Extracranial–Intracranial Bypass for Cerebral Ischemia

Bypass for Atherosclerotic Steno-occlusive Disease

In 1977, a prospective randomized trial was initiated to evaluate whether EC-IC bypass, specifically in the form of STA-MCA anastomosis in conjunction with best medical therapy, was superior to best medical therapy alone in patients with cerebrovascular ischemia of the anterior circulation. The EC-IC bypass trial included 1377 patients with ischemic symptoms within 3 months of enrollment in the setting of either proximal middle cerebral artery (MCA) stenosis or occlusion, internal carotid artery (ICA) stenosis above the C2 vertebral body, or extracranial internal carotid artery occlusion (ICAO). The results, published in 1985, did not show a statistically significant difference in outcome between the two cohorts in the peri-operative period (30 days) or during long-term follow-up (mean 55.8 months). Follow-up revealed a recurrent stroke risk of 31% versus 29% in the EC-IC bypass group and the medically managed group, respectively. The results of this study led to a steep decline in the performance of EC-IC bypasses. However, subsequent analysis of the study methodology highlighted limitations with the study’s design and implementation that suggested universal abandonment of EC-IC bypass in treatment of ischemic cerebrovascular disease was premature. Such limitations included potential for selection bias in the enrollment process, questions about the adequacy of bypass flow provided by STA-MCA bypass, and, most importantly, lack of hemodynamic evaluation in patient selection for study enrollment. With regard to selection bias, a large number of patients underwent surgery outside of the trial, raising the possibility that subjects may have been selectively offered surgery versus enrollment in the trial based on the perceived benefit of intervention. This type of bias can skew study outcomes if patients more likely to benefit from bypass were systematically selected for surgery, rather than enrolled into the trial. Furthermore, despite the 96% patency rate of the bypasses reported in the study, the adequacy of STA-MCA bypasses as opposed to higher flow bypasses, via larger caliber conduits, was questioned. Ultimately, the most substantial drawback of the EC-IC bypass trial, which laid the groundwork for further studies, was the lack of hemodynamic evaluation in selecting patients for enrollment. Given that STA-MCA bypass is aimed at improving blood flow, it can be surmised that only individuals with hemodynamic impairment stand to realize any substantial benefit from surgery. Furthermore, since intrinsic routes of collateral blood supply vary substantially among patients, some individuals may compensate hemodynamically in the setting of cerebrovascular occlusive disease, whereas others experience hemodynamic compromise. It was, therefore, hypothesized that revascularization in a more targeted subgroup of patients with ICAO (i.e., those with poor hemodynamic reserve and resultant ischemia) may result in a decreased risk of subsequent stroke. Future studies set out to evaluate the efficacy of bypass in this subgroup.

Unilateral atherosclerotic ICAO is seen in 3% of the elderly population. Up to 10% of transient ischemic attacks (TIAs), as well as 15%–20% of ipsilateral carotid territory strokes, can be attributed to ICAO. Despite maximal medical management, the 2-year risk of ipsilateral ischemic stroke in patients following carotid occlusion is 5%–8% per year. The prognostic value of hemodynamic assessment in ICAO has been examined in a number of prospective studies, using a variety of modalities for measurement of cerebral hemodynamic reserve. Chronic cerebrovascular insufficiency states, such as ICAO, can be categorized into three stages of hemodynamic failure: in stage I, autoregulatory compensatory vasodilation is still able to maintain normal CBF, and oxygen extraction fraction (OEF); stage II hemodynamic failure, also termed misery perfusion, is characterized by exhaustion of autoregulatory compensation with resultant reduction in CBF, and with cerebral metabolic compensation by increased OEF; and in stage III hemodynamic failure, both CBF and OEF are reduced to the point of ischemia and ultimately infarction.

To capture the hemodynamic state of patients with chronic cerebrovascular insufficiency states, imaging modalities capable of distinguishing CBF, compensatory flow, and vascular anatomy are required. Such imaging includes positron emission tomography (PET), single-photon emission computed tomography (SPECT), xenon computed tomography (Xe CT), computed tomographic perfusion (CTP), magnetic resonance imaging (MRI) techniques, and cerebral angiography. ^{, ,} With these imaging techniques, the degree of cerebrovascular reserve impairment can be further characterized by the use of vasodilatory challenge during testing. The response to a vasodilatory challenge can be categorized into three main categories of cerebrovascular reactivity (CVR) that are felt to correlate with stages of hemodynamic failure. The first, correlating to stage I impairment, demonstrates reduced response to challenge when compared to the contralateral hemisphere (which is presumed to have a normal response), or when compared to previously determined normal values. The second shows absent flow augmentation response to vasodilatory challenge and correlates with stage II failure. The third shows a paradoxical reduction in flow also known as the steal phenomenon, which has been considered consistent with stage III hemodynamic failure.

In 2003, a blinded, prospective, longitudinal cohort study of 81 patients with symptomatic ICAO, the St Louis Carotid Occlusion Study (STLCOS), demonstrated that stage II hemodynamic failure demonstrated by ipsilateral increased OEF on PET was an independent risk factor for subsequent stroke. The 2-year risk of ipsilateral ischemic stroke was shown to be 5.3% in 42 patients with normal OEF and 26.5% in 39 patients with increased OEF. STLCOS appeared to identify the subgroup of patients who were at high risk of subsequent stroke and would thus be candidates for a revascularization procedure in the hopes of improving hemispheric OEF and reducing stroke risk.

In 2006, in a multicenter, prospective, randomized controlled trial of 196 patients with symptomatic cerebral artery occlusive disease and hemodynamic cerebral ischemia, the Japanese EC-IC Trial (JET) Study Group published preliminary data supporting the role of revascularization in decreasing subsequent strokes at 2 years. ^23a Hemodynamic compromise was assessed using a paired CBF SPECT study done before and after an acetazolamide challenge. Of the 196 patients, 98 patients were randomized to the maximal medical therapy arm and 98 to the EC-IC surgical bypass plus maximal medical therapy group. The interim results revealed incidence of stroke recurrence and death in the surgically treated group was lower than that of the medical group alone (23.1% in the medical arm and 15.2% in the surgical arm; P = .046), but the final results of the trial were not published in the English-language literature.

In 2011, the Carotid Occlusion Surgery Study (COSS), a prospective, randomized controlled trial of patients with complete ICAO and elevated OEF in the ipsilateral cerebral hemisphere, was halted early due to futility analysis showing no benefit to surgery and no significant difference in events rates between the medical and surgical group (22.7% vs. 21%). The study compared STA-MCA bypass to medical therapy alone, with a primary endpoint of 30-day stroke and death and ipsilateral ischemic stroke within 2 years. Despite outstanding long-term bypass graft patency rates reaching 96%, a significantly reduced incidence of stroke after the initial postoperative period, and improved cerebral hemodynamics with reduced OEF on follow-up PET, STA-MCA bypass failed to provide an overall advantage when considering the ipsilateral 2-year stroke recurrence. This was in part due to the effectiveness of medical management in preventing recurrent strokes (better than the expected 22.7% in the medical arm vs. a 40% projected event rate) and a relatively high 14.4% 30-day peri-operative event rate.

In response to the COSS trial, the Cerebrovascular Section of the American Association of Neurological Surgeons (AANS) and Congress of Neurological Surgeons (CNS) published a commentary to expand upon the methodology and design limitations, as well as utility of STA-MCA bypass for cerebral ischemic disease. Therein it was highlighted that PET imaging criteria used in COSS had been modified compared to the criteria reported in the original STLCOS study, which poses a potential issue in appropriately selecting the most hemodynamically compromised patients. In COSS, a total of 93 bypasses were performed at the hands of 30 surgeons. When considering the elevated perioperative stroke risk, the need for bypass surgery to be performed in the hands of high-volume surgeons with sufficiently low perioperative morbidity becomes evident; if the peri-operative event rates in COSS had approximated 8%, surgery would have demonstrated clear benefit in the COSS population. Despite the overall results, the stroke event rate after the 30-day peri-operative period was significantly lower in the bypass patients, at approximately 6%, compared to approximately 20% in the medical group. Thus, longer follow-up (e.g., 5 years) in COSS may have shown benefit of surgery. Finally, it was pointed out that COSS failed to evaluate patients with severe hemodynamic compromise whose symptoms are refractory, postural, or blood pressure dependent. In their conclusion, the AANS/CNS CV section provided support for EC-IC bypass surgery as a viable option in a select patient population when performed at institutions with low peri-operative morbidity. This opinion was also echoed by the Cerebrovascular Section of the European Association of Neurological Surgeons (EANS) in a published commentary.

Following the series of publications on utility of EC-IC bypass after COSS, Kuroda et al. published a prospective but nonrandomized study in 2014 to evaluate the benefits of STA-MCA “double” anastomosis in a select subgroup of patients with reduced CBF, CVR, and elevated OEF as a result of occlusive carotid disease. ^, Patients were evaluated via SPECT to delineate the subgroup with reduced CBF and CVR in the ipsilateral MCA territory, followed with PET imaging to evaluate OEF. Double anastomosis was performed in surgical patients, using both branches of the STA to two recipient MCA vessels, to maximize flow augmentation. In 25 surgical patients, compared with 11 medically treated, the annual rate of ipsilateral stroke was 0.7% in the surgical arm versus 6.5% in the medical arm ( P = .0188). Based on these results, the authors concluded that STA-MCA double anastomosis could potentially reduce the risk of recurrent ipsilateral stroke in hemodynamically compromised patients.

In studying more inclusive endpoints of EC-IC bypass, the 2014 Randomized Evaluation of Carotid Occlusion and Neurocognition (RECON) trial was carried out as a COSS ancillary study. RECON aimed to determine whether neurocognition could be improved over 2 years following EC-IC bypass surgery, when compared to best medical therapy alone in patients with symptomatic ICAO and increased OEF as measured by PET. In this ancillary study, 89 patients from COSS were recruited. Of the cohort of patients with increased OEF, 13 patients were randomized to the surgical group and 16 to the medical group. After controlling for age, education, and depression, no difference in 2-year cognitive changes between the medical and surgical arms was seen ( P = .9). RECON concluded there was no significant neurocognitive improvement after bypass surgery.

Overall, with the currently available evidence, only a subgroup of patients with recurrent ischemic events and extensive cerebrovascular hemodynamic disease who have failed maximal medical therapy may potentially benefit from EC-IC bypass ( Fig. 77.1 ). This benefit is hypothesized to be seen in centers where EC-IC bypass surgery can be performed with sufficiently low perioperative morbidity.

Bypass for Moyamoya

Moyamoya is a progressive disease affecting the intracranial vasculature, causing stenosis and occlusion of the ICA over time, leading to TIAs or strokes. Suzuki and Takaku defined the stages of progression of moyamoya disease in their 1969 article as follows: stage I, narrowing of the ICA bifurcation; stage II, dilatation of the ACA and MCA and narrowing of the ICA bifurcation with formation of moyamoya vessels; stage III, narrowing of the ACA, MCA, and the ICA bifurcation with progression of moyamoya vessel collaterals; stage IV, occlusion of the ICA with tenuous ACA and MCA; stage V, continued occlusion of the ACA, MCA, and ICA with reduction of the moyamoya vessels; stage VI, disappearance of the ICA with progression of ECA collaterals and disappearance of the moyamoya vessels. This progressive ischemic picture often presents as ischemic stroke in children and ischemic or hemorrhagic stroke in adults.

For ischemic moyamoya disease, the indications for bypass can include the presence of ischemic changes, TIA, and presence of hemodynamic impairment even in the absence of clinical symptoms. Assessment of hemodynamic impairment and identification of reduced/absent CVR can be performed using any of the imaging techniques discussed earlier. Evidence for the effectiveness of EC-IC bypass in moyamoya is derived from multiple observational series. Revascularization has been shown to have benefits in preventing future strokes and neurologic decline ( Fig. 77.2 ). There are distinctions to be made in timing of treatment and modes of bypass between the pediatric and adult populations. Given the progressive nature of the disease within the pediatric population, early intervention in the setting of hemodynamic impairment without clinical symptoms is supported. Furthermore, within the pediatric population, use of indirect bypass in addition to direct bypass can be utilized to provide alternative flow to the ischemic region. Indirect bypass entails placement of surrounding tissue, such as temporalis muscle, dura, or periosteum, over the brain surface, which allows angiogenesis of new vessels into the underlying brain tissue from ECA feeders, such as meningeal arteries. Depending on the tissue utilized, this procedure can be referred to as encephaloduroarteriosynangiosis or encephalomyosynangiosis . Indirect bypass can be performed in addition to direct STA-MCA bypass, or, in the pediatric population, is often performed in lieu of direct bypass. Indirect bypass alone in the pediatric population has been shown to have long-term benefits. In adults, direct bypass or the combination of direct and indirect bypass offers the most robust revascularization strategy. ^,

In the case of hemorrhagic moyamoya, which is encountered in the adult population, patients have been shown to benefit from direct STA-MCA bypass. The data demonstrating reduction in recurrent hemorrhagic stroke from bypass stems primarily from the Japan Adult Moyamoya (JAM) trial published in 2014. In this multicenter, prospective, randomized study, 80 patients with moyamoya who had experienced intracranial hemorrhage within a year were randomized to conservative treatment or EC-IC bypass with 5 years follow-up. The primary endpoint was defined as all adverse events, with re-hemorrhage as a secondary endpoint. The surgically managed group did significantly better pertaining to adverse events with 14.3%, compared to 34.2% in the nonsurgical group. Kaplan-Meier cumulative curves for primary endpoints revealed the surgical group to be superior with 3.2% per year risk of adverse events versus 8.2% per year. The same analysis for secondary endpoints revealed a re-hemorrhage rate of 2.7% per year for the surgical group versus 7.6% per year for the nonsurgical group. These results have since been replicated by more recent studies, further highlighting the benefits of EC–IC bypass in the moyamoya population, particularly that of the adult population.