Introduction

For many decades, neurosurgical treatment (clipping) has been the sole treatment for intracranial aneurysms. From the early 1990s on, endovascular treatment (coiling) has progressively become an alternative for clipping, which was enforced by the publication of the initial results of the International Subarachnoid Aneurysm Trial (ISAT) in 2002 [10].

The ISAT randomized patients with a ruptured intracranial aneurysm, in which both clipping and coiling were deemed valid, were compared for clinical outcome after either treatment option. Its results after a mean 1-year follow-up were in favor of coiling, showing an absolute risk reduction in dependency or death of 6.9%. However, the recently published 5-year follow-up data demonstrate that the differences in outcome of the two treatment modalities have vanished over the years [11]. In a modified intent-to-treat analysis, disregarding the mortality occurring before any treatment was instituted, all statistical differences have even disappeared [1]. Nevertheless, nowadays a substantial quantity of intracranial aneurysms, both ruptured and unruptured, are treated by endovascular techniques, also in vascular territories that were clearly underrepresented in ISAT, such as the middle cerebral artery (MCA). It is remarkable how quickly, particularly in Europe, the attitude has changed in favor of coiling of intracranial aneurysms, despite the lack of evidence. This tendency towards coiling has led to a significant decrease of the neurosurgical caseload, which has serious implications for the vascular neurosurgeons in each neurosurgical center, and for the training of neurovascular fellows.

The University Medical Center Groningen (UMCG) is a large academic and community-based hospital with a high-volume neurovascular department, as a tertiary referral center delivering 24/7 emergency care for patients suffering an aneurysmatic subarachnoid hemorrhage (aSAH). The hospital covers a population of 1.7 million people in the northern part of The Netherlands, with a constant number of around 120 patients treated for an intracranial aneurysm each year. In the UMCG neurosurgical center the majority of MCA aneurysms are preferably treated by clipping, since MCA aneurysms are very well accessible for surgery, and because of the specific local angioarchitecture that often necessitates vascular remodeling of the MCA bifurcations, which is better coped with by surgical than by endovascular techniques.

In this paper a contemporary consecutive series of the preferred surgical strategy for MCA aneurysms is presented and balanced against the relevant literature.

Methods

A retrospective review of a prospectively kept neurovascular database was performed for patients with an MCA aneurysm in the period from January 2001 until December 2006. In this 6-year period 151 consecutive patients with an MCA aneurysm were treated (Fig. 1). For each patient, the choice of the preferred treatment strategy (observation, clipping, or coiling) was made by a multidisciplinary team of neurosurgeons, neurologists and interventional neuroradiologists after clinical assessment and initial computed tomographic angiography (CTA) or conventional cerebral angiography. If treatment options were considered equal, in MCA aneurysms priority was given to surgery.

Fig. 1
figure 1

Distribution of the 151 MCA aneurysm patients, according to their treatment

Full size image

The records of all consecutive surgically treated patients were searched to obtain patient characteristics, characteristics of the aneurysm, treatment details, complications and follow-up. The clinical condition of all patients was classified according to the World Federation of Neurosurgical Societies scale (WFNS) [7, 17]. Clinical outcome was graded according to the modified Rankin scale [2, 15].

Patients with an unruptured MCA aneurysm (WFNS 0) had a full assessment in the outpatient clinic and treatment was only instituted after thorough multidisciplinary counseling. Patients with a ruptured aneurysm of a good grade (WFNS 1–3) were, as a rule, treated within 72 h after the aSAH. In a minority of symptomatic patients, the treatment of the ruptured aneurysm was postponed because of co-morbidity or late referral. Patients with a ruptured aneurysm in a poor clinical condition (WFNS 4–5) were treated for acute hydrocephalus before making the final decision about the timing of the aneurysm treatment [18]. In case of an aSAH, treatment was at first selectively aimed at the ruptured aneurysm; additional unruptured aneurysms (coincidental aneurysms) were not treated in the same session, unless in the same surgical field.

All procedural complications and perioperative events were registered. To complete missing long-term follow-up, the family doctors of the patients were contacted.

Results

The patient characteristics are outlined in Table 1. The majority of the surgical patients with a ruptured MCA aneurysm were women (75%); the mean age was 52.3 years (range 33–79). The WFNS grade at admission was 1 or 2 in 67% of the cases. The aneurysm characteristics are summarized in Table 2. Most MCA aneurysms were larger than 5 mm (70%).

Table 1 Characteristics of surgical patients
Full size table
Table 2 Aneurysm characteristics
Full size table

The treatment strategies of all consecutive patients are delineated in Fig. 1. The majority of MCA aneurysms was treated by surgery (105 patients). In 13 patients, the MCA aneurysm was coiled. Fourteen aSAH patients with a poor WFNS grade died before an attempt to treat the ruptured aneurysm could be made. Nineteen patients with an asymptomatic aneurysm were left untreated: 17 patients had an aneurysm judged too small for treatment according to the recommendations of the ISUIA study [19]; two patients rejected the advised treatment.

Surgical clipping of an MCA aneurysm was performed following an aSAH in 77 patients, of which the majority (83%) was treated within 72 h following the hemorrhage. Thirteen patients had postponed clipping of the ruptured aneurysm. Two patients harbored a coincidental (contralateral) MCA aneurysm and had a second surgical procedure. Twenty-three surgically treated patients presented with a significant hematoma, requiring an immediate evacuation of the clot in 20 patients. In 95 of 107 procedures (89%) the surgeon reported a full occlusion of the aneurysm.

Procedural and perioperative events are depicted in Table 3. Since both clinical and non-clinical (TCD) vasospasm was recorded as a postoperative event, only 32 of 105 patients (30.5%) experienced a fully uneventful course. Preoperative subarachnoid rebleeding occurred 11 times (in two cases twice in the same patient). Postoperative rebleeding of a clipped MCA aneurysm occurred once, leading to a poor outcome. The other postoperative rebleeding was from a ruptured anterior communicating artery aneurysm in a patient with a coincidental MCA aneurysm.

Table 3 Perioperative complications
Full size table

Follow-up of the surgically treated symptomatic patients at 2 months was complete for all but one patient (99%). The mean follow-up was 4.7 years, obtained in 96% of the patients. The outcome was graded mRankin 0–2 in 80% of the cases (Fig. 2). During follow-up, four patients died from an unrelated cause, all more than 1 year after the aSAH and treatment.

Fig. 2
figure 2

Outcome of the surgically treated MCA aneurysms after 2 months and after 4.7 years’ follow-up. Good = mRankin 0–2, fair = mRankin 3–4, poor = mRankin 5–6

Full size image

Patients with a significant intracerebral hematoma (ICH) (n = 23) presented with a worse WFNS grade than the average study population (WFNS 1–2 in only 39% of cases), but nevertheless in 69% of cases their outcome was good (mRankin 0–2). If these patients had been excluded in the outcome analysis, outcome would have been calculated as mRankin 0–2 in 88% of the patients.

All 19 patients with a surgically treated asymptomatic (not coincidental) MCA aneurysm had a good outcome at the latest follow-up (mRankin 0–2). In the other patients with asymptomatic aneurysms, follow-up was discontinued in five cases because of patient-related factors, e.g., co-morbidity or age. Two patients refused to have a follow-up. Eleven patients harboring an asymptomatic aneurysm had a radiological follow-up after 1 year; no enlargement of their aneurysm was found on digital subtraction angiography (DSA) or magnetic resonance angiography (MRA) examination.

Discussion

Literature

Surgical series from the era before ISAT (before 2002) have demonstrated good surgical results for MCA aneurysms. Nevertheless, in many institutions the surgical experience has diminished during the last decade. This is confirmed by a search of the PubMed internet database from 2001 on, yielding very few surgical reports. In an extensive review of the Finnish experience with MCA aneurysms, Dashti et al. stated that these aneurysms are still best treated surgically [4–6]. Prat and Galeano [13] described 12 patients with a significant intracranial hematoma due to an MCA aneurysm that presented with WFNS grade 4 or 5, of which only five patients had a good or moderate outcome. The literature search also revealed a recent publication by Quadros et al. [14] on the coiling of MCA aneurysms in 55 patients. They concluded that coiling was effective in preventing a rerupture of the symptomatic aneurysm (mean follow-up 14 months); however, with a total occlusion of the ruptured MCA aneurysm in merely 42% of their cases. In addition, only 57% of patients with a ruptured MCA aneurysm in their study had a good outcome after 1 year and six patients (18%) died. These endovascular results are clearly unfavorable in comparison with the results of surgical studies, including the present study showing 89% intraoperative total occlusion and 80% good outcome.

UMCG surgical series

After a mean 4.7-year follow-up, the clinical outcome of 77 patients with a ruptured aneurysm was graded good (mRankin 0–2) in 80% of the cases. Excluding the patients with a significant ICH would even lead to a good outcome in 88% of the patients. Also, all patients treated for an unruptured MCA aneurysm were graded with a good outcome after clipping; on the whole unchanged in comparison with the preoperative condition. These results are far better than the related endovascular series for MCA aneurysms in the literature.

All perioperative events are outlined in Table 3. Most events were related to the aSAH and not to the surgical procedure. Postoperative rebleeding of an MCA aneurysm occurred in one patient with an incompletely clipped giant aneurysm that had presented with a large ICH and a WFNS grade 4 SAH. Postoperative epidural or subcutaneous hematoma occurred in four out of 105 surgically treated patients. Seizures were encountered in four surgically treated patients. Ventriculitis occurred in seven patients, without exception related to an external cerebrospinal fluid (CSF) shunt.

Scientific, ethical and political reflections

In ISAT, the MCA aneurysms were noticeably underrepresented. This is almost certainly due to the inclusion criteria that required that both clip and coil were considered equivalent in leading to a definite cure. As a consequence, MCA aneurysms presenting with a hematoma or having vascular branches originating from its neck were not randomized, whereas these features often occur. Also, a significant accompanying hematoma that needs surgical evacuation is frequently encountered after MCA aneurysm rupture. Nevertheless, particularly in Europe, the ISAT results are applied to the whole spectrum of cerebral aneurysms, including MCA aneurysms, despite the fact that good results of aneurysm clipping are still reported [3, 12]. In addition, recent publications on the long-term results of the ISAT have shown that after 5-year follow-up the clinical outcome after coiling is no longer superior to a surgical strategy, due to persistent higher rebleeding rates and a wider need for re-treatment after coiling [8, 9, 16].

Everyday clinical practice requires that choices between the different treatment modalities for patients with an aSAH are being made on a scientific basis. ISAT has proven its value in this decision-making process, particularly for aneurysms of the carotid artery and the anterior cerebral artery, but its recommendations cannot be used as a license to frankly extrapolate in favor of coiling of MCA aneurysms. Also, it has to be taken into account that to overcome the angioarchitectural disadvantages of the MCA anatomy there has been over the years an increased use of remodeling balloons and stents in addition to coiling. These added technical novelties add to the morbidity and mortality, and have never been tested in a randomized study to be better or safer than coiling alone or surgery. It is remarkable that the devices are frequently innovated and taken into clinical use without any proof of their superiority. In contrast, surgical treatment of MCA aneurysms is straightforward, with the possibility to remodel the aneurysm in order to obtain a perfect neck occlusion and, if necessary, to remove an associated hematoma. As such, surgical clipping should be recognized as the preferred treatment of MCA aneurysms.

Apart from the technical aspects, there is another serious concern. Particularly in Europe, the notable rise in endovascular procedures has led to a significant decline of the surgical caseload. This raises questions about the maintenance of surgical skills and the ability to train future neurovascular surgeons. A loss of surgical neurovascular expertise will certainly lead to a downward spiral in the quality of the overall treatment of intracranial aneurysms. It is extremely important to ensure that the high-quality standards of today are also guaranteed tomorrow. Caseloads should be high, in order to preserve neurovascular experience at the highest level. Obviously, regional and nationwide centralization of neurovascular care and training will be the only way to secure this.

Conclusion

This contemporary study confirms that good results are achieved with clipping of MCA aneurysms. All attempts to treat MCA aneurysms endovascularly, often with the use of novel (not clinically tested) endovascular devices, are unjustified in a situation where an excellent surgical solution is at hand. Also, the treatment of cerebral aneurysms should be restricted to specialized regional centers with a high caseload, which are able to provide full-time high-level care by a team of experienced vascular neurosurgeons and interventional neuroradiologists, to guarantee optimal care for this category of patients.