Bypass for Acute and Chronic Ischemic States
Pearls
Do not treat embolic stroke with a bypass!
Do not base treatment solely on the presence of anatomic occlusion. Instead, focus on treating defined hemodynamic abnormalities in symptomatic individuals.
Careful planning of the incision is critical to allow adequate vessel dissection and reduce the risk of superficial temporal artery (STA) injury. The Doppler probe is used to map and mark the course of the STA. Very gentle handling of the donor artery is critical. Overly aggressive maneuvers can either damage the artery or lead to vasospasm of the graft. Adventitial irrigation with papaverine solution can help to reduce spasm of the artery. Care should also be taken in planning the anastomosis to avoid kinking or twisting of the graft.
Improper orientation or twisting of vein grafts is an important pitfall in executing a vein graft bypass. Orientation should ensure that the vein valves are aligned to allow the flow of blood in the intended direction. Every effort should be made to ensure that the graft does not twist or kink as it is passed through the subcutaneous tunnel from the cervical to cranial incision, as twisting and kinking have been shown to be common reasons for graft thrombosis and failure.
♦ Indications
The Extracranial-to-Intracranial Bypass Trial
After the introduction of extracranial-to-intracranial (EC-IC) arterial bypass in 1967 by Donaghy and Yasargil,1 the techniques were further developed and refined by numerous surgeons over the following years. Initially envisioned as a strategy to prevent ischemic stroke in patients with cerebrovascular arterial occlusion, bypass later found utility as flow replacement in the treatment of complex cerebral aneurysms and skull base tumors as well. However, following the publication of the results of the international EC-IC bypass trial in 1985, the utility of bypass for cerebrovascular ischemic disease was cast into serious doubt.2 The study demonstrated that patients randomized to surgical treatment did no better than patients with medical treatment in terms of subsequent risk for stroke. However, subsequent evaluations of the study design have pointed to serious flaws that may compromise the validity of the study findings.3 , 4 Most significantly, beyond demonstration of anatomic cerebrovascular occlusive disease, there was no hemodynamic evaluation performed in these patients, principally because the study preceded effective cerebral blood flow testing methods. Therefore, patients with embolic strokes and otherwise relatively normal cerebral blood flow may have been treated by bypass, which is not the preferred management in this setting.
Other factors also compromise the findings of the study. The study group was quite heterogeneous, enrolling patients with various types of cerebrovascular lesions including internal carotid artery (ICA) occlusion, ICA siphon stenosis, middle cerebral artery (MCA) stenosis, and MCA occlusion. Only anterior circulation disease was treated. There also appeared to be a selection bias in that some patients were treated by bypass without randomization and not included in the study, suggesting that those patients who were perceived to be most in need of bypass, and therefore most likely to benefit, were not enrolled.4 Interestingly, both medical and surgical treatment groups, had a low rate of major ipsilateral stroke following enrollment, suggesting that most patients had adequate collateral circulation and probably did not need surgery. Although some of these patients may have been symptomatic briefly when the acute occlusion occurred, due to acute embolization or infarction of areas with inadequate circulation, the remaining circulation following occlusion had or developed adequate native collateral flow.
The subsequent St. Louis Carotid Occlusion Study assessed risk factors for stroke in patients with asymptomatic and symptomatic carotid occlusion.5 – 7 Patients were studied with oxygen-15 positron emission tomography (PET) for evidence of hemodynamic failure (Fig. 14.1). Elevated oxygen extraction fraction (OEF) as measured by this method (so-called stage II hemodynamic failure) is a marker of hemodynamic failure, indicating increased extraction of oxygen by ischemic brain tissue. It was found in this study that not all patients with carotid occlusion have hemodynamic failure. Over 50% of patients with carotid occlusion had normal OEF. It was also shown that patients with carotid occlusion and ipsilateral elevation in OEF had a significantly higher risk for subsequent stroke than patients without elevated OEF (2-year ipsilateral stroke rates of 26.5% and 5.3%, respectively). This stroke rate in patients with hemodynamic failure as demonstrated by O-15 PET was much higher than the overall stroke rate of patients in the EC-IC bypass trial, suggesting that many patients in the prior study did not have hemodynamic failure. In fact, the stroke rate in the EC-IC trial was close to those who had carotid occlusion without hemodynamic failure.
Despite its shortcomings, the EC-IC bypass trial remains the only large prospective randomized trial to assess cerebrovascular bypass for the prevention of stroke. Therefore, bypass cannot be generally recommended as a routine treatment for stroke prevention in patients with ICA or MCA occlusion or stenosis. However, because of these shortcomings of the EC-IC bypass trial, the use of bypass in the treatment of cerebrovascular ischemic disease has continued, albeit with stricter attention paid to patient selection to avoid the problems with the trial. Most notably, cerebrovascular bypass is currently being reevaluated by both the Carotid Occlusion Surgery Study (COSS)6 and the Japanese EC-IC Bypass Trial (JET).8 These studies are utilizing modern techniques to evaluate hemodynamic parameters indicative of cerebrovascular insufficiency and only randomizing patients with hemodynamic failure thought to be at highest risk for subsequent stroke. At present, COSS has randomized close to 200 patients to bypass or medical treatment, with an ultimate target of 372 randomized patients. The results of this study are not likely to be available for several years.
Patient Selection
Embolic Stroke Versus Hemodynamic Compromise
Although we do not yet have the results of these ongoing studies to establish guidelines that support current treatment, with the information we already have we can begin to plan a patient selection and treatment algorithm to determine the suitability of bypass in the treatment of these diseases for individual patients. One of the first considerations when evaluating a patient with ischemic symptomatology, including strokes, transient ischemic attacks (TIAs), cognitive decline, and other manifestations of focal or widespread ischemia, is to determine whether the underlying pathology is related to embolism or hypoperfusion. Embolic stroke, due to embolism from an extracranial source such as cervical internal carotid atherosclerotic stenosis, valvular or intracardial thrombus, or MCA stenosis, does not represent an ongoing failure of cerebrovascular blood flow but rather a discrete blockage of blood flow in one distribution due to a clot. Therefore, bypass for these patients does not treat the underlying problem and does not prevent subsequent embolic stroke, as was seen in the EC-IC bypass trial. Once identified, embolic sources of cerebral infarction are best treated by removal of the embolic source (e.g., carotid endarterectomy, endovascular angioplasty and stenting, treatment of atrial fibrillation) or medical management (antiplatelet medications or anticoagulation). Transcranial Doppler may be useful to directly document spontaneous microemboli.
Alternatively, if a patient with ischemic symptomatology does not have findings consistent with embolic disease but has anatomic evidence of cerebrovascular occlusive disease (either cervical or intracranial), a workup for hemodynamic compromise and consideration of a bypass are warranted. The decision to treat these patients by bypass should be based on these blood flow considerations and not entirely based on the demonstration of stenotic or occlusive disease. For example, even patients with complete carotid occlusion may not be at significant risk for stroke if their collateral circulation is adequately supplying the needs of that hemisphere.7 Asymptomatic patients and those with adequate cerebral blood flow and without markers of hemodynamic compromise are not likely to benefit from bypass. As stated previously, the St. Louis Carotid Occlusion Study demonstrated a low annual risk for stroke in patients with occlusion but adequate perfusion (by O–15 PET parameters).5 – 7 Less than 50% of patients with carotid occlusion were found to have stage II hemodynamic failure (elevated ipsilateral OEF).
With stenosis or occlusion, if there is a mild to moderate decrease in blood flow, autoregulatory compensation occurs, in which vasodilation maintains a normal cerebral blood flow (CBF). These patients have a reduced cerebrovascular reserve capacity. At more severe levels of compromise, blood flow drops below the level that can be regulated by vasodilation, and CBF will begin to decrease, marked by an increased extraction of oxygen from the blood, often termed “misery” perfusion or stage II hemodynamic failure. It has been shown that this subgroup of patients with a severe compromise in cerebrovascular reserve capacity or frank misery perfusion are at significantly higher risk of stroke.9 – 11 Demonstration of misery perfusion by PET has been associated with a 2-year stroke risk of approximately 30%.5 This subgroup has been chosen for randomization to bypass in the COSS trial.
There are a couple of clinical situations that warrant special mention when considering bypass for patients with an occlusive lesion. Some patients with occlusion may present with episodic symptoms related to orthostatic hypotension despite adequate baseline collateral circulation. Some patients treated for hypertension are on excessive doses of antihypertensive medications, which result in severe postural hypotension. A careful history and physical can usually identify this as the cause of a patient’s symptoms. In these cases, medication adjustments may be adequate to prevent hypotension, and the resulting ischemic episodes as their blood flow may be adequate under normal blood pressure parameters.
A minor stroke or TIA at the time of acute ICA or MCA occlusion may occur due to transient embolization or obstruction of small perforator branches. However, some of these patients stabilize after the acute occlusion if collateral circulation is adequate. Despite the initial symptoms, these patients do not necessarily require a bypass if they do not have ongoing TIAs and their hemodynamic assessment does not reveal hemodynamic failure.
Hemodynamic Assessment
There are several current methods to assess the hemodynamic status of a patient with cerebrovascular occlusive disease, including O-15 PET, xenon computed tomography (CT), single photon emission computed tomography (SPECT), transcranial Doppler (TCD), and magnetic resonance imaging (MRI) perfusion protocols.12 Currently, O–15 PET is considered the gold standard because of its ability to demonstrate both misery perfusion, as indicated by an increased oxygen extraction fraction (OEF), and reduced cerebrovascular reserve capacity, as indicated by increased cerebral blood volume (CBV) or CBV/CBF ratio, which reflect autoregulatory vasodilation. Although patients with misery perfusion (stage II hemodynamic failure) are at the highest risk of stroke, patients with reduced reserve may also be at risk of stroke, particularly if the causative occlusive lesion is not stable and progresses.
Because O–15 PET imaging is not always readily available, other indicators of hemodynamic compromise have been used. In cases of severely reduced perfusion, CT- and MRI-based perfusion studies can be useful to demonstrate ischemia. However, these studies do not demonstrate reduced reserve capacity while autoregulatory vasodilation is compensating for reduced blood flow directly. If PET is not available to assess CBV, provocative testing along with SPECT, xenon CT, or MRI perfusion can be used to measure the degree of reduction in reserve. Blood flow measurements are taken at baseline and then following a vasodilatory stimulus (acetazolamide [Diamox] administration, hypercapnia, or physiologic tasks such as hand movement).12 Currently, xenon CT is not readily available and rarely used due to problems with Food and Drug Administration (FDA) approval. Therefore, currently MRI perfusion with and without Diamox is often used to assess hemodynamic failure. In patients with severe hemodynamic compromise, no improvement in hypoperfusion is seen with Diamox. Likewise, TCD can be used in cases of ICA occlusion to evaluate flow before and after carbon dioxide inhalation. TCD has the added advantage of allowing for detection of spontaneous microembolization. Several small studies have suggested that assessing cerebrovascular reserve vasoreactivity by SPECT, xenon CT, MRI, or TCD with Diamox or hypercapnea challenge can predict patients at highest risk for subsequent stroke.9 – 11 , 13 , 14
A relatively new technique for measuring flow velocities in individual intracranial arteries is available—quantitative magnetic resonance angiography (MRA). This software (noninvasive optimal vessel analysis [NOVA], VasSol, Inc., Chicago, IL) allows the measurement of absolute blood flow velocities along with flow direction for all major intracranial arteries and has been validated in both in vitro and in vivo studies.15 , 16 This emerging technology can play an important role both in assessing preoperative blood flow volume to identify patients for whom bypass may be beneficial as well as in following bypasses postoperatively in a noninvasive manner. Additionally, because it has the capability to show the direction of flow, it can help elucidate very complex flow patterns and suggest the most efficient way to improve that patient’s perfusion. For example, it has been used to demonstrate subclavian steal and reverse vertebral flow as the offender in a patient with anterior circulation hypoperfusion.17 This finding suggested correctly that treatment of the subclavian stenosis was the best management, not bypass to the MCA. It may also, along with PET or other perfusion techniques, help to demonstrate normal flow that does not require a bypass. It therefore suggests the correct treatment target and can help avoid unnecessary bypasses.
Hemodynamic studies can also be used to follow patients post-bypass. Some small studies have already demonstrated improvement in misery perfusion as measured by PET following bypass18 , 19 and we have observed this in patients of our own (Fig. 14.2).
Indications for Bypass
Considering both patient-specific factors and the hemodynamic assessment of cerebral blood flow, the following general indications for bypass have been suggested20:
Poor cerebrovascular reserve or misery perfusion as demonstrated by hemodynamic imaging (xenon CT, PET, SPECT, MRI perfusion, TCD, quantitative MRA, etc.)
Failure of maximal medical treatment
Symptomatic patient with radiographic findings demonstrating an occlusive lesion consistent with symptoms
Lack of other major medical comorbidity that would be a contraindication to bypass surgery
Chronic States
The most common ischemic indications for bypass are chronic conditions that result in decreased cerebral blood flow. The most common underlying pathologies include cervical carotid occlusion without adequate distal collateral circulation and intracranial atherosclerosis. As mentioned previously, occlusion alone is not indication enough for bypass. If adequate collateral circulation exists, stenosis or occlusion of cervical or intracranial blood vessels does not in and of itself warrant bypass. Moyamoya disease can be another less common cause of decreased cerebral blood flow chronically.
Anterior Circulation Ischemia
Occlusive disease of the carotid circulation, either cervical or intracranial, can be a cause of hemodynamic failure and cerebral ischemia. Although the COSS trial is still underway and there is no evidence to support it, the finding that hemodynamic failure, as measured by increased OEF on PET or vasodilatory challenge with TCD or MRI perfusion, is highly associated with subsequent stroke risk supports the aggressive treatment of these patients. The COSS trial is randomizing patients to superficial temporal artery (STA)-MCA bypass if they have symptomatic carotid disease with hemodynamic failure.
Patients with chronic ICA or MCA occlusion should be considered for bypass only if they have a documented occlusive lesion with impaired hemodynamic reserve and continue to have refractory ischemic symptoms (TIAs or strokes) despite maximal medical therapy with antiplatelet or anticoagulant medications and correction of hypotension. Currently, we assess hemodynamic failure in these patients by MRI perfusion with and without Diamox or with TCD with carbon dioxide reactivity.
♦ Treating Stenosis
It should be noted that there is even more controversy in regard to treating intracranial atherosclerotic stenotic disease by bypass. Some patients with intracranial occlusive lesions, particularly if they have complete intracranial occlusion with hemodynamic failure, may benefit from bypass. The ongoing JET trial is testing the safety and efficacy of bypass in these patients with intracranial arterial occlusive disease.
However, the EC-IC bypass trial suggested that patients with stenosis actually fared worse after bypass. The competitive flow from the bypass changes the flow dynamics such that it can precipitate flow stasis and thrombosis at the location of the stenosis. In some cases this may cause a thrombus that can occlude the lenticulostriate arteries, propagate, or embolize, resulting in infarction. Therefore, stenosis without occlusion is a contraindication for bypass and should be treated with antiplatelet or anticoagulant therapy and endovascular techniques (angioplasty and stenting) when possible. In rare cases, symptomatic stenoses not amenable to endovascular and medical treatment can be considered for indirect revascularization procedures. These procedures provide additional collateral without the sudden hemodynamic change associated with a direct bypass that may result in acute occlusion.
♦ Moyamoya Disease
Moyamoya disease represents a unique case of intracranial stenosis or occlusion, and the indications for revascularization have not been definitively established by a randomized trial. We consider revascularization in any patient with significant, symptomatic moyamoya disease because it is usually progressive. In the case of ICA or MCA stenosis, we strongly prefer the use of indirect revascularization (encephaloduroarteriosynangiosis with or without bur holes) to avoid competing flows precipitating occlusion and stroke. In the case of complete obstructive lesions, bypass or indirect revascularization has been shown to be of benefit in preventing future ischemic events.
Posterior Circulation Ischemia
Symptomatic vertebrobasilar occlusive disease carries a high stroke risk, estimated at 10 to 15% per year despite maximal medical management.21 , 22 Patients with stenosis are generally treated by anticoagulation with or without endovascular angioplasty or stenting.23 Most patients who present with complete occlusion of bilateral vertebral arteries or the basilar artery have acute, severe, and often life-threatening, strokes at the time of occlusion. However, some patients survive with chronic bilateral vertebral artery occlusion because they develop collaterals through posterior communicating arteries that are present but inadequate or through other unusual collateral pathways such as the anterior spinal artery. These patients may be disabled by recurrent, frequent TIAs or strokes. In the case of complete bilateral occlusion, there are no endovascular options for treatment. In these patients with medically refractory TIAs or recurrent strokes and demonstrated occlusive lesions (by MRA, computed tomography angiography [CTA] or angiography), posterior circulation bypass may be a treatment option to consider. However, because of the higher rate of complications, morbidity, and mortality associated with the more technically demanding posterior circulation bypasses, caution must be used in considering these treatments. Of note, cerebrovascular reactivity and perfusion studies are technically ineffective and therefore are rarely performed. However, quantitative MRA (NOVA) demonstrating low distal flow has shown some utility in predicting those patients at highest risk for subsequent stroke in preliminary studies.24
Acute States
There is only rarely an indication for surgical revascularization in the setting of acute ischemic disease. The large majority of cerebral infarctions that present acutely, within hours of the onset of symptoms, are best treated by rapid interventions including IV administration of recombinant tissue-type plasminogen activator (rtPA), intraarterial chemical thrombolysis, or mechanical clot retrieval. Those that present outside of the 3- to 6-hour window will generally only be treated with anticoagulation and treatment of the embolic source (carotid disease, cardiac disease, etc.), if applicable. These patients may also benefit from treatment with partial aortic occlusion as evaluated in the Safety and Efficacy of NeuroFlo Technology in Ischemic Stroke (SENTIS) trial.25 , 26 The results of this trial are promising, but the mechanism of action has not been completely elucidated. Bypass performed acutely after an occlusion with a large acute infarction is contraindicated as it can induce hyperperfusion and cerebral swelling or hemorrhagic transformation or intracerebral hemorrhage in the area of the infarction.
Occasionally, patients with underlying chronic atherosclerosis may present acutely with a rapidly progressive deterioration in perfusion, resulting in an unstable pattern of symptoms such as crescendo TIAs, waxing and waning neurologic deficit, or gradually progressive “stuttering” neurologic deficits. In these clinical circumstances, if hemodynamic assessment determines that hypoperfusion, and not embolic occlusion, is the culprit, and symptoms are not stabilized rapidly with anticoagulation or induced arterial hypertension in the intensive care unit (ICU), acute bypass may be an appropriate line of action.
♦ Types of Bypass
How Much Blood Volume Do You Need and Where to Put It?
Flow Augmentation Versus Flow Replacement
When considering a bypass for any pathology, planning of the appropriate bypass graft should first take into consideration the flow needs of the patient. In patients with some native flow, flow augmentation aims to supplement the existing blood flow to a vascular territory. In other words, it adds blood flow to a territory that has blood flow but not enough to fully supply the demands of that tissue territory. Because less flow is needed, flow-augmentation procedures do not require bypasses with high flow rates. Native scalp arteries, the superficial temporal artery and occipital artery, almost always supply enough supplemental flow in these cases. Flow replacement, on the other hand, aims to completely replace the blood flow to a particular vascular territory. Because a large amount of blood flow is needed immediately following bypass, larger, high-flow grafts are sometimes required in this setting. What type of graft is needed to supply this flow depends on the size and normal flow of the artery in question. Although in smaller terminal branches (distal MCA, posterior inferior communicating artery [PICA], etc.) flow can be replaced by an STA or occipital artery graft, larger arteries (complete vertebrobasilar system with atretic posterior communicating arteries, entire internal carotid artery, etc.) may require higher-flow grafts such as a saphenous vein graft or radial artery graft.
True flow replacement is generally only required when there is a planned, intentional occlusion of a blood vessel without adequate collateral circulation to supply the area irrigated by that artery. This most commonly occurs in cases of complex or giant aneurysms or invasive tumor surgery with planned blood vessel sacrifice.27 Some large or complex aneurysms, including fusiform aneurysms, cannot be treated by standard clipping or coiling techniques. Occlusion of the artery may require proximal hunterian ligation or completely trapping the aneurysm, occluding it and the parent artery from the circulation. In some cases there will still be distal blood flow through collateral circulation. However, if distal collateral circulation is very poor or absent (determined by angiogram or balloon test occlusion), a flow replacement bypass may be required. In the treatment of ischemic disease, flow replacement is rarely required.
Patients with ischemic disease due to hemodynamic compromise have some level of basal blood flow hovering near the ischemic threshold (otherwise they would have infarcted the entire territory already) and supplementing that blood flow is generally all that is required to prevent further ischemic injury. In fact, a high-flow bypass in a patient with some native flow may create a hyperperfusion syndrome or precipitate hemorrhage into an infarcted area of brain.28 A high-flow graft is almost never required to treat ischemia because if patients have such a profound decrease in blood flow, far below the ischemic threshold, they generally have a large stroke quickly and are not candidates for bypass. After bypass, if the patient’s underlying collateral flow continues to become progressively worse, this generally occurs chronically and the flow through the bypass can increase over time to accommodate it. Therefore, flow augmentation is almost always sufficient for the treatment of cerebrovascular ischemic disease.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
