Giant cerebral aneurysm

Giant aneurysms account for approximately 2–5% of all intracranial aneurysms and are more frequent among aneurysms within the pediatric population. They most commonly arise from the internal carotid artery (ICA), particularly the cavernous and paraclinoid segments, followed by the vertebrobasilar system, MCA and anterior cerebral artery.

The clinical presentation is variable ^{Reference Lownie, Drake, Peerless, Ferguson and Pelz2,Reference Hahn, Nicolle, Lownie and Drake3} : the majority (60–80%) present with mass effect or distal embolic events, while a smaller proportion (20–30%) present with subarachnoid hemorrhage (SAH), intracerebral hemorrhage or both. Less commonly, patients experience perforator infarction, seizures or remain asymptomatic.

The natural history is poor due to their high rupture risk, significant mass effect, intraluminal thrombus and distal embolization leading to cerebral ischemia. This grave prognosis underscores the imperative for curative treatment, regardless of rupture status. The International Study of Unruptured Intracranial Aneurysms ^{Reference Wiebers, Whisnant and Huston4} reported a five-year rupture incidence of 40% for anterior circulation and 50% for posterior circulation giant aneurysms (annual rupture risk 8–10%). Early data from Peerless and Drake ^{Reference Peerless and Drake5,Reference Peerless, Hernesniemi, Gutman and Drake6} at the University of Western Ontario (UWO) further underscore the poor prognosis: among patients without hemorrhagic presentation, mortality exceeded 60% within two years, and all survivors exhibited significant neurological deficits. Outcomes were worse for those presenting with SAH.

Posterior circulation aneurysm

Posterior circulation aneurysms comprise approximately 8–15% of all intracranial aneurysms with a poorer natural history than anterior counterparts due to an even higher rupture risk. ^{Reference Wiebers, Whisnant and Huston4} This elevated risk – estimated at 3–5% annually – often warrants definitive surgical treatment, particularly in younger patients. In a case series by Luzzi et al., ^{Reference Luzzi, Maestro and Galzio7} the majority (137/157) of posterior circulation aneurysms were located proximally. Of the total cohort, 14% were classified as giant and 12% as large. Proximal aneurysms most frequently involved the vertebral artery, basilar artery and proximal segments of the posterior cerebral artery (PCA). Distal aneurysms were typically located in the distal PCA, superior cerebellar artery (SCA) and anterior inferior cerebellar artery.

“Complex aneurysm” typically refers to giant aneurysms but may also include smaller aneurysms located in technically challenging regions, such as those in the posterior circulation. ^{Reference Hanel and Spetzler8} Additional features that confer complexity include unfavorable location, prior treatment, compromised collateral circulation, intraluminal thrombus, neck calcification, complex neck morphology and aneurysmal wall calcification. Collectively, aneurysm size, location and anatomical or technical considerations contribute to the complexity and treatment risks.

Aneurysm sac relaxation

An aneurysm presents as a pulsating, pressure-tense yet fragile mass. Reducing inflow (or intramural pressure) can soften the aneurysm sac, decrease the risk of rupture and facilitate permanent clipping and/or reconstruction. Historically, extreme induced hypotension – targeting an initial mean arterial pressure of 40 mmHg, and in critical cases as low as 20 mmHg – was employed, guided by an aneroid manometer (See Supplementary Information 2 for the interview transcript with Emeritus Professor Dr Gord Sellery of UWO, recounting his experience with the use of aneroid manometer-guided induced hypotension in the late 1960s and 1970s). ^{Reference Sellery, Aitken and Drake15} In contemporary practice, for small or straightforward anterior circulation aneurysms, temporary afferent artery clipping and supplemented with controlled hypotension is often sufficient. However, ICA, posterior circulation and giant aneurysms pose greater challenges. ICA control may require exposure of cervical, petrous (via Glasscock’s triangle) or clinoid segments (via anterior clinoidectomy) of the ICA, while posterior circulation aneurysms may necessitate endovascular balloon occlusion due to inaccessibility. In complex cases, limited exposure may hinder safe placement of temporary clips, especially if multiple clips are needed for trapping or if deep-seated aneurysms preclude direct access. Temporary clipping also carries risks of thromboembolism, vessel injury and intraoperative rupture.

Alternative flow arrest techniques address these limitations. In the literature, these techniques are often referred to as “cardiac standstill” or “circulatory arrest,” terms that imply global circulatory arrest; however, a more precise description is “aneurysm blood flow arrest.” Approaches can be categorized into local (e.g., temporary clip), cerebral (e.g., retrograde suction decompression (RSD)) and systemic flow arrest (e.g., adenosine-induced asystole, rapid ventricular pacing (RVP) or deep hypothermic total circulatory arrest (DH-TCA)). Each carries distinct ischemic risks. Selection should depend on aneurysm location and morphology, anticipated dissection difficulty and institutional experience (Table 2).

Table 2.

Summary of flow arrest strategy

ICA = internal carotid artery; AV = atrioventricular.

Retrograde suction decompression

Proximal retrograde suction decompression, first described by Batjer and Samson in the 1990s, ^{Reference Batjer and Samson16} provides rapid indirect aneurysm decompression and minimizes optic nerve manipulation. ^{Reference Mattingly, Kole, Nicolle, Boulton, Pelz and Lownie17} It is especially valuable for paraclinoid ICA aneurysms, which often exhibit brisk backflow from the ophthalmic artery and cavernous ICA – even when trapped. Historically, such aneurysms were managed with cervical carotid ligation.

The technique of RSD is illustrated in Figure 2. ^{Reference Batjer and Samson16,Reference Eliava, Filatov and Yakovlev18–Reference Srinivasan, Zhang, Scherschinski, Whiting, Labib and Lawton20} A 16-gauge angiocatheter is inserted into the common carotid artery (CCA) and connected via sterile extension tubing and a three-way stopcock to a 60-mL syringe and heparinized circuit. Temporary clamps are applied to the CCA and external carotid artery, and a temporary clip is placed on the intracranial ICA distal to the aneurysm to achieve trapping. Blood is aspirated via the syringe to collapse the aneurysm sac, with aspirated blood collected and reinfused intravenously. If distal clipping is not feasible, rapid aspiration alone may collapse the sac, albeit with greater transient blood loss. After definitive clipping, all temporary clips and angiocatheter are removed.

Figure 2.

Illustrative figures of open and endovascular retrograde suction decompression. The above figure illustrates (left) open surgical retrograde suction decompression technique, where internal carotid artery is exposed surgically and a balloon catheter is inserted directly for suction decompression and (right) endoscopic technique using double balloon catheter to occlude the external carotid artery and common carotid artery, respectively. The aspirated blood, in both surgical and endoscopic techniques, is transfused back to the patient directly or collected via cell-saver.

A modified endovascular technique, first reported by Scott in 1991 ^{Reference Scott, Horner and Leipzig21} and subsequently by other authors, ^{Reference Parkinson, Bendok and Getch22–Reference Pojskić, Arnautović, Essayed and Al-Mefty25} uses a transfemoral 5-Fr double-lumen balloon catheter advanced into the ICA about 2 cm above the bifurcation. Balloon inflation achieves proximal control, while a 20-mL syringe provides aspiration to soften the aneurysm. Additional advantages include rapid access to the paraclinoid ICA, immediate balloon occlusion if intraoperative rupture occurs and control angiography after clipping is completed.

A systematic review ^{Reference Flores, White, Batjer and Samson26} of 26 studies (1990–2016) involving 525 patients compared surgical and endovascular RSD techniques. The average duration of temporary flow arrest was 13 minutes and the aneurysm obliteration rates reaching 95% in both approaches. Procedure-specific complication rates were low (1.3%), with good or fair neurological outcomes reported in over 91% of patients. Surgical RSD was associated with slightly lower morbidity. In a case series of 18 patients, one ICA dissection required emergency endarterectomy. ^{Reference Batjer and Samson16} Additional complications include thromboembolism and hemispheric ischemia. Overall, both surgical and endovascular RSD have demonstrated high aneurysm obliteration rates, favorable long-term neurological outcomes and low procedure-related morbidity and mortality. However, RSD does not prevent distal brain ischemia. Intermittent reperfusion via distal clip release and continuous neuromonitoring are essential. RSD is also contraindicated in heavily calcified or thrombosed aneurysms and in significant ICA stenosis, which may impede effective decompression.

In our experience, RSD reliably decompresses the aneurysm and facilitates safe clipping. We prefer endovascular RSD in the recent years and use a cell saver to centrifuge and filter the aspirated blood before reinfusion to reduce thromboembolic complications. However, reinfusion may be delayed by the minimum volume (∼1 L) needed for processing, and the heparinized salvage blood may increase coagulopathy risk. Hypotension due to acute blood loss may occur, while awaiting the centrifugation of the salvaged blood, which can be managed with vasopressors or blood transfusion as needed.

Total circulatory arrest

TCA, often achieved via induced asystole, provides a bloodless field and complete aneurysm decompression. Shorter durations of cardiac arrest – induced by adenosine or RVP – are better tolerated and preferred for less complex cases. To mitigate prolonged global ischemia, TCA together with profound hypothermia (i.e., DH-TCA) necessitates the use of extracorporeal circulation; however, its use has significantly declined due to high morbidity and the emergence of safer alternatives.

Extracorporeal circulation

The first recorded case of total cerebral flow arrest was performed by Dott et al. in 1955, involving bilateral carotid and vertebral artery occlusion for 47 minutes without any neuroprotection strategies. ^{Reference Wright, Huang and Sharma27,Reference Todd, Howie and Miller28} The patient succumbed four days later due to extensive vertebral artery thrombosis. ^{Reference Wright, Huang and Sharma27,Reference Lougheed, Sweet, White and Brewster29} The first substantive series was reported by the Toronto General Hospital group, who employed moderate hypothermia (28–30°C) using surface cooling with temporary carotid and/or vertebral artery occlusion in 73 patients. ^{Reference Botterell, Lougheed, Morley and Vandewater30} Favorable outcomes were reported in 50 patients, supporting the role of hypothermia in extending tolerable durations of circulatory arrest. ^{Reference Botterell, Lougheed, Morley and Vandewater30}

By the early 1960s, cardiac surgeons began using profound hypothermic TCA with extracorporeal circulation for valve surgery, which laid the groundwork for its adoption in cerebrovascular procedures. Drake and colleagues ^{Reference Drake, Barr, Coles and Gergely31} from UWO were among the first to report 10 cases treated with deep hypothermia (15–18°C) and arrest durations of 2–18 minutes. Deep hypothermia can afford cerebral protection for 30–45 minutes, allowing ample time for complex dissection and aneurysm clipping. TCA provides a motionless, bloodless field and a completely decompressed aneurysm sac. However, the technique carries substantial risk, including coagulopathy, vasospasm and ischemic brain injury. Subsequent refinements came through multiple case reports and small series. The largest series, reported by Lawton et al. (University of California, San Francisco) and Spetzler et al. (Barrow Neurological Institute), described 60 patients undergoing 62 DH-TCA procedures, predominantly for posterior circulation aneurysms. ^{Reference Lawton, Raudzens, Zabramski and Spetzler32} Morbidity and mortality rates were 13.3% and 8.3%, respectively. All these experienced authors consistently concluded that DH-TCA should be reserved for complex aneurysms not amenable to conventional or endovascular treatments or for recurrent aneurysms following endovascular coiling.

Extracorporeal circulation techniques can be performed via open-and closed-chest approach. The open-chest approach mirrors standard cardiac surgery, utilizing aortic and right atrial venous cannulation. Cardiac arrest is achieved through profound hypothermia-induced standstill (<18°C), extreme low-flow bypass and/or cardioplegia administered via aortic cross-clamping. In contrast, the closed-chest approach avoids sternotomy by employing femoral arterial and venous cannulation. Cardiac arrest can be induced by RVP (150–180 beats per minute). Some centers utilize an endo-aortic catheter with an inflatable balloon to occlude the proximal ascending aorta, followed by pacing at a reduced rate to maintain coronary perfusion pressure (20–30 mmHg) or retrograde administration of cardioplegia. ^{Reference Campkin and Dallas33} Some authors described the use of high-concentration potassium infusion via the pulmonary artery catheter venous port or electric shock to induce cardiac arrest. ^{Reference McMurtry, Housepian, Bowman and Matteo34} In closed-chest approach, external defibrillator pads should be placed for potential synchronized cardioversion or defibrillation.

Regardless of technique, precise control of circulatory arrest is critical, as even occasional escape beats can lead to dramatic and abrupt aneurysm motion. Implementation of DH-TCA is logistically demanding and requires coordinated multidisciplinary planning. A recommended checklist and perfusion circuit set-up are provided in Supplementary Information 3 and Figure 3, respectively.

Figure 3.

Illustrative figures of bypass circuit for deep-hypothermic total circulatory arrest and selective brain cooling. This schematic shows different bypass configurations used in complex aneurysm surgery. (A) Full cardiopulmonary bypass circuit performed via median sternotomy with aortic cannulation and bicaval venous access. Venous blood is drained into a reservoir, oxygenated by an oxygenator, and returned to the body via a centrifugal pump. System deep hypothermia is achieved with a heat exchanger, and cardioplegia is delivered through a dedicated cardioplegia pump. Standard left ventricular venting and aortic cross-clamping are applied. (B) Selective brain cooling. Blood is withdrawn from the femoral artery through an outflow cannula, cooled by a heat exchanger, and infused – mixed with cardioplegia – directly into the internal carotid artery (ICA). (C) Femoral-femoral bypass. Venous blood is drained from the femoral vein to the oxygenator and returned via the femoral artery. (D) Endovascular circuit for minimally invasive cardiac surgery that can also be used to avoid sternotomy in complex aneurysm surgery. Venous return to the extracorporeal circuit is achieved through a femoral venous cannula advanced to the junction of the inferior vena cava and right atrium. The endoaortic catheter, an inflatable balloon device, functions as an aortic cross-clamp and allows antegrade cardioplegia infusion into the aortic root and coronary arteries; its lumen can also be used for aortic root venting. Alternatively, endocoronary sinus catheter is inserted via the internal jugular or another neck vein and advanced in the coronary sinus for retrograde cardioplegia delivery. An endopulmonary vent catheter is placed in the pulmonary artery for decompression. CP = cardioplegia pump; R = reservoir; P = centrifugal pump; HE = heat exchanger; O = oxygenator.

A recent review ^{Reference Nager, Pontes and Udoma-Udofa35} summarized pooled outcomes in 233 patients undergoing DH-TCA: 30-day mortality was 12%, postoperative neurological deficit rate was 35%, and 69% of patients achieved good neurological recovery. Although DH-TCA is now reserved for the most extreme cases, these results support its efficacy when employed judiciously; nevertheless, its application has become rare and is no longer employed in many institutions.

Hypothermia

Hypothermia remains one of the most reliable neuroprotective strategies during prolonged cerebral ischemia in aneurysm surgery. By reducing cerebral metabolic rate and oxygen consumption, it extends neuronal survival during ischemic periods. This protective effect is dose-dependent and well-documented at profound levels (18–26°C). ^{Reference Chawla and Tempe47} Routine use of mild hypothermia (32–35°C), as investigated in the IHAST trial, ^{Reference Todd, Hindman, Clarke and Torner48} has shown no neuroprotective benefit in predominantly good-grade SAH patients; however, these findings may not necessarily extrapolate to situations like prolonged aneurysm flow arrest.

Selective brain cooling

Selective brain cooling is an emerging investigational approach targeted at cooling part of the brain while maintaining normothermia in the rest of the body. Verdura et al. first demonstrated this concept in dogs using bilateral carotid perfusion to lower brain temperature to 15°C, while systemic temperature remained above 29°C. ^{Reference Verdura, White and Albin49} In humans, Williams et al. ^{Reference Williams and Turner50} applied unilateral carotid perfusion with occlusion of the contralateral carotid and both vertebral arteries to achieve total cerebral circulatory arrest, reaching brain temperatures as low as 11.6°C. However, the technique was abandoned due to high mortality. Demian et al. ^{Reference Demian, White, Yashon and Kretchmer51} later introduced bi-hemispheric cooling via a heat exchanger without a perfusion pump during frontal tumor resection. However, this technique renders cerebral perfusion fully dependent on the patient’s cardiac output.

Our institution has been employing selective (ipsilateral) brain cooling using a simplified extracorporeal perfusion technique in highly selected cases. ^{Reference Lownie, Menkis, Craen, Mezon, MacDonald and Steinman52} The key procedural steps utilized at our center are outlined in Supplementary Information 4. After preparing the patient for full cardiopulmonary bypass, temperatures are monitored at two sites (e.g., nasopharyngeal and rectal, or via pulmonary artery catheter). A crystalloid-primed cardioplegia circuit is connected to a cannulated CCA, with a distal catheter in the ICA for pressure monitoring. The proximal CCA is clamped to direct perfusate into both internal and external carotid territories. This setup – leveraging external carotid branches to cool the scalp, muscle and dura – promotes early cerebral hypothermia and minimizes rewarming via convection. As the arterial inflow originates from an already oxygenated femoral outflow, no oxygenator is needed, simplifying the circuit.

Extracorporeal perfusion is titrated to 150–180 mL/min to maintain mean ICA pressure ≤100 mmHg or lower. Cooling is achieved by circulating ∼20°C blood, reducing brain temperature to 28°C within 10–12 minutes. Flow adjustments may be required to maintain hemodynamic stability and minimize bleeding. The target temperature (22–25°C) is then maintained. Once the aneurysm is secured, active rewarming is unnecessary. Brain temperature typically rises from ∼22°C to 26°C within 5 minutes and normalizes over 30 minutes. Systemic temperature remains near normothermic with the use of convective warming blankets to counteract cerebral venous return cooling. Hemodynamic stability is generally preserved, and in uncomplicated cases, most patients can be extubated at the end of surgery.

Choice of neuromonitor

Intraoperative neuromonitoring should be tailored to the brain region at risk of ischemia. While EEG and near-infrared spectroscopy offer real-time cortical monitoring, their effectiveness may be limited when sensor placement overlaps with the surgical field. A multimodal approach is often necessary. Evoked potentials (EP) – somatosensory (SSEPs), motor (MEPs) and brainstem auditory (BAEPs) – assess the functional integrity of sensory, motor and auditory pathways via far-field responses, making them especially valuable for detecting perfusion deficits beyond the exposed cortex. However, interpretation is complex due to the distinct anatomical pathways involved, requiring experience to distinguish ischemic patterns. Effective use demands a strong understanding of cerebral vascular anatomy, modality-specific limitations and the functional implications of injury to eloquent brain regions. A detailed discussion of neuromonitoring strategies specific to each aneurysm type or location is beyond the scope of this review.

Respiratory monitoring

Respiratory monitoring is an additional, anecdotal yet practically useful adjunct for detecting brainstem ischemia; however, published reports to date have been primarily confined to our institution. Sellery et al. ^{Reference Sellery, Aitken and Drake15} from UWO observed that changes in respiratory pattern were often the earliest sign of impending brainstem ischemia, preceding alterations in blood pressure or cardiac rhythm. Respiration has therefore been used as an intraoperative monitor to ensure adequate cerebral perfusion, particularly during concurrent induced hypotension. A subsequent report by Manninen et al. ^{Reference Manninen, Cuillerier, Nantau and Gelb60} from UWO reported the monitoring of respiration in 50 patients during temporary occlusion of the vertebrobasilar artery system. In this cohort, 13 patients exhibited reversible respiratory changes: six developed apnea, two of these also showed simultaneous hypertension (defined as >30% above baseline) and only one patient had changes in all three parameters – respiratory, cardiovascular and EP. Of the 13 patients monitored with EPs, only two demonstrated concurrent changes in SSEPs and BAEPs, highlighting the complementary nature of respiratory monitoring alongside neurophysiologic modalities.

The technique for respiration monitoring is usually straightforward but could be challenging in selected patients such as those with history of drug abuse, as it involves maintaining a narrow anesthetic window that permits spontaneous respiration while suppressing all catastrophic movements. Our typical regimen includes total intravenous anesthesia using propofol, dexmedetomidine and remifentanil infusion. Remifentanil is discontinued approximately 10 minutes before monitoring begins. In the series by Manninen, naloxone (median dose 0.2–0.3 mg) was used to reverse the respiratory depressant effects of fentanyl or sufentanil. The ventilator is transitioned from synchronized intermittent mandatory ventilation to pressure support ventilation (PSV) if respiratory triggering persists and finally to full spontaneous breathing. Respiratory function is monitored for signs such as irregular breathing patterns, decreased frequency, reduced tidal volume and/or onset of apnea. A more detailed respiratory changes by injury to specific brainstem centers are listed in Supplementary Information 5. Significant hypertension may also accompany respiratory arrest. If severe hypercapnia leads to brain swelling, PSV may be reintroduced, although monitoring can only focus on respiratory rate rather than tidal volume/pattern. Respiratory arrest is usually clinically obvious, even on PSV mode. If anesthesia depth is excessive, reducing the propofol dose may help restore respiratory effort, and low-dose ketamine (0.1–0.2 mg/kg/hr) can be added to maintain adequate anesthetic depth without respiration suppression. No diagnostic accuracy data are available for respiratory rate monitoring.

In addition to functional monitoring, anatomical assessment of cerebral blood flow is critical for assessing adequacy of blood flow. Compared with intraoperative Doppler ultrasound and indocyanine green angiography, a hybrid operating room equipped with conventional digital subtraction angiography is ideal for managing complex aneurysms. At our institution, we routinely perform baseline conventional angiography for all complex cases in our hybrid OR. Intraoperative angiography is repeated if the anatomy remains uncertain or if a conversion to a hybrid microsurgical and endovascular techniques is considered. A final angiographic assessment is performed after vascular reconstruction to confirm the adequacy of repair. We have found that combining anatomical and functional neuromonitoring assessments is particularly effective and advantageous in managing complex aneurysms. However, such a setup requires substantial institutional resources, including dedicated biplane angiography equipment, specialized operating room infrastructure and the coordinated availability of neuroradiology personnel and technical staff.