of 49 years (range 46 to 63). Subarachnoid hemorrhage was the initial presentation in 78% (7/9) of patients. (Table 1). Aneurysms’ most common location was the middle cerebral artery bifurcation (5/9) followed by the anterior communicating artery (3/9). One aneurysm (1/9) was located at the internal carotid artery bifurcation.

All patients were initially treated by simple coiling. Mean aneurysm diameter was 6.14 mm (range 4 to 17.4 mm). Considering their morphologic characteristics, 7/9 aneurysms had regular whereas 2/9 had irregular shapes. Mean time interval from initial endovascular treatment to surgical clipping was 25.5 months. In every case, the operation was scheduled in a 15-day period from the discovery of the recurrence. Indications for surgery were recanalization alongside coil compaction in 2 IAs, regrowth in 2 IAs, formation of a new bleb in 1 IA, and residual neck in 4 IAs.

Recanalization was defined as the reappearance of contrast filling in a previously successfully occluded aneurysm sac. Residual neck was defined as the persistent post-embolization contrast filling of the original aneurysmal neck without opacification of the aneurysmal sac [28,44]. Regrowth was defined as enlargement of the initial aneurysm sac or the formation of a new daughter sac.

The clinical presentation of our patients before and after microsurgery was evaluated using the modified Rankin Scale (mRS). Mean clinical follow-up after clipping was 22.7 months (ranging from 1 to 78 months). Pre- and postoperative mRS score is presented in Table 1. All patients had good clinical outcomes with preserved or improved mRS scores. Procedure-related morbidity and mortality were zero whereas complete aneurysm occlusion was achieved in all cases after surgical clipping. Neurological assessment at the final follow-up indicated that all patients had minimal symptoms or were asymptomatic (mRS 0-1). Neurological status improved for 6 patients whereas the rest (3/9) remained stable (Fig. 1).

We conducted a retrospective review of 48 studies that met our criteria of selection and are presented in Table 2. For the statistical data that follow we used only studies from the table that provided the needed data. Based on the studies above, a total number of 969 patients underwent microsurgical treatment for 976 previously coiled aneurysms. The male-to-female ratio (M:F) was approximately 1:2.

The mean interval between the initial embolization and the microsurgical retreatment in 889 patients (studies with provided data) was 16.12 months. Furthermore, we categorized the surgical indications into 5 major categories: (a) Recurrence, which includes coil compaction, recanalization, and regrowth, (b) Mass Effect (ME), (c) Coil Migration (CM) which includes Coil Protrusion and Coil extrusion, (d) Rebleeding, (e) Residual Aneurysm. For the majority of patients, the surgical indication was recurrence of the aneurysm (59.14%), followed by residual aneurysm (28.63%) (Fig. 2). The treatment strategy was divided into 3 categories: (a) clipping with or without coil extraction, (b) bypass, (c) other treatment strategies such as Parent Artery Occlusion (PAO) and wrapping. As expected, the first choice of retreatment was clipping (91.3%). Fig. 3 presents patients’ data concerning retreatment strategies.

Most of the studies (42 out of 48) provided patients’ postoperative outcome data (Table 3). According to these, (data available for 868 patients), a good outcome (GOS 4-5, mRS 0-2) was observed in 88.9%, a bad outcome (GOS 2-3, mRS 3-5) in 7.5%, whereas 3.6% died (GOS 1, mRS 6), as illustrated in Fig. 4. Furthermore, 19 out of 48 studies additionally provided the preoperative patients’ clinical status (Table 4). According to the data available for 420 patients, the majority of patients remained stable postoperatively and maintained their good preoperative clinical status (GOS 4-5, mRS 0-2) (87.9% preoperatively vs 87.6% postoperatively), as illustrated in Fig. 5.

A 47-year-old female patient presented with SAH due to a ruptured narrow neck anterior communicating artery (ACoA) aneurysm (A, white arrow). The aneurysm had an irregular shape, and the Dome/Neck ratio was 2.33. The patient was initially treated with an endovascular approach using Guglielmi Detachable Coils (GDCs) and full occlusion was achieved (B, white arrow). In the angiographic follow-up nineteen months later, recanalization of the neck alongside with coil compaction (C, red arrow) was observed, and the patient was admitted for microsurgical retreatment. Right frontotemporal craniotomy was performed, and the aneurysm neck was carefully dissected (D, white arrow). A 45° angled clip was placed at the aneurysm neck beneath the coil mass (E, white arrow). A 100% obliteration rate was accomplished, as demonstrated in postoperative DSA (F, G, white arrow) and follow-up MRA (H). The patient had an uneventful recovery (post-clipping mRS 0).

A 46-year-old male patient presented with a ruptured supraophthalmic right ICA aneurysm (A, white arrow) that was emergently embolized by balloon-assisted technique (B, white arrow). During the follow-up period, a newly formed (de novo) aneurysm was observed at the right ICA bifurcation (C, D, red arrow). The Dome/Neck ratio was 2,8 with an irregular shape. Complete endovascular occlusion of the aneurysm by simple coiling was performed (E, red arrow). Five months later, regrowth of the ICA bifurcation aneurysm sac was observed in follow-up DSA (F, red arrow). Intraoperative view after right pterional craniotomy shows the neck of the regrowing aneurysm (G, red arrow). Direct clipping of the ICA bifurcation aneurysm was performed. Postoperative DSA confirmed the optimal placement of the clip at the neck of the ICA bifurcation aneurysm (I, red arrow) and complete angiographic aneurysm obliteration (H, red arrow), while the initially ruptured supraophthalmic aneurysm remained stable (H, white arrow). No perioperative complications occurred, while the neurologic status improved from mRS 1 to mRS 0.

A 63-year-old man was admitted lethargic due to a 4 mm ruptured wide-neck ACoA aneurysm (A, black arrow). During embolization using balloon-assisted technique, re-rupture of the aneurysm occurred after deployment of the first coil (B, black arrow). Further embolization with coils, keeping the microcatheter in place, resulted in complete aneurysm occlusion from outside to inside (C, black arrow). The patient experienced a full recovery and was discharged after 15 days. Nevertheless, a follow-up angiography revealed a newly formed aneurysm projection (C, D, black arrow), which was treated by microsurgical occlusion. Intraoperative view shows the new aneurysm bleb (E, black arrow). A mini-clip was used to occlude the new bleb (F, G, black arrow) followed by aneurysm wrapping. Postoperative angiogram demonstrated complete occlusion of the aneurysm (H, black arrow). No perioperative complications occurred.

Aneurysms’ recurrence rate among patients receiving endovascular treatment is estimated at roughly 20%, a percentage that varies between patients depending on several risk factors [7,12,37]. Aneurysms with complete occlusion exhibit lower recurrence rates as opposed to inadequately occluded or those with residual neck. Aneurysm sac >10 mm and neck size >4 mm are also significant risk factors. Although some studies report that aneurysm location is not a risk factor, other studies suggest that posterior circulation aneurysms have a higher rate of recurrence in relation to anterior circulation aneurysms. Previously ruptured aneurysms tend to correlate with an increased probability of recurrence [7,11,12,42,56]. According to data extracted from our review, we classified the mechanisms of recurrence after endovascular treatment into 4 types based on follow-up angiographies (Fig. 9). Type I: coil loosening; presence of contrast fluid in a diffuse, spotted pattern among intra-aneurysmal coils, caused by a combination of gradual coil loosening and hemodynamic factors. Type II: coil compaction; compression and subsequent volume reduction of coils inside the aneurysm without enlargement of the aneurysmal sac or neck. Type III: aneurysm regrowth; it can be subclassified as follows: IIIa: neck growth with or without coil compaction, IIIb: aneurysmal sac growth with or without coil compaction, IIIc: a new aneurysmal sac formed from the original aneurysmal neck. Type IV: coil migration, which is further subclassified as Type IVa: coil extrusion outside of the aneurysmal sac, and Type IVb: coil protrusion in the parent artery.

Even though the recurrence rate is about 20%, retreatment is pursued at roughly half of those, i.e. 10% of all embolized aneurysms [12,42]. On the contrary, ISAT and BRAT trials demonstrated retreatment rates of 17.4% and 20% respectively [4,47]. Recurrence is the main indication for retreatment, as it may lead to either re-rupture or an increased risk of treatment failure [17]. The Cerebral Aneurysm Rerupture After Treatment Trial (CARAT), proved that the obliteration rate of the aneurysmal sac has a direct correlation to the re-rupture risk [19]. More specifically overall re-rupture risk was reported to be: 1.1% for complete occlusion, 2.9% for 91% to 99% occlusion, 5.9% for 70% to 90%, and 17.6% for <70% [19].

In comparison to clipping, coiling comes with a higher risk of rebleeding, rerupture and retreatment. More specifically, the rebleeding rate after endovascular treatment as presented in the CARAT study (3.4%) is higher than that of clipping (1.3%) [19]. Similar results are presented by Hulsbergen et al. with a rebleeding rate of 2.1% and 1.1% in initial endovascular and clipping treatment respectively [17]. Furthermore, ISAT demonstrates a 2.1% rebleeding rate after initial embolization and 1% after surgical clipping [4]. According to the BRAT trial, 20% of patients who received endovascular treatment required retreatment in the long term, in comparison to less than 1% of patients who went through surgical treatment [47]. Retreatment rates of cerebral aneurysms increased over time after endovascular treatment, thus indicating the need for long-term follow-up after the aforementioned procedure [4].

A reasonable retreatment strategy is the endovascular approach. Li et al. in their study, report that out of 1385 aneurysms that were initially treated endovascularly, 5.6% were retreated by further re-embolization. This percentage represents 51.6% of the total number of recurrent aneurysms [27]. Comparable results are seen in a study conducted by Dorfer et al, in which 59.1% of recurrent aneurysms received endovascular treatment as the second approach [11].

The endovascular retreatment strategy is considered an effective approach, especially in cases that are presented with coil compaction and recanalization, however, it does not show similar effectiveness when it comes to aneurysm regrowth [11,58]. Although morbidity rates in endovascular retreatment are lower compared to the microsurgical approach, this advantage may be offset by the higher rates of incomplete aneurysm occlusion and recanalization [39,42]. Endovascular retreatment techniques include standard coiling, stent-assisted coiling, balloon-assisted coiling, covered stents, and flow diversion.

Despite the fact that microsurgical clipping of recurrent, previously embolized aneurysms may be a technically demanding technique, it may be a safer and more effective definitive treatment option when compared to endovascular retreatment [42]. Pirayesh et al. suggest that major indications for microsurgical retreatment are aneurysm recanalization and regrowth, while another indication worth mentioning is neck remnant after partial endovascular coiling [39]. Wang et al. mention to their paper that while recoiling is preferred because is minimally invasive, surgical clipping offers definitive closure of the aneurysm and can be used in cases presented with mass effect, broad neck aneurysms and cases with recanalization after repeated coiling. To achieve surgical clipping adequate neck for the placement of the clip is needed. However, no method can evaluate and predict sufficiently the success of microsurgical treatment in a previously embolized aneurysm [53]. Moreover, according to Yu et al., microsurgical treatment is suggested for type III-V aneurysm recurrence according to their classification. These types include aneurysms with formation of a new neck remnant, with or without coil compaction in the initial aneurysmal sack (III, IV), and aneurysms with a newly regrown aneurysmal sac from the initial aneurysmal neck (V). Contrariwise, for large and giant recurrent aneurysms and specifically for Types I and II, (I, pure recanalization inside the aneurysm sac; II, pure coil compaction without aneurysm growth), surgical clipping is not suggested, due to high rates of morbidity [56].

Neck width to compaction height ratio (N:H) <0.5, coil width to compaction height ratio (C:H) <2.5 and compaction height (H) >2 mm, seem to be positive predictive factors of aneurysmal suitability to clipping, with C:H being the index with the highest predictive value [52,58].

There are several reported microsurgical techniques for the treatment of previously coiled aneurysms. The main is direct clipping, with the clip being placed underneath the coils [36,52]. When this technique is not feasible, other techniques may be applied. Aneurysm trapping or proximal occlusion of the parent artery combined with bypass may be an alternative option in order to exclude dysplastic recurrent aneurysms, not amenable to direct clipping, from brain circulation [36,52,58]. Parent Artery Occlusion alone could be also used in case of adequate retrograde flow distal to the aneurysm. In order to convert an unclippable aneurysm, due to inadequate neck remnant, into a clippable one, temporary clips may be placed to trap the aneurysm, followed by opening of the aneurysmal sac, removal of the formed thrombus and mobilization of the coils away from the aneurysmal neck [52,58]. Rubino et al. suggest that the coil removal should be avoided due to the unwieldiness and the possibility of surrounding structures’ injury, as even in cases of absolute indication, coil extraction is associated with an increased chance of poor outcome [45]. Wrapping with muslin or cotton aims to reinforce the aneurysm wall through inflammation and scar tissue formation. This technique may also be combined with clipping [36,52].

Retrospectively reviewing 9 cases that have been treated in our department, the surgical indication was aneurysm recurrence in 55.6% (5/9 pts) and residual aneurysm after initial embolization in 44.4% (4/9 pts). These percentages are comparable to those extracted from the literature, with recurrence reported to be the leading indication for surgical treatment, followed by residual aneurysm. Complete aneurysm occlusion was achieved for all our patients after additional surgical clipping (100%).

No mortality or morbidity related to the surgical procedure was observed in our series. Neurological assessment at the final follow-up indicated that all patients had minimal symptoms or were asymptomatic (mRS 0-1). All patients either remained stable or showed improvement as compared to their neurological status prior to retreatment (Fig. 1).

The findings derived from our study correspond to the literature’s data, indicating that microsurgery is a safe and efficient retreatment option for recurrent IAs.

Some factors may have introduced bias to our institutional study’s results. First and foremost, it is composed of a small relatively young patient population, while 88.9% (8/9) of our patients were in good clinical status preoperatively. Only one patient was preoperatively in a poor clinical condition with an mRS score of 5 but fortunately ended up with major improvement presenting a postoperative mRS score of 1. Another factor that may have favored the surgical procedure is that all aneurysms were located in the anterior circulation. Moreover, none of our cases exhibited giant-sized aneurysms (mean diameter 6.1 mm; range 4-17.3 mm). Last but not least, none of the patients presented with rebleeding, thus decisions regarding the need for retreatment were based on the systematic clinical and angiographic follow-up. However, technical operative difficulties were not absent from this study. In 55.6% (5/9) of the patients, a branch was originating from aneurysm’s neck whereas a wide neck aneurysm (≥4 mm in diameter) was present in 77.8% (7/9) of cases.

Considering our literature review, although a large number of studies are included, several of them are constituted of a small number of patients and the majority are retrospective studies without control groups. Therefore, they are vulnerable to selection bias and statistically significant conclusions about the ideal treatment modality cannot be exported. The same limitations also apply to our institutional cases. Moreover, aneurysms that were admitted for microsurgical retreatment were accessible and selected as feasible for direct clipping, making the use of the aforementioned technique an attractive option. Thus, direct comparison of its efficacy and effectiveness with the other techniques is not possible. Last, due to the lack of standardized protocols and criteria for the ideal retreatment technique, the treatment decision is usually made by the treating neurosurgeon and neuroradiologist on a case-by-case basis. Hence, it may vary between different centers.