Microsurgery of Giant Intracranial Aneurysm: A Single Institution Outcome Study
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.325355
Source of Support: None, Conflict of Interest: None
Keywords: Endovascular therapy, giant intracranial aneurysm, modified Rankin scale, microsurgeryKey Message: GIAs are difficult lesions to manage due to their varied morphology, significant morbidity and mortality on intervention, and poor natural history. A judicious approach toward the management of these lesions considering microsurgery or endovascular therapy or if need be combined treatment should be considered.
Giant intracranial aneurysms (GIAs) have, by definition, a minimum diameter of 25 mm. GIAs are treacherous lesions with grave prognosis and their management results in a combined surgical morbidity and mortality in the range of 20%–30%. Yet the GIAs need treatment as these often have a downhill course without treatment, with the mortality rate at 2 and 5 years after diagnosis being 68 and 85%, respectively.
Over the last few decades, there has been refinement in both microsurgical techniques and endovascular treatment (EVT). While small saccular aneurysms are optimally excluded from circulation by EVT, there is a high failure rate after EVT of GIAs even with flow diverters. Advances in microsurgery, including improvements in instrumentation and hardware, intraoperative indocyanine green (ICG) angiography, skull base surgical techniques, revascularization procedures, advances in anesthetic techniques such as cerebral protection, adenosine-induced cardiac standstill, and rapid ventricular pacing, have made microsurgery of GIAs safer.,,, In the current study, the surgical outcome of GIAs between 1996 and 2019 by a single surgeon at a single institute is presented to justify continuing microsurgery.
During a 24-year period, 134 patients with 147 GIAs underwent microsurgical intervention [Table 1]. The GIAs were classified based on the aneurysm that bled or was symptomatic. The medical and diagnostic imaging records of the patients were retrospectively reviewed. The patient outcome was determined by modified Rankin scale (mRS); ≤3 was considered as a good outcome and 4 and beyond was a poor outcome. Outcome data were collected from our records and were compared to the Glasgow outcome scale.
Aneurysm occlusion was classified as complete (no residual aneurysm), minimal residual aneurysm (small neck remnant), or incomplete (>5% of the original aneurysm lumen remaining). Aneurysm treatment failure was defined as post-treatment regrowth of residual aneurysm as documented by angiography or post-treatment aneurysm rupture.
Statistical analysis was done using the SPSS program. Preoperative, intraoperative, and postoperative univariate and multivariate predictors of surgical outcome were assessed using an exact logistic regression. Odds ratios and their 95% confidence intervals were computed, and a probability value of ≤0.05 was considered significant.
Demographics and aneurysm characteristics
The median age of the 134 patients was 47.15 years (range: 2–76 years). The total number of males and females was 57 and 77, respectively. Patients presented with a constellation of symptoms which included SAH, mass effect, or a combination of symptoms. ICA was the most common location to harbor GIAs, the carotico-ophthalmic segment (35 cases) being the most common ICA location [Table 1].
Decision-making depends on factors such as aneurysm location, morphology, size, presenting clinical and neurological condition, comorbidities, predicted treatment risk, preference of patient and family members, and the preference of the multidisciplinary team. Direct clipping was our preferred surgical technique for aneurysm occlusion. Novel clipping techniques were employed to satisfactorily occlude most aneurysms [Figure 1]. Pterional approach was the workhorse for anterior circulation aneurysms and an orbitozygomatic (OZ) extension was added when required to reduce retraction. The preferred approaches for posterior circulation aneurysms were OZ, transpetrosal, and farlateral for upper basilar, basilar trunk, and vertebral-posterior inferior cerebellar artery (PICA) aneurysms, respectively. Cervical carotid artery exposure for proximal control and clinoidectomy was performed for ophthalmic segment aneurysms. Surgical decompression of aneurysm was performed in 42 cases. Thrombectomy and decompression were done in 39 cases to facilitate clipping. Extracranial-intracranial (ECIC) bypass was employed in 40 cases. High-flow bypass (saphenous vein/radial artery) was used in patients who failed BTO immediately, and low-flow bypass (STA-MCA) was used in patients who failed BTO after hypotensive challenge (lowering mean arterial pressure with nitroprusside drip by 20 mmHg or 20% of the mean arterial pressure, whichever was greater). Bypass helped in treating GIAs that otherwise were deemed impossible to clip. Postoperative angiography was done routinely in post-clipping and post-flow-diversion procedures. The microsurgical strategy employed depended on the individual aneurysm [Table 2].,, There was a total of 29 high-flow (external carotid artery to middle cerebral artery bypass (M2) and 11 low-flow (superficial temporal to middle cerebral artery) bypasses. The radial artery graft was the commonest high-flow bypass conduit while the saphenous vein graft was used in only one case.
Postoperative angiography data were available for 81 out of 134 patients; 77 showed complete non-opacification of the aneurysm. Four cases showed residual remnant. One underwent a redosurgery, 1 was referred for EVT, and 2 were observed; 1 remained stable and the other was referred to EVT after 10 years for growth of the residual aneurysm on follow-up.
Overall 103 out of 134 (76.8%) patients had a good outcome postoperatively defined as mRS score of ≤3 [Table 3]. The remaining 31 patients (including the 6 who expired) had a poor outcome. Preoperative neurological status significantly influenced the eventual outcome. Out of 117 patients who presented with a good preoperative score, 99 (84.61%) remained good postoperatively, while out of 17 patients who presented with a poor preoperative score, only 5 (29.4%) had a good outcome. Patients with preoperative mRS score of 0–1 had the best outcome, while those with poor preoperative mRS score had a far worse outcome. Good preop mRS score (i.e., ≤3) had an overall good prognosis in the postoperative period as shown by both univariate and multivariate analysis and the values were statistically significant. P was 0.000 with the odds ratio being 0.036 (95% CI: 0.008–0.171). Presence of SAH showed statistical significance with good clinical outcome in univariate analysis; P was 0.04 with the odds ratio being 2.898 (95% CI: 1.051–7.991). Patients older than 50 years had a poorer outcome than patients of age 50 years or less though it did not reach statistical significance; P was 0.073, the odds ratio was 2.296 (95% CI: 0.925–5.699)
Aneurysms located in posterior circulation have been identified as an independent predictor for poor outcome and it has been found to be applicable in the treatment of very large and giant aneurysms. In our series, patients with posterior circulation aneurysms presenting with mRS scores of ≤3 and >4 were 19 and 5, respectively, and postoperative mRS scores of ≤3 and > 4 were 20 and 4, respectively. Thus, a good preoperative mRS score was directly proportional to a good postoperative mRS score. However, statistical analysis was not possible because of the small sample size of 24. GIAs of PCA were found to have a better outcome, the probable reason being the rich leptomeningeal collateral circulation of the PCA from the MCA and ACA territories.,
Mortality within 30 days of treatment was 6 out of 134 patients (4.47%). Four patients died of ischemic complications, 1 died of postoperative hemorrhage, and 1 died of sepsis. Nineteen patients deteriorated because of ischemic injury, 7 of which needed decompressive craniectomy. Three patients were reexplored, 2 for extradural hematoma and one for subdural collections. VP shunt for hydrocephalus was performed in 4 patients. The overall incidence of treatment-related morbidity was 19.4% and mortality was 4.47%.
Representative Case 1: A 25-year-old gentleman had presented in a coma from an acute subdural hematoma (SDH) in an outside facility and underwent decompressive craniotomy. He was referred to us for definitive treatment after recovery. He was found to have a 77-mm partially thrombosed aneurysm involving the whole of right ICA from the petrous to the supraclinoid segment (the ruptured aneurysm) and a large wide-necked unruptured basilar apex aneurysm. He underwent decompression of the ICA aneurysm mass through middle cranial fossa, had ECA-M2 bypass with RAG, and trapping of ICA (just after bifurcation in the neck and clipped intracranially proximal to the ophthalmic artery). The recovery was uneventful with good bypass patency and he was referred for EVT of the basilar apex aneurysm. However, the EVT failed and at the second stage; after 3 months, he underwent FTOZ, extradural anterior clinoidectomy, intradural posterior clinoidectomy, and clipping of the basilar apex aneurysm. The patient had no postoperative deficit, and DSA confirmed patent ECIC bypass and complete occlusion of both the aneurysms [Figure 2].
Representative Case 2: A 50-year-old lady presented with progressive visual loss of 4 months duration. On evaluation, she was found to have an unruptured 4.1 × 3.8 cm GIA of the right carotico-ophthalmic segment without any thrombus. After a frontotemporal craniotomy, a STMC bypass was carried out to avoid ischemic insult. The aneurysm was decompressed after temporarily trapping the ICA (ligature in cervical ICA and temporary clipping of ICA distal to the aneurysm) and the aneurysm was secured successfully with multiclipping technique. The patient had an uneventful recovery and recovered her vision completely [Figure 3].
Representative Case 3: A 35-year-old patient presented to us with progressive quadriparesis, gait ataxia of 6 months duration, and requiring support for daily activities. MRI/MRA Brain and 3D cerebral DSA revealed a heavily thrombosed giant basilar top aneurysm. He underwent FTOZ, extradural anterior clinoidectomy, intradural posterior clinoidectomy, and primary clipping of aneurysm at the neck with two large multicurved Yasargil aneurysm clips. The aneurysm neck was free of thrombus and hence could be clipped without decompression. ICG angiography demonstrated complete occlusion of the aneurysm and preservation of normal branches and perforators. Postoperative third-nerve paresis improved completely over a couple of months and the patient had no deficit at 3 months follow-up [Figure 4].
Representative Case 4: A 7-year-old boy presented to us in altered sensorium (GCS: E2M5V2). On evaluation, he had a Fisher grade 4 SAH with bifrontal hypodensities on CT. CTA and 3D DSA demonstrated a giant multilobulated wide-necked basilar top aneurysm with all four terminal branches arising from the aneurysm. He underwent obliteration of the aneurysm under hypothermic circulatory arrest. A FTOZ craniotomy and extradural anterior clinoidectomy were performed. He had a stormy postoperative course and at 1-year follow-up, he had improved to GCS E4M6V2 [Figure 5].
The goal of the treatment of GIAs is to exclude the lesion from circulation while preserving function in the neural tissues. Treatment of GIA is a challenge given the giant size, partial thrombosis, calcification, and the typical presence of a wide neck. GIAs and large (20–24 mm) aneurysms have a dismal natural history, with neurological deterioration or death seen in 65% to 100% of cases over two years and hence need treatment.,,,, In the International Study of Unruptured Intracranial Aneurysms, the rate of SAH for very large or giant aneurysms was found to be as high as 10% per year. The decision-making depends on multiple factors like age, clinical presentation, neurological status, aneurysm size, location, and morphology.,,
GIAs present many surgical challenges. Size of sac >15 mm in diameter blocks access to afferent and efferent vessels preventing dissection and gaining control of the parent artery. Puncture followed by suction or suction decompression through cervical ICA softens the aneurysm facilitating access. The authors have also utilized adenosine-induced transient cardiac arrest to help dissection. Transient trapping of the parent artery and decompression of thrombosed aneurysm by ultrasonic surgical aspirator helps, though not always. Protective low-flow bypasses were performed to prevent ischemic injury when prolonged trapping was anticipated. Previously coiled aneurysms prevent mobilization and remodeling of the sack by their resistance. Partial extraction of coils before clipping as well as novel multiclipping techniques help in complete occlusion. Excision of the residual neck close to the coils to completely isolate them and reconstruct the neck has also been advocated.
The postoperative outcome in the current series was directly proportional to the preoperative mRS score at presentation. This correlation was also noted in other studies., Mass effect has been identified by some authors as a predictor of poor outcome, but it is probably due to the poor preoperative neurological status of many patients with this presentation., Of the 56 patients in our series presenting with mass effect, 8 had fair to poor outcome. Postoperative clinical deterioration was seen in 5 while the remaining 3 had poor mRS score at presentation. Good outcomes following surgery are seen in 61%–87% of patients.,, Patients who become symptomatic with SAH and mass effect have the worst surgical outcomes. Younger patients with GIAs seem to have a better prognosis. This was also true in our series. In patients older than 70 years, the risk of surgery exceeded the lifetime risk of rupture. A presentation of SAH in our series showed a favorable outcome in 71.8% of cases, and a preoperative mRS score of ≤3 was shown to have a good outcome in 76.8% of cases. Those patients presenting with TIAs, poor baseline condition, aneurysm ≥25 mm and age ≥50 years have been shown to have a poorer outcome; 75% of poor outcomes were due to postoperative strokes.,
The morbidity and mortality rates of 20% and 4.5%, respectively, in our series compare well with the morbidity and mortality reported in the literature of 11% to 35% and 6% to 22%, respectively [Table 4].,,,
Size: Morley and Barr in the first cooperative aneurysm study designated aneurysms 25 mm or larger as “giant” and associated them with poor outcomes. Since then, there have been multiple reports in the literature that confirm the same.,,,,,
The literature suggests posterior circulation location as a well-established risk factor in the surgical treatment of aneurysms and relatively better outcome in the anterior circulation GIAs, specifically in the cavernous and paraclinoid locations.,, There was also a significantly higher good outcome in aneurysms involving cavernous and carotico-ophthalmic segments of ICA (84% and 82.5%, respectively) in our series. Posterior circulation GIAs are the most frequent indication for hypothermic circulatory arrest., Among our 24 posterior circulation GIAs, 17 with good preoperative scores continued to have a good outcome. Of the 7 patients with poor preoperative scores, good and poor outcomes were observed for 3 and 4 patients, respectively.
Cantore et al. in their study concluded that surgical clipping remains the gold standard for GIAs, and if this fails or is contraindicated, the high-flow ECIC bypass is a viable surgical option. Ensuring adequate flow in the vessels at risk and perforators is the key to success. Sughrue et al. proposed indirect aneurysm occlusion (proximal occlusion, distal occlusion, or trapping) with or without a bypass as a more acceptable alternative than hypothermic cardiac arrest. Campos et al. in their recent article recommended a multimodal approach. Results of our experience with microsurgery of GIAs justify its continuing use in GIAs.
The treatment of GIAs poses a formidable challenge. Yet, these need treatment because of their dismal natural history. Any decision to intervene should carefully consider the various options like microsurgery, EVT, and observation. The treatment should be individualized taking all the patient factors, pathological factors, and expertise available. Any intervention should be expected to better the natural history. EVT is an attractive option. Unfortunately, the high incidence of incomplete treatment, delayed complication, recurrence, and inadequate long-term follow-up data still make microsurgery the treatment of choice in most GIAs. However, the treatment of GIA is a major endeavor and should not be taken up by the occasional aneurysm surgeon. Committed groups in specialized centers with all available resources are best suited to tackle these life-threatening GIAs.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]
[Table 1], [Table 2], [Table 3], [Table 4]