Role of and Indications for Bypass Surgery After Carotid Occlusion Surgery Study (COSS)?
- Carotid Occlusion Surgery Study (COSS)
- cerebral hemodynamics
- cerebral revascularization
Bypass surgery falls into 2 distinct categories: flow augmentation and flow preservation. Flow augmentation aims to restore flow to hypoperfused brain territories in patients with steno-occlusive diseases.1 Flow preservation aims to replace the blood flow provided by a major intracranial vessel, the occlusion of which is necessary for treating an underlying disease, such as an aneurysm or a tumor.2,3
Flow augmentation bypass has been critically studied in randomized clinical trials (RCTs),4–6 most recently the Carotid Occlusion Surgery Study (COSS)5 and the Japanese Adult Moyamoya (JAM) trial,6 whereas flow preservation bypass remains a niche procedure because of its rare indication.2,3,7,8
In this review, we aim both to critically summarize the current state of knowledge on the role of cerebral bypass surgery because the publication of COSS and to present possible future directions for surgical cerebral revascularization.
Beyond the underlying disease and the consequent aim that defines the 2 categories of bypass (see above), several other criteria are used to classify bypass constructs. A well-known classification is the distinction into direct versus indirect revascularization procedures or the combination of both.9,10 Direct bypasses consist of direct anastomosis between a donor artery and an intracranial recipient artery. A direct bypass has the advantage of instantly delivering blood flow to the brain.1,2,7 Indirect techniques rely on the overlay of vascularized tissue (ie, muscle, dura, pericranium, omentum) onto the cerebral cortex to promote neoangiogenesis over time and achieve a delayed revascularization.10,11 Combined procedures consist of the combination of direct and indirect techniques in the same surgical session.10
According to the origin of the donor artery, direct bypass is further categorized into extracranial-to-intracranial (EC–IC) versus intracranial-to-intracranial (IC–IC). Furthermore, the donor and the recipient artery can be anastomosed with versus without graft interposition, depending on the interposition or not of a vascular conduit (ie, arterial or venous graft).3,7 Traditionally, the bypass will be named according to the donor and the recipient vessels (ie, superficial temporal artery-to-middle cerebral artery [STA–MCA] bypass).2
The type (end-to-side, end-to-end, and side-to-side) and the number of microanastomosis also vary depending on patient-specific indications and angioanatomy.2 The surgeon can further use the conventional occlusive microanastomosis technique or apply a nonocclusive technique (eg, excimer laser-assisted non-occlusive anastomosis [ELANA] technique).12
Direct bypass procedures can be also categorized according to the amount of flow (the capacity) provided by the construct: low (<50 mL/min) versus intermediate (50–100 mL/min) versus high-capacity bypass (>100 mL/min).3 Figures 1 and 2 schematize the most commonly used bypass procedures.
The choice of the ideal bypass depends on several factors, the most important of which are the indication and the aim for the bypass, as well as the match between the flow demand of the revascularized brain territory and the flow capacity of the bypass. Indications will be discussed in the next sections.
Indications for Cerebral Bypass
Complex Intracranial Aneurysms
Complex intracranial aneurysms are not always amenable to selective clipping, coiling, or other endovascular procedures. Treatment of such lesions may require trapping, that is, exclusion of the aneurysm with sacrifice of the artery bearing the aneurysm.2,7 Because the goal of any aneurysm treatment is both aneurysm exclusion and preservation of blood flow to the brain, flow preservation bypass strategies represent an essential treatment option to divert blood flow to downstream vascular territories (territory supplied by the occluded vessel/s).2,3,7,12 For a detailed description of types of and indication for trapping strategies, we refer to the pertinent literature.2,3,7,8,13,14
Matching the bypass flow to the demand of the brain territory perfused by the occluded artery is the key element of decision-making when performing a flow preservation bypass.2,7,8 Preoperative and intraoperative quantitative flow measurements are necessary to predict the flow and to confirm that the capacity of the bypass matches the flow demand of the vascular territory.2,7,8
No RCTs to test the value of bypass surgery for treating cIAs have been performed. Because of the rarity and variety of cIAs, it has not been feasible to perform large-scale trials. Furthermore, the evolution of endovascular treatments for cIAs tends to decrease the indication for flow preservation bypass.15 Nevertheless, bypass is established by a multitude of case series, which document the usefulness of revascularization and demonstrate that bypass surgery plays an important role for managing cIAs.2,3,8,13,14
Cerebral Tumors Involving the Proximal Vasculature
Benign skull base tumors tend more to encase than invade major arteries. However, in case of previous surgery or irradiation, tumors can be densely adherent to the arterial wall. Malignant skull base tumors are more prone to invade major arteries. Radical tumor removal can therefore be impossible in some skull base tumors without sacrificing a major artery.16
The risk/benefit ratio for complete resection combined with a bypass versus partial resection has evolved in favor of partial resection and adjuvant radiotherapy or chemotherapy.16,17 Thus, flow preservation bypass for skull base tumors has declined in frequency during the last decades. Bypass surgery is currently performed only in selected and rare cases, in which it is of importance to consider whether the benefit of radical resection plus arterial sacrifice and bypass improves survival with good quality of life and outweighs the risks.3,16–18
Moyamoya vasculopathy is a rare steno-occlusive condition characterized by idiopathic intimal thickening of the internal carotid artery (ICA) and its proximal branches.9 Moyamoya progressively compromises cerebral perfusion and hemodynamics, predominantly in the middle cerebral artery (MCA) territories and in the frontal areas, and patients develop compensatory collateral vascular networks.1,9
Symptoms are brain ischemia (stroke and transient ischemic attacks [TIAs]) or brain hemorrhage because of either the insufficiency or the rupture, respectively, of the compensatory collateral vessels under hemodynamic stress.1,9
Although the value of bypass surgery for prevention of stroke and of cognitive deterioration in Moyamoya patients has not been studied with RCTs, all observational studies indicate the benefit of cerebral revascularization.10,19,20 The literature consistently documents: (1) unfavorable annual ischemic stroke rate in untreated patients (13.3%)21; (2) high rate of disease progression with subsequent symptoms occurrence in the nonsurgically treated hemispheres22; (3) favorable result of revascularization both in children and adults.10 Therefore, it seems unlikely that RCTs will be performed to test the efficacy of revascularization surgery for prevention of stroke recurrence and of cognitive deterioration.10,19,20
Surgery is recommended for children and adults with ischemic symptoms and compromised hemodynamics.1,10,19,20,23 Careful observation is justified in asymptomatic patients with normal cerebral hemodynamics.10 Figure 3 describes the stages of cerebral hemodynamic impairment.
There is no consensus on what type of revascularization surgery should be performed. Direct, indirect, and combined bypass procedures are used for treating Moyamoya.9,10 The choice of technique remains debated in the literature, both in adults and in children.10,11 The most commonly used direct revascularization procedure is the STA–MCA bypass (Figure 2, left panel). Other extracranial donor vessels can also be used (for instance the occipital or the posterior auricular artery).1,9
Several indirect techniques have been proposed using the overlay of vascularized tissue (periosteum, muscle, and dura) onto the cerebral cortex to promote neoangiogenesis over time.1,10 (Figure 2, right panel; online-only Data Supplement). Combined revascularization procedures combining direct and indirect bypass provide the advantages of both techniques, but at the risk of a somewhat more complex procedure.1,10
Adult patients are generally treated by means of direct bypass techniques (typically STA–MCA bypass), in comparison with children for whom indirect or combined revascularization strategies are preferred9,10
Although most techniques aim to revascularize the MCA territory, augmentation of cerebral blood flow of the frontal areas is of importance especially in the pediatric population. Bifrontal hypoperfusion plays a deleterious role in intellectual development and cognitive performance and in lower extremity and sphincter function.24,25 Therefore, it is important to consider timely revascularization of the frontal areas to prevent neurocognitive decline.24,25 Besides the direct STA to anterior cerebral artery (STA-ACA) bypass,26 indirect and combined bypass techniques have been proposed for bifrontal reinforcement of blood supply.1
The role of surgical treatment in Moyamoya presenting with intracerebral bleeding has been recently demonstrated with the JAM trial,6 the first prospective RCT focused on Moyamoya disease (MMD). Eighty adult patients (between 16 and 65 years) with hemorrhagic MMD were enrolled and randomized, 42 to the surgical and 38 to the nonsurgical group. In the surgical group, patients underwent bilateral direct bypass. Indirect bypass alone was prohibited in the study protocol, as indirect revascularization thought to be associated with insufficient neoangiogenesis in adult patients.23 Although statistically marginal, the JAM trial revealed that direct bypass surgery for adult patients with hemorrhagic MMD reduces the rebleeding rate and improves the patient’s prognosis during the 5 years after enrollment.6 This trial showed that improvement of the hemodynamic state of the revascularized hemisphere reduces the hemodynamic overload of the rupture-prone fragile moyamoya collateral vessels.6 Recurrent bleeding can however take place >10 years after the initial attack.27 Therefore, these patients warrant longer term observation, and the JAM Trial Executive and Steering Committee has already decided to continue patients’ follow-up and report the 10-year results when available.6
All institutions participating in the trial had vast experience with treatment of patients with MMD. Only registered surgeons were allowed to perform the operations. The rate of perioperative complications, including transient events, was 9.5%. No permanent severe disability was reported.6
Symptomatic Cerebrovascular Atherosclerotic Steno-Occlusive Disease
Indirect cerebral revascularization methods are felt to be largely ineffective for non-moyamoya vasculopathy28,29 because of the presumed absence of the angiogenic milieu associated with MMD, although small case series have suggested some limited success with indirect strategies.30 Currently, a RCT, the EDAS (Surgical) Revascularization in patients with Symptomatic Intracranial Arterial Stenosis (ERSIAS) is underway in the United States, which may provide some insights into the role of indirect bypass for atherosclerotic disease (ClinicalTrials.gov NCT01819597; EDAS: encephalo-duro-arterio-synangiosis). Beyond this, for atherosclerotic occlusions, the primary focus has been on examining the role of direct cerebral bypass.28,31
The EC–IC bypass trial published in 1985 was the first prospective RCT aimed at investigating whether EC–IC bypass was superior to best medical therapy alone for stroke prevention.4 The study was conducted in 1377 symptomatic patients with (1) 1 or more TIAs or minor ischemic strokes within 3 months of enrollment, and 1 of the following: (2) stenosis or occlusion of the MCA proximal trunk, (3) stenosis of the ICA above the C-2 vertebral body (inaccessible to carotid endarterectomy), or (4) atherosclerotic ICA occlusion.4 The average follow-up duration was 55.8 months. Bypass patency rate was excellent (96%). However, the trial showed no significant difference between the 2 randomized groups: 29% of medically treated patients experienced 1 or more strokes compared with 31% in the surgical group. No significant difference in the incidence of fatal and nonfatal ischemic strokes was reported.4,28 This study was fiercely debated.32 One of the primary criticisms related to the lack of hemodynamic criteria used to identify and select those high-risk patients who might benefit most from bypass28,33,34
A Cochrane review31 published in 2010 reported the results of 21 trials (2 RCTs and 19 nonrandom studies) for total of 2591 patients with symptomatic carotid occlusion. This review showed bypass was neither better nor worse than medical care alone.31 Again most of the analyzed studies lacked hemodynamic criteria for patients’ inclusion.28
After the EC-IC bypass trial, methods and criteria for quantification and assessment of cerebral hemodynamic impairment were validated (Figure 3). The St Louis Carotid Occlusion Study (STLCOS)33 was a prospective study showing that symptomatic patients with carotid occlusion and presenting stage II hemodynamic impairment (increased oxygen-extraction-fraction with positron emission tomography [OEF-PET]) were at significantly increased risk of ipsilateral stroke at 2 years compared with patients without stage II hemodynamic impairment (26.5% versus 5.3%).28,33 During the same period, several other studies demonstrated that in stage II patients bypass was able to improve or even normalize the hemispheric OEF ratios postoperatively.28 Therefore, all the conditions were present to justify a new RCT to test the hypothesis that EC–IC bypass surgery would benefit patients with recently symptomatic atherosclerotic carotid occlusion when selected according to strict validated hemodynamic criteria.28,33,35,36
The COSS5 was a prospective, multicenter RCT designed to assess whether STA–MCA bypass (plus best medical therapy) was superior to best medical therapy alone in stroke prevention in patients with (1) complete ICA occlusion; (2) TIA or ischemic stroke in the hemispheric territory of an occluded ICA in the preceding 120 days.5 The participants underwent PET at COSS-approved sites. Patients’ neurological deficits were required to be stable for 72 hours before the performance of PET. Those presenting with ipsilateral-to-contralateral hemispheric OEF ratios >1.13 (derived from retrospective STLCOS subgroups analysis identifying a high-risk patient cohort) were selected and randomized.28,32,33,35 The study was prematurely stopped in June 2010 by the Data Safety Monitoring Board after enrollment of 195 randomly assigned patients because of interim futility: 97 patients were randomized to the surgical group and 98 to the medical group.32,34 The 2-year rates for ipsilateral stroke were 21% for the surgical group versus 22.7% for the medical group (P=0.78). Perioperative (within 30 days of surgery) ipsilateral stroke rates were 14.4% in the surgical group and 2.0% in the medical group, a significant difference of 12.4%.32,34 STA–MCA bypass surgery was shown to provide no clinical benefit over medical therapy.5 Notably, the medical therapy cohort fared better overall than the historical control group.37
Subsequently published COSS data showed: (1) high rates of bypass patency (98% and 96% at 30-day and at last follow-up, respectively); (2) bypass surgery improved, but did not normalize, cerebral hemodynamics (OEF-PET analysis 30- to 60-day follow-up); (3) the OEF improvement greatly reduced the risk of subsequent stroke in these patients; (4) the surgical group had much lower rates of recurrent ipsilateral ischemic stroke after postoperative day 2 when compared with the medical group (9% versus 22.7%); and (5) no patient characteristics or intraoperative variables were able to predict the occurrence of ipsilateral, periprocedural ischemic stroke.38
Further reports from COSS investigators studied the mechanisms of perioperative ischemic stroke. In patients (n=14), who developed ipsilateral perioperative ischemic stroke, 14% of stroke mechanisms (2 patients) were found to be bypass-related (ischemic infarct in the territory of the recipient artery, likely related to technical performance of the anastomosis) and 86% (12 patients) were nonbypass-related mechanisms (ischemic infarct attributable to embolism, hypoperfusion, etc).39 These perioperative events were proposed to be likely attributable to the fragility of individuals with symptomatic carotid occlusion and severely impaired hemodynamics, rather than to poor surgical techniques.28,39
COSS faced nourished criticism after its publication in 2011, focused on clinical eligibility criteria, PET eligibility criteria, selection of surgeons, duration of follow-up, power, and end points.32,34 These were addressed by Powers et al36,40 in subsequent commentaries. In particular, it has been argued that patients with symptom recurrence despite medical therapy after a carotid occlusion represent a subgroup of patients that could ultimately benefit from surgery. In fact COSS only required patients to have a single ischemic event to be eligible for the trial, whereby the single ischemic event at the time of carotid occlusion might represent a single embolic event.32 However Powers et al36 pointed out that (1) neither earlier enrollment nor recurrent ischemia identified patients at high risk of recurrent stroke in COSS; (2) these findings are consistent with data from the EC–IC bypass trial in which neither the subgroup with earlier surgery nor the subgroup with recurrent symptoms showed benefit from surgery; (3) similarly in STLCOS, neither recurrent symptoms nor time from qualifying event to enrollment were associated with subsequent stroke occurrence. Furthermore, it has been speculated that the continuation of the event rate observed during the first 2 years would have been responsible for a significant difference if the study had been continued for 5 years. As stated by COSS investigators, such an assumption would be inconsistent with data from other trials of medically treated symptomatic large artery atherosclerosis that show a major decrease in stroke rate after 2 years. For instance in the EC–IC bypass trial, the stroke rate at 2 years was 20%, but at 4 years the stroke rate had only increased by an additional 6%.4 A 2% to 3% per year-rate of stroke in the nonsurgical group of COSS for an additional 3 years would not result in a statistically significant benefit for surgery, even if no additional strokes occurred in the surgical group. The COSS final results, even with early termination for futility, excluded with >95% confidence, a benefit for surgery.28,39
Not-addressed criticisms remain: the lack of selection requirements for specialized neuroanesthesia, dedicated neurointensive care, and specialized nursing. Similarly, no recommendations were established for a perioperative management protocol.32 Nevertheless, the practicality of performing such large-scale studies has also been debated, considering the high number of patients and high cost needed to demonstrate the superiority of one treatment on the other.32
The Randomized Evaluation of Carotid Occlusion and Neurocognition (RECON) trial41 is an ancillary study of the COSS. Given the evidence from prior studies that cerebral hemodynamic is a determinant of cognitive function, RECON aimed to test if bypass could improve or preserve neurocognition at 2 years in COSS patients, in comparison with best medical therapy alone.41 Eighty-nine patients were enrolled; 41 had increased OEF and were randomized. Two died, 2 were lost to follow-up, and 2 refused retesting. Of the 35 remaining, 6 had ipsilateral stroke or death, leaving only 13 surgical and 16 medical patients. Because of the early termination of the parent study (COSS), RECON was not able to complete enrollment, although 2-year follow-up for the already randomized individuals was completed. Controlling for age, education, and depression, RECON showed that for patients with symptomatic carotid occlusion and stage II hemodynamic impairment bypass provides no benefit on neurocognition after 2 years compared with medical therapy alone.41 The small numbers however limit the power of this study.41
The Japanese EC–IC Bypass Trial (JET) was a multicenter RCT designed to assess whether STA–MCA bypass (plus best medical therapy) is superior to best medical therapy alone in reducing subsequent ischemic events in patients with recently symptomatic hemodynamic (at least stage-I on single-photon emission computed tomography; Figure 3) cerebral ischemia from chronic ICA or MCA occlusion.42–44 One hundred ninety-six patients were enrolled and randomized 50:50.28,42,44 An interim analysis with a mean follow-up of 15 months reported a statistically significant reduction of major stroke and death (primary outcome) in the surgical arm (5.1%) when compared with the medical one (14.3%).43,44 The published Kaplan–Meier curve shows, however, no end points within the first month in the surgical group: it is not mentioned if the results include the morbidity and mortality rate of the first postoperative month.28,43,44 As commented by Powers et al5, it seems unlikely that this rate was 0%, given that it was 12% in the original EC–IC Bypass Trial and 15% in the COSS. JET final results have not yet been published in a peer-reviewed, English written journal.43,44 Therefore, it is difficult to include JET study results into the general evidence base regarding bypass in atherosclerotic disease.
Currently, a large-scale multicenter RCT, the Carotid and Middle Cerebral Artery Occlusion Surgery Study (CMOSS), is underway in China, which may provide further insights into the role of bypass for atherosclerotic disease (ClinicalTrials.gov NCT01758614).
Considerations in Patients With Atherosclerotic Steno-Occlusive Disorder
Level I evidence from both the EC–IC Bypass Trial and COSS indicates that bypass does not have proven benefit in patients with recently symptomatic carotid artery occlusion with or without stage II hemodynamic failure.4,5,28 RECON furthermore failed to demonstrate that for these same patients with stage II hemodynamic failure, bypass improves cognitive function after 2 years.28,41
Advances in medical management and lifestyle modification seem to have reduced stroke risk in these patients and made the proof of benefit of surgery more difficult to achieve.34 The Stenting and Aggressive Medical Management for Preventing Recurrent Stroke in Intracranial Stenosis (SAMMPRIS) trial, for example, showed that an aggressive strategy, including low-density lipoprotein target <70 mg/dL, as well as vigilant targeting of blood pressure, diabetes mellitus, smoking, excessive weight, and inactivity, resulted in substantially lower stroke rates in patients with intracranial stenosis.45 Nonetheless, although best medical therapy is more effective than in the past, it is still not curative and many patients with severe hemodynamic insufficiency fare poorly.32
The results of these trials significantly narrow the indications for flow augmentation bypass for atherosclerotic steno-occlusive cerebrovascular disease with chronic hemodynamic insufficiency. However, rather than a blanket rejection of bypass surgery for all patients with atherosclerotic cerebrovascular disorders, we want to address 2 challenging questions in this review: (1) Are there methods to reduce perioperative complications in the early postoperative period? (2) Are there methods to identify subgroups of patients who could benefit from bypass surgery?
With regard to the first question, a significant reduction of perioperative ischemic complications could change the current statement of no benefit from bypass.28 The COSS eligible patient might still benefit from flow augmentation bypass if perioperative morbidity can be sufficiently lowered, much lower than reported in COSS and EC–IC bypass trial.28 Low perioperative morbidity is therefore a key development to aim for.43 COSS patients have been considered fragile and at high risk for perioperative ischemic events. Although surgeons underwent certification for participating into COSS, no selection requirements were established for specialized neuroanesthesia, dedicated neurointensive care, and specialized nursing. Similarly, no recommendations were established for a perioperative management protocol.32,43 Despite lower perioperative complication rates have been reported in several case series,43,46 the most reliable perioperative morbidity rates are considered to be from RCTs. Indeed, as cited by Powers et al,36 self-reported case series have been shown to consistently underreport operative complications in comparison with independent adjudication,47 and observational studies with historical controls from case series have been repeatedly shown to overestimate benefit and underestimate risk in comparison with RCTs.36 Because of the similarity of perioperative morbidities in the EC–IC bypass trial (12%)4 and COSS (15),5 it will need a thorough analysis of the comprehensive perioperative management—surgical, anesthesiological, and medical—of these fragile patients to prove that the reduction of perioperative complications is possible. A RCT, however, is not the only mechanism to establish improvements in perioperative risk. Prospective adjudicated observational data, as can be achieved with well-administered and audited registries, can serve to provide such data. The Society of Thoracic Surgeons (STS) National Database represents an example of prospective registry data on outcomes.
Finally, technical innovation such as the systematic use of minicraniotomy (2–2.5 cm) may represent an opportunity to lower the perioperative complications rate.48
The second question aims to identify subgroups of patients that would benefit from bypass.37 There are distinct subgroups of patients, for whom COSS was not specifically designed: patients with chronic retinal ischemia resulting in progressive visual loss and patients presenting with ongoing hemodynamic symptoms despite optimal medical therapy.28,36 These are patients who develop ischemic symptoms with postural changes or blood pressure variation (for instance, patients with debilitating orthostatic hypoperfusion syndrome or limb-shaking TIAs).28 Furthermore, there are patients with symptomatic carotid occlusion and particularly marked hemodynamic impairment (more severe than the OEF ratio >1.13 used in COSS), who may have a significant risk for subsequent stroke.28,36 In this context, particular attention should be given to patients harboring multiple extracranial arterial occlusions, not amenable to carotid endarterectomy or stenting, who are symptomatic despite best medical therapy. All these patient subgroups above have hemodynamic conditions indicating exhausted brain vascular reserve capacity with symptomatic oligoemia exacerbated by any hemodynamic challenge. These patients were included in neither trial and represent possible bypass candidates, if surgery can be performed with low enough morbidity. The eventual benefit of bypass surgery over medical therapy in these individuals will most likely not be testable in RCTs.
Finally, a different subgroup that may benefit from bypass are patients having acute stroke with brain tissue at risk of infraction because of persistent oligoemia in the acute phase (=penumbra), despite optimal medical and interventional management. In patients with acute or evolving stroke, outcome is known to be dependent on the urgent reestablishment of cerebral perfusion.49 Several different reperfusion methods are available: intravenous thrombolysis, intra-arterial thrombolysis or mechanical thrombectomy, carotid endarterectomy, angioplasty/stenting, surgical embolectomy, and EC–IC bypass.50 Level I evidence now exists for endovascular interventions51–53 in an emergency setting. Only few descriptive case series exist describing the role of urgent EC–IC bypass in treating patients with acute and progressive ischemic symptoms despite optimal medical therapy because of acute atherosclerotic steno-occlusive event; these report good results in term of patients’ outcome.50,54,55 These case series all have a retrospective design and are therefore vulnerable to selection bias. Furthermore, they have been published before the recent RCTs reinforcing the role of endovascular therapy for stroke.51–53 On one hand, one may state that the efficacy and safety of emergent EC–IC bypass would need to be proven by studies using adequate analytic multicenter designs. On the other hand, given the rarity of the bypass procedures performed worldwide nowadays, it is of importance to assure that these patients are referred to specialized neurovascular centers, with sufficient expertise. This stands against the call for RCTs on emergent bypass surgery as a treatment option for acute revascularization.
We however hypothesize that there is a small number of patients having acute stroke with persistent penumbra that cannot benefit from other acute revascularization interventions and may benefit from emergent EC–IC bypass, if the procedure can be performed with low enough morbidity.55 One of the key elements is to define the method to select these patients having acute stroke and who have persistent ischemic penumbra with limited core infarct, indicating ischemic tissue still viable and salvageable if local perfusion is efficiently restored.54 The concept of ‘mismatch’ is an attempt to define ischemic penumbra by neuroimaging49,53 A detailed analysis and definition of the mismatch concept is beyond the scope of this review, and we refer to the relative literature.49,53 Among individuals having acute or evolving stroke, patients who could benefit from bypass surgery might be the ones presenting with the 3 following criteria: (1) acute stroke or stroke in progress (fluctuating or worsening symptoms) despite maximal applicable medical and interventional treatment; (2) major cerebral artery occlusion, with documented region of penumbra, and (3) only a small area of infarction (to avoid hemorrhagic conversion of an acute infarction).49,53,54 Without an adequately designed analytic study to test this hypothesis, this remains an unproven but intriguing potential indication for revascularization using bypass.
We essentially distinguish 2 types of bypass, according to function: flow preservation and flow augmentation bypass. Flow preservation bypass surgery plays an important role in the management of complex intracranial aneurysms not amenable to selective clipping or endovascular procedures, when vessel occlusion is required for definitive treatment. Matching the bypass flow capacity to the flow demand of the territory that needs to be revascularized is the key element of flow preservation bypass. Technical variations in the bypass construct allow the surgeon to customize the bypass to the patient’s need. The bypass will always be a direct bypass to deliver the flow instantly to the involved territory.
Flow augmentation bypass is the only effective treatment modality for symptomatic Moyamoya patients with hemodynamic insufficiency. Revascularization has been shown to decrease both ischemic and hemorrhagic stroke rates, as well as neurocognitive decline. Revascularization surgery for Moyamoya comprises both direct and indirect techniques depending on patient age, the vascular regions to revascularize and the individual angioanatomy. Not infrequently, the surgeon will decide to perform a combination of direct and indirect revascularization techniques.
COSS and RECON trials showed no benefit of bypass over medical therapy for patients with atherosclerotic ICA occlusion with severe hemodynamic impairment because of the increased perioperative morbidity. These results narrow the indications for flow augmentation EC–IC bypass in the setting of ischemic cerebrovascular disease.
The following observations deserve, however, attention for future developments. COSS results showed a reduction in subsequent (after postoperative day 2) event rates in the bypass group, despite the failure to show an overall benefit from surgery. These data confirm the basic concept of EC–IC revascularization and indicate that if perioperative complications could be lowered, benefit is likely.
Patients with severe steno-occlusive disease continue to have significant event rates, despite medical therapy has become more effective. Further studies and improvement in perioperative management may benefit these brittle patients.
Furthermore, COSS was not designed to study 2 particular categories of patients: (1) patients presenting with ongoing hemodynamic symptoms (postural or with blood pressure variations) resistant to best medical treatment; (2) patients having acute stroke with evidence of persistent oligemic brain tissue at risk of infarction (penumbra) despite optimal medical and interventional management. The general concept of nonsurgical brain revascularization for penumbra salvation is proven in the clinical setting, through endovascular recanalization trials, and could be extended to the concept of urgent bypass in highly select patients. The benefit of interventions has to be weighed against its risks, and further prospective studies are necessary to prove the eventual benefit of bypass in selected patients with atherosclerotic steno-occlusive disorders.
We thank Mr Peter Roth (Neurosurgery, University Hospital Zurich) for the drawings in Figures 1 and 2.
Dr Amin-Hanjani (last 12 months) received material research support (no direct funds) from GE Healthcare and VasSol, Inc and also received grant support from National Institute of Health (NIH)/National Institute of Neurological Disorders and Stroke (NINDS). The other authors report no conflicts.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.115.008220/-/DC1.
- Received July 19, 2015.
- Revision received October 13, 2015.
- Accepted October 20, 2015.
- © 2015 American Heart Association, Inc.
- Esposito G,
- Kronenburg A,
- Fierstra J,
- Braun KP,
- Klijn CJ,
- van der Zwan A,
- et al
- Esposito G,
- Durand A,
- Van Doormaal T,
- Regli L.
- Powers WJ,
- Clarke WR,
- Grubb RL Jr,
- Videen TO,
- Adams HP Jr,
- Derdeyn CP
- Miyamoto S,
- Yoshimoto T,
- Hashimoto N,
- Okada Y,
- Tsuji I,
- Tominaga T,
- et al
- van Doormaal TP,
- van der Zwan A,
- Verweij BH,
- Han KS,
- Langer DJ,
- Tulleken CA.
- Briganti F,
- Leone G,
- Marseglia M,
- Mariniello G,
- Caranci F,
- Brunetti A,
- et al
- Roach ES,
- Golomb MR,
- Adams R,
- Biller J,
- Daniels S,
- Deveber G,
- et al
- Kuroda S,
- Ishikawa T,
- Houkin K,
- Nanba R,
- Hokari M,
- Iwasaki Y.
- 23.↵Research Committee on the Pathology and Treatment of Spontaneous Occlusion of the Circle of Willis; Health Labour Sciences Research Grant for Research on Measures for Infractable Diseases. Guidelines for diagnosis and treatment of moyamoya disease (spontaneous occlusion of the circle of Willis). Neuro Med Chir. 2012;52:245–266.
- Park JH,
- Yang SY,
- Chung YN,
- Kim JE,
- Kim SK,
- Han DH,
- et al
- Reynolds MR,
- Derdeyn CP,
- Grubb RL Jr,
- Powers WJ,
- Zipfel GJ.
- Gonzalez NR,
- Dusick JR,
- Connolly M,
- Bounni F,
- Martin NA,
- Van de Wiele B,
- et al
- Fluri F,
- Engelter S,
- Lyrer P.
- Amin-Hanjani S,
- Barker FG II,
- Charbel FT,
- Connolly ES Jr,
- Morcos JJ,
- Thompson BG
- Derdeyn CP,
- Gage BF,
- Grubb RL Jr,
- Powers WJ.
- Carlson AP,
- Yonas H,
- Chang YF,
- Nemoto EM.
- Powers WJ,
- Clarke WR,
- Adams HP Jr,
- Derdeyn CP,
- Grubb RL Jr
- Reynolds MR,
- Grubb RL Jr,
- Clarke WR,
- Powers WJ,
- Zipfel GJ,
- Adams HP Jr,
- et al
- Powers WJ.
- Ogasawara K,
- Ogawa A.
- Low SW,
- Teo K,
- Lwin S,
- Yeo LL,
- Paliwal PR,
- Ahmad A,
- et al
- Wegener S.