Читать книгу Vascular Medicine - Thomas Zeller - Страница 8
Оглавление1 Supra-aortic vessels
1.1 Extracranial stenoses and occlusive processes
Anatomy of the extracranial arteries: Reinhard Putz
Introduction: Jennifer Franke and Horst Sievert
Doppler/duplex ultrasonography: Tom Schilling
Conservative treatment: Horst Sievert and Jennifer Franke
Endovascular treatment: Horst Sievert and Jennifer Franke
Surgical treatment: Farzin Adili and Thomas Schmitz-Rixen
1.1.1 Anatomy of the extracranial arteries
The external arterial supply to the head is provided by branches of the external carotid artery, which arises bilaterally from the common carotid artery at approximately the level of the fourth cervical vertebra or upper margin of the thyroid cartilage (Fig. 1.1-1). Its course is highly variable; it usually runs ventral and superficial to the internal carotid artery until it is behind the mandible. Its anterior branches pass to the thyroid gland, tongue and facial skin. A deep branch passes to the pharynx. Posteriorly, arteries branch off to the occiput and auricle. Behind the temporomandibular joint, the external carotid artery divides into its two terminal branches, the superficial temporal artery and the maxillary artery (Fig. 1.1-2). These two branches, like the occipital artery, may serve as collaterals to supply the brain in cases of occlusion of the internal carotid artery or vertebral artery. The lingual artery arises just above the hyoid bone and enters the tongue in an anterior direction behind the hyoglossus. One of its branches passes under the sublingual gland to the gingiva and mandibular muscles. The facial artery sometimes arises from a common trunk with the lingual artery and turns around the submandibular gland anteriorly to the edge of the mandible, through which it passes in a small ostium in front of the insertion of the masseter muscle. After giving off the inferior and superior labial arteries, it winds through the facial muscles lateral to the nose as far as the medial angle of the eye, where it connects with the superficial terminal branches of the ophthalmic artery—although this is rarely relevant in practice.
The posteriorly-directed occipital artery is important in that after crossing under the sternocleidomastoid and splenius muscles it supplies a dense vascular network rising at the occiput at the posterior margin of the trapezius muscle; this network has connections with the neighboring arteries. The superficial temporal artery is often visible on the surface at the temple; it divides at the level of the upper margin of the auricle into an often noticeably tortuous frontal branch and a parietal branch (Fig. 1.1-2).
The maxillary artery is the most important branch for the supply of the external parts of the head (Figs. 1.1-2, 1.1-3). After its origin (the mandibular part), it passes anterior or posterior to the lateral pterygoid muscle (pterygoid part) into the depth of the infratemporal fossa (pterygopalatine part), where the pterygoid venous plexus is also located. From it, branches pass to the teeth in both jaws, to the masticatory muscles, to the gingival and nasal mucosa, and—via the medial meningeal artery, which passes through the foramen spinosum—to the dura mater in the middle cranial fossa and part of the anterior cranial fossa.
Fig. 1.1–1 Variants in the branching of the common carotid artery (60%, according to Lippert and Pabst 1985).
Fig. 1.1–2 Overview of the branches of the external carotid artery.
Fig. 1.1–3 Course of the maxillary artery (after Lippert and Pabst 1985).
1.1.2 Clinical picture (carotid artery, vertebral artery)
Cerebral ischemic events are the most frequent cause of stroke (80–85% of cases) and currently represent the third most frequent cause of death in the industrialized countries. Approximately 50% of the patients have significant mental or physical disturbances after a stroke. Stenoses of the internal carotid artery are the cause of nearly one-third of all cases of ischemic stroke. In stenoses of the internal carotid artery, nearly 80% of the clinical symptoms arise due to cerebral emboli from arteriosclerotic plaques. The remaining 20% of strokes arise due to hemodynamic compromise of cerebral circulation.
In approximately one-quarter of all cases of ischemic stroke, there is involvement of the posterior or vertebrobasilar circulation. Stenoses of the vertebral artery are the cause of up to 20% of all cases of ischemic stroke in the posterior flow region. Proximal stenoses of the vertebral artery are the second most frequent cause of ischemic stroke after stenoses of the internal carotid artery. As in the pathogenesis of internal carotid artery stenoses, stroke in the posterior flow region most often results from emboli from arteriosclerotic plaques. Hemodynamic compromise must be regarded as comparatively rare in the posterior flow region, as the basilar artery is supplied by two vertebral arteries. In contrast to the internal carotid artery, the vertebral artery also has many collateral vessels that are able to compensate cerebral perfusion when there is proximal constriction of a vertebral artery.
Good familiarity with the extracranial and intracranial anatomy is decisive for treatment of the relevant vascular occlusion processes. The left common carotid artery arises from the aortic arch, and the right one from the bifurcation of the brachiocephalic trunk. The common carotid artery does not have any side branches. At the level of the upper edge of the thyroid cartilage, it divides into the internal and external carotid arteries. The external carotid artery supplies the temporomandibular joint, the face, the neck, and the meninges. It has two terminal branches—the superficial temporal artery and the maxillary artery. These two branches may also serve, along with the occipital artery, as collaterals if there is occlusion of the internal carotid artery or vertebral artery. The internal carotid artery ascends laterally behind the hypopharynx, where it can be palpated. It branches into the anterior and medial cerebral arteries. It has no visible side branches in its extracranial course up to the branching point. The first major intracranial branch is the ophthalmic artery. The vertebral artery arises from the subclavian artery and travels to the sixth cervical vertebra, where it passes through the transverse foramen in the transverse process of the sixth vertebra and courses in a sharply cranial direction in the same way through the corresponding foramina of cervical vertebrae C5 to C1. In this segment, the vertebral artery courses more or less parallel to the carotid artery. At C1, the vertebral artery turns posteriorly and arches around the posterior part of the vertebral arch. It then passes through the foramen magnum into the cranial cavity. At the lower margin of the pons, the left and right vertebral arteries unite to form the basilar artery, which in turn flows into the circle of Willis. Although generally the vertebral arteries make only a minor contribution to the overall cerebral circulation, they can be of significant importance when occlusion or stenosis of the internal carotid artery develops. The circle of Willis is the most important intracranial collateral pathway. Via the anterior and posterior communicating arteries, the anterior, middle, and posterior cerebral arteries are connected in a circuit. In situations in which a proximal atherosclerotic process progresses slowly, the circle of Willis is able to compensate for vascular occlusions. However, the circle is not fully developed in all individuals. In patients with an incomplete circle of Willis, even brief occlusion of a vessel can lead to a stroke.
1.1.3 Clinical findings
The clinical symptoms of stenosis of the internal carotid artery or vertebral artery are characterized by ischemia in the corresponding region of cerebral flow. Symptomatic stenosis of the internal carotid artery is defined by syndromes in the ipsilateral hemisphere during the previous 6 months:
Transitory ischemic attack (TIA) = focal neurological deficit for less than 24 h
Amaurosis fugax = monocular amaurosis for less than 24 h
Stroke = focal neurological deficit for more than 24 h
Symptomatic stenosis of the vertebral artery can become manifest in the form of:
Syncope
Vertigo
Ataxia
Ipsilateral TIA or stroke in the posterior region of flow
1.1.4 Differential diagnosis
Etiological differential diagnosis of ischemic stroke:
Macroangiopathy of the supra-aortic vessels:
– Atherosclerotic
– Non-atherosclerotic (dissection, fibromuscular dysplasia, giant cell arteritis, Takayasu syndrome, bacterial or viral vasculitides)
Cerebral microangiopathy:
– Atherosclerotic
– Non-atherosclerotic (vasculitides)
Sources of cardiac emboli:
– Atrial fibrillation, atrial myxoma, atrioseptal aneurysm, acute myocardial infarction, endocarditis, status post-cardiac valve replacement, dilated cardiomyopathy
Paradoxical embolism:
– Atrial septal defect, persistent oval foramen, in deep vein thrombosis in the lower extremity or pelvis, thrombophilia
Atherosclerosis of the aortic arch
Coagulation disturbances:
– Genetic—e.g., in antithrombin III deficiency, protein S/protein C deficiency, activated protein C (APC) resistance, factor V Leiden mutation
– Acquired in disseminated intravascular coagulation—e.g., in multiple trauma, sepsis
Hematological diseases:
– e.g., polycythemia, hemoglobinopathies, iron-deficiency anemia, leukemia, thrombocythemia
Differential diagnosis of acute focal neurological deficit:
Intracerebral hemorrhage
Subarachnoid hemorrhage
Sinus thrombosis/cerebral venous thrombosis
Migraine with aura
Postictal hemiparesis (Todd paralysis)
In cases of suspected stenosis of the vertebral artery with no evidence of cerebral ischemia in the posterior region of flow, all of the differential diagnoses of syncope, vertigo, or ataxia also need to be considered.
1.1.5 Diagnosis
Carotid stenosis is diagnosed using color duplex ultrasonography, which is widely available and is low-cost. In addition to assessment of the grade of stenosis, the method also makes it possible to identify the plaque morphology (e.g., with an ulcerated surface).
Patients who should undergo duplex evaluation include those who have suffered a focal neurological deficit or amaurosis fugax. Cerebral computed tomography (CT) or magnetic resonance imaging (MRI) can also help exclude other causes of neurological symptoms, such as hemorrhage or tumor. In asymptomatic patients, there are no standard diagnostic recommendations, except when a bypass operation is planned. In this special situation, a duplex ultrasound examination is recommended in patients ≥ 65 years, patients with stenosis of the left main coronary artery, peripheral arterial occlusive disease, nicotine abuse, or a history including cerebral ischemia. Duplex ultrasound screening should also be carried out in asymptomatic patients who have a bruit over the internal carotid artery if the patient’s general condition allows surgical or interventional treatment. When duplex ultrasound results are unclear, the diagnostic accuracy can be increased using supplementary computed-tomographic angiography or magnetic resonance angiography.
In the diagnosis of extracranial vertebral artery stenosis, a noninvasive duplex ultrasound examination is also the procedure of choice. More than 80% of vertebral arteries can be detected using duplex ultrasonography. Digital subtraction angiography is still the “gold standard” for assessing stenosis of the vertebral artery. However, it is associated with a certain rate of morbidity and mortality, albeit low.
1.1.5.1 Doppler/duplex ultrasonography
Examination technique
The examination is best carried out with the examiner positioned behind the patient’s head, as this allows both hands to be used and functional tests can be carried out more easily. The examination is carried out at two levels using a high-frequency (≥ 7.5 MHz) linear probe, including the proximal vascular segments near the aortic arch (to detect any stenoses at the orifices for the common carotid artery, brachiocephalic trunk, subclavian artery, and vertebral artery). Supraclavicular and jugular insonation should be carried out, preferably using a microconvex probe (≥ 7.5 MHz). If this is not available, adapted programming of a low-frequency convex probe can be used, as well as insonation of carotid segments near the base of the skull. Although the bifurcation region is the most frequent site for pathological carotid processes, it is not the only one.
Examination of the carotid system
In addition to demonstration of any extracranial pathology, attention should be given to the Doppler velocities, as an indirect sign of proximal or distal pathologies. Side-to-side Vmean differences in the internal carotid artery showing ≥ 10% pulsatility differences or bilateral pulsatility changes should be assessed as follows:
Increased pulsatility:
– Distal circulatory pathway obstruction (stenosis, occlusion)—possibly unilateral in such cases
– Cerebral microangiopathy (Binswanger disease)
– Raised intracranial pressure (edema, bleeding)
– Bradycardia
– Aortic insufficiency
Reduced pulsatility:
– Proximal circulatory pathway obstruction (prolonged acceleration time)
– Distal arteriovenous malformation or angioma (normal acceleration time)
– Aortic stenosis, reduced ejection fraction
Transcranial Doppler/duplex ultrasonography should be carried out in cases of cerebral pathology with evidence of hemodynamically relevant extracranial processes. If it is not available, insonation of the supratrochlear artery with compression of the facial artery at the mandibular angle and of the superficial temporal artery in front of the tragus should be carried out (preferably at the same time). The findings at the supratrochlear artery reflect the cerebral hemodynamics along with the quality of intracerebral collateralization and can be predictive for watershed infarction.
Flow differences in the supratrochlear artery greater than 1:2 ratio may suggest:
Proximal internal carotid artery stenosis on the side with lower flow
External carotid artery stenosis on the side with higher flow
Distal intracranial internal carotid artery stenosis after the origin of the ophthalmic artery or higher-grade T-fork/middle cerebral artery main trunk stenosis on the side with higher flow
False-negative compression tests may occur due to:
Common carotid artery processes with similar compromise of flow in the internal and external carotid arteries
Simultaneous internal carotid and external carotid processes
Hypoplasia or aplasia of the supratrochlear artery
False-positive compression tests may occur due to:
Wobbling: the most frequent error
Anatomic variants (rare)
Examination of the subclavian/vertebral artery system
Demonstration of the orifice of the vertebral artery is obligatory, as this is the most frequent site for arteriosclerotic stenoses. Identification of vessels at the orifice using undulation at the mastoid is also necessary. Demonstration of the distal course (V2) allows better visualization of hypoplasia and aplasia, occlusions and dissections. The vertebral arteries are of equal size in only one-quarter of cases; when there are differences in diameter between the two sides, differences in pulsatility and amplitude between the two vertebral arteries are often seen and the Doppler velocities often differ due to an absent or hypoplastic connection between the vertebral and basilar arteries. It is incorrect to conclude from a bilaterally identical Dopplers of the distal vertebral artery that conditions are normal, although this frequently happens.
In cases of proximal occlusion, there is often distal collateral filling in the region of the atlas via occipital branches or other collaterals to the external carotid artery—and insonation should be carried out there as well, at least in these cases.
Differential diagnosis
Arteriosclerosis:
Typical plaque morphology
Classically localized at the carotid bifurcation and at the orifices of the vertebral artery, subclavian artery, and brachiocephalic trunk
Embolism:
Hypoechoic to isoechoic occlusion material
Otherwise unremarkable vascular system with no major atheroma
Arterial thrombosis:
In addition to the hypoechoic occlusion, evidence of prior arteriosclerosis with sonographically complex plaques and structures creating acoustic shadows, possibly “older” echogenic occlusion components
Involvement particularly of proximal vascular segments and vascular segments near the aortic arch and of branches of the external carotid artery (superficial temporal artery)
Dissection:
There may initially be a floating dissection membrane.
The true lumen may be alternately compromised as far as the occlusion—possibly with a sharply tapering occlusion pattern (the string sign).
The false lumen may thrombose.
When there is a floating membrane, there may be a pathognomonic triphasic “splash signal.”
Circumscribed mural hematomas may occur (“carotidynia”).
Variants:
– The vertebral artery in particular often shows hypoplasia and also aplasia.
– Definition of vertebral artery hypoplasia: vascular diameter < 2.5 mm or diameter ratio > 1:1.7.
Large-vessel vasculitis:
– Typical homogeneous, hypoechoic and widened intima–media complex (macaroni sign)
Caution: there are often respiratory-modulated buckling stenoses in the subclavian artery, particularly on the left side, which regress on inspiration.
Table 1.1–1 The reaction pattern in the supratrochlear artery during compression of the external branches.
| Increased flow | Evidence of normal orthograde flow |
| Reverse flow | Evidence of pathological retrograde flow |
| Reduced flow | Evidence of pathological retrograde flow; intracranial perfusion pressure probably poorer than with reverse flow |
| No reaction | If there are no notable external–internal carotid artery anastomoses or maxillary/ethmoidal artery collaterals → contralateral compression, and then alternatively a compression test in the supraorbital region → an increase in outflow resistance leads to reduced flow velocities, particularly diastolic → pulsatility increases |
| Zero flow, with evidence of flow only during compression | Pathological → equalized extracranial/intracranial pressure |
Fig. 1.1–4 Carotidynia: hypoechoic focal wall thickening, representing mural hematoma.
Specific findings
Common carotid artery
The width of the boundary zone reflection (synonymous with the intima–media thickness, IMT) in the common carotid artery (CCA, and other vessels) has been found to be a parameter for assessing the atherosclerotic risk, and it correlates with the incidence of vascular events.
Fig. 1.1–5a, b Dissection of the internal carotid artery. (a) A visible dissection membrane in the area of the common carotid artery. (b) The typically altered polyphasic pulsed-wave Doppler signal.
There is no established classification based on maximum velocities for grades of stenosis in common carotid artery stenoses. In addition to the general characteristics of a stenosis (see under internal carotid artery stenoses), the peak velocity ratio (PVR) can be used.
Internal carotid artery
Various angiographic grading methods are available, with differing percentage figures. The residual diameter may refer either to the distal diameter of the internal carotid artery, with the local stenosis grade based on the North American Symptomatic Carotid Endarterectomy Trial (NASCET) criteria; or to the original proximal diameter, with the local stenosis grade based on the European Carotid Surgery Trial (ECST) criteria. An approximate formal conversion can be carried out:
Local stenosis grade (based on ECST): ECST % = 0.6 × NASCET % + 40%
Distal stenosis grade (based on NASCET): NASCET % = (ECST – 40%)/0.6
The traditional duplex ultrasound criteria used by the German Society for Ultrasound in Medicine (Deutsche Gesellschaft für Ultraschall in der Medizin, DEGUM) correlated with the ECST local stenosis grade (although the NASCET method was mainly used in radiography). The DEGUM criteria were revised in 2010 to establish comparability and transferred to NASCET (Table 1.1-3). The method used for duplex ultrasound classification must be clearly stated.
N.B.: Confusion may arise when the old and new stenosis grades for the internal carotid artery are compared or used in parallel. Stenoses in other locations (external carotid artery, vertebral artery, common carotid artery, etc.) are generally continuing to be classified according to the local stenosis grade on the basis of previously customary hemodynamic criteria. Supra-aortic stenoses with the same local stenosis grade in other vascular territories may therefore be given as “higher” percentages than internal carotid artery stenoses, which are classified according to the distal stenosis grade.
Fig. 1.1–6 NASCET–ECST: the principles of stenosis grading based on NASCET and ECST (B flow imaging).
Table 1.1–2 Age-dependent and gender-dependent normal values for intima–media thickness (IMT) in the common carotid artery (CCA), showing means, standard deviations, and 95% confidence intervals (Temelkova-Kurktschiev et al. 2001).
In addition to hemodynamic criteria, new B-image optimization techniques and digital subtraction ultrasonography (B flow) provide effective assistance in demonstrating the morphology. Bifurcation stenoses in particular are often missed or incorrectly classified in angiographic procedures.
Cross-sectional planimetry correlates more with the local stenosis grade and is associated with error due to vascular remodeling of the internal carotid artery. A compensatory increase in the terminal diameter of the vessel occurs with increasing grades of stenosis, so that the percentage grade of stenosis is incorrectly raised. These values therefore need to be treated with caution.
Fig. 1.1–7a, b Bifurcation stenosis of the internal carotid artery. (a) Imaging of a high-grade, eccentric internal carotid artery stenosis with apparently hypoechoic plaque material. (b) The bifurcation stenotic process is detected on the B-image/B-flow.
Table 1.1–3 Stenosis grading in the internal carotid artery based on NASCET (adapted from Arning et al. 2010).
Notes on criteria 1–10 (see text for further explanations): stenosis grade based on NASCET (%): the figures given refer in each case to a 10% range (± 5%). Criterion 2: Evidence of low-grade stenosis (local aliasing effect) distinct from nonstenotic plaque, demonstration of flow direction in moderate to highgrade stenoses and evidence of vascular occlusion. Criterion 3: The criteria apply to stenoses with a length of 1–2 cm and only to a limited extent to processes affecting multiple vessels. Criterion 4: Measurement well distally, outside of the zone with jet stream and flow disturbances. Criterion 5: Possibly only one of the collateral connections may be affected: if only an extracranial examination is carried out, the value of the findings is lower. Criterion 9: Confetti sign can only be recognized when the pulse repetition frequency (PRF) is set low. ACA, anterior cerebral artery; ICA, internal carotid artery; CCA, common carotid artery.
Fig. 1.1–8 Internal carotid artery planimetry. The principle of planimetric stenosis grading. The limitations due to a compensatory increase in the external diameter should be noted.
Additional information on plaque status and on the prognostic assessment is particularly desirable for treatment decision-making in patients with asymptomatic stenoses. The aim is to detect stenoses that are associated with an increased risk of embolism, since at this stage it is usually an embolic source rather than a hemodynamically compromising structure that is removed (otherwise there is a very high number needed to treat in therapy for asymptomatic stenoses). Prognostic significance has not been conclusively evaluated for all of the parameters.
Plaque echogenicity: hypoechoic plaques are prognostic for a 4–5-fold increase in the risk of stroke (Mathiesen et al. 2001).
Plaque perfusion: the presence and extent of neovascularization of a plaque—demonstrated using ultrasound contrast enhancement with a low mechanical index (MI) technique—correlates positively with the risk of stroke and the general rate of cardiovascular events.
Rate of spontaneous cerebral embolism (high-intensity transitory signals, HITS; see section A 1.2): for example, when there is evidence of HITS in asymptomatic 60% internal carotid artery stenosis, there is an approximately 15-fold increase in the risk of stroke in comparison with negative HITS (Spence et al. 2005).
Intracerebral collateralization/autoregulation reserve/CO2 reactivity (see section A 1.2)—limited autoregulation reserve/CO2 reactivity in intracerebral vessels correlates with hemodynamically caused watershed infarction.
There is a risk of overestimating stenoses in the internal carotid artery:
When there is contralateral internal carotid artery occlusion or very high-grade stenosis with collateral function in the ipsilateral ICA (only with collateralization via the anterior communicating branch (transcranial Doppler examination required)
When there are collaterals in the ICA via the posterior communicating branch in the flow area of the posterior cerebral circulation
In cases of distal arteriovenous malformation
Possibly in cases of elongation and kinking at the site of the stenosis
Measurement of the relevant parameters during cardiac irregularities—e.g., after extrasystoles with a compensatory pause, or in absolute arrhythmia after a longer RR interval, or with an increased ejection volume in cases of aortic insufficiency or marked bradycardia
There is a risk of underestimating stenoses in the internal carotid artery:
In cases of tandem stenoses of the internal carotid artery or highgrade flow obstructions in the area of the carotid “T” or main trunk of the middle cerebral artery
With hyperventilation (intracerebral vasoconstriction)
In cases of marked cerebral microangiopathy or raised intracranial pressure with disturbed distal outflow
When measurements are carried out during hemodynamically compromising tachycardic heart action
In cases of upstream high-grade flow obstructions
Occlusion can be differentiated from “pseudo-occlusion” or filiform stenosis by:
Low flow–optimized setting of the device (PRF low, line thickness increased, etc.)
Increasing the color enhancement appropriately (caution: artifacts)
Using alternative color scales (mixed mode), color procedures (power mode), and imaging modes (B flow, color B flow)
Optimizing the transducer head position and beam position
Using ultrasound contrast enhancement
Including the intracranial findings
Check-up examinations postoperatively or after stent angioplasty in the internal carotid artery:
In both procedures: residual stenosis, recurrent stenosis, intimal hyperplasia in the internal carotid artery, intraluminal thrombi, flow conditions at the external carotid artery orifice?
After surgery: intimal flap, proximal/distal step formation, aneurysmal dilation in the patch area (TEA)
After internal carotid artery stenting: is the stent well opposed or is there flow behind it; is the stenotic area completely covered?
In-stent restenosis:
Slightly altered velocity criteria have been described (by Armstrong et al. 2007, among others), partly due to reduced wall compliance in the stent area and loss of the luminal dilation (“bulb”) often (but not always) located in the internal carotid artery orifice area.
External carotid artery
There is no established classification of stenosis grades for external carotid artery stenoses based on maximum velocities. In addition to general stenotic characteristics (see under internal carotid artery stenoses), the peak velocity ratio can be used.
The differential diagnosis should include distal arteriovenous malformations in the afferent areas for branches of the external carotid artery, as well as dural fistulas; flow acceleration/aliasing is then seen over long stretches and not only focally in the “stenotic area.”
Table 1.1–4 Stenosis grades after stent angioplasty in the internal carotid artery.
| > 70% | > 50% | |
| PSV (cm/s) | > 350 | > 225 |
| ICA–CCA ratio | > 4.75 | > 2.5 |
CCA, common carotid artery; ICA, internal carotid artery; PSV, peak systolic velocity.
Fig. 1.1–9 An aneurysm in the internal carotid artery. A typical “coffeebean” fragmented color pattern in an internal carotid artery aneurysm, with relative stenosis in the outflow area.
Fig. 1.1–10 Stenosis of the external carotid artery. A typical undulation phenomenon is seen in the external carotid artery, with evidence of moderate stenosis.
In cases of occlusion of the external carotid artery, side branches are often collateralized via the corresponding external carotid artery branches on the contralateral side or branches of the thyrocervical trunk or costocervical trunk.
Subclavian artery
See section A 4.1.
Vertebral artery
For findings involving the subclavian steal effect/syndrome, see section A 4.1.
There is no established classification of stenosis grades based on maximum velocities for vertebral artery stenoses. The criteria mentioned in connection with the internal carotid artery can be used, but it should be noted that minor flow disturbances in the obtusely angled orifice area of the vertebral arteries must be regarded as physiological. The velocities given should be lower, as the physiological maximum systolic flow velocity is approximately 60–100 cm/s. Values above 100 cm/s must be regarded as suspicious. The critical velocity after which the presence of a > 50% stenosis must be assumed is 120 cm/s (Baumgartner et al. 1999). The peak velocity ratio may also be used.
Differential diagnoses in cases of flow acceleration in the vertebral artery, with a risk of overestimating stenoses:
Inadequate perfusion of the contralateral vertebral artery (e.g., due to aplasia, hypoplasia, stenosis, occlusion, dissection) with hyperperfusion of the artery being examined
Hyperperfusion of the vertebral arteries when collaterals via the posterior communicating branch are compensating for obstructive carotid processes
Arteriovenous malformations in the vertebrobasilar flow area
Measurement of the relevant parameters, e.g., after extrasystoles, with a compensatory pause, or in cases of absolute arrhythmia after a longer RR interval, or with an increased ejection volume in patients with aortic insufficiency or marked bradycardia
Differential diagnoses in cases of flow deceleration in the vertebral artery, with a risk of underestimating stenoses:
Distal vertebrobasilar flow obstruction (high pulsatility, and reduced diastolic flow in particular)
Hypoplasia or dilatory arteriopathy in the vertebral artery
Marked cerebral microangiopathy or raised intracranial pressure with disturbed distal outflow (with identical bilateral findings)
When measurements are carried out during hemodynamically compromising tachycardic heart action
In cases of proximal high-grade flow obstructions
Findings in occlusions of the extracranial vertebral artery:
Vessel not identifiable (in continuous-wave mode, occlusion may only be suspected, however)
Possible hyperperfusion of the contralateral vessel
Arterial lumen definitely demonstrated on duplex ultrasound, without PW Doppler or color signals
Demonstration of distal inflow collaterals possible
Findings in occlusions of the intracranial vertebral artery:
Markedly increased extracranial pulsatility, with reduced diastolic flow in one vertebral artery that may amount to pendular flow (caution: outflow into a dilated PICA may lead to a normal Doppler signal when there is a distal vertebral artery occlusion)
Possible hyperperfusion of the contralateral artery
Course of the vertebral artery not capable of being imaged transnuchally
Vertebral artery compression syndrome (often suspected, but actually a rare finding) characterized by:
Rotatory vertigo and nystagmus
Capable of being reproducibly triggered by head rotation
Can be stopped by a return movement
Clear hypoplasia in one vertebral artery (duplex ultrasound evidence)
Movement-dependent, hemodynamically compromising compression of a normal-lumen vertebral artery (Doppler and duplex ultrasound evidence possible), so that with relevant hypoplasia of a vertebral artery and suspicious clinical findings, an attempt should be made to provoke symptoms during Doppler/duplex ultrasound examination of various segments of the normal-lumen vertebral artery
1.1.6 Treatment
1.1.6.1 Conservative treatment
Medical treatment is indicated in both internal carotid artery stenosis and vertebral artery stenosis, in order to limit atherosclerotic progression and reduce the risk of a neurological event. This treatment recommendation is independent of the decision on whether to offer interventional or surgical revascularization therapy. Treatments currently available include inhibition of platelets using acetylsalicylic acid (ASA), dipyridamole plus acetylsalicylic acid, or clopidogrel. In addition, treatment with statins is advised due to their anti-inflammatory and thus plaque-stabilizing effect in other vascular territories. Medical treatment alone is recommended in patients with stenosis of the internal carotid artery who either have a low risk of stroke (symptomatic stenoses < 50%, asymptomatic stenoses < 60%) or who are at high perioperative or peri-interventional risk due to comorbid conditions, or who have a limited life expectancy.
Drug treatment was the only form of therapy available for cerebral ischemia in the posterior flow region prior to the advent of endovascular approaches. Unfortunately, there is still a lack of data from randomized studies comparing drug treatment with surgical or interventional therapy for extracranial stenoses of the vertebral artery. The results of the Vertebral Artery Stenting Trial (VAST), which is currently still recruiting, will probably be able to provide important information. At present, 180 patients with symptomatic vertebral artery stenosis > 50% have been randomly assigned either to endovascular treatment (stent implantation) or to the conservatively treated group in the study.
1.1.6.2 Endovascular treatment
Patient preparation
General patient history
Medication and allergy history
Complete neurological evaluation, plus National Institutes of Health Stroke Scale (NIHSS)
Cranial CT or MRI examination
Duplex ultrasonography to exclude fresh thrombus formation
ASA 100–300 mg/day and clopidogrel 75 mg/day, starting at least 5 days before a planned intervention, or bolus administration (ASA 500 mg, clopidogrel 600 mg) on the day before the procedure
Peri-interventional therapy
Heparin (70–100 IU/kg) with an activated clotting time (ACT) of 250–300 seconds
Electrocardiographic (ECG) monitoring due to potential bradycardia
Blood pressure monitoring for possible hypotension related to carotid sinus stimulation by balloon inflation
Intravenous administration of 1 mg atropine 2–3 min before implantation of the carotid stent, to prevent possible bradycardia or asystole (to be used with caution in patients with narrow-angle glaucoma)
Infusions for marked or prolonged bradycardia/hypotension
Technique of carotid artery stenting (CAS)
Access route
It is important to establish a safe vascular access route in order to minimize complications during carotid stent implantation, and access via the femoral artery is the approach most often employed. The common femoral artery is punctured using a Seldinger needle, and a 12-cm long 5–6F sheath is placed. This initial sheath is then exchanged during the procedure for a 90-cm long sheath (e.g., Cook Shuttle sheath.). If a guiding catheter is to be used, a 12-cm long 8–9F sheath is needed. In patients in whom the pelvic arteries are occluded or who have high-grade stenosis, or in situations in which the access route via the femoral artery is unavailable for other reasons, access via the brachial or radial artery is obtained (Fig. 1.1-11). The right brachial artery is preferable for interventions on both the right internal carotid artery and the left internal carotid artery. If neither access route is possible, direct cervical common carotid access (percutaneous or open surgical) can be considered.
Engaging the common carotid artery
Angiography of the aortic arch is generally performed prior to any selective carotid angiography in order to identify possible difficult anatomic conditions that might make it necessary to exchange the typically employed diagnostic catheters (e.g., Berenstein, Judkins Right, Head Hunter, IMA, JB-1) for an alternative one (e.g., Simmons or Vitek catheter) (Fig. 1.1-12). To engage the common carotid artery, the 5F diagnostic catheter (e.g., Berenstein, Right Judkins, Head Hunter, IMA) is positioned over a 0.035-inch hydrophilic guidewire in the ascending aorta with the catheter tip pointing downward. This technique reduces the likelihood of embolization from aortic plaque or traumatic injury to the intima of the aortic arch and prevents the catheter from becoming caught in a vascular ostium. As soon as the catheter reaches the ascending aorta, it is rotated 180°. This places the tip of the catheter in a vertical, upright position on fluoroscopy. The catheter is then carefully withdrawn until it slides into the brachiocephalic trunk, and the hydrophilic wire is then advanced into the right common carotid artery and the catheter is advanced over this wire into the common carotid artery. To intubate the left common carotid artery, the catheter is slowly withdrawn from the ostium of the brachiocephalic trunk. It should be rotated 20° counterclockwise, so that the catheter tip points slightly anteriorly. When the aortic arch becomes unwound with advancing age, the origin of the left common carotid artery is located slightly further posterior. In these cases, it may be necessary to rotate the catheter posteriorly instead.
Fig. 1.1–11a, b Access via the brachial artery.
Fig. 1.1–12a, b (a) Simple anatomic conditions. (b) More difficult anatomic conditions. The catheter is capable of prolapsing into the ascending aorta.
Once the left common carotid artery has been entered, the catheter should be rotated back 20° clockwise, so that the tip is once again pointing vertically or slightly posteriorly. The catheter position is checked by administering a small amount of contrast. This can exclude subintimal contrast flow or reduced blood flow. The hydrophilic wire is advanced to the distal common carotid artery, followed by the catheter.
Exploring the common carotid artery in difficult anatomy
If engagement of the common carotid artery is unsuccessful with the standard catheter, then a switch to a Simmons catheter is usually made. This type of catheter has a large reverse curve which must be re-shaped in the aorta after wire removal, usually in the ascending aorta. Moving the catheter backward slightly guides the tip into the brachiocephalic trunk, then into the left common carotid artery, and finally into the left subclavian artery. In contrast, Vitek or Mani catheters have smaller pre-formed curves and do not require shaping so these catheters are advanced from, rather than withdrawn toward, the distal aortic arch selecting the left subclavian artery first, and so on. Once the desired vessel is selected, the wire is advanced followed by the catheter. Advancement of the catheter is carried out slowly and with assistance from the pulsating blood flow. At the same time, the wire is withdrawn slightly, so that its position is maintained proximal to the carotid bifurcation. Advancement of the catheter and withdrawal of the guidewire are carried out alternately several times until the catheter is safely positioned in the vessel (push-and-pull technique) (Fig. 1.1-13).
Fig. 1.1–13a, b The push-and-pull technique.
Visualizing the vessels
Injections of contrast medium should be carried out manually or with a small amount of automated contrast administration (a maximum of 6 mL per injection). Larger amounts would lead to mixing of the arterial, intermediate, and venous phases, potentially leading to masking of early venous filling or other types of pathology. Some operators carry out four-vessel angiography to show the status of the collateral arteries. However, as this represents an additional procedural risk, the need for it is questionable, particularly in patients in whom magnetic resonance angiography has previously been carried out. During balloon dilation, the absence of collaterals may cause short periods of cerebral ischemia due to brief occlusion of the internal carotid artery. However, this reaction is reversible after deflation of the balloon and has no influence on the completion of the procedure. Once the anatomy of the target vessel has been defined, a hydrophilic guidewire is advanced into the external carotid artery so that the diagnostic catheter can be exchanged for a sheath or guide sheath. Bony landmarks can be used for guidance instead of road mapping to mark the origin of the external carotid artery during wire placement.
Vascular kinking
If the vessel is very tortuous, it can be straightened using a wire. It is also helpful to ask the patient to inhale deeply and hold the breath. An acute vessel angle can be negotiated by careful rotation and advancement of the catheter until it has reached the desired position (Fig. 1.1-14). If it is still not possible to advance the catheter, a Simmons III catheter should be used to introduce the guidewire into the external carotid artery. The Simmons III catheter can then be exchanged for a 4F multipurpose catheter. After this, the hydrophilic wire is exchanged for a 0.035-inch Amplatz wire or a softer wire. Finally, the 4F catheter is exchanged for a 5F catheter.
Fig. 1.1–14a-c (a) The technique of probing the left common carotid artery (CCA) and placing the guidewire in the external carotid artery (ECA) in the presence of a common trunk. (b) Placement of an IMA diagnostic catheter at the origin of the CCA. (c) Probing the ECA with a 0.035” wire with antifriction coating (Terumo) and advancing the diagnostic catheter with slight right–left rotation as far as the ECA. Replacement with a more rigid uncoated wire (e.g., Supracor).
Placement of the guiding catheter
An 8F guiding catheter (e.g., a right coronary guiding catheter) is introduced into the ascending aorta via a hydrophilic 0.035-inch wire. In cases of difficult or abnormal anatomy, aortography of the aortic arch can be used to assist in selective exploration. Following angiography of the aortic arch and assessment of the anatomy, the guiding catheter is introduced into the common carotid artery. This should be followed by careful aspiration and flushing with saline to clear any possible atherosclerotic particles out of the catheter.
Placement of the long sheath
Engagement of the common carotid artery is carried out with a 5F diagnostic catheter and access to the external carotid artery obtained with an angled hydrophilic guidewire, and the diagnostic catheter introduced into the external carotid artery as described above. The wire is exchanged for a 0.035-inch wire, typically a stiff Amplatz wire. The diagnostic catheter is removed and a 6F 90-cm sheath placed using an over-the-wire technique into the common carotid artery below the bifurcation. The sheath should be handled very carefully, as trauma to the common carotid artery ostium or release of atherosclerotic deposits may occur leading to neurologic sequelae. The sheath should be meticulously aspirated and flushed to eliminate possible air or atherosclerotic debris.
Carotid access in occluded external carotid artery or common carotid artery stenosis
When the external carotid artery is occluded, or there is significant stenosis below the bifurcation, or a stenosis at the ostium of the common carotid artery, placing the 6F 90-cm sheath in the common carotid artery may represent a considerable challenge. If possible, crossing the stenosis with a stiff wire should be avoided, as this may dislodge necrotic plaque material and cause distal embolization. If necessary, the 5F diagnostic catheter is advanced over a 0.035- or 0.038-inch guidewire for placement further distally, slightly proximal to the stenosis. It can then be exchanged over a 0.035-inch Amplatz wire (extra stiff). If there is an ostial/proximal stenosis of the common carotid artery, it may be necessary to treat this stenosis first in order to obtain access to the distal stenosis. However, if this stenosis is not severe, the bifurcation stenosis should be treated first and then the proximal stenosis on the “way back.”
Predilation
Some operators predilate the stenosis using a small angioplasty balloon and a short inflation time of 5–10 seconds. This provides for better passage and positioning of the stent. The present authors would only recommend predilation if primary stent implantation has failed. In our view, primary stent implantation has a protective effect against distal embolization by fixing deposits on the vascular wall.
Protection against emboli
The possibility of procedural cerebral embolization is an important concern in carotid angioplasty. Balloon dilation, stent implantation, and manipulation of the vessels by the catheter and wire can easily release emboli, which if large enough can in turn cause severe cerebral damage. For this reason, emboli protection systems are routinely used in most centers. There are currently three different underlying principles on which protection against cerebral embolism is based: filter systems, distal occlusion balloons, and proximal occlusion balloons.
Distal occlusion balloons
Distal occlusion balloons (Fig. 1.1-15a) were the first embolism protection systems to become available, and were widely used in the initial carotid stent experience. It consists of a 0.014-inch guidewire with an occlusion balloon in the distal section, which is inflated and deflated through a very small channel in the guiding catheter (Guardwire® Temporary Occlusion and Aspiration System, Medtronic Vascular; TriActiv® ProGuard™ Embolic Protection System, Kensey Nash). After the guiding catheter is placed, the occlusion balloon is positioned distal to the stenosis and the balloon inflated until blood flow into the internal carotid artery stops. Stent implantation then follows. After the intervention, an aspiration catheter is introduced up to the occlusion balloon, and the blood in the occluded artery is aspirated. Any particles released during the intervention are thus removed. The advantages of the distal occlusion system are its low profile (2.2F), flexibility and good steerability. Disadvantages include the fact that balloon occlusion is not tolerated in 6–10% of patients, and that the vascular segment distal to the occlusion balloon cannot be imaged using contrast during the balloon occlusion procedure.
Filter systems
Most filter systems (Fig. 1.1-15b) consist of a metal framework that is covered with a polyethylene membrane or a nitinol mesh. The pore size can vary between 80 and 200 μm in diameter depending on the specific device. Filters are usually attached to the distal section of a 0.014-inch guidewire. In its closed state, the filter is sheathed by an introducer catheter, and it is introduced into the vascular segment distal to the stenosis. Once the stenosis has been crossed, the filter is opened by withdrawing the outer catheter. Following stent implantation, the filter is closed by withdrawing it into a recovery catheter, and then removed from the vessel.
Fig. 1.1–15a-c Embolism protection system.
A wide range of second-generation and third-generation filter systems are currently available. The technical characteristics of a good filter consist of a low profile (< 3F), adequate steerability for maneuvering through highly tortuous vessels, and—when the filter is opened—good apposition to the vessel wall to allow the best possible protection against emboli.
Proximal occlusion systems
All distal protection systems, occlusion balloons and filters have the potential disadvantage that the stenosis has to be crossed before the system can be deployed and protection established. This unavoidable step carries a risk of distal embolization during the initial unprotected phase of the procedure. Proximal protection systems (Fig. 1.1-15c), such as the Gore Neuro Protection System (Gore) and the MO.MA System (Invatec), provide protection against cerebral embolism even before crossing the stenosis. This is particularly important in the case of stenosis with fresh thrombi where embolization with a distally placed system may be problematic. The use of a proximal protection system allows the operator to use any wire to negotiate difficult stenoses. These two systems consist of a long main sheath with a balloon on its distal end that is inflated in the common carotid artery to occlude forward carotid flow. A second balloon, which is inflated in the external carotid artery, prevents retrograde external flow, thus establishing complete arrest of antegrade flow into the internal carotid artery. The principle of proximal embolic protection systems takes advantage of the cerebral collateral system of the circle of Willis (Fig. 1.1-16). Following balloon occlusion of the external and common carotid artery, collateral flow via the circle of Willis produces what is known as reserve pressure. This prevents antegrade flow into the internal carotid artery. After stent implantation and before deflation of the occlusion balloon, blood in the internal carotid artery, which might contain released particles, is aspirated and removed. One disadvantage of the proximal protection system is that a small percentage of patients are unable to tolerate balloon occlusion due to incomplete intracranial collateralization.
Fig. 1.1–16 The principle used in the proximal protection system (e.g., Gore Neuro Protection System), with reversal of flow in the internal carotid artery and continuous diversion of arterial blood via the protection system (with femoral venous return). This requires an intact anterior circle of Willis or other collateral support.
Stent implantation
Self-expanding stents are usually implanted in carotid stenting. Balloon-expandable stents are recommended in ostial stenoses of the common carotid artery, stenoses located in the distal internal carotid artery, and sometimes in severely calcified stenoses. The disadvantages of balloon-expandable stents are the repeated balloon dilations that are needed to implant the stent adequately, and stent compression that can occur during the long-term follow-up in areas vulnerable to external manipulation.
In vessels with the potential to bend or be manipulated, self-expanding nitinol stents are the best choice. They are designed to adapt to the shape of the vessel and therefore have little tendency to straighten it (Fig. 1.1-17). Stent-induced straightening of the vessel can give rise to a new stenosis distal to the stent due to kinking or folding of the vessel. Stents with a strong radial force are recommended for treatment of severely calcified stenoses. “Closed-cell” carotid stents usually have stronger radial force. Their cell structure may also provide better plaque coverage, which may be theoretically advantageous in stenoses with a high embolic risk. The clinical value of “open-cell” vs. “closed-cell” designs and the importance of the stent cell size is currently still unclear.
Fig. 1.1–17a, b (a) Elongation of the internal carotid artery, with vascular kinking distal to the stenosis before stent implantation. (b) After implantation of a nitinol stent.
The authors recommend a stent diameter 1–2 mm larger than the largest vascular diameter to be stented. Carotid stents with a diameter of 6–8 mm are usually used if the stent is being implanted exclusively in the internal carotid artery, or with a diameter of 8–10 mm if the stent is to cross the bifurcation. Stenting across the external carotid artery is not a problem and priority should be given to the stent covering the entire stenosis, which in most cases will mean crossing the bifurcation to cover the distal common carotid artery plaque. While there are no data suggesting stent length is a determinant of restenosis in the carotid, a stent 20–30 mm longer is usually selected for discrete lesions, while for tandem stenoses, 40-mm stents are recommended.
Postdilation
Postdilation is usually carried out using a balloon with a diameter of ~5 mm, matched but not larger than the diameter of the internal carotid artery, but not referenced to the common carotid artery. A balloon with an unnecessarily large diameter might force particles through the stent cells and cause distal embolization. To prevent dissections, postdilation should be carried out at nominal pressure, and within the stent borders. A residual stenosis ≤ 30% is acceptable, since an adequate blood flow is established and the potentially emboligenic atherosclerotic deposits are compressed sufficiently to induce neointimal formation and eliminate the embolic potential of the lesion. The stent expands further during the following few hours. If contrast-enhancing ulcerations occur outside the stent edge, they do not need to be obliterated and can be left without any untoward effects. Postdilation of the stent segment in the common carotid artery is not necessary. If significant stenosis or occlusion of the external carotid artery develops following postdilation, it does not require treatment.
Following postdilation of the stent, angiography of the carotid arteries and intracranial vessels is carried out. Imaging of the intracerebral vessels should always include the venous phase, to allow objective comparison of conditions before and after stent implantation. For assessment of the intracerebral vessels and in preparation for possible intracranial emergency intervention in case of cerebral embolism, angiography should be carried out with a lateral and anteroposterior 30° cranial (Towne) projection.
Fig. 1.1–18a, b (a) Outlet stenosis of the left vertebral artery before stent implantation. (b) After implantation of an expandable balloon stent.
Technique of vertebral artery stent implantation
The vascular access route for vertebral artery interventions is the same as for carotid artery stent implantation, via the femoral or brachial artery. A contralateral oblique projection is best for demonstrating the ostium of the vertebral artery. The intracranial vertebrobasilar vascular system is best demonstrated in lateral and steep anteroposterior projections. A multipurpose 6F guiding catheter is suitable for the vertebral artery procedure. Balloon-expandable coronary stents should be used for ostial stenoses of the vertebral artery, and self-expanding stents can be used in extracranial vertebral artery stenoses that are located distal to the ostium in the body of the vessel. A 4-mm coronary balloon is usually used for dilation (Fig. 1.1-18). The embolic protection systems currently available cannot usually be recommended for vertebral artery procedures, as they still have relatively large diameters for this application.
Postinterventional follow-up and medication
Following the intervention, blood pressure has to be closely checked for at least 6 h. A neurological evaluation including the National Institutes of Health Stroke Scale (NIHSS) must be carried out before the patient is discharged. Lifelong ASA treatment (100 mg) and clopidogrel during the first month are recommended.
Clinical results
Clinical series/carotid stent implantation registry
A summary of the results in 12,392 carotid stent implantations in a total of 11,243 patients from 53 centers worldwide was published by Wholey et al. in 2003. Complications during the first 30 days included: TIA (3.1%), minor stroke (2.1%), major stroke (1.2%), and death (0.6%) (Wholey et al. 2003). In 2001, Roubin et al. published a series of 528 patients who had undergone carotid stent implantation. The major stroke rate was 1% (n = 6) and the minor stroke rate was 4.8% (n = 29). Overall, the rate of stroke/death after 30 days was 7.4% (Roubin et al. 2001).
In 2003, Cremonesi et al. published a series of 442 consecutive patients treated with carotid stent implantation with embolic protection. Stroke or death within the first month after the procedure occurred in 1.1% of these patients (Cremonesi et al. 2003).
Biasi et al. reported on the use of the echogenicity index, known as the gray scale median, as an indicator of the risk of stroke during carotid stent implantation. The authors concluded that low echogenicity in the carotid plaque, measured as a gray scale median ≤ 25, increased the risk of peri-interventional stroke (Biasi et al. 2004). The German Association for Angiology and Radiology has developed a prospective registry for carotid stent implantations. The results for the first 48 months, from a total of 38 participating centers, were published in 2004. Carotid stent implantation was carried out in 3267 patients. The procedure was successful in 98% of the interventions. The peri-interventional mortality was 0.6%, the major stroke rate was 1.2%, and the minor stroke rate was 1.3% (Theiss et al. 2004).
In 2005, Bosiers et al. published the ELOCAS Registry, compiled retrospectively and prospectively from the results of four “highvolume centers.” A total of 2172 consecutive patients were treated, and 99.7% of the procedures were technically successful. The stroke/death rate was 4.1% after 1 year, 10.1% after 3 years, and 15.5% after 5 years (Bosiers et al. 2005).
The CaRESS study was a nonrandomized multicenter study including 143 patients treated with carotid stent implantation and 254 patients who underwent carotid endarterectomy. No significant differences were observed with regard to the stroke/death rates either after 30 days (2.1% stent, 3.6% surgery) or after 1 year (10.0% stent, 13.6% surgery) (CaRESS 2005).
The ARCHeR study was published by Gray et al. in 2006 and consisted of three sequential multicenter studies. In ARCHeR 1, only the use of the Acculink (Guidant) carotid stent was evaluated; in the two subsequent studies (ARCHeR 2 and 3), adjuvant use of the Accunet embolic protection system (Guidant) was also tested. A total of 581 patients with high surgical risk from 48 centers were included between 2000 and 2003. The combined stroke/death/myocardial infarction rate was 8.3% after 30 days. The ipsilateral stroke rate after the first month and up to 1 year was 1.8%. The repeat stenosis rate was 2.2% within the first year (Gray et al. 2006).
The CAPTURE Registry (Carotid Acculink/Accunet Post-Approval Trial to Uncover Unanticipated or Rare Events) was published in 2007. A total of 3500 patients with high surgical risk and a stenosis grade > 50% (symptomatic) or > 80% (asymptomatic) were included. The stroke/death/myocardial infarction rate was 6.3% after 30 days. The major stroke/death rate after 30 days was 2.9%.
Embolism protection systems have been evaluated in several large, unrandomized multicenter studies. These include the European DESERVE study (Diffusion-Weighted MRI-Based Evaluation of the Effectiveness of the Mo.Ma System), which demonstrated that the Mo.Ma system, using diffusion-weighted magnetic resonance imaging (DW-MRI) during carotid stent implantation, is effective for protection against embolism. The cranial MRI examination was carried out before the procedure and 3–12 hours after carotid stent implantation in order to identify new ischemic lesions. In a total of 127 patients treated, new cerebral lesions were identified on DWMRI in 38 (29.9%). However, clinically relevant stroke was only present in three patients (2.4%) (Rubino et al., paper presented at EuroPCR 2011).
In the EMPIRE Study, the Gore Flow Reversal System was evaluated in 29 centers with a total of 245 patients. The primary end point was the rate of myocardial infarction/stroke/death after 30 days. Thirty-two percent of the patients were treated for symptomatic carotid stenosis. Intolerance for balloon occlusion was noted in a total of six patients in the study (2.4%). The primary end point occurred in 3.7% of the patients (Clair et al 2011).
In the EMBOLDEN Study, the Gore Embolic Filter was evaluated in a total of 250 patients in 35 centers in 2009–2010. Fifteen percent of the patients included in the study were suffering from symptomatic carotid stenosis. After 30 days, 10 patients (4.0%) reached the primary end point, defined as myocardial infarction, stroke, and/or death (Gray et al., paper presented at the 23rd Annual International Symposium on Endovascular Therapy, 2011).
Randomized studies on carotid stent implantation
The Carotid and Vertebral Transluminal Angioplasty Study (CAVA-TAS I) was the first large-scale study in which carotid angioplasty (largely without stenting and embolic protection) was compared with carotid endarterectomy. The early stroke/death rate was 10% in both the carotid angioplasty and carotid endarterectomy groups. In the SAPPHIRE Study, carotid stent implantation using embolic protection was compared with carotid endarterectomy both in a randomized study and in a registry. The 30-day stroke/death rate was 4.5% in the group with randomized carotid stent implantations and 6.6% in the group of randomized patients who underwent surgical treatment. The registry also included patients who did not meet the inclusion criteria for the randomized study. The 30-day stroke/death rate in the registry among patients who underwent interventional treatment was 6.9%.
The EVA-3S study compared carotid stent implantation with carotid endarterectomy in patients with symptomatic stenoses ≥ 60%. The participation criteria for interventionalists were set very low; a participating interventionalist only had to have implanted 12 carotid stents beforehand. The stroke/death rate after 30 days was lower in the group of patients treated with embolic protection than in the group without embolic protection (18 of 227, 7.9% versus 5 of 20, 25%; P = 0.03). Randomization to the group with stent implantation without embolic protection was therefore prematurely stopped. The stroke/death rate after 30 days in the group with stent implantations with embolic protection was significantly higher (9.6%) in comparison with endarterectomy (3.9%). The relative risk was 2.5.
The SPACE (Stent-Protected Percutaneous Angioplasty of the Carotid versus Endarterectomy) study included patients with symptomatic stenoses ≥ 70% on duplex ultrasonography or ≥ 50% according to the NASCET criteria. The use of embolic protection systems was optional in this study. Many interventional centers had difficulties in meeting the participation criteria—at least 25 carotid stent implantations had to have been carried out previously. The 30-day ipsilateral stroke rate was 6.84% in the group of patients treated with carotid stent implantation, and in the surgical group, the rate was 6.34%. Although using embolic protection systems had already become a standard part of the procedure in most centers, 73% of the interventions in this study were carried out without embolic protection. Complications such as myocardial infarction, contralateral stroke, and cranial nerve lesions, which are almost exclusively complications of endarterectomy, were not evaluated. The study was designed as a non-inferiority study with a non-inferiority margin of < 2.5% which was based on a calculated complication rate of 5%. The one-sided P value for non-inferiority was 0.09. The study was terminated prematurely after inclusion of 1200 patients due to low patient recruitment and lack of finance when an analysis determined that including more than 2500 patients would have been necessary to reach a statistical power of at least 80%. According to the investigating physicians in the study, the study results were not capable of definitively demonstrating the non-inferiority of carotid stent implantation in comparison with carotid endarterectomy since the trial was not completed.
Overall, the results of the large randomized studies indicate that the differences between surgery and intervention were usually very small.
The Carotid Revascularization Endarterectomy vs. Stenting Trial (CREST) study started recruiting patients in 2000. Preliminary data from the introductory phase of the interventional arm of the study were published in 2004. The data so far indicate that the periprocedural risk increases significantly with increasing age. The stroke/death rate was 1.7% in the patient cohort under the age of 60; 1.3% in patients aged 60–69; 5.3% in those aged 70–79; and 12.1% in those aged 80 or over. The significant difference was independent of the patients’ neurological status, grade of stenosis, or use of embolic protection systems.
Studies for which recruitment has already started include the International Carotid Stenting Study (ICSS) and the Asymptomatic Carotid Stenosis, Stenting versus Endarterectomy Trial (ACT-1). Studies already in the planning stage that are expected to provide new findings on the success of treatments for asymptomatic stenoses—e.g., with regard to neurocognitive function—include the Transatlantic Asymptomatic Carotid Intervention Trial (TACIT) and the Asymptomatic Carotid Surgery Trial 2 (ACST-2).
The Carotid Revascularization Endarterectomy vs. Stenting Trial (CREST), including a total of 2502 recruited patients, is the largest randomized study so far carried out to compare carotid stent implantation with carotid surgery. Data from the lead-in phase of the interventional arm of the study were published in 2004 (Hobson et al. 2004). In that report, the rates of stroke/death in the patient cohort were 1.7% in those under the age of 60, 1.3% in those aged 60–69, 5.3% in those aged 70–79, and 12.1% in patients aged 80 or over. The significant difference was independent of the patients’ neurological status, the grade of stenosis, or the use of embolism protection systems. The full study results were published in 2010. No significant difference was seen between the two treatment methods with regard to the combined primary end point of periprocedural stroke, myocardial infarction or death, or ipsilateral stroke during the follow-up period. The event rates were 7.2% in the group of patients who were treated with stent implantation and 6.8% in the patients who underwent surgery (Brott et al. 2010). Studies still currently recruiting include the International Carotid Stenting Study (ICSS), the Asymptomatic Carotid Stenosis Stenting Versus Endarterectomy Trial (ACT-1), and the Asymptomatic Carotid Surgery Trial 2 (ACT-2).
An interim analysis of the ICSS study has been published for the first 120 days of follow-up. Patients with > 50% symptomatic carotid stenosis were randomly assigned either to endovascular or surgical treatment methods. It was not yet possible to analyze the study’s primary end point (the 3-year stroke rate). Instead, the results of the analysis focused on the 120-day rate of stroke, death, or procedure-related myocardial infarction. In a total of 1713 randomized patients, a significantly higher rate was seen after carotid stent implantation at 8.5%, in comparison with 5.2% after carotid surgery (Ederle et al. 2010).
Clinical results—vertebral artery stent implantation
Sundt et al. first reported successful treatment of the vertebrobasilar system using angioplasty in 1980 (Sundt et al. 1980). Since then, many clinical series have been published on vertebral artery angioplasty/stent implantation, with high rates of technical success (98–100%) (Malek et al. 1999; Piotin et al. 2000; Mukherjee et al. 2001). However, there is still a lack of data from large clinical series and from controlled randomized studies to allow more precise assessments of the complication and restenosis rates during long-term follow-up. A subanalysis of the CAVATAS study compared a small group of 16 patients with symptomatic vertebral artery stenosis who received either endovascular treatment (PTA or stent implantation) or conservative drug therapy. Two of the eight patients who received endovascular treatment suffered a periprocedural TIA. No cases of stroke in the vertebrobasilar flow area occurred in either treatment arm during the long-term follow-up (mean 4.7 years). Three patients each in the drug-treated arm of the study and also in the group with endovascular treatment died due to myocardial infarction or stroke in the carotid flow area (Coward et al. 2007). In the VAST study, currently still in progress, 180 patients with symptomatic vertebral artery stenosis > 50% have been randomly assigned to groups receiving either endovascular treatment (stent implantation) or conservative therapy. The planned follow-up period is one year (Compter et al. 2008).
Prospects
The rapid developments in interventional treatment for carotid and vertebral artery stenoses, such as new stents with greater flexibility and smaller diameters, improved embolic protection systems with secure apposition on the vascular wall, and other technical innovations, are anticipated to improve the results of carotid and vertebral artery stent implantation in the near future.
1.1.6.3 Surgical treatment
Surgical removal of atherosclerotic obstruction in the carotid artery is the most frequently conducted vascular operation worldwide. In terms of the criteria of evidence-based medicine, it is also the best-studied surgical intervention that exists, with tens of thousands of patients documented in international prospective, randomized studies. Carotid endarterectomy (CEA) was first reported by Eastcott and colleagues in the treatment of a patient who had had 33 transitory ischemic attacks (TIAs) (Eastcott et al. 1954). The operation is now carried out in Germany, for example, over 25,000 times per year (BQS-Bundesauswertung 2007).
Indications for surgery
CEA is able to reduce the risk of stroke sevenfold in patients with TIAs, and in patients with 60–90% asymptomatic stenoses of the internal carotid it can achieve an absolute risk reduction of 5% over 5 years (Biller et al. 1998). The highest incidence of perioperative stroke is associated with the presence of high-grade bilateral internal carotid artery stenosis.
On the basis of large prospective, randomized multicenter studies (Anon. 1995; Anon. 1998; Ferguson et al. 1999), the American Heart Association formulated the following generally accepted guidelines for establishing the indication for carotid endarterectomy in an interdisciplinary consensus taking into account the natural history without surgery and thus the maximum acceptable perioperative rates of stroke and mortality (Biller et al. 1998):
Asymptomatic stenoses of the internal carotid artery with a stenosis grade > 60%, with a combined perioperative stroke and mortality rate (major adverse cardiovascular event rate, MACE) of < 3% (evidence level 1b; recommendation grade A)
Asymptomatic internal carotid artery stenosis > 60% and contralateral stenosis > 75% or occlusion, with a MACE of < 5% (evidence level 4; recommendation grade C)
Symptomatic internal carotid artery stenoses > 50%, with a complication rate of < 6% (evidence level 1a; recommendation grade A)
The following groups of patients are particularly able to benefit from the operation:
Those with hemispheric TIAs
Those with crescendo TIAs—i.e., with the number and/or length of transitory ischemic attacks continually increasing
Those with progressive stroke—i.e., patients with low-grade symptoms initially who show marked clinical deterioration within 6 h
Those with stroke during the previous few weeks, as the risk of recurrent cerebral ischemia is particularly high in these cases
Those with a subtotally occluded internal carotid artery but persistent residual flow (pseudo-occlusion)
CEA should be carried out without delay in these cases (Eckstein et al. 2004).
Despite the lack of level 1a evidence, differential treatment consideration for carotid artery stenting (CAS) appears to be justified, above all in the presence of a “high carotid bifurcation” (bifurcation of the carotid artery higher than C2); for repeat operations in the neck; when there is paralysis of the contralateral recurrent laryngeal nerve; and after cervical radiotherapy, as it avoids local complications (e.g., nerve injuries).
Contraindications
Patients in poor general condition (American Society of Anesthesiologists IV, V) or with limited life expectancy (< 6 months)
Fresh large cerebral infarction (major stroke), with no tendency to show clinical improvement (< 4 weeks)
Following earlier disabling stroke (Rankin scale 5)
Patient preparation
Specialist neurological examination to determine prior neurological symptoms or establish the patient’s neurologic status
B-imaging and duplex ultrasonography to determine the level of the carotid bifurcation, show whether the internal carotid artery is patent, and assess the grade of stenosis
Cranial CT or MRI to identify older or more recent cerebral hemorrhage, infarct areas, and tumors
Angio-CT, angio-MRI, or intra-arterial digital subtraction angiography only if there is a suspicion of supra-aortic multiple-vessel disease, poor definition of the stenosis on duplex ultrasound, contradictory duplex-ultrasound findings regarding the grade of stenosis, suspected intracranial tandem stenoses, or extreme kinking or coiling of the vessel with ambiguous duplex findings
Prophylaxis against thrombosis with low-molecular-weight heparin s.c. on the evening before the operation
Fasting for at least 6 h before the operation; long-term medication may be taken with a little water on the morning of the operation. Caution: antidiabetic agents, clopidogrel, Coumadin
Preparations for the operation
The operation can be carried out either with regional anesthesia or general anesthesia, although a recent prospective randomized multicenter study (the GALA study) showed some advantages for regional anesthesia (Lewis et al. 2009).
Positioning: modified beach-chair position—trunk raised, legs slightly lowered, head reclining/hyperextended and turned to the contralateral side, head positioned on a rubber ring, both arms juxtaposed (in regional anesthesia, the contralateral arm is laid free and a squeezing toy is placed in the patient’s hand, which has to be rhythmically squeezed by the patient when requested and produces a loud sound)
Sterile cleaning of the side of the neck that is being operated on to beyond the midline, laterally including the shoulder (acromion), caudally as far as the nipples, and cranially to include the mandible, chin, earlobe, and mastoid
Surgical access
The incision is made at the anterior margin of the sternocleidomastoid; preoperative assessment of the level of the carotid bifurcation can provide good guidance. The level of the palpable cricoid can also be used for guidance. The skin incision is usually approximately 7 cm long, and it is carried cranially toward the inferior margin of the earlobe. As far as possible, the incision should be made as little cranially or ventrally from the earlobe as possible, as injury to the oral branch of the facial nerve could occur, either directly or due to a subsequent retractor movement, leading to pareses in the ipsilateral corner of the mouth postoperatively.
After division of the skin, subcutaneous tissue, and platysma, the common carotid artery, which is usually easily palpated, is dissected and looped with a vascular sling. Following exposure of the common carotid artery, the internal carotid artery is exposed above the carotid bifurcation. Unnecessary manipulations of the vessel are avoided to avoid triggering embolizations (“no-touch” technique). During further dissection of the bulb in the direction of the internal carotid, the ansa cervicalis of the hypoglossal nerve is spared as much as possible. However, it may also be transected if necessary, usually without sequelae. Following it cranially leads to the hypoglossal nerve as it crosses the internal carotid artery. Particularly when there is a high carotid bifurcation, the hypoglossal nerve has to be mobilized, and the sternocleidomastoid branches of the occipital artery and vein attached to it have to be ligated. Circular exposure of the carotid bifurcation is only carried out after clamping of the internal carotid artery, in order to prevent embolization (Fig. 1.1-19a, b).
Surgical procedure
There are basically two procedures that can be used for plaque removal and reconstruction of the internal carotid artery or carotid bifurcation:
Thromboendarterectomy (TEA) with a patch graft
Eversion endarterectomy (EEA)
Fig. 1.1–19a, b (a) Exposure of the surgical site in the carotid triangle. The common carotid artery, internal carotid artery, external carotid artery from its first branch, and superior thyroid artery have been exposed. The facial vein has been transected. Lateral to the stenotic carotid artery lies the ansa cervicalis, which courses cranial to the horizontally crossing hypoglossal nerve. (b) Photograph of the original surgical site. The common carotid artery has been looped with a yellow vascular sling. A hypoplastic external carotid artery is branching off in the medial direction (top). The internal carotid artery is marked with a black arrow (<). The hypoglossal nerve (*) crosses the internal carotid artery.
Thromboendarterectomy (TEA)
In classic TEA of the internal carotid artery, following clamping and dissection of the carotid bifurcation (Fig. 1.1-20a), a longitudinal arteriotomy is carried out on the anterolateral wall of the carotid bifurcation and is extended into the common and internal carotid artery. Using a dissector, the endarterectomy is then carried out, and any distal intima layers that may be present are fixed with a single or continuous attachment suture (polydioxanone 7–0). If there is significant stenosis of the external carotid artery, an EEA can also be carried out. The disengaged plaque is further smoothly transected proximally, in the common artery, preferably with scissors. After copious rinsing of the endarterectomized vascular segment and checking for possible residual tissue and thrombus, an alloplastic patch (polyester, Dacron, or expanded polytetrafluoroethylene) is trimmed to size and sutured in with a continuous nonresorbable monofilament suture (polypropylene, Prolene 5–0) (Fig. 1.1-20b, c). In delicate vessels, a patch obtained from autologous great saphenous vein can also be used. Comparative studies have not shown that autologous venous patches confer any significant advantages in comparison with alloplastic materials in routine use (Bond et al. 2004; Naylor et al. 2004) (Fig. 1.1-20c).
When there is symptomatic coiling or kinking, shortening can be carried out either with segmental resection of the internal carotid artery or proximalization (transposition) of the origin of the internal carotid. The advantage of TEA is that it provides a good overview of the operative vascular segment and of the distal layer of the intima. This allows safe treatment, particularly for long stenoses. In addition, it is technically easier to introduce a temporary shunt. However, the procedure is more time-consuming and usually leaves foreign material (the patch) in the artery.
Fig. 1.1–20a-c (a) Positioning of clips on the carotid artery before the arteriotomy. (b) Longitudinally arteriotomized common and internal carotid artery. An intraluminal shunt has been introduced. A Dacron patch is being sewn in with a continuous suture. (c) The Dacron patch has been sewn in with a continuous suture.
Eversion endarterectomy (EEA)
In EEA, after clamping of the carotid bifurcation, transection of the internal carotid is carried out directly at its origin in the bifurcation. The correct dissection level is identified at the level of the external elastic lamina between the media and the intima before the internal carotid is rolled from inside out and the stenotic plaque is removed (Fig. 1.1-21a). At the distal end of the plaque, which usually terminates smoothly, the plaque cylinder normally breaks off without leaving a flap (Fig. 1.1-22). The internal carotid artery is rolled back into its original position and is then reattached with a resorbable monofilament continuous suture (polydioxanone 5–0) (Fig. 1.1-21b). To prepare a longer anastomosis, the arteriotomy can be extended cranially to a length of 15–30 mm in the internal carotid or proximally at the common carotid. Plaque removal of the bulb or of the common and external carotid artery can be carried out in the form of an open TEA or as an EEA at the external carotid artery. If a relevant distal intimal flap appears in the internal carotid, it can be secured with single fixation sutures stitched from the outside (tack sutures).
The technical advantage of EEA lies in the much shorter operating time required and the fact that there is no foreign material in the artery, provided resorbable suture material is used for reinsertion (Fig. 1.1-21c). As carotid stenoses often only appear in the bulb and in the area of the origin of the internal carotid, the limited extent of the vascular segment opened during eversion is usually adequate. EEA also makes it possible to correct symptomatic coiling or kinking of the internal carotid artery much more easily using a shortening operation than with TEA; either a segment of the internal carotid is resected, or a longer, proximalized anastomosis can be created if there is only slight coiling. The EEA procedure only becomes more difficult in longer lesions or in extracranial tandem stenoses. Introducing a temporary shunt is also slightly more complicated. In this case, it is best to start with the suture in the internal carotid artery and to introduce the shunt only after the side further away from the surgeon has been completed.
The current data show that there are no relevant differences between the TEA and EEA procedures with regard to postoperative mortality, morbidity, or recurrent stenoses (Cao et al. 2004; Crawford et al. 2007).
Intraoperative neuromonitoring and temporary intraluminal shunt
To prevent procedure-related cerebral ischemia during CEA, systolic blood pressure should be raised to > 150 mmHg if possible and systemic heparinization with 50–100 IU heparin/kg body weight is obligatory.
Fig. 1.1–21a-c (a) The surgical site in eversion endarterectomy of the internal carotid artery. The internal carotid artery has been transected obliquely at the common carotid artery. A dissection cylinder has been spatulated at the dissection level in the area of the external elastic membrane. The adventitia is now rolled up cranially until the cranial end of the dissected part tears spontaneously. (b) Reinsertion of the internal carotid artery into the common carotid artery using a continuous suture. (c) The surgical site after completion of the eversion endarterectomy.
Fig. 1.1–22 The dissection specimen after eversion endarterectomy of the internal carotid artery. The arrow tip (<) marks the residual lumen of the artery before disobliteration.
The most reliable method of intraoperative neuromonitoring involves assessment of changes in the state of consciousness and in motor activity in cooperative, conscious patients under regional anesthesia. These two checks can be carried out before and after clamping of the carotid artery by speaking to the patient and using rhythmic activation of a toy in the contralateral hand. If clouding of consciousness or loss of motor control immediately after clamping of the carotid artery occurs at high normal blood-pressure values, a temporary intraluminal or extraluminal shunt (e.g., a Javid or Pruitt shunt) should be placed. However, if the symptoms only occur after a latency period of several minutes, it is advisable—depending on the progress of the operation—either to complete the surgery quickly or to introduce a shunt. Optional shunt placement is needed in 5–10% of cases (BQS-Bundesauswertung 2007).
For placement of a temporary shunt in patients undergoing surgery with general anesthesia, there are basically two different approaches that have developed:
Obligatory shunt placement
Optional shunt placement after neuromonitoring
The advantages of general shunt placement have been reported to include in particular the safe maintenance of perfusion to the internal carotid artery, enlargement of the artery by the shunt, and educational and training considerations. Arguments in favor of optional shunt placement, by contrast, include the relatively infrequent need for shunting, the greater effort involved, and the risk of injury to or dissection of the distal internal carotid artery. However, the results with the two procedures in relation to perioperative mortality and morbidity are probably similar (Girn et al. 2008; Woodworth et al. 2007).
Methods of evaluating cerebral perfusion intraoperatively include measuring pressure in the stump of the internal carotid artery, assessment of somatosensory evoked potentials (SSEPs) or electroencephalography (EEG), and transcranial Doppler ultrasonography. However, there is controversy regarding the validity and practicality of these methods in the operating room. The use of such methods therefore depends on the preferences and experience of each surgical team.
Intraoperative quality control
Intraoperative quality control allows timely identification and correction of lesions capable of causing a cerebral insult. Such lesions include free-floating flaps > 2 mm and dissections with residual stenoses of more than 25%. Quality control can be carried out using the following methods:
Duplex ultrasonography
Continuous-wave (CW) Doppler
Angioscopy
Electromagnetic flowmetry
Angiography at two levels
Postoperative follow-up
Frequent checking of blood pressure and heart rate is needed. Normotensive blood-pressure values should be achieved before the patient is discharged from the hospital.
Attention should be given to ipsilateral headaches/hyperperfusion (hemicephalalgia).
A neurological status check by personnel managing postoperative care should be carried out to survey for defects.
The patient should be allowed to drink clear fluids immediately. A normal diet can be resumed postoperatively after 6 h (or 3 h with regional anesthesia).
The patient should be mobilized after 6 h—e.g., sitting in a chair or walking in the corridor (if this is possible in view of prior neurological deficits).
A postoperative neurological examination should be carried out.
Long-term inhibition of platelet function should be managed with aspirin (100 mg/d), plus clopidogrel (75 mg/d) for a period of 6 weeks.
Plaque stabilization should be achieved with statins.
Risk factors should be checked (hypertension, diabetes, etc.).
For the period of the hospital stay, low-molecular-weight heparin should also be administered for prophylaxis against thrombosis (medium risk or higher due to other diseases).
Duplex ultrasonography check-up examinations of the carotid vessels are performed before discharge, after 4 weeks, after 6 months, and then annually.
Clinical results
High-grade hemodynamically relevant recurrent stenoses > 70% occur much less frequently in the carotid artery (2%) than in other arteries such as the superficial femoral artery (BQS-Bundesauswertung 2007). If they develop within 1 year, they are most often caused by intimal hyperplasia, while luminal strictures occurring later are usually due to progression of the atherosclerosis. Since recurrent stenoses and those secondary to radiation injury are rarely embolic, invasive therapy is only required if they have hemodynamic effects. However, surgical correction and prior cervical surgery are associated with a much higher risk of neural injury (5%) due to adhesions in the wound area (Mozes 2005). Paralysis of the contralateral vocal cord significantly increases the perioperative morbidity. Carotid artery stenting (CAS) therefore offers theoretical advantages given the lack of cranial nerve injury with the percutaneous procedure.
False aneurysms are extremely rare surgical sequelae. They may be caused by incorrect surgical technique, such as an incorrect dissection level during endarterectomy, or by tearing or breakage of the patch suture. Very rarely, deep wound infection needs to be taken into consideration in aneurysms following alloplastic patch placement (with an incidence < 0.5%) (BQS-Bundesauswertung 2007). Treatment consists of complete removal of the patch and replacement with an autologous patch or bridging graft (from the great saphenous vein).
General complications
Death (< 1%)
Cardiovascular complications (decompensated cardiac insufficiency, severe cardiac dysrhythmia, cardiac infarction) (1.9%)
Perioperative stroke (2.2%); 5–10% of cases are caused by intracerebral hemorrhage and the remainder are ischemic (BQS-Bundesauswertung 2007)
Local complications
Postoperative hemorrhage requiring surgery (2.5%)
Peripheral nerve lesions, mainly temporary (hypoglossal nerve, facial nerve, recurrent laryngeal nerve) (1.5%)
Carotid occlusion (0.3%)
Postoperative wound infection (0.2%)
Hyperperfusion syndrome (hemicephalalgia) < 1% (BQS-Bundesauswertung 2007)
Internal carotid artery strictures can also be caused by fibromuscular dysplasia, in which constriction of the artery results from fibrous transformation of the media. Depending on the length of the vessel segment affected, the artery can be either patched or replaced. Segmental replacement of the carotid artery may be necessary in some cases for aneurysms, injuries, lesions resulting from prior radiation therapy, and after TEA with limited residual vascular wall. Autologous greater saphenous vein should be used for this purpose if possible. Replacement with a thin-walled polytetrafluoroethylene (PTFE) prosthesis is also possible in exceptional cases.
Prospects
CEA is currently the gold standard in the treatment of high-grade symptomatic and asymptomatic carotid stenoses. CAS has now entered clinical practice as a competitive procedure that is attractive for patients since it is less invasive, and may have some advantages over the established method of CEA in individual cases. While the trial data comparing the procedural safety of CAS to CEA are mixed due largely to probable methodological issues in individual trial conduct, there nevertheless is ample and growing data that long-term stroke prevention is equal between the two procedures out to at least 4 years and likely beyond. Several current studies nearing completion (the CREST, SPACE-2, and CAVATAS-2 trials) should provide a clearer picture as to the place CAS has in the management of carotid bifurcation disease. However, an indication for CAS may be considered after interdisciplinary consultation in individual cases in patients who are at high risk (e.g., with recurrent stenoses, radiogenic stenoses, contralateral recurrent nerve paralysis, etc.), or in the context of controlled studies.
References
ACCF/SCAI/SVMB/SIR/ASITN 2007 clinical expert consensus document on carotid stenting: a report of the American College of Cardiology Foundation Task Force on Clinical Expert Consensus Documents (ACCF/SCAI/SVMB/SIR/ASITN Clinical Expert Consensus Document Committee on Carotid Stenting). J Am Coll Cardiol 2007 Jan 2; 49 (1): 126–70. Erratum in: J Am Coll Cardiol 2007 Feb 27; 49 (8): 924.
Adams HP Jr, Bendixen BH, Kappelle LJ, et al. Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in Acute Stroke Treatment. Stroke 1993 Jan; 24 (1): 35–41.
Al-Mubarak N, Roubin GS, Vitek JJ, et al. Effect of the distal-balloon protection system on microembolization during carotid stenting. Circulation 2001; 104: 1999–2002.
American Heart Association. Heart Disease and Stroke Statistics—2004 Update. www.americanheart.org.
Anonymous. Carotid endarterectomy for patients with asymptomatic internal carotid artery. JAMA 1995; 273: 1421–8.
Anonymous. Randomised trial of endarterectomy for recently symptomatic carotid stenosis: final results of the MRC European Carotid Surgery Trial (ECST). Lancet 1998: 1379–87.
Armstrong PA, Bandyk DF, Johnson BL, Shames ML, Zwiebel BR, Back MR. Duplex scan surveillance after carotid angioplasty and stenting: a rational definition of stent stenosis. J Vasc Surg 2007 Sep; 46 (3): 460–5; discussion 465–6.
Arning C, Widder B, von Reutern GM, Stiegler H, Görtler M. [Revision of DEGUM ultrasound criteria for grading internal carotid artery stenoses and transfer to NASCET measurement]. Ultraschall Med 2010 Jun; 31 (3): 251–7. Epub 2010 Apr 22. Review. German.
Bamford J, Sandercock P, Dennis M, Burn J, Warlow C. Classification and natural history of clinically identifiable subtypes of cerebral infarction. Lancet 1991; 337: 1521–6.
Bartels E, Flugel KA. Advantages of color Doppler imaging for the evaluation of vertebral arteries. J Neuroimaging 1993; 3: 229–33.
Baumgartner RW, Mattle HP, Schroth G. Assessment of >/=50% and <50% intracranial stenoses by transcranial color-coded duplex sonography. Stroke 1999 Jan; 30 (1): 87–92.
Berteloot D, Leclerc X, Leys D, Krivosic R, Pruvo JP. Cerebral angiography: a study of complications in 450 consecutive procedures. J Radiol 1999; 80: 843–8.
Biasi G, Froio A, Diethrich E, et al. Carotid Plaque Echolucency Increases the Risk of Stroke in Carotid Stenting: The Imaging in Carotid Angioplasty and Risk of Stroke (ICAROS) Study. Circulation 2004; 110: 756–62.
Biller J, Feinberg WM, Castaldo JE, et al. Guidelines for carotid endarterectomy: a statement for healthcare professionals from a Special Writing Group of the Stroke Council, American Heart Association. Circulation 1998; 97 (5): 501–9.
Bogousslavsky J, van Melle G, Regli F. The Lausanne Stroke Registry: analysis of 1,000 consecutive patients with first stroke. Stroke 1988; 19: 1083–92.
Bond R, Rerkasem K, Naylor R, et al. Patches of different types for carotid patch angioplasty. Cochrane Database Systematic Reviews 2004 (2).
Bosiers M, Peeters P, Deloose K, et al. Does carotid artery stenting work on the long run: 5-year results in high-volume centers (ELOCAS Registry). J Cardiovasc Surg (Torino) 2005 Jun; 46 (3): 241–7.
BQS-Bundesauswertung 2006 Karotis-Rekonstruktion. BQS-Qualitätsreport 2006. Düsseldorf: 2007: 1–75.
Cao P, De Rango P, Cieri E, et al. Eversion versus conventional endarterectomy. Semin Vasc Surg 2004; 17(3): 236–42.
Caplan LR, Amarenco P, Rosengart A, Lafranchise EF, Teal PA, Belkin M, DeWitt LD, Pessin MS. Embolism from vertebral artery origin occlusive disease. Neurology 1992; 42: 1505–12.
Caplan LR. Brain embolism, revisited. Neurology 1993; 43: 1281–7.
Caplan LR. Posterior Circulation Disease: Clinical Findings, Diagnosis, and Management. Cambridge MA, Blackwell Scientific Publishers: 1996.
CaRESS Steering Committee Carotid Revascularization Using Endarterectomy or Stenting Systems (CaRESS) phase I clinical trial: 1-year results J Vasc Surg 2005; 42: 213–9.
Clair DG, Hopkins LN, Mehta M, Kasirajan K, Schermerhorn M, Schönholz C, Kwolek CJ, Eskandari MK, Powell RJ, Ansel GM; EMPiRE Clinical Study Investigators. Neuroprotection during carotid artery stenting using the GORE flow reversal system: 30-day outcomes in the EMPiRE Clinical Study. Catheter Cardiovasc Interv 2011 Feb 15; 77 (3):420–9. doi: 10.1002/ccd.22789. Epub 2010 Nov 3.
Crawford RS, Chung TK, Hodgman T, et al. Restenosis after eversion vs patch closure carotid endarterectomy. J Vasc Surg 2007; 46 (1): 41–8.
Cremonesi A, Manetti R, Setacci F, et al. Protected carotid stenting: clinical advantages and complications of embolic protection devices in 442 consecutive patients. Stroke 2003 Aug; 34 (8): 1936–41.
Eastcott HH, Pickering GW, Rob CG. Reconstruction of internal carotid artery in a patient with intermittent attacks of hemiplegia. Lancet 1954; 267 (6846): 994–6.
Eckstein H-H, Heider P, Wolf O. Chirurgische Therapie extrakranieller Karotisstenosen. Schlaganfallprophylaxe auf höchstem Evidenzniveau. Deutsches Ärzteblatt 2004; 101 (41): A2753–62.
Endovascular versus surgical treatment in patients with carotid stenosis in the carotid and vertebral artery transluminal angioplasty study (CAVATAS): a randomized trial. Lancet 2001; 357: 1729–37.
Ferguson GG, Eliasziw M, Barr HW, et al. The North American Symptomatic Carotid Endarterectomy Trial: surgical results in 1415 patients. Stroke 1999; 30: 1751–8.
Girn HR, Dellagrammaticas D, Laughlan K, et al. Carotid endarterectomy: technical practices of surgeons participating in the GALA trial. Eur J Vasc Endovasc Surg 2008; 36 (4): 385–9.
Gray WA, Hopkins LN, Yadav S, et al. Protected carotid stenting in high-surgical-risk patients: the ARCHeR results. J Vasc Surg 2006 Aug; 44 (2): 258–68.
Gray WA , Yadav JS, Verta P, et al. The CAPTURE registry: predictors of outcomes in carotid artery stenting with embolic protection for high surgical risk patients in the early post-approval setting. Catheter Cardiovasc Interv 2007 Dec 1; 70 (7): 1025–33.
Hass WK, Fields WS, North RR, Kircheff II, Chase NE, Bauer RB. Joint study of extracranial arterial occlusion. II. Arteriography, techniques, sites, and complications. JAMA 1968; 203: 961–8.
Hobson RW 2nd, Howard VJ, Roubin GS, et al. Carotid artery stenting is associated with increased complications in octogenarians: 30-day stroke and death rates in the CREST lead-in phase. J Vasc Surg 2004 Dec; 40 (6): 1106–11.
Lewis SC, Warlow CP, Bodenham AR, et al. General anaesthesia versus local anaesthesia for carotid surgery (GALA): a multicentre, randomised controlled trial. Lancet 2009; 372 (9656): 2132–42.
Malek AM, Higashida RT, Phatouros CC, Lempert TE, Meyers PM, Gress DR, Dowd CF, Halbach VV. Treatment of posterior circulation ischemia with extracranial percutaneous balloon angioplasty and stent placement. Stroke 1999; 30: 2073–85.
Mas JL, Chatellier G, Beyssen B et al. Endarterectomy versus stenting in patients with symptomatic severe carotid stenosis. N Engl J Med 2006 Oct 19; 355 (16): 1660–71.
Mathiesen EB, Bønaa KH, Joakimsen O. Echolucent plaques are associated with high risk of ischemic cerebrovascular events in carotid stenosis: the tromsø study. Circulation 2001 May 1; 103 (17): 2171–5.
Mathiesen EB, Johnsen SH, Wilsgaard T, Bønaa KH, Løchen ML, Njølstad I. Carotid plaque area and intima-media thickness in prediction of first-ever ischemic stroke: a 10-year follow-up of 6584 men and women: the Tromsø Study. Stroke 2011 Apr; 42 (4): 972–8.
Mozes G. High-risk carotid endarterectomy. Semin Vasc Surg 2005; 18 (2): 61–8.
Mukherjee D, Roffi M, Kapadia SR, Bhatt DL, Bajzer C, Ziada KM, Kalahasti V, Hughes K, Yadav JS. Percutaneous intervention for symptomatic vertebral artery stenosis using coronary stents. J Invasive Cardiol 2001; 13: 363–6.
Naylor R, Hayes PD, Payne DA, et al. Randomized trial of vein versus dacron patching during carotid endarterectomy: long-term results. J Vasc Surg 2004; 39 (5): 985–93.
Piotin M, Spelle L, Martin JB, Weill A, Rancurel G, Ross IB, Rufenacht DA, Chiras J. Percutaneous transluminal angioplasty and stenting of the proximal vertebral artery for symptomatic stenosis. AJNR Am J Neuroradiol 2000; 21: 727–31.
Rabe K, Sugita J, Sievert H et al. Flow-Reversal Device for Cerebral Protection During Carotid Artery Stenting—Acute and Long-Term Results. J Interv Cardiol 2006 Feb; 19 (1): 55–62.
Roubin GS, New G, Iyer SS, et al. Immediate and late clinical outcomes of carotid artery stenting in patients with symptomatic and asymptomatic carotid artery stenosis: a 5 year prospective analysis. Circulation 2001; 103: 532–7.
SPACE Collaborative Group. 30 day results from the SPACE trial of stent-protected angioplasty versus carotid endarterectomy in symptomatic patients: a randomised non-inferiority trial. Lancet 2006 Oct 7; 368 (9543): 1239–47.
Spence JD, Tamayo A, Lownie SP, Ng WP, Ferguson GG. Absence of microemboli on transcranial Doppler identifies low-risk patients with asymptomatic carotid stenosis. Stroke 2005 Nov; 36 (11): 2373–8.
Spence JD, Pelz D, Veith FJ. Asymptomatic Carotid Stenosis: Identifying Patients at High Enough Risk to Warrant Endarterectomy or Stenting. Stroke 2011 Jul 28. [Epub ahead of print] No abstract available.
Sundt TM Jr, Smith HC, Campbell JK, et al. Transluminal angioplasty for basilar artery stenosis. Mayo Clin Proc 1980; 55: 673–80.
Temelkova-Kurktschiev T, Fischer S, Koehler C, Mennicken G, Henkel E, Hanefeld M. Intima-Media. Dicke bei Gesunden ohne Risikofaktoren. DMW 2001; 126: 1–5.
Theiss W, Hermanek P, Mathias K, et al. Pro-CAS: a prospective registry of carotid angioplasty and stenting. Stroke 2004 Sep; 35 (9): 2134–9.
Wholey MH, Al-Mubarek N. Updated review of the Global Carotid Artery Stent Registry. Catheter Cardiovasc Interv 2003 Oct; 60 (2): 259–66.
Wolf PA, Kannel WB, McGee PC. Epidemiology of strokes in North America. In: Barnet HJM, Stein BM, Mohr JP, Yatsu FM. Stroke: Pathophysiology, Diagnosis and Management Vol 1, New York: Churchill Livingstone, 1929 (1986).
Woodworth GF, McGirt MJ, Than KD, et al. Selective versus routine shunting during carotid endarterectomy: a multivariate outcome analysis. Neurosurgery 2007; 61 (6): 1170–6.
Yadav JS, Wholey MH, Kuntz RE, et al. Protected carotid-artery stenting versus endarterectomy in high-risk patients (the SAPPHIRE study). N Engl J Med 2004; 351: 1493–501.
1.2 Intracranial stenoses and occlusive processes
Basic anatomy of the intracranial arterial system: Reinhard Putz
Conservative treatment: Sophia Göricke, Marc Schlamann, Isabel Wanke
Doppler/duplex ultrasonography: Tom Schilling
Endovascular treatment: Sophia Göricke, Marc Schlamann, Isabel Wanke
Surgical treatment: Peter Horn
1.2.1 Basic anatomy of the intracranial arterial system
The interior of the cranium is supplied by two large paired arteries, in addition to several smaller afferents from the external carotid artery. The internal carotid artery, which flows without branches as far as the base of the skull, passes into the cranium through the carotid canal. Dorsally, the vertebral artery, after following a tortuous course in the vertebral artery groove, enters the vertebral canal through the atlanto-occipital membrane just under the foramen magnum. Four segments of the internal carotid artery are distinguished along its course: a cervical part (this is described in more detail in section A 1.1), a petrous part, a cavernous part, and a cerebral part (Fig. 1.2-1).
The petrous part has a diameter of around 5 mm, is approximately 3–4 cm long; accompanied by a venous plexus and an autonomic nerve plexus, it courses in the double-curved carotid canal without giving off any branches.
The cavernous part starts as the artery enters the cavernous sinus; this part of the artery also has a double S-curve, although it is highly variable. A few very small arterial branches are given off by the internal carotid artery in the sinus and pass to the trigeminal nerve ganglion, the pituitary gland, and the dura mater. Embedded in the spongy sinus, the artery is partly also enclosed by endothelium externally. Laterally, it is obliquely crossed by the abducent nerve.
Fig. 1.2–1 The parts of the internal carotid artery.
Fig. 1.2–2 The intracranial arterial supply.
The cerebral part starts as it passes through the diaphragma sellae (Fig. 1.2-2). From this point on, the wall structure of the internal carotid artery and of all its intracranial parts changes. The elastic fibroreticulate structures in particular are less developed. The first branch after the artery emerges from the cavernous sinus is the ophthalmic artery, which—covered by the optic nerve—passes obliquely forward to the base of the optic canal. It then gives off a few very small branches to the upper part of the pituitary gland and the anterior choroid artery, as well as giving off the anterior cerebral artery, and after a very short course the artery then passes into the middle cerebral artery.
The middle cerebral artery bends in its sphenoid part (the M1 segment) almost at a right angle into the lateral cistern. Lenticulostriate branches from the M1 segment pass through the anterior perforated substance to enter the base of the telencephalon and supply large parts of the telencephalic nuclei. The insular part (the M2 segment), located in the lateral fossa, gives off multiple branches over a tortuous course and supplies the adjoining parts of the frontal and temporal lobes (Fig. 1.2-3a).
Fig. 1.2–3 Arterial supply to the telencephalon. Left: lateral view of the left hemisphere; right: medial view of the left hemisphere.
Fig. 1.2–4 Parts of the vertebral artery.
The right and left anterior cerebral artery are connected through the usually very short, thin anterior communicating artery, which rests directly in front of the pituitary stalk on the skull base. From the precommunicating part (the A1 segment), fine branches pass to the adjoining parts of the frontal lobe and to the hypothalamus and thalamus. The postcommunicating part (the A2 segment) follows the contour of the corpus callosum as far as its splenium, and branches arising from it supply the medial surface of the cerebral hemispheres to above their superior margin—with the exception of the corpus callosum itself.
Four parts of the vertebral artery are also distinguished (Fig. 1.2-4). The first segment, the prevertebral part, is generally given off dorsally as the last branch from the subclavian artery. Infrequently it arises directly from the aortic arch. In its transverse part, it enters the transverse foramen of the C6 vertebra from the caudal direction in 90% of cases and runs upward in the series of foramina as far as the transverse foramen of the atlas. Between C1 and C2, it always forms a loop, which often extends far laterally—a sign of the mobility of the atlantoaxial joint. The short atlantic part bends sharply at the posterior arch of the atlas and embeds itself there into the vertebral artery groove, surrounded by a dense venous plexus and lying tightly on the suboccipital nerve. At the medial end of the groove, which is sometimes closed to form a canal, it bends ventrally and—after penetrating the atlanto-occipital membrane and spinal dura mater—it reaches the subarachnoid space as the intracranial part. From the vertebral artery, which has a diameter of approximately 2.5 mm intracranially, the basilar artery forms in the pontocerebellar cistern, resting on the clivus, after the posterior inferior cerebellar artery has been given off (Figs. 1.2-2, 1.2-5). It is originally paired during embryonic development, and this explains some rare variations. From it emerge—in addition to the anterior inferior cerebellar artery and the superior cerebellar artery—numerous branches to the brain stem and inner ear, as well as to the meninges. Just before reaching the dorsum sellae, the basilar artery finally divides into the two posterior cerebral arteries, which in turn are connected to the two middle meningeal arteries via the posterior communicating arteries.
Fig. 1.2–5 The arterial circle.
From the precommunicating part (the P1 segment), small branches enter the posterior part of the mesencephalon, parts of the hypothalamus and internal capsule, as well as the posterior part of the thalamus. Branches from the postcommunicating part (the P2 segment) pass to the anterior part of the mesencephalon and to the thalamus. Finally, the terminal branch of the artery (the P3 and P4 segments) supplies the medial surface of the parietal lobe and the inferior surface of the temporal lobe, as well as the posterior pole of the telencephalon (Fig. 1.2-3b).
The arterial circle lies centrally at the base of the skull or brain (Figs. 1.2-2, 1.2-5). It is formed bilaterally from the anterior cerebral artery, the trunk of the internal carotid artery, the middle cerebral artery, the posterior communicating artery, and the posterior cerebral artery. The ring is closed anteriorly by the anterior communicating artery. The sizes of the individual arteries involved differ widely even in normal conditions. The communicating arteries are usually very thin, and bilateral circulatory compensation is thus hardly possible. Almost every conceivable variant is also found, from a different origin of the cerebral arteries to a complete absence of individual arteries.
1.2.2 Clinical picture
In Europe, stroke is the third most frequent cause of early invalidity, after cardiovascular and malignant diseases. Ischemic stroke is the most frequent etiological and pathologic cause, representing 85% of cases. Arteriosclerosis in the intracranial vessels is the cause of approximately 8–10% of all cases of ischemic stroke. The annual risk of stroke in patients with intracranial stenosis of more than 50% who have already suffered stroke or a transient ischemic attack (TIA) lies in the range of 12–14% even with drug treatment. In high-risk patients (those with higher-grade stenoses > 70%, those with current symptoms, and women), the annual risk may be as high as up to 23%. Clinically asymptomatic stenoses are associated with a low annual risk of stroke (< 3.5%). Due to the narrow caliber of the intracranial vessels, the exact percentage of stenoses is difficult to assess. Strictures of approximately 50% of the vascular lumen may already lead to clinical symptoms. The findings of the Warfarin-Aspirin Symptomatic Intracranial Disease (WASID) study showed that stenosis greater than 70% markedly increased the relative risk of suffering a subsequent ischemic stroke. Three percent of patients with low-grade stenoses (50–69%) suffered a TIA within 1 year, in comparison with 14% of those with higher-grade stenoses (> 70%). Cases of manifest stroke within 1 year occurred in 8% of patients with low-grade stenoses and in more than 23% of those with high-grade stenoses. The most frequent locations for intracranial stenoses are the internal carotid artery at the level of the siphon, the main trunk of the middle cerebral artery, the distal vertebral artery and the middle segment of the basilar artery. The significance of tandem stenoses and multiple intracranial stenoses is as yet unclear, but there is considerable evidence to suggest that each individual stenosis carries the corresponding risk of infarction and that the risk is then increased.
Intracranial vascular occlusions often arise due to emboli that originate in the heart or aortic arch or from stenotic cervical arteries. Rarer causes include thrombotic occlusions of existing stenoses. The site of an intracranial occlusion is often influenced by wall changes and stenoses that are already present.
1.2.2.1 Clinical symptoms
Vascular occlusion can lead to loss of neurological function (hemiplegia, speech disturbance, visual disturbance, ataxia, and unconsciousness), but it can also produce nonspecific symptoms such as headache and vertigo. The clinical symptoms depend on the capacity of the intracranial collateral circulation and the extent and location of the occluded vascular segment.
Ischemic stroke is classified according to its temporal course. If the symptoms completely resolve within 24 hours, it is called a transient ischemic attack (TIA). Otherwise it represents ischemic stroke syndrome. However, magnetic resonance imaging shows diffusion disturbances as the correlate of cerebral ischemia in up to 50% of cases that are clinically classified as TIAs. Additional classifications into prolonged reversible ischemic neurological deficit (PRIND) and reversible ischemic neurological deficit (RIND) have been abandoned.
1.2.3 Differential diagnosis
Intracranial stenoses and occlusions need not always be caused by arteriosclerosis. Other causes may include dissection, vasculitis, vasculopathy (e.g., fibromuscular dysplasia, moyamoya), vasospasm and vascular compression by extravascular space-occupying lesions.
Dissection is the most frequent cause of ischemic stroke in young adults. However, it is not always possible to confirm intracranial dissection beyond doubt using imaging diagnosis, as intramural hematoma is difficult to confirm in these small vessels. This remains a diagnosis of exclusion, when there is no arteriosclerosis or no other risk factor for cerebral ischemia are present. Sites of predilection are the distal vertebral artery and main trunk of the middle cerebral artery.
In cases of vasculitis, magnetic resonance imaging usually shows multiple smaller subcortical ischemic areas in the white matter in both cerebral hemispheres. In some cases, typical irregularities in the vessel wall may already be seen on imaging (MRA, CTA, DSA), with stenotic and dilated segments. The final diagnosis is established by evidence of antibodies in serum and cerebrospinal fluid; a cerebral vessel biopsy is sometimes necessary. Treatment is determined by the underlying disease and may involve anti-inflammatory therapy or even immunosuppressive therapy.
One example of vasculopathy is moyamoya disease (moyamoya is Japanese for “foggy” or “smoky”), in which extremely fine, complex collateral networks form in the area of the perforating arteries when major cerebral arteries are occluded (see the section on surgical therapy below). Inflammatory diseases (e.g., meningitis) or neoplasia (e.g., meningioma at the base of the skull) may also lead to narrowing and occlusion of intracranial vessels. However, the main diagnosis in these cases is usually so obvious that differential-diagnostic considerations are secondary.
1.2.4 Imaging diagnosis
Ischemic cerebrovascular events present as transient, fluctuating, or permanent neurological deficits. Transient ischemic attack (TIA) has so far been defined clinically as a neurological deficit that lasts for up to a maximum of 24 hours. TIA leads to frank stroke within 90 days in over 10% of cases. In contrast to earlier approaches, rapid clarification and treatment within 24 hours are now recommended, and this reduces the risk of stroke to 2%. Imaging diagnosis must include either computed tomography (CT) or magnetic resonance imaging (MRI), or a combination of the two modalities. In patients with acute non-fluctuating deficit and significant loss of function, imaging should be carried out as quickly as possible in order to limit the damage caused by the ischemia.
To establish an indication for intravenous or intra-arterial thrombolysis or thrombectomy, noncontrast CT followed by CT angiography is usually sufficient. This makes it possible to distinguish hemorrhagic from ischemic stroke. In addition, treatment-relevant vascular occlusions can be quickly demonstrated (with an examination time of < 2 min). MRI is preferable to CT diagnosis in individual cases (e.g., unclear time window, young patient) in an acute setting, but it involves a longer examination time (approximately 15–20 min).
In addition to imaging of the area with diffusion disturbance or infarction and the underlying vascular pathology on MRI and CT, a perfusion study (CT and MRI) after bolus contrast administration may reveal underperfusion of the vascular territory involved. Intracranial stenoses may lead to fluctuating neurological deficits as a result of recurrent emboli or underperfusion. When there is a drop in systemic blood pressure, stenoses often have hemodynamic effects as a sign of an inadequate collateral supply. Clarification of the hemodynamic components before a treatment decision is taken (for endovascular or surgical treatment) thus requires perfusion measurements in order to assess the quality of the cerebral collateral supply (described in the section on surgical therapy below) and the perfusion reserve.
1.2.4.1 Doppler/color duplex ultrasonography
Examination technique
Transtemporal insonation of the distal internal carotid artery, carotid T, middle cerebral artery (M1/2), anterior cerebral artery (A1), and posterior cerebral artery (P1/2) is obligatory. The latter may be confused with the superior cerebellar artery when pulsed-wave Doppler is used, therefore visual stimulation should be carried out as a check (the posterior cerebral artery shows increased flow in response to inert optic stimuli after prior closing of the eye, due to supply of the visual cortex; the superior cerebellar artery shows no reaction).
Optionally, transtemporal imaging of the internal carotid artery may be carried out at cerebral levels and toward the base of the skull, as well as in the basilar artery and clivus if needed. Transnuchal insonation of the vertebral artery bilaterally, as well as of the basilar artery, is also obligatory, with imaging of the posterior inferior cerebellar artery (PICA) if needed. Transtemporal insonation is difficult in approximately 30% of cases and impossible due to a widened calvaria in 20% of cases; an ultrasound contrast medium may then be used → initial saturation artifact (“blooming”) can be corrected by reducing the transmission power (but not the gain; with a lower mechanical index, the contrast medium has a longer intravascular persistence). One milliliter of ultrasound contrast medium is often sufficient.
Differential diagnosis
Arteriosclerosis: classic focal flow acceleration.
Occlusion: no flow can be demonstrated even with ultrasound contrast medium, despite good imaging of other vessels at the skull base.
Typical indirect preocclusive/postocclusive criteria.
Vasculitis: there may be multifocal and serial focal or long flow acceleration.
Vasospasm: there may be multifocal and serial focal or long flow acceleration.
Findings may change over time.
Dissection
In cases of floating membrane, there may by a pathognomonic triphasic “splash signal”—otherwise, see stenosis/occlusion.
Fibromuscular dysplasia.
Beaded appearance.
AVM feeder.
Typical excess flow signal, particularly in diastole.
Long increase in flow velocity.
Possibly corresponding changes in the afferent extracranial vessels.
Possibly pulsatile venous flow.
Congenital variants.
The vertebral artery often shows hypoplasia and also aplasia.
The anterior/posterior communicating branch is often hypoplastic or aplastic.
The P1 segment of the posterior cerebral artery is often hypoplastic (in up to 20% of cases)—then there is supply from P2 via the posterior communicating branch—i.e., from the internal carotid artery (caution: stenosis of the internal carotid artery may then be the cause of posterior infarction).
Specific findings
Middle cerebral artery
Unremarkable findings: maximum systolic velocity < 120 (–140) cm/s, with no flow disturbances detectable acoustically or in the spectral analysis.
Borderline findings: circumscribed maximum velocities around 140 cm/s (> 120 cm/s), but still with no flow disturbances.
Hyperperfusion: during recanalization after a stroke, there may be a phase of hyperperfusion. Signs arguing in favor of hyperperfusion and against a stenotic process in the middle cerebral artery are:
Long increased flow velocities without circumscribed turbulence phenomena.
Increased flow velocities, possibly also in medial branches.
Flow velocities not exceeding 2 m/s.
Quotients for maximum velocities in the middle cerebral artery/internal carotid artery < 2; values > 3 tend to suggest stenoses.
The intra-individual course shows a tendency toward normalization.
Similar findings are seen after carotid surgery, with hyperperfusion (an increased flow velocity of > 50% in the middle cerebral artery on side-to-side comparison) also being observed in some patients—particularly those with poor preoperative collateralization and long clamping times (Widder 1999)—as a sign that autoregulation is still disturbed.
Low-grade stenosis (50 to < 70%): circumscribed increase in the maximum systolic flow velocity to values of 140–200 cm/s, as well as in middle and possibly diastolic flow velocities. The critical velocity after which the presence of a stenosis of more than 50% can be definitely assumed is 220 cm/s (Baumgartner et al. 1999). Incipient flow disturbances, spectral widening, signs on the color Doppler signal of localized increases in flow velocity and turbulences, possibly a mixed signal with a normal main trunk signal and a stenotic signal in M2/middle artery stenoses.
Moderate stenosis (approximately 70% to < 80%): a focal increase in the maximum systolic flow velocity to values of 200–280 cm/s and also diastolic flow velocities. Clear flow disturbances, spectrum widening, blurred systolic window, retrograde flow components, marked signs on the color Doppler signal of a localized increase in flow velocity, reduced lumen size possibly visible in velocity mode and/or power mode.
High-grade stenosis (> 80%): a circumscribed increase in maximum systolic flow velocity to values of > 280 cm/s and of the diastolic flow velocities, marked flow disturbances, spectrum widening, blurred systolic window, marked retrograde flow components, clearly reduced post-stenotic flow velocity, marked signs on the color Doppler signal of a localized increase in flow velocity, also an increase in diastolic flow velocity visible particularly with variance coding (candleflame phenomenon), lumen reduction visible in velocity mode and/or power mode, possible occurrence of musical murmurs.
Middle cerebral artery main trunk occlusion (M1): no demonstrable M1 signal despite good ultrasound imaging conditions, reduced flow velocities in the proximal vessels, possibly hyperperfusion of collaterals—e.g., the anterior and posterior cerebral arteries as afferent vessels for leptomeningeal anastomoses; in cases of occlusion distal to the origins of the lenticulostriate arteries, there may be a typical prestenotic highly pulsatile signal with low flow velocity in the proximal main trunk—confirmation using ultrasound contrast enhancement.
Middle cerebral artery branch occlusion (M2): lower flow velocity in the M1 segment in side-to-side comparison, possibly higher pulsatility in M1, absent M2 imaging, possible hyperperfusion of collaterals—e.g., the anterior and posterior cerebral arteries as afferent vessels for leptomeningeal anastomoses; possible retrograde perfusion of collaterals, occlusions of small vessels not detectable; flow differences in the proximal middle cerebral artery can be identified by detecting absolute systolic/diastolic velocities and by detecting pulsatility differences.
Carotid-T processes: these are often combined with distal internal carotid artery and outflow regions of the middle and anterior cerebral arteries; carotid-T occlusion processes are the hemodynamically most severe findings, since all collateralization pathways from the middle cerebral artery are blocked with the exception of leptomeningeal anastomoses. The anterior cerebral artery may be collateralized from the contralateral side via the anterior communicating branch.
Assessment of hemodynamic consequences:
Demonstration of post-stenotic and postocclusive flow, assessment of the spectrum (steepness of increase, acceleration time, Vmax systolic/diastolic)
Demonstration of collateralization pathways (ideally using duplex ultrasonography)
Possible compression test if the information provided is expected to have implications—e.g., for:
– Clarifying the indication for surgery in multiple-vessel disease (e.g., internal carotid artery stenosis with contralateral occlusion)
– Assessment of the afferent components of the various collateral pathways
– Assessment of the hemodynamic consequences of a potential vascular occlusion (e.g., progressive asymptomatic internal carotid artery stenosis)
– Testing the quality of the collateralization
– Detecting any collateral pathways that cannot be spontaneously imaged (e.g., posterior communicating branch) by inducing hyperperfusion
– Assessment of the risk of intraoperative clamping (see below) before or during carotid thromboendarterectomy or proximal embolic protection. Evidence for sufficient residual perfusion includes:
– In cases of ≤ 80% stenosis of the internal carotid artery: mean residual flow velocity in the middle cerebral artery > 30–40% of the resting value
– In cases of ≤ 80% stenosis of the internal carotid artery: mean residual flow velocity ≥ 30 cm/s
Testing of CO2 reactivity/autoregulation reserve:
– Correlates with the risk of hemodynamic watershed infarction—corresponds to the remaining CO2-induced dilation capacity in the intracerebral vessels
– Methods:
– Breath-holding index
– Doppler CO2 testing
– Apnea–hyperventilation testing
– Acetazolamide (Diamox) testing
Fig. 1.2–6 Apnea test. Top: the power Doppler profile of the middle cerebral artery bilaterally, with massively reduced amplitude on the right, reduced pulsatility and prolonged acceleration time as signs of poor intracerebral collateralization in subtotal internal carotid artery stenosis. Bottom: inverse steal phenomenon in the apnea test as a sign of an absent autoregulation reserve and steal in a vascular area that is still capable of reacting.
The apnea–hyperventilation test is the fastest exploratory test in routine work: resting Vmax is measured, followed by respiration for at least 30 seconds once a second or once per pulse and measurement of Vmax, and finally an apnea phase induced at mid-respiration for a maximum duration, during which Vmax is measured again. Standard: increase (apnea) and decrease (hyperventilation) by at least 15% in comparison with the resting values (Widder 1999).
Anterior cerebral artery
There are as yet no validated Doppler or duplex ultrasound criteria for grading stenoses of the anterior cerebral artery. The critical velocity from which a > 50% stenosis must be assumed to be present is 155 cm/s (Baumgartner et al. 1999). The criteria mentioned in connection with the middle cerebral artery can be used as an approximation. It is sometimes difficult to differentiate organically fixed stenoses from functional stenoses in collateral function: circumscribed flow accelerations argue more for localized stenoses, while longer flow accelerations—particularly in combination with other signs of collateralization (such as retrograde perfusion in the contralateral anterior cerebral artery, compression tests) suggest relative stenoses with collateral function.
Posterior cerebral artery
There are as yet no validated Doppler or duplex ultrasound criteria for grading stenoses of the posterior cerebral artery. The critical velocity from which a > 50% stenosis must be assumed to be present is 145 cm/s (Baumgartner et al. 1999). The criteria mentioned in connection with the middle cerebral artery can be used as an approximation. Particular sites of predilection for arteriosclerotic stenoses are the start of the P2 segment, the posterior arch, and more rarely the P1 outflow region. In the P1 segment, relative stenoses due to hyperperfusion in collateral function of the posterior cerebral artery via the posterior communicating branch or stenotic signals from the hyperperfused posterior communicating branch must be taken into consideration (caution: risk of possible confusion); color-coded imaging can be helpful for differentiation here.
Vertebral artery
See under extracranial occlusion processes.
Basilar artery
The head of the basilar artery is a site of predilection for arteriosclerotic lesions. There are as yet no validated Doppler or duplex ultrasound criteria for grading stenoses of the basilar artery. The critical velocity from which a > 50% stenosis must be assumed to be present is 140 cm/s (Baumgartner et al. 1999), but suspicion should already be raised at flow velocities of 100–120 cm/s. The criteria mentioned in connection with the middle cerebral artery can also be used as an approximation.
Basilar artery hypoplasia must be assumed when there is extracranial evidence of bilateral vertebral artery hypoplasia, particularly if the total diameter of the two vertebral arteries is less than 5 mm. Occlusion of the basilar artery (basilar thrombosis) must be assumed with:
High pulsatility (low or absent diastolic flow) in the extracranial segments of both vertebral arteries
High pulsatility in the transnuchally visible vertebrobasilar pathway
Inability to image the basilar artery on color-coded duplex ultrasound (signal enhancement may be needed)
Noticeable postocclusive Doppler signal in the posterior cerebral arteries
Possible collateral flow via the posterior communicating branch
Basilar occlusion cannot be definitively excluded only by evaluating the findings from the extracranial vertebral artery (particularly with older, collateralized occlusions that have developed gradually).
Other applications for transcranial Doppler and duplex ultrasonography
Evidence of spontaneous cerebral emboli/HITS analysis
High-intensity transient signals (HITS) with a relevant signal intensity and temporal latency in their occurrence in two sample volumes (multigating procedure) in the main trunk of the middle cerebral artery represent high-intensity signal peaks within the Doppler spectrum of blood components—i.e., spontaneous cerebral emboli. There is a 15-fold increase in the risk of stroke when there is evidence of HITS—e.g., in patients with 60% asymptomatic internal carotid artery stenosis—in comparison with negative HITS findings (Spence et al. 2005).
Testing for persistent patent foramen ovale (PFO)
An ultrasound contrast medium that will not enter the capillaries is injected into a large antecubital vein or the common femoral artery; after approximately 8 seconds, a Valsalva maneuver is carried out for approximately 4 seconds, possibly supported by compression in the abdominal area, with further ultrasound imaging for 5–10 seconds. When there is an intermittent cardiac right–left shunt due to PFO, there is a mean contrast appearance time of 9 ± 6 s (< 15 s) in comparison with 24 ± 9 s (> 15 s) with transpulmonary passage. A relevant shunt is present if more than 10 emboli appear at rest and/or more than 25 emboli appear after the Valsalva maneuver within the time stated. Larger persistent foramina lead to showers of emboli that can no longer be detected individually. Lower emboli rates are probably not relevant. The advantage of TCD in comparison with transesophageal echocardiography is that the procedure is not invasive, the patient can still cooperate (with the Valsalva), evidence of noncardiac shunts is also possible (explaining occasional differences), and the sensitivity is comparable for relevant shunts.
Intracranial pressure monitoring
Diastolic flow/pulsatility correlates with intracerebral pressure/outflow resistance; the parameters along the course are highly sensitive and measurement of absolute values is of course not possible (with the exception of diastolic zero flow, in which case intracranial pressure corresponds to the diastolic pressure, with phasic flow when there is a further rise in intracranial pressure).
Diagnosis of cerebral death
Transcranial Doppler ultrasound is an approved procedure used to shorten the waiting time before cerebral death is diagnosed. Prerequisites include not only availability of an examiner with the relevant expertise, but also confirmation that ultrasonography can be carried out in the patient and appropriate adjustment of the device settings (high gain, maximum transmission power, high reception speed, low wall filter, large measurement volume ≥ 15 mm). Typical Doppler findings in cerebral perfusion standstill:
Phasic flow (biphasic flow with backflow components representing > 30% of the antegrade flow)
No systolic peaks (maximum amplitudes 50 cm/s, duration < 200 ms)
Passive breath-regulated signal amplitudes
Absence of a diastolic signal
No evaluable signal →caution! Check insonability, examination technique → possible use of ultrasound contrast enhancement
Prerequisites:
Systemic blood pressure > 80 mmHg systolic
Heart rate < 120/min
No relevant calvarial defects
Criteria for brain death: evidence of phasic flow and/or small systolic peaks twice at an interval of at least 30 minutes, on each occasion bilaterally at the middle cerebral artery or intracranial internal carotid artery, or with duplex ultrasound in the internal carotid artery extracranially and in the basilar or vertebral artery intracranially, or with duplex ultrasound in the vertebral artery extracranially.
Doppler ultrasound can be used independently of the type of cerebral damage—i.e., even in cases of toxic injury. When an appropriate acoustic window is used, a false-positive finding cannot occur when an experienced examiner uses the above criteria. False-negative findings are possible:
With cerebral arteriovenous malformations with shunt flow
In the absence of raised intracranial pressure
When perfusion re-starts (with intracranial pressure declining
again)
In infants with open fontanelles and sutures
In the case of the proximal vertebral artery, due to branches not supplying the brain
In a purely extracranial examination, due to residual perfusion of the internal carotid artery via the ophthalmic artery
Vasospasm monitoring in subarachnoid bleeding
The vasospasm that mainly occurs from days 4 to 10 can be recorded and monitored using transcranial Doppler (TCD) follow-up observations and by measuring velocities particularly in the middle cerebral artery main trunk:
Borderline: Vmean ≥ 120 cm/s
Pathological: Vmean ≥ 160 cm/s
Critically raised: Vmean ≥ 200 cm/s, Vmax ≥ 300 cm/s
Suspicious: increases in flow velocities ≥ 50% or ≥ 40 cm/s/d during the first 6–7 days (Grosset 1993)
Pulsatility index (PI) > 1, resistance index (RI) > 0.6 (Klingelhöfer et al. 1991)
Ratio of the maximum velocities in the middle cerebral artery and internal carotid artery (the middle cerebral artery/internal carotid artery index) ≥ 3
Caution:
False-negative findings may occur when nonspastic vascular segments are examined.
False-negative findings may occur with simultaneous raised intracranial pressure and consequently reduced flow velocities (pulsatility parameters should therefore also be used).
Prior arteriosclerotic stenoses.
Hyperperfusion (middle cerebral artery/internal carotid artery index < 2).
1.2.5 Treatment
1.2.5.1 Conservative treatment
The treatment strategy depends in principle on multiple factors and should be decided on an individual basis. Risk factors such as nicotine consumption, lipid disorders, hypertension, and diabetes should be eliminated or treated.
In patients with asymptomatic stenoses that are not hemodynamically relevant, invasive treatment is not currently recommended, but drug therapy with platelet inhibitors is recommended, possibly in combination with lipid-lowering agents.
Patients with symptomatic stenoses that are hemodynamically relevant initially receive drug treatment, and invasive therapy is only considered if symptoms recur. The WASID study found no benefit with warfarin administration in comparison with aspirin in patients with symptomatic intracranial stenoses. In comparison, the risk of stroke during the first year was 12% in the aspirin group and almost as high in the warfarin group at 11%; in the second year, the figures were 15% and 13%, respectively. However, as major hemorrhage only occurred in 3% of the patients in the aspirin group in comparison with 8% in the warfarin group, treatment with vitamin K antagonists is now obsolete. Treatment with dual platelet inhibition (e.g., aspirin and clopidogrel) should therefore be considered, particularly when an ischemic event has occurred during single antiplatelet treatment. Individual testing of drug efficacy is advisable in any case, as there is a high percentage (up to 30%) of low responders and nonresponders, although no data in this population that correlates responder rates and clinical events.
When ischemic symptoms recur during medication, the indication for PTA and, if appropriate, stent placement should be considered. A recently published study (the SAMMPRIS study) did not observe any clear benefits with an invasive approach using stenting—a finding that requires further research after optimization of patient selection and concomitant drug therapy.
As mentioned above, it is therefore absolutely imperative to differentiate between embolic and hemodynamically relevant stenoses. In stenoses that have hemodynamic effects, with clinical symptoms, conservative therapy is only appropriate in combination with endovascular therapy (PTA, possibly with stent assistance) in order to improve perfusion. If this is not technically possible, attention should be given to ensure that blood pressure is not reduced too much, to avoid negative effects on the collateral supply.
The treatment of intracranial vascular occlusions also depends on clinical symptoms. Conservative treatment for acute stroke includes normalization of general parameters (cardiovascular and pulmonary function, as well as fluid balance and metabolic parameters) and oxygenation. If a patient reaches hospital within the “thrombolysis window” (usually < 6 hours after the start of symptoms), systemic intravenous thrombolysis therapy (up to 4.5 hours after the initial symptoms) or local intra-arterial embolectomy or thrombolytic therapy can be carried out. The 4.5-hour time interval currently applies to systemic intravenous thrombolytic therapy (with recombinant tissue plasminogen activator, rt-PA). In all cases, the earlier the patient is treated, the better the clinical outcome is. In what is known as the “bridging approach,” intravenous treatment (two-thirds of the total dosage, with 10% of that as a bolus) is combined with intra-arterial therapy. The initial intravenous thrombolysis allows rapid initiation of treatment and may optimize the efficacy of the endovascular intra-arterial therapy.
1.2.5.2 Endovascular intra-arterial therapy
Endovascular intra-arterial therapy is indicated and in most cases possible for hemodynamically relevant intracranial stenoses, or for stenoses that cause recurrent arterioarterial embolism in spite of adequate platelet function inhibition. The endovascular procedure consists of dilation (percutaneous transluminal angioplasty, PTA) (Fig. 1.2-7), which can be combined with stent implantation (Figs. 1.2-8, 1.2-9, 1.2-11). To prevent thromboembolic events during possible stent implantation, the patients should receive dual platelet inhibition (e.g., aspirin and clopidogrel), starting if possible several days before the procedure. It is advisable here to have the efficacy of platelet inhibition tested in the laboratory in order to identify nonresponders and poor responders, who may represent up to 30% of the patients. These patients then have to be treated with a correspondingly higher dosage. Following stent implantation, long-term monotherapy with aspirin or clopidogrel is required after temporary dual therapy.
Using expanding balloon stents to treat intracranial arteriosclerotic stenoses requires precise measurement of the vascular diameter in order to avoid possible overexpansion and rupture of intracranial arteries. However, precise measurement is not always technically reliable, and self-expanding stents were therefore developed that have markedly reduced the complication rate, as they are more flexible and no balloon remodeling is needed. Self-expanding stents can adapt to varying vascular diameters and adjustment of critical stent radial forces allows secondary remodeling of the vessels after the vascular plaque has been broken through using PTA.
Fig. 1.2–7a-d A 71-year-old patient with unilateral symptoms on the right side in a case of left-sided carotid T-occlusion (a). During thrombus aspiration with a Penumbra® catheter, the left carotid flow area was recanalized (b). Multiple aspiration of small thrombus fragments (c). The vessels were successfully reopened (d).
Fig. 1.2–8a-c A 57-year-old patient with unilateral brachiofacial symptoms on the left side in a case of right-sided occlusion of the main trunk of the middle cerebral artery (a). During intra-arterial abciximab (ReoPro®) administration, with stenting required for a subtotal internal carotid artery ostial stenosis on the left side and supplementary rt-PA administration, patency of the middle cerebral vessels was restored (b). Small areas of infarction are demarcated on the follow-up CT after 24 hours (c, arrows).
Fig. 1.2–9a-e A 56-year-old patient with high-grade stenosis of the vertebral artery on the left, on time-of-flight magnetic resonance angiography (MRA) (a, arrow), with recurrent TIAs and a hypoplastic vertebral artery on the right ending at the PICA. The corresponding digital subtraction angiography (DSA) image (b). A Pharos® stent at the level of the stenosis, with the balloon inflated (c). Imaging of the stent in the bone window (d). Normalization of the vascular caliber after complication-free stenting (e).
Fig. 1.2–10a, b A 72-year-old patient who suddenly became comatose, with distal occlusion of the basilar artery (a). Recanalization of the posterior flow area during thrombus aspiration using a Penumbra® catheter (arrow: catheter tip) (b).
The intervention should be carried out with the patient under general anesthesia. Access is obtained via the route of least risk, usually the femoral artery. A microballoon catheter is introduced via a guide catheter positioned in the internal carotid artery or vertebral artery, the balloon positioned at the site of the stenosis, and percutaneous transluminal angioplasty (PTA) is carried out. In a second step, a self-expanding microstent catheter is advanced using a wire exchange maneuver. After optimal placement at the previously stenotic vascular segment, the stent is released by withdrawing the microcatheter. With balloon-mounted stents, prior PTA is not required, as this takes place during the stent deployment (Fig. 1.2-15).
A multicenter prospective study (the Wingspan study) reported good treatment results, with a fatal ipsilateral stroke rate of 7% within 6 months. During the same period, the repeat stenosis rate (with a stenosis grade of > 50%) was low at 7.5% and the patients affected remained without neurological symptoms. A multicenter study in the United States has also reported successful treatment, with technical success and a low periprocedural risk. The primary end point of stroke or death within 30 days after the intervention or ipsilateral stroke after 30 days was observed in 15.7% of the patients and in most cases was associated with withdrawal of platelet inhibition or recurrent stenosis.
In cases of acute vascular occlusion, local therapy nowadays consists of mechanical extraction of the embolus/thrombus.
The mechanical procedures mainly used today include aspiration via wide-lumen, highly flexible catheters that are advanced as far as the vascular occlusion (Fig. 1.2-10a) and a stent retriever system (Figs. 1.2-13, 1.2-14). The efficacy of these systems depends on the consistency of the thrombus/embolus and the arterial access. The systems are sometimes used in combination. Rapid mechanical recanalization of occluded intracranial arteries can be achieved in more than 90% of cases. All of the mechanical procedures can be combined with intra-arterial thrombolysis therapy or platelet inhibition with administration of glycoprotein receptor antagonists.
Fig. 1.2–11a-d A 77-year-old patient with high-grade stenosis of the middle cerebral artery trunk on the left side (a), with recurrent infarction in the middle cerebral artery flow area during dual medication. A Wingspan® stent at the level of the stenosis with an exchange wire in place in an M2 branch (b). Normalization of the vascular caliber after complication-free stenting (c). Three-dimensional imaging of the stent on Xper-CT (d).
Fig. 1.2–12a-f An 82-year-old patient with multiple cervical and intracranial vascular stenoses, with recurrent TIAs in the posterior flow area and a principal finding of high-grade basilar artery stenosis (a, MPR on CTA, arrow). The patient had declined endovascular therapy. Three months later, there was acute hemiplegia on the right side, with dysarthria and bilaterally disturbed eye movements during dual medication. The cause was a proximal basilar artery occlusion (b) following earlier stenosis. It was only possible to pass the stenosis after balloon dilation, with contrast imaging of the distal occluded basilar artery (c). During local thrombolytic therapy with rt-PA, the posterior flow area reopened distally (d). Due to persistent relevant stenosis (e), stenting was carried out at the same time. The Wingspan® stent is seen at the level of the stenosis, with the exchange wire in place (f). In these conditions, there was normalization of the vessels in the posterior flow area and a tolerable occlusion of the right posterior cerebral artery with an anterior fetal supply.
Fig. 1.2–14a, b A 51-year-old patient with acute right-sided hemiparesis and aphasia, with occlusion of the main trunk of the middle cerebral artery on the left (a). After the thrombus had been passed with a microcatheter and a Solitaire® stent had been introduced, the thrombus was successfully extracted from the main trunk and the flow area was revascularized (b).
Fig. 1.2–13a-d A 72-year-old patient with right-sided occlusion of the main trunk of the middle cerebral artery (a) and acute left-sided hemiparesis. The thrombus was passed with a microcatheter (b) and introduction of a Solitaire® stent (c; arrow: distal stent marker), with revascularization with the opened stent lumen. Removal of the stent and successful thrombus extraction from the main trunk of the middle cerebral artery (d). It was not possible in this case to reopen an M2 branch, despite supplementary rt-PA administration and attempts at mechanical thrombolysis.
If recanalization is successful and an underlying stenosis is identified (Fig. 1.2-12), it is beneficial due to the high rate of recurrent occlusion to treat the stenosis during the same session (with PTA alone or in combination with a stent). Unfortunately, the clinical outcome for the affected patients does not depend only on the recanalization of the vessels. The duration and location of the vascular occlusion, the collateral blood supply, the side of the affected brain area that has already suffered infarction, and concomitant diseases that the patient may have also play an important role.
Fig. 1.2–15a-c A 52-year-old patient with high-grade stenosis of the vertebral artery on the right side (a, arrow), with recurrent TIAs and occlusion of the vertebral artery on the left. A Pharos® stent was placed at the level of the stenoses, with an inflated balloon (b). Normalization of the vascular caliber following complication-free stenting (c).
In rare cases when there is persistent intermittent perfusion, recanalization treatment for non-acute, chronic vascular occlusion is indicated.
1.2.5.3 Surgical treatment
Surgical treatment for stenotic and occlusive processes in the region of the extracranial and intracranial circulation of the arteries supplying the brain represents a complementary strategy for preventing ischemia-related neurological deficits. In this approach, a further critical underperfusion event in circulation regions that are at risk for ischemia is avoided by carrying out surgical improvement in the exhausted cerebral collateral supply. This strategy is based on the results of studies in which symptomatic patients with signs of inadequate cerebral collateral supply—i.e., hemodynamic insufficiency or exhausted cerebral perfusion reserve—had a markedly higher annual risk of stroke, at 18–46%, in comparison with patients with an intact collateral supply. Patient selection—i.e., the choice of which patients should undergo targeted revascularization—therefore focuses on assessing what is known as the cerebral perfusion reserve. This can be measured using various functional examinations of regional cerebral blood flow (rCBF).
Determining cerebral perfusion reserve
This is based on carrying out paired examinations of the rCBF, with a baseline measurement being carried out in resting conditions. Following a vasodilatory stimulus—e.g., inhalation of carbon dioxide or administration of acetazolamide (15 mg/kg body weight i.v.)—the rCBF is measured again. After appropriate evaluation and application of various calculation algorithms, the perfusion reserve (cerebrovascular reserve capacity, CVRC, expressed as a percentage; also known as the vasomotor reserve) is calculated. For direct measurements of rCBF, methods that can be used include PET, stable Xe-CT, and quantitative SPECT; transcranial Doppler (TCD) ultrasonography only allows indirect estimates.
In physiological conditions, rCBF stimulation can be expected to produce at least a 30% increase in rCBF (CVRC > 30%—i.e., normal). Depending on the extent of hemodynamic restriction, there is then a reduction in reactive vasodilation, so that restricted CVRC is described as being present at a CVRC < 30%. At values < 10%, CVRC is regarded as having been eliminated. When there is a paradoxical reduction in rCBF (CVRC < –5%)—i.e., maximum vasodilation before stimulation and consequent nonreaction of the resistance vessels—it is assumed that an intracranial steal phenomenon is present. This is the most severe grade of hemodynamic impairment and is associated with the highest risk of secondary ischemia (Fig. 1.2-16).
Indications for revascularization
The following criteria generally arise:
Age < 70 years
Clinical symptoms: recurrent TIA/PRIND
Watershed infarction or normal findings in morphological diagnosis (MRI)
Stenotic occlusive lesions (stenosis and/or occlusion) in the area of the anterior circulation that are not accessible to primary interventional treatment or vascular surgery/intervention
Confirmed hemodynamic cerebrovascular insufficiency
With regard to the underlying pathology, there is considerable variability in the pathogenesis. In most cases, the patients have localized or systemic atherosclerosis. In a far smaller proportion of the patients, there is an indication for surgery due to inadequate collateral supply after carotid dissection or progressive tumor growth with resulting constriction or occlusion of the internal carotid artery, mostly in the area of the skull base. Patients with moyamoya disease or moyamoya syndrome represent a special case, which is discussed separately below.
Surgical technique
The aim of the surgical procedure is to normalize the CVRC and thus to “restore” physiological perfusion conditions. The technique used is a standard extracranial–intracranial (EC/IC) bypass. A donor vessel in an extracranial site, usually the superficial temporal artery (STA), is anastomosed with a cortically located branch of the middle cerebral artery (MCA) in the area of the lateral sulcus (sylvian fissure). The procedure is carried out with the patient under general anesthesia, with endotracheal intubation and neuroprotective measures to prevent ischemic complications. After preparation of the donor vessel (the STA), a craniotomy with a diameter of approximately 3 cm is carried out in the area of a defined target point over the lateral sulcus. After opening of the dura mater and exposure of a suitable recipient vessel (the middle cerebral artery in the M2/M3 segment), the standard bypass is placed using an end-to-side technique with 10–12 interrupted sutures (Fig. 1.2-17). The patency of the bypass that has been created can usually be immediately documented intraoperatively using indocyanine green (ICG) video angiography. Thanks to improved perioperative treatment, more sophisticated surgical techniques, and the ability to check the success of the procedure immediately, it is now possible to carry out this procedure with only minor surgical morbidity (< 5%). The operation is only carried out after adequate inhibition of platelet aggregation (100 or 325 mg ASA/d p.o.). Among other things, this makes it possible to achieve a bypass patency of > 98%. Platelet aggregation inhibition starts before the operation and continues on a lifelong basis.
Fig. 1.2–16a, b A standard extracranial–intracranial (EC/IC) bypass in a patient with atherosclerotic occlusion of the right internal carotid artery. (a) Conventional angiography reveals insufficient collateral supply to the right hemisphere. Angiographic demonstration of the STA-MCA bypass using the right superficial temporal artery (STA), which was anastomosed with a distal branch of the middle cerebral artery (MCA). (b) Anteroposterior view.
In addition, selective DSA and MRI are carried out postoperatively. Specific follow-up treatment for the patients is not necessary. When the indications described above are observed, the principles of microsurgery are rigorously applied, and the relevant expertise is present, it is possible to reduce the risk of secondary ischemia significantly in comparison with conservative treatment.
Special case: moyamoya disease and moyamoya syndrome
This is a rare, progressive steno-occlusive disease in which, due to unexplained causes, slowly progressive occlusion of the cerebral arteries in the region of the circle of Willis occurs. In parallel with this, spontaneous intracerebral and also extracranial–intracranial formation of collateral vessels is seen. These are regarded as outstanding examples of the way in which complex natural collaterals can develop on the basis of chronic ischemia. Both steno-occlusive changes with hemodynamically significant impairment of the cerebral blood supply along with spontaneous compensation mechanisms in the form of neoangiogenesis and arteriogenesis are therefore seen in patients with moyamoya disease. In addition to the impressive angiographic findings, the disease is distinct from other steno-occlusive diseases in relation to its epidemiology and clinical course (Fig. 1.2-18). The diagnostic characteristics of moyamoya disease (bilateral lesions, spontaneous collateral network) are more frequently associated with other diseases, sometimes systemic ones. In these cases, the condition is known as moyamoya syndrome. However, this does not generally lead to any changes in the treatment strategy. In central Europe, moyamoya disease and moyamoya syndrome occur sporadically and affect both children and adults. In contrast to the Asian form, which usually becomes manifest in adults in the form of intraparenchymal hemorrhage, cerebral ischemia is the major symptom in both age groups in Europe. Although it is mainly adult patients who suffer TIAs or stroke, the disease can also occur in children, often in the context of focal epilepsy or a cerebral organic psychological syndrome, resulting in inadequate formation of spontaneous collaterals and cerebral compensation mechanisms.
Fig. 1.2–17a, b Intraoperative view. (a) Positioning of the patient for planned placement of a superficial temporal artery (STA)–middle cerebral artery (MCA) anastomosis. The planning incision line (blue) is shown, along with the course of the superficial temporal artery (red) and the planned craniotomy. (b) Completed bypass in the area of the lateral sulcus (sylvian fissure).
Fig. 1.2–18a-c Typical angiographic findings in adult moyamoya disease. There are bilateral steno-occlusive changes in the area of the intracranial carotid bifurcation, with simultaneous formation of extensive basal collaterals.
Indications for revascularization
In view of the poor results of conservative therapy and the underlying pathogenesis of the disease, revascularization surgery is an option for both children and adults. Surgical treatment is indicated in symptomatic adult patients when the diagnosis is confirmed, a reduced CVRC is demonstrated in the symptomatic hemisphere, and there is no severe focal neurological deficit. In children, surgical treatment in both hemispheres is recommended, due to the generally unfavorable natural course of the condition. This allows a marked improvement in the overall prognosis.
Surgical technique
The aim of surgical treatment is to improve the cerebral collateral supply. Various techniques are available for the purpose. In addition to the standard EC/IC bypass described above (STA–MCA anastomosis or STA–anterior cerebral artery anastomosis), which is preferred in adults, there are also a number of indirect procedures. These take advantage of the capacity for spontaneous collateralization. For example, extracranial “donor tissue” can be placed in contact with the arachnoid on the surface of the brain in order to induce spontaneous vascularization with new collateral vessels. This leads to effective revascularization approximately 3 months after the operation. Overall, more than 30 surgical modifications have been described, with indirect revascularization techniques being particularly effective in pediatric patients. Examples of potential “donor tissue” include the temporalis muscle, the epicranial aponeurosis, and the parietal side of the dura mater. In practice, a combination of direct and indirect techniques is used when possible, since in addition to immediate stabilization of perfusion conditions this can also allow delayed synergistic stabilization of regional perfusion. To improve bypass function and for secondary prophylaxis, lifelong medication with platelet aggregation inhibitors is also administered in moyamoya patients.
When combined revascularization is planned, the principles used in direct bypass surgery apply, and these can then be extended, depending on the planned procedure. Combining an STA–MCA anastomosis with encephalomyosynangiosis (EMS) requires an extended craniotomy (Fig. 1.2-19). In this procedure, the temporalis muscle is placed on the cerebral surface after a standard STA–MCA bypass has been created. Regular DSA and MRI examinations are then carried out during the postoperative follow-up. These are useful not only for checking the quality of the surgical procedure but also to assess the subsequent clinical course.
Fig. 1.2–19a, b (a) Positioning of the patient for planned combined revascularization. The planned incision line (blue), the course of the superficial temporal artery (red) and the planned craniotomy have been marked. (b) The intraoperative view after dissection of the temporal muscle, completion of the craniotomy, and resection of the dura mater before placement of the direct bypass.
References
Adams et al. Preview of a new trial of extracranial-to-intracranial arterial anastomosis: the carotid occlusion surgery study. Neurosurg Clin N Am 2001; 12 (3): 613–24, ix–x.
Baker WL, Colby JA, Tongbram V, Talati R, Silverman IE, White CM, Kluger J, Coleman CI. Neurothrombectomy devices for the treatment of acute ischemic stroke: state of the evidence. Ann Intern Med 2011; 154: 243–52.
Baumgartner RW, Mattle HP, Schroth G. Assessment of ≥ 50% and < 50% intracranial stenoses by transcranial color-coded duplex sonography. Stroke 1999; 30 (1): 87–92.
Bose A, Hartmann M, Henkes H, Liu HM, Teng MM, Szikora I, Berlis A, Reul J, Yu SC, Forsting M, Lui M, Lim W, Sit SP. A novel, self-expanding, nitinol stent in medically refractory intracranial atherosclerotic stenoses: the Wingspan study. Stroke 2007; 38: 1531–7.
Bose A, Henkes H, Alfke K, Reith W, Mayer TE, Berlis A, Branca V, Sit SP. For the Penumbra phase 1 stroke trial investigators. The penumbra system: a mechanical device for the treatment of acute stroke due to thromboembolism. AJNR 2008; 29: 1409–13.
Brekenfeld C, Schroth G, Mordasini P, Fischer U, Mono ML, Weck A, Arnold M, El-Koussy M, Gralla J. Impact of retrievable stents on acute ischemic stroke treatment. AJNR 2011; 32: 1269–73.
Chimowitz MI, Lynn MJ, Derdeyn CP, Turan TN, Fiorella D, Lane BF, Janis LS, Lutsep HL, Barnwell SL, Waters MF, Hoh BL, Hourihane JM, Levy EI, Alexandrov AV, Harrigan MR, Chiu D, Klucznik RP, Clark JM, McDougall CG, Johnson MD, Pride GL Jr, Torbey MT, Zaidat OO, Rumboldt Z, Cloft HJ; SAMMPRIS Trial Investigators. Stenting versus aggressive medical therapy for intracranial arterial stenosis. N Engl J Med 2011 Sep 15; 365 (11): 993–1003. Epub 2011 Sep 7.
Derdeyn et al. Cerebral hemodynamic impairment: methods of measurement and association with stroke risk. Neurology 1999; 53 (2): 251–9.
Derdeyn et al. Re: Stages and thresholds of hemodynamic failure. Stroke 2003; 34 (3): 589.
Fiorella DJ, Turk AS, Levy EI, Pride GL Jr, Woo HH, Albuquerque FC, Welch BG, Niemann DB, Aagaard-Kienitz B, Rasmussen PA, Hopkins LN, Masaryk TJ, McDougall CG U.S. Wingspan Registry: 12-month follow-up results. Stroke 2011 Jul; 42 (7): 1976–81.
Gröschel K, Schnaudigel S, Pilgram SM, Wasser K, Kastrup A. A systematic review on outcome after stenting for intracranial atherosclerosis. Stroke 2009; 40: e340–7.
Gross et et al. Use of transcranial Doppler sonography to predict development of delayed ischemic deficit after subarachnoidal hemorrhage. J Neurosurg 1993; 78: 183–7.
Grundmann K, Jaschonek K, Kleine B, Dichgans J, Topka H. Aspirin non-responder status in patients with recurrent cerebral ischemic attacks. J Neurol 2003; 250: 63–6.
Hacke W, Kaste M, Bluhmki E, Brozman M, Dávalos A, Guidetti D, Larrue V, Lees KR, Medeghri Z, Machnig T, Schneider D, von Kummer R, Wahlgren N, Toni D; ECASS Investigators. Thrombolysis with alteplase 3 to 4.5 hours after acute ischemic stroke. N Engl J Med 2008 Sep 25; 359 (13): 1317–29.
Horn et al. Moyamoya-like vasculopathy (moyamoya syndrome) in children. Child’s nervous system: ChNS: official journal of the International Society for Pediatric Neurosurgery 2004; 20 (6): 382–91.
Horn et al. Risk of intraoperative ischemia due to temporary vessel occlusion during standard extracranial-intracranial arterial bypass surgery. J Neurosurg 2008; 108 (3): 464–9.
Kasner SE, Lynn MJ, Chimowitz MI, Frankel MR, Howlett-Smith H, Hertzberg VS, Chaturvedi S, Levine SR, Stern BJ, Benesch CG, Jovin TG, Sila CA, Romano JG; Warfarin Aspirin Symptomatic Intracranial Disease (WASID) Trial Investigators. Warfarin vs aspirin for symptomatic intracranial stenosis: subgroup analyses from WASID. Neurology 2006; 67: 1275–8.
Kasner SE. Natural history of symptomatic intracranial arterial stenosis. J Neuroimaging 2009; 19 (Suppl 1): 20S–1S.
Khan et al. Moyamoya angiopathy in Europe. Acta Neurochir Suppl 2005; 94: 149–52.
Klingelhöfer J, Sander D, Holzraefe M. Cerebral vasospasm evaluated by transcranial Doppler ultrasonography at different intracranial pressures. J Neurosurg 1991; 75: 752–8.
Kraemer et al. Moyamoya Disease in Europeans. Stroke 2008; 39: 3193.
Kuroda et al. Incidence and clinical features of disease progression in adult moyamoya disease. Stroke 2005; 36 (10): 2148–53.
Peña-Tapia et al. Identification of the optimal cortical target point for extracranial-intracranial bypass surgery in patients with hemodynamic cerebrovascular insufficiency. J Neurosurg 2008; 108 (4): 655–61.
Rothwell PM, Giles MF, Chandratheva A, Marquardt L, Geraghty O, Redgrave JN, Lovelock CE, Binney LE, Bull LM, Cuthbertson FC, Welch SJ, Bosch S, Alexander FC, Silver LE, Gutnikov SA, Mehta Z; Early use of Existing Preventive Strategies for Stroke (EXPRESS) study. Effect of urgent treatment of transient ischaemic attack and minor stroke on early recurrent stroke (EXPRESS study): a prospective population-based sequential comparison. Lancet 2007 Oct 20; 370 (9596): 1432–2.
Schmiedek et al. Improvement of cerebrovascular reserve capacity by EC-IC arterial bypass surgery in patients with ICA occlusion and hemodynamic cerebral ischemia. J Neurosurg 1994; 81 (2): 236–44.
So et al. Prediction of the clinical outcome of pediatric moyamoya disease with postoperative basal/acetazolamide stress brain perfusion SPECT after revascularization surgery. Stroke 2005; 36 (7): 1485–9.
Spence JD, Tamayo A, Lownie SP, Ng WP, Ferguson GG. Absence of microemboli on transcranial Doppler identifies low-risk patients with asymptomatic carotid stenosis. Stroke 2005; 36 (11): 2373–8. Epub 2005 Oct 13.
Weerakkody GJ, Brandt JT, Payne CD, Jakubowski JA, Naganuma H, Winters KJ. Clopidogrel poor responders: an objective definition based on Bayesian classification. Platelets 2007; 18: 428–35.
Widder B. Doppler- und Duplexsonographie der hirnversorgenden Arterien. Springer, 1999.
Woitzik et al. Intraoperative control of extracranial-intracranial bypass patency by near-infrared indocyanine green videoangiography. J Neurosurg 2005; 102 (4): 692–8.
1.3 Intracranial aneurysms, AV malformations, and other vascular malformations
Introduction and conservative treatment: Werner Weber, Uwe Dietrich
Interventional treatment: Werner Weber
Surgical treatment: Maximilian J.A. Puchner
Intensive care: Hans-Georg Bone
1.3.1 Aneurysms
1.3.1.1 Clinical picture
Intracranial aneurysms are bulges, usually saccular, in the arteries at the base of the brain that occur at vascular branching points, which represent a site of predilection for this disease. The vascular branching points must be regarded as a weak point at which the fibers of the tunica media—which is thinner in the intracerebral vessels than in the extracranial ones—diverge in order to divide into two vessels. This histological aspect means that the wall structure of aneurysms differs from that in normal cranial vessels. The tunica media is absent and the internal elastic lamina is very thin. The pathogenesis of the disease has not been fully explained. In addition to congenital factors such as hereditary connective-tissue diseases, factors such as degenerative vascular diseases, infection (in what are known as mycotic aneurysms, which are typically located mainly in the peripheral cerebral vessels) and hemodynamic factors (hypertension and downstream arteriovenous malformations) also need to be taken into account. In addition, there is an association of aneurysm with cystic renal degeneration.
The circle of Willis is the vascular circuit that connects the posterior and anterior circulation as well as linking the right and left halves of the brain circulation, and the circle has many variants. The connecting segments linking the two carotid flow areas are called the anterior communicating branch (usually unpaired, but with many variants, including even aplasia) and the posterior communicating branch, linking the carotid flow area with the posterior circulation (paired, with widely varying caliber and even paired or unpaired aplasia; hypoplasia of the segment of the posterior cerebral artery between the tip of the basilar artery and the orifice of the posterior communicating branch may also be seen here). It is not possible to describe all of the possible variants here, but to understand the frequency of aneurysms in these locations it is important to know that due to hypoplasia and aplasia, individual vascular segments in the circle of Willis may be subject to greater pressure at the remaining points; this is one reason for the frequency of aneurysms there. The most frequent location for aneurysms is in the anterior communicating branch, followed by the internal carotid artery/posterior communicating branch and the middle cerebral artery (together 60–70%). Aneurysms in the posterior circulation are much rarer; the most frequent there are aneurysms at the tip of the basilar artery, at about 10% (Fig. 1.3-1).
In addition to saccular aneurysms, there are also fusiform ones that affect the entire circumference of the vessel segment involved. These are caused by vascular wall dissection with bleeding into the wall and formation of what are known as “false aneurysms.” Vascular dissections occur either spontaneously or due to trauma (Fig. 1.3-6). On the basis of autopsy studies, the incidence of aneurysms is estimated at around 3–5% of the population. The incidence of intracranial aneurysmal bleeding varies regionally, and in Germany it is estimated at 10 cases per 100,000 population per year. The probability of bleeding increases with age; the patients’ average age is approximately 50, and women are more often affected. The morbidity and mortality rates for these initial bleeds are very high. The outcome for the affected patients can be roughly estimated using the “rule of thirds”: one-third of the patients die due to the acute effects of hemorrhage and complications during the course (spasm and hydrocephalus); one-third become handicapped as a result of the hemorrhage and are unable to return fully to normal everyday life; and one-third of the patients are able to return to their normal lives. The indications for the treatment of ruptured or space-occupying aneurysms (cranial nerve paralysis or other compression syndromes, e.g., in the brainstem with large fusiform basilar aneurysms) are based on the symptoms. Apart from emergency operations when there are acute symptoms of constriction, ruptured aneurysms should be treated promptly (within 24 h) in order to prevent an early second bleed. The indication for treatment of unruptured and asymptomatic aneurysms is much less clear. The bleeding risk in unruptured aneurysms is approximately 1–2% per year and is higher in patients with multiple aneurysms, large aneurysms, and those with aneurysms that have already ruptured. A growth tendency in the aneurysm, an irregular shape with daughter aneurysms, and a family history of subarachnoid bleeding can also be used as criteria for establishing the indication for treatment. The patients’ psychological burden due to the diagnosis and their age and general condition, as well as comorbid conditions, should be taken into consideration when establishing the indication. Research on this topic is incomplete and unclear.
Fig. 1.3–1 Sites of predilection for saccular intracranial aneurysms. 1, anterior communicating artery (25%); 2, origin of the posterior communicating artery (20%); 3, division point of the middle cerebral artery (15%); 4, supraclinoid internal carotid artery (10%); 5, tip of the basilar artery (< 10%).
1.3.1.2 Clinical findings
Patients are at risk from intracranial aneurysms, as they may rupture, usually leading to what is known as subarachnoid bleeding. This may also be associated with intraparenchymal bleeding. The acute clinical symptoms of subarachnoid bleeding (SAB) are highly varied, depending on its extent. They range from mild headache to sudden coma. Typically, there is a sudden-onset headache (thunderclap headache), accompanied by nausea and stiff neck. Neurological deficits and various degrees of disturbance of consciousness may also occur. The Hunt and Hess scale (Table 1.3-1) describes the severity of SAB and the surgical mortality risk. Associated vascular spasm with consequent circulatory disturbances and imminent infarction, as well as disturbed cerebrospinal fluid resorption with resultant hydrocephalus, can aggravate the clinical findings as well as the outcome. In the worst case, the bleeding may be so extensive that immediate relief of cerebral pressure becomes absolutely necessary.
Unruptured aneurysms are asymptomatic, or may become apparent as a result of their space-occupying character (known as paralytic aneurysms). These are usually very large or medium-sized aneurysms in the corresponding locations; typical examples include aneurysms in the posterior communicating artery with compression of the oculomotor nerve and consequent paresis of the ocularmuscles, and large aneurysms in the anterior circulation in contact with cranial nerves. Figure 1.3-10 shows an aneurysm in the anterior communicating branch with space-occupying effects on the optic nerve. Figure 1.3-5 illustrates the state after surgical treatment. In extremely rare cases, stroke may occur in association with very large, partly thrombosed aneurysms, due to thrombi embolizing of the aneurysm after forming at its neck when stenosis develop there, or simply due to the aneurysms’ space-occupying effect.
Table 1.3–1 The Hunt and Hess (HH) scale for assessing the severity and surgical risk of subarachnoid bleeding, correlated with the Glasgow Coma Scale (GCS) and World Federation of Neurologic Surgeons scale (WFNS).
Table 1.3–2 The Fisher CT classification of subarachnoid bleeding (SAB).
| Fisher 1 | No bleeding |
| Fisher 2 | SAB < 1 mm thick |
| Fisher 3 | SAB > 1 mm thick |
| Fisher 4 | SAB of any thickness with intraventricular spread or intracranial bleeding |
1.3.1.3 Diagnosis
SAB is usually demonstrated using a noncontrast CT (Fig. 1.3-2). Typically, reversal of contrast is seen between the brain parenchyma and the subarachnoid space. The Fisher classification of SAB describes its extent in the brain and postulates a connection with the occurrence of vasospasm (large amounts of blood = extensive spasm). In questionable cases, and particularly when the bleeding was a considerable time before, lumbar puncture may be necessary in order to demonstrate blood or siderophages in the cerebrospinal fluid.
In the acute setting, the vessels are usually imaged using CT angiography (CTA) or digital subtraction angiography (DSA). Many incidental aneurysms are identified using magnetic resonance angiography (MRA).
However, the gold standard is still DSA, in which the aneurysm is displayed directly with a catheter, usually introduced via the femoral access route. Very small aneurysms in particular can be identified in this way with greater certainty. Modern DSA workstations are equipped with solid-state detectors that allow direct digital imaging for three-dimensional rotation angiography or for measuring the aneurysm. Biplanar examination systems that allow simultaneous radiography at two levels are usually used (Fig. 1.3-3). DSA also has the advantage that endovascular treatment can be added or vasospasm can be treated in the same session. In some cases (approximately 10%), an aneurysm is not identified as the bleeding source in patients with clearly confirmed SAB. Two patterns need to be distinguished: in a typical case of perimesencephalic SAB (also known as a prepontine SAB), a little blood is seen on the CT immediately in front of the brainstem/mesencephalon.
Fig. 1.3–2a, b Subarachnoid bleeding in a patient with an aneurysm in the anterior communicating artery. (a) The CT shows subarachnoid bleeding as a hyperdensity in the basal cisterns, with a spherical gap in the position of the anterior communicating branch (arrow). There is a typical reversal of contrast between the cerebrospinal fluid and brain parenchyma. (b) The gap corresponds to the angiographically demonstrated aneurysm in the anterior communicating artery (anteroposterior projection) (arrow).
Fig. 1.3–3a-f Imaging of an aneurysm using three-dimensional techniques in digital subtraction angiography (DSA) and computed-tomographic angiography (CTA). (a) DSA imaging of a middle cerebral artery aneurysm on the right side. (b) CTA of the same aneurysm. (c, d) AP and lateral radiographic images of an aneurysm at the tip of the basilar artery. (e, f) 3D reconstruction of the aneurysm at the tip of the basilar artery before coiling (e) and after (f).
With this type of finding, no aneurysm is identified later either and the follow-up examination can be scheduled at a longer interval after the hemorrhage. When there is extensive SAB and an aneurysm is not identified, a repeat examination should be carried out at an early stage and further examinations should follow at appropriate intervals in order to exclude aneurysm with certainty.
1.3.1.4 Differential diagnosis
An SAB is considered to be excluded if the CT is negative and there is no evidence of erythrocytes on lumbar puncture. Differential-diagnostic considerations should then turn toward other causes of SAB, if there is no evidence of an SAB in connection with craniocerebral trauma.
Other vascular causes of spontaneous SABs include other types of cerebral or spinal vascular malformation (arteriovenous malformations, dural fistulas), vasculitides or dissections of intracranial vessels, rare brain tumors, coagulation disturbances, and thromboses in the venous sinuses.
When SAB occurs in connection with craniocerebral trauma, the following question always needs to be raised: did the accident occur because the patient had an SAB, or was the SAB exclusively a consequence of the trauma? In unclear cases, CTA, MRA, and/or DSA should be carried out.
Larger aneurysms may resemble a contrast-absorbing tumor of the skull base on imaging morphology and may be confused with meningioma or pituitary adenoma. This can usually be unambiguously clarified with modern diagnostic methods.
1.3.1.5 General treatment considerations
Ruptured aneurysms are usually treated in the first 24 hours after the hemorrhage, except when there is an accompanying space-occupying bleeding in which immediate decompression is required. Both treatment options—clipping and coiling—are usually now available in neurovascular centers. While it is beyond the scope of this chapter, decision-making on how to treat the aneurysm is best done in an interdisciplinary discussion that should include several variables: clinical findings, comorbidities, medication, spasm, location, anatomy of the aneurysm neck, and calcification in the aneurysm wall.
In patients with asymptomatic aneurysms, the indication for treatment is not clear and cases need to be considered on an individual basis. There have been no validated studies on this topic. To provide counseling for patients, data are available from large case series and in particular from a quite controversial study, the International Study of Unruptured Intracranial Aneurysms (ISUIA). When the indication is being established, the following criteria need to be taken into account: patient’s age, comorbidities, medication, size and shape of the aneurysm, family history, and history of smoking. Space-occupying aneurysms and what are known as complicated aneurysms require particular individualized consideration. In addition to establishing the indication, the treatment technique—endovascular, surgical, or combined (e.g., with bypass)—also needs to be intensively discussed as well in these cases. New approaches to endovascular treatment have been developed recently (e.g., with flow diverters) and have allowed treatment of complicated aneurysms, which had previously been associated with unsatisfactory results. While treatment considerations include vascular reconstruction or occlusion of the aneurysmal vessel, the aim should be to consider how the space-occupying effect of these usually large and/or fusiform and sometimes thrombosed aneurysms can be relieved.
Conservative treatment
As the clinical findings are usually clear, patients with an SAB are admitted to hospital as emergency cases. These patients should be in an intensive-care ward or, if the clinical findings are less severe, a monitoring ward. In the initial phase of the disease, the patients are at risk of secondary bleeding and possible hydrocephalus. Independently of the treatment for the aneurysm with a clip or coil, conservative treatments should also be started: pain treatment, possible sedation, and prophylaxis against vasospasm are the cornerstones when ventilation is not required. Additional information on this topic follows in the section on intensive-care medicine below.
Endovascular treatment
Endovascular treatment for cranial artery aneurysms has become established since the introduction of Guglielmi detachable coils (GDCs) in the early 1990s as an alternative to neurosurgical clipping operations. The preferred indications for interventional or endovascular procedures are aneurysms in the posterior cranial fossa. All accompanying conditions (advanced age, comorbidities) that may make an open surgical procedure problematic, including a prolonged recovery, tend to favor an endovascular intervention. There are also sequelae of SAB, such as extensive vascular spasm and cerebral edema, which argue against brain surgery. Endovascular approaches are usually more appropriate for aneurysms with its neck located in the bony base of the skull, and aneurysms of the anterior communicating branch that are large (approximately 1 cm in diameter) and with the dome directed dorsally. By contrast, surgery is better for very large, broad-based aneurysms, sometimes including branch orifices in the aneurysmal neck. Many aneurysms can be treated equally well using either technique, particularly as they have both undergone considerable further development in recent years. The following advances and innovations should be mentioned here in connection with the endovascular technique:
Improvement and development of new coils
Considerable improvements in catheters and wire systems
Refinement of the balloon-assisted coil technique
Introduction of highly flexible stents
A general increase in interventionalists’ degree of experience
Without an in-depth analysis of the International Subarachnoid Aneurysm Trial (ISAT), the following conclusions can be made: endovascular treatment has become established as an alternative and complementary form of treatment to surgery for cerebral artery aneurysms. A modern center for neurovascular treatment should have both capacities available. The prospective and randomized ISAT study showed a poor outcome (modified Rankin scale > 2) in approximately 24% of patients who received endovascular treatment, in comparison with 31% of those who underwent surgery for ruptured intracranial aneurysms. Generally speaking, the advantages of endovascular therapy lie in the fact that the intervention is less invasive, and this usually allows more rapid mobilization of the patient. The general disadvantage, particularly with large, broad-based aneurysms, is that recanalization is possible as a result of aggregation of the platinum coils and that it may not be possible to achieve complete occlusion of the aneurysm in this anatomic situation. It is not clear whether such recanalization is associated with a relevant risk of bleeding. In general, the rate of secondary bleeding from aneurysms after SAB is extremely low both after surgery and after coiling, as demonstrated in the ISAT study.
Fig. 1.3–4a-f Endovascular treatment of an aneurysm at the tip of the basilar artery. (a-f) DSA and road map images of the several steps involved in endovascular treatment of an aneurysm of the basilar artery tip. (a) AP projection before treatment. (b) Introduction of a catheter into the aneurysm. (c) During application of the first coil. (d, e) Imaging of the next treatment steps. (f) Final image after occlusion of the aneurysm with coils.
Endovascular treatment for an aneurysm (known as “coiling,” Fig. 1.3-4) can be described in outline as follows:
Femoral arterial access with a 6F catheter system.
Imaging of the aneurysm using 3D rotation angiography.
Establishment of a suitable treatment position.
Probing with a microwire and microcatheter.
Selection of a coil with the appropriate size and length relative to the morphology of the aneurysm.
After introduction of the first basket coil, the basket is filled with spirals of a suitable size and shape and the appropriate stiffness until no further flow into the aneurysm can be seen.
In most centers, the interventions are carried out with the patient under general anesthesia and with heparin prophylaxis. Premedication with platelet inhibitors is increasingly being recommended for incidental aneurysms.
Stents and balloons can support the treatment as permanent or temporary implants to protect the aneurysm-bearing vessel.
The typical complications of endovascular treatment consist of thrombus formation in the aneurysmal neck, with subsequent infarction, and perforation of the aneurysm. In experienced hands, both of these complications now occur only extremely rarely and they have also been minimized with premedication for the patients and as a result of improved devices. The combined mortality and morbidity rate in experienced centers is now at a low single-figure percentage.
In most centers, follow-up examinations with catheter angiography are carried out after 3 months or 6 months, and then with MRA at 1-year or 2-year intervals. Secondary bleeding is unusual with coiled aneurysms, but approximately 10% of these aneurysms receive secondary treatment when there is asymptomatic recanalization, in order to ensure long-term protection against bleeding.
Surgical treatment
Definitive occlusion of the aneurysm using a vascular clip is carried out with neurosurgery in order to prevent repeat bleeding. The timing of the operation depends on the patient's condition, on the interval between the SAB and diagnosis, and on the extent of vasospasm. To forestall repeat bleeding and provide thorough treatment for vasospasm, an attempt is usually made to obliterate the aneurysm as early as possible.
Space-occupying intracranial bleeding is an absolute indication for surgical treatment of an aneurysm. Relative indications include broad-based aneurysms (usually of the middle cerebral artery) and those that can only be treated interventionally with difficulty or at high risk. In patients with broad-based aneurysms or giant aneurysms, the vessel can be reconstructed by placing multiple aneurysm clips. Concomitant medical conditions may argue in favor of surgical treatment in cases of renal insufficiency and hyperthyroidism, and poor general condition due to various comorbidities is an argument against surgical therapy.
For achieving a definitive occlusion, the surgical approach is the most effective treatment method; a surgical vascular clip is placed directly on the neck of the aneurysm above the vessel in order to permanently occlude the aneurysm without impairing blood flow in the affected vessel (Fig. 1.3-5). Particular attention should be given here to ensure that blood flow definitely remains unaffected by the clip, both in the vessel bearing the aneurysm and also in vascular branches that originate from it. Intraoperative fluorescence angiography with indocyanine green (ICG), a fluorescent dye, allows reliable checking of blood flow after clip application. Any intracranial hematoma that is present is removed in the same session. An external ventricular drain is often placed before the actual occlusion of the aneurysm in order to treat coexistent hydrocephalus and relieve intracranial pressure. In highly selected cases, surgical placement of an extracranial-intracranial bypass may be indicated in advance of the neurosurgical aneurysmal occlusion.
Fig. 1.3–5a-f (a, b) Status after surgical treatment for an aneurysm of the anterior communicating artery (see Fig. 1.3-10 for the preoperative image) and arteriovenous malformation (AVM) in the left frontal area. (c, d) Appearance after endovascular treatment of the AVM. (e, f) Appearance after surgical removal of the AVM.
The surgical mortality rate for unruptured aneurysms is approximately 2%, with a morbidity rate of around 7%. These complication rates increase dramatically with giant aneurysms. Following SAB, the outcome depends to a great extent on the patient’s initial clinical state. The poorest postoperative results can be expected in patients who start with a Hunt and Hess grade of 4 or 5. The poor clinical outcome here is due to the preoperative condition and is not necessarily specific to the surgical procedure.
The course and prognosis after any form of treatment for aneurysms are affected by the following variables: secondary bleeding before occlusion of the aneurysm (5-10%), treatment-related complications (4-8%), occurrence of vasospasm-related infarction (27-31%), malabsorptive hydrocephalus (10-45%), and medical complications (10-15%). The long-term course is determined by the extent of neurological losses, as well as neuropsychological deficits and psychosocial environment.
Intensive-care medicine
Patients with acute intracerebral or subarachnoid bleeding due to intracranial aneurysms, arteriovenous malformations, and other vascular malformations should receive intensive-care treatment or monitoring until their vital parameters have stabilized until there is no further acute risk of secondary bleeding or vasospasm. In addition to the specific therapy for these patients in collaboration among neurologists, neurosurgeons, neuroradiologists, and intensive-care specialists, excellent basic intensive care is decisive for the prognosis. The intensive care includes optimization of respiration and oxygenation, hemodynamics, and of the blood glucose level, fluid and electrolyte balance.
Respiratory therapy. The aim should be to achieve adequate oxygenation of arterial blood (Sao2 > 92%), which can be critically important for metabolism in the critically perfused brain tissue. If adequate oxygenation cannot be achieved by supplying oxygen via a nasal probe or face mask, or if the patient has limited protective reflexes due to the bleeding, or a pathological respiratory pattern, he or she should be intubated and given controlled ventilation.
Cardiac treatment. Particularly after subarachnoid bleeding, cardiac arrhythmia and changes on ECG that meet the criteria for acute myocardial infarction are not rare, even in patients who do not have coronary heart disease. Cardiac enzymes may also be raised in these patients. Close monitoring during the first days after the bleeding is therefore obligatory in these patients.
Blood-pressure adjustment. Optimizing cardiac output, with systemic blood-pressure values in the high normal range, is also important. The patient should receive adequate volume substitution, and additional administration of catecholamines may also be required. Blood-pressure spikes must be fastidiously avoided in patients who have suffered subarachnoid bleeding with an aneurysm that has not yet been treated. However, drastic drops in blood pressure must also be avoided in these patients.
Glucose metabolism. In patients with acute cerebral pathology, hyperglycemia is an independent risk factor for a poor outcome. The raised glucose values that are often found in these patients should not be regarded merely as a stress response by the body but should be actively treated.
Control of fluid and electrolyte metabolism. Both disturbances of fluid metabolism and severe disturbances of electrolyte homeostasis are frequently observed after intracerebral and subarachnoid hemorrhage. In patients with acute cerebral pathology, hypernatremia may occur in cases of diabetes insipidus, as well as hyponatremia in patients with syndrome of inappropriate secretion of antidiuretic hormone (SIADH) or cerebral salt-losing nephritis.
Prophylaxis and treatment for vasospasm after SAB. The calcium antagonist nimodipine (Nimotop®) significantly reduces the risk of secondary neurological injury triggered by vasospasms after SAB and it should therefore be administered prophylactically. Adequate data are only available for oral administration of nimodipine (60 mg p.o. every 4 h, daily dosage 360 mg in all patients after day of admission for approximately 20 days). In patients who are unconscious and those with unclear enteral absorption, nimodipine can be started i.v. at a dosage of 1 mg/h (5 mL/h) in the first 6 h and after blood-pressure controls can be initially raised to 1.5 mg/h and after a further 6 h to the maintenance dosage of 2 mg/h (10 mL/h). Ensuring adequate (130–150 mmHg systolic) and stable blood pressure takes priority over nimodipine administration. Other pharmacological approaches for prophylaxis against cerebral vasospasm after SAB, such as administration of magnesium sulfate, statins, and endothelin receptor antagonists are still controversial. Hypertensive hypervolemic hemodilution (triple-H therapy) can be carried out when vasospasms are present after SAB, although the value of this form of treatment has so far not been demonstrated in larger randomized studies. For triple-H therapy, blood pressure and blood volume are raised by administering crystalloid or colloid infusion solutions and catecholamines. Triple-H therapy is not recommended for prophylactic treatment against cerebral vasospasms.
1.3.1.6 Special types of aneurysm
As mentioned initially, there are also special types of aneurysm—i.e., vascular abnormalities with a specific pathogenesis that does not fit the contexts described above. These special aneurysms may be caused by:
Connective-tissue diseases, arteriosclerosis, and hypertension (dilatory arteriopathy)
Inflammatory emboli
Dissections
Trauma and iatrogenic injury
Connective-tissue diseases, arteriosclerosis and hypertension, alone or in combination, lead to various degrees of dilatory arteriopathy that may lead to massive vascular changes.
The most frequent of these are large fusiform aneurysms—i.e., long, spindle-shaped dilation of the basal cranial arteries. The basilar artery is often affected (megadolichobasilar artery), or more rarely the internal carotid artery or middle cerebral artery (Fig. 1.3-6). These changes usually become apparent clinically as a result of thromboembolic ischemia or neurological deficits due to local pressure damage, or rarely through rupture and intracranial bleeding. Treatment therefore usually consists of drug-based secondary prophylaxis. Surgical and endovascular treatment approaches are associated with a high level of procedural risk. They are usually attempted if local pressure damage has led to disability. Recently, flow diverters—extremely closely-woven stents—have been used in these conditions in order to modulate flow in the diseased vessel and eventually seal off the aneurysm.
The old term “mycotic aneurysms” (Fig. 1.3-7) covers aneurysms with an infectious and embolic pathogenesis. These are usually located on peripheral branches of the intracranial arteries and may be either saccular or fusiform. Treatment is indicated in order to prevent intracranial bleeding. If selective occlusion of the aneurysm is not possible, the affected vessel has to be occluded endovascularly or surgically together with the aneurysm.
Dissecting aneurysms occur intracranially only rarely. Due to the narrow vascular caliber, dissection as the cause of this type of aneurysm can usually not be confirmed on MRI and it can only be postulated. Here again, the location and shape of the aneurysm suggest the diagnosis. Morphologically, these aneurysms tend to be fusiform rather than saccular. The last segment of the vertebral artery (V4) is frequently affected. In this location, endovascular occlusion is the treatment of choice if the contralateral vertebral artery is adequate. In other cases (Fig. 1.3-8), treatment with a flow diverter or stent may be considered.
Traumatic intracranial aneurysms are also rare. The “trauma” usually involves iatrogenic injury to the internal carotid artery near the base of the skull, or more rarely a skull base fracture (Fig. 1.3-9). As these aneurysms may cause life-threatening bleeding intracranially or into the nasopharynx, they should be identified and treated at an early stage. Due to their position inside the skull base, surgical treatment is not usually possible and an endovascular approach is preferable. Here again, vascular occlusion is the treatment of choice when there is an adequate collateral supply, especially if endovascular reconstruction of the vascular lumen is not possible. As mentioned above, high-density and conventional neurostents are available for the purpose. This group of aneurysms also includes aneurysms of the internal carotid artery that rupture into the cavernous sinus, as well as iatrogenic vascular injuries—e.g., during pituitary surgery.
Fig. 1.3–6a-e A fusiform aneurysm in the basilar artery on digital subtraction angiography (a, b) and on a magnetization-prepared rapid gradientecho (MP-RAGE) sequence after intravenous contrast administration, in three spatial directions (c-e).
Fig. 1.3–7a, b “Mycotic aneurysm.” Peripheral fusiform aneurysm in the left middle cerebral artery; a “mycotic aneurysm” may have this appearance.
Fig. 1.3–8a-f Dissection-related aneurysm. (a) Digital subtraction angiography (DSA) of the left vertebral artery after subarachnoid bleeding, with evidence of slight wall irregularities. (b) At the follow-up examination, magnetic resonance angiography reveals dissection of the left vertebral artery. (c, d) A fusiform aneurysm at the level of the dissection, on DSA with 3D reconstruction and subtraction. (e, f) Findings after implantation of a flow diverter (f); 3D reconstruction of the check-up DSA (e).
1.3.2 Arteriovenous malformations and dural fistulas
1.3.2.1 Clinical picture in AVM
Vascular malformations in the brain and meninges are pathological shunts between the afferent arteries to the brain or dura and the efferent veins or venous sinuses.
AVMs are located in the cerebral parenchyma and are primarily supplied by afferent cerebral vessels; dural vessels may be recruited secondarily via transdural anastomoses. Drainage into the large venous sinuses takes place via parenchymal veins. The shunt (nidus) is located in the brain parenchyma and may have a plexiform, fistulous, or mixed structure. In the fistulous type, a very strong afferent artery opens directly into a vein without an intermediate vascular plexus. AVMs occur much more rarely than aneurysms. They are congenital vascular malformations with a tendency to grow larger during the course of life. They can occur in any location in the brain. They are usually classified using the criteria included in the Spetzler and Martin score, which was developed in order to assess operability, with the risk being classified relative to the size, location and type of venous drainage (Table 1.3-3).
Fig. 1.3–9a, b Traumatic (false) aneurysm. (a) Fractures in the middle and anterior cranial fossa (arrowheads) were diagnosed on the right side on a CT 6 months previously. (b) Due to recurrent epistaxis, a “false” aneurysm on the right internal carotid artery was in the meantime occluded with coils. One month later, massive bleeding from the nasopharynx occurred. The DSA shows the aneurysm, with extravasation into the sphenoid sinus (black arrow). The coils can be seen at the lower edge (white arrow). The internal carotid artery was occluded with additional coils.
Table 1.3–3 Spetzler-Martin grading of surgical risk in arteriovenous malformations (1 point = no risk, 5 points = high risk).
| Size Small | (< 3 cm) | 1 point |
| Medium (3–6 cm) | 2 points | |
| Large(> 6 cm) | 3 points | |
| Site | Not “eloquent” | 0 points |
| “Eloquent” | 1 point | |
| Venous drainage | Superficial veins | 0 points |
| Deep cerebral veins | 1 point |
1.3.2.2 Clinical findings in AVM
AVMs usually become clinically manifest in middle age. The bleeding risk is estimated at approximately 2–4% per year. The leading diagnostic syndrome is intracranial bleeding (ICB; 50%), followed by headache and seizures at almost equal frequency. Neurological deficits without ICB as the first symptom are rare with these malformations. In children, ICB is the initial symptom more often. In contrast to aneurysmal bleeding, the course of bleeding from an AVM is clinically less severe and the risk of early secondary bleeding is much lower. Seizures occur particularly with large AVMs in supratentorial and cortical locations. Headache often occurs when the external vessels are involved. The clinical findings in patients with AVM are much more heterogeneous than in those with aneurysms and they depend much more on the size and position of the AVMs.
1.3.2.3 Diagnosis of AVMs
Starting from a size of at least 1 cm in diameter, there is no difficulty in diagnosing an AVM. Modern tomographic imaging procedures are all able to detect the lesions. A node of convoluted vessels with blood flowing through them is usually found, along with one or several large veins and hypertrophic afferent vessels. On CT, AVMs may show dilated vessels, calcification and hemorrhage. After contrast administration, there is strong enhancement of the vascular structures. MRI shows pathological vascular structures even before contrast administration (Fig. 1.3-10), particularly the angioma nidus and dilated efferent veins. For treatment planning, however, DSA with complete intracranial angiography is necessary, even after previous CTA or MRA. DSA not only identifies the afferent and efferent vessels in the vascular malformation, but also delineates the normal vascular supply to the brain, allowing assessment of the extent of the arteriovenous shunt and detection of aneurysms and stenoses in the vessels involved. In addition, DSA makes it possible to classify the malformation, and important treatment decisions usually depend on this. When there is spontaneous intracerebral bleeding, the following criteria should suggest a vascular malformation as the bleeding source: an “atypical” bleeding location (i.e., not in the white matter, basal ganglions, or cerebellum), age under 50, and no risk factors present (such as hypertension, coagulation disturbances, or amyloid angiopathy). Smaller AVMs showing bleeding or thrombosis can sometimes not be demonstrated during the initial diagnosis when there is intracerebral bleeding. In such cases, repeating CTA, MRA, or DSA after blood resorption can be recommended.
1.3.2.4 Indication for invasive therapy for AVMs
There is no conservative treatment approach for AVMs. Only supportive treatment, with anticonvulsant medication and symptomatic treatment for possible headache, is possible. A direct indication for invasive treatment is present when bleeding has taken place. The indication for invasive treatment in unruptured AVMs is a matter of debate.
1.3.2.5 General treatment considerations
There are three invasive treatment approaches for intracranial AVMs, which in principle can be used either alone or in combination in the framework of a multimodal concept. The decision for or against the various approaches is heavily dependent on local availability and expertise. In principle, endovascular occlusion, surgical removal, and stereotactic radiotherapy are available. The aim in all these procedures is to definitively eliminate the arteriovenous shunt in the AVM. With all of them, the difficulty of achieving this goal increases relative to the size and location of the AVM. It is certainly true to say “anyone can do the small ones.” The neuroradiologist can occlude the vessels using embolization; the surgeon can expose the AVM, identify all of the arterial afferents and can remove the AVM completely; and the radiotherapist can induce sclerosis of the AVM vessels. Most medical centers will use a multimodal approach in which the first step consists of endovascular reduction in the size of the AVM, followed by removal of operable AVMs and radiotherapy for AVMs in inoperable locations. Endovascular treatment can only lead to obliteration of an AVM to a limited extent. Depending on the location, radiotherapy or surgery are thus usually required. The advantage of radiotherapy is the lack of direct invasiveness, but the disadvantage is that the full effect of the treatment only follows after a latency period of 2-3 years and success rates of only 80-90% are possible, whereas incomplete surgical removal of AVMs tends to be the exception.
Fig. 1.3–10a-c MRI and DSA in the patient shown in Fig. 1.3-5 before treatment of either the aneurysm or the AVM. (a) T2-weighted and (b) T1-weighted images after intravenous contrast administration, showing the AVM (white arrow) and the aneurysm (black arrow). (c) DSA showing the aneurysm (white arrow) and the AVM in AP and lateral projections.
1.3.2.6 Endovascular treatment
In principle, endovascular treatment for AVMs is carried out in the same way as for aneurysms. However, the catheter materials are slimmer and different embolic agents are used. Corpuscular and liquid embolic agents are used.
Particulate embolic agents:
Polyvinyl alcohol particles, etc.
Very slim coils for injection
Liquid embolic agents:
Ethylene vinyl alcohol copolymer (Onyx™)
Acrylates (Histoacryl™, Glubran™)
The materials differ with regard to the level of occlusion (capillary or precapillary), the duration of the occlusion, and in their physical properties. For example, acrylates as a liquid embolic agent create a capillary occlusion that is permanent and seals the vessels using a polymerization process. Particulate embolic agents create a precapillary occlusion that is not permanent. Onyx™ behaves like acrylate at the occlusion level, but it is not a glue and has other different properties. With all of the embolic agents, the initial aim is to advance the catheter as close as possible to the AVM. When embolizing with Onyx™, it is best to position the catheter tip intranidally so that the nidus can be filled from the center. The aim of embolization is to achieve compact occlusion of all the angioma structures (Fig. 1.3-11).
Fig. 1.3–11a-d Endovascular treatment of an AVM. (a) Superselective imaging of the AVM via the posterior cerebral artery. (b) Positioning of the microcatheter in the posterior cerebral artery. (c, d) Step-by-step embolization of the AVM.
1.3.2.7 Surgical treatment
The indications for surgical removal of an AVM are prior bleeding, difficult-to-treat seizures, and prophylaxis against cerebral hemorrhage (Fig. 1.3-5). In small and easily accessible AVMs (Spetzler grades 1-3), this is the method of choice for complete obliteration, and it can often be done without prior interventional treatment. However, surgery or radiotherapy can also be carried out at lower risk after previous—and if necessary multiple—sessions of interventional treatment. In large AVMs (Spetzler grades 4–5), the risk of postoperative neurological deficits needs to be weighed against the natural course (influenced by variables such as the bleeding risk and the patient’s age and condition). Treatment is therefore contra-indicated in older patients with multimorbid conditions with no history of bleeding but with extensive AVMs occupying large parts of the cerebral hemisphere. The view has become generally accepted in recent years that Spetzler–Martin grade 5 AVMs should only be treated in exceptional cases.
The aim of the operation is to excise the AVM completely. Partial removal is not useful and even increases the risk of bleeding. During the operation, all of the vessels leading to the AVM are initially coagulated and transected until finally the AVM is dissected free in a circular fashion. The mobilized AVM is then removed at the efferent vessel. Coexisting hemorrhage is controlled during the operation. Neuronavigation and intraoperative fluorescence angiography can be used as technical aids. The risks and complications depend on the Spetzler grade, with a global mortality rate of 1–5% and morbidity rate of 2–20%. If the postoperative control angiography shows a residual AVM, it must be removed promptly in a subsequent operation.
1.3.2.8 Radiotherapy
As mentioned above, radiotherapy is usually carried out when AVMs are in an inoperable location. Following the appropriate planning, the treatment is carried out using a linear accelerator or gamma knife with a stereotactic technique. One disadvantage of radiotherapy is the long latency period until the onset of effect—1–3 years are usually required before complete thrombosis takes place. In long-term follow-up, the rate of complete occlusions is much lower compared with surgery. The size of the AVM or residual AVM is a limitation for radiotherapy: good results are achieved with AVMs < 3 cm in size. Undirected and noncompact embolization before radiotherapy can actually have a negative effect on the treatment results. Embolization before radiotherapy is only useful if it compactly obliterates a defined part of the nidus.
1.3.2.9 Clinical picture in DAVFs
Dural arteriovenous fistulas (DAVFs) are acquired arteriovenous shunts on the wall of the dural sinus. In contrast to intracerebral pial AVMs, dural fistulas are mainly supplied by dural branches of the afferent cerebral vessels (“dural AVMs”). Pathophysiologically, it is assumed that there is a prior thrombosis in the affected sinus that has recanalized “incorrectly”—i.e., when the body attempts to recanalize the occluded vessel, arteries sprout into the sinus wall and the fistula arises when the sinus reopens.
The age at manifestation is 40–60 years. The symptoms depend on the location and type of venous outflow. A basic distinction should be made between three groups: fistulas at the large sinuses (transverse sinus and superior sagittal sinus); tentorial fistulas and ethmoidal fistulas; and thirdly, cavernous sinus fistulas. The Cognard classification (Table 1.3-4) grades dural fistulas in relation to the type of venous outflow, from which the risk of bleeding is inferred depending on whether or not there is reflux of arterial blood into the cerebral veins. Such reflux is usually associated with obstruction of regular venous outflow, with stenosis or occlusion of the sinuses.
Tentorial and ethmoidal fistulas represent a rare and special form where the arterial supply and forms of venous drainage need to be managed on a case by case basis, since this type of DAVF is associated with a higher risk of bleeding.
The Barrow classification (Table 1.3-5) grades cavernous sinus fistulas in relation to their arterial afferents. Type A is a special form in which the pathogenesis mentioned above in connection with sinus thrombosis does not apply. Barrow type A fistulas arise as a result of rupture of an aneurysm in the cavernous course of the internal carotid artery or injury to the vessel in that area. They are also referred to as direct fistulas, in contrast to indirect fistulas, which are caused by the thrombosis described above.
Table 1.3–4 Cognard classification of dural arteriovenous fistulas relative to venous drainage (orthograde: no symptoms or only slight symptoms; reflux: raised intracranial pressure, bleeding risk 10% in type IIb, 40% in type III, 65% in type IV).
| Type I | Orthograde drainage into the sinus |
| Type II | Retrograde drainage (reflux) |
| Type IIa | Reflux into the sinus |
| Type IIb | Reflux into cortical veins |
| Type IIc | Reflux into the sinus and cortical veins |
| Type III | Direct drainage into cortical veins |
| Type IV | Drainage into cortical veins with venous ectasia |
| Type V | Reflux into spinal veins |
Table 1.3–5 Barrow classification of cavernous sinus fistulas.
1.3.2.10 Clinical findings in DAVFs
In dural fistulas, the symptoms depend on the fistula’s position and the extent to which venous drainage is compromised. Tinnitus and a pulse-synchronous bruit occur with lateral dural fistulas, and visual disturbances or double vision with cavernous sinus fistulas. Headache, hydrocephalus, neurological deficits, and even dementia have been attributed to chronic venous hypertension due to disturbance of intracranial venous drainage. However, the most severe clinical manifestation is spontaneous intracranial bleeding. As mentioned above, in dural fistulas the risk of spontaneous intracranial bleeding depends on the extent and type of venous reflux intracranially, as well as on the location of the fistula.
1.3.2.11 Diagnosis of DAVFs
Dural fistulas are rarely asymptomatic, incidental findings; as with cavernous sinus fistulas, it is more often the case the clinical findings tend to be incorrectly assessed. The typical situation is a patient with a pulse-synchronous bruit behind the ear that stops when the occipital artery is compressed, or which is position-dependent. The clinical examination already points the way here. DSA is the method of choice for diagnosis and classification. With cavernous sinus fistulas, the emphasis is on ocular symptoms, the leading ones being exophthalmos, chemosis, ciliary injection and pareses of the extraocular muscles.
Fig. 1.3–12a-c Dural fistula in the transverse sinus (MRA and DSA). (a) The T1-weighted MRI after intravenous contrast administration shows extensive vascularization in the left occipital lobe, with an urgent suspicion of cortical drainage in the presence of a dural arteriovenous fistula. (b) The DSA shows a view over the left common carotid artery, with confirmation of the suspected fistula on the left transverse sinus and extensive cortical drainage (arrows). (c) Appearance after occlusion of the fistula site in the transverse sinus using platinum coils.
If vision is impaired as a result of venous drainage disturbances, there is an urgent indication for treatment. CT or MRI of the orbit can often reveal dilation of the superior ophthalmic vein. Here as well, however, DSA is the method of choice for diagnosis and classification. If the patients develop conspicuous symptoms as a result of intracranial bleeding or symptoms resembling infarction, a differential diagnosis of dural fistula needs to be considered on the basis of CT and/or MRI. Indicative signs here may include visible pathological vessels, signs of congestion in unusual locations such as the edge of the tentorium, or multiple collaterals, which can be well imaged using time-of-flight magnetic resonance angiography (TOF-MRA) (Fig. 1.3-12).
1.3.2.12 Indication for invasive treatment of DAVFs
Invasive therapy is indicated:
When there are spontaneous fistulas at the transverse sinus and superior sagittal sinus in the venous drainage pattern, treatment is indicated when there is drainage into cortical veins. If there is no cortical drainage, the extent to which the patient is affected by the bruit in the ear may be an indication for therapy.
Tentorial and ethmoidal fistulas represent an increased bleeding risk.
In cavernous sinus fistulas with ocular symptoms, cosmetic effects and pareses of the eye muscles. Deteriorating vision represents an emergency indication.
1.3.2.13 General treatment considerations in DAVFs
In the most frequent dural arteriovenous fistulas—in the transverse sinus and cavernous sinus—treatment considerations are concerned with locating the fistulous segment on the wall of the sinus. Transvenous treatment of the fistula-bearing segment is the most widely used form of endovascular therapy. It is based on the assumption that this sinus segment is no longer required for normal drainage of the brain or eye. This is usually not critical in treatment of the cavernous sinus, but when fistulas of the transverse sinus are being treated, the anatomy of cerebral venous drainage needs to be analyzed carefully. All of the cerebral vessels have to be imaged with a long venous phase in order to delineate the anatomy of the intratentorial and supratentorial veins. Once the decision has been taken that the fistula-bearing segment can be occluded, then the location up to which this is to be carried out also has to be carefully assessed. Sinus segments that take up blood from the brain and are required for drainage must not be closed. Usually, coils are used to occlude the sinus. Recent developments have been based on a different approach: the aim is to preserve the sinus by inflating a balloon at a suitable point and then, via an arterial access route, injecting a liquid embolic agent that is modeled to the shape of the sinus wall by the balloon.
Surgical therapy and treatment of dural fistulas using stereotactic radiotherapy are unusual and are not discussed here in greater detail. Spontaneous healing has only rarely been reported with AVMs, but it does occasionally occur with dural fistulas. For example, manual compression of the occipital artery in patients with lateral dural fistulas and mild symptoms can be carried out in an attempt to reduce the shunt or even occlude the fistula by thrombosis. This procedure can also be used to achieve complete occlusion of fistulas that have already been treated but show a slight degree of residual flow.
1.3.2.14 Endovascular therapy for DAVFs
Tentorial and ethmoidal DAVFs
Tentorial and ethmoidal DAVFs are usually treated with arterial injection of liquid embolic agents in one or several sessions. This diverges from the considerations discussed above, because these fistulas are typically located in a circumscribed site and transvenous access is usually not possible due to the anatomic conditions. A surgical approach should also always be considered for these fistulas, and ethmoidal fistulas in particular are well accessible for surgical treatment.
Fig. 1.3–13a-c Endovascular treatment of a traumatic fistula in the cavernous sinus. (a) DSA of the internal carotid artery, showing a fistula between the internal carotid artery and the cavernous sinus. (b) DSA after occlusion of the fistula with a detachable balloon (arrow). (c) Enlarged image of the balloon after opening at the fistula site in the cavernous sinus outside of the internal carotid artery (arrow).
Cavernous sinus fistulas (spontaneous, indirect, Barrow C-D)
These lie in the domain of endovascular therapy. In most cases, the fistula-bearing segment of the cavernous sinus is probed via the inferior petrosal sinus. If this is not possible, access via the facial vein is possible, with or without surgical exposure. The sinus is occluded with coils after arterial demonstration of the fistula; to secure the eye against any recurrence, the initial segment of the orbital veins should also be occluded.
Cavernous sinus fistulas (trauma or aneurysmal rupture, direct, Barrow A)
The fistulous connection consists of a hole in the internal carotid artery resulting from rupture of an aneurysm at that location, or due to direct injury. Endovascular therapy involves occluding the hole by coiling the aneurysm or using a detachable balloon that is opened at the rupture site outside of the carotid lumen, i.e. in the cavernous sinus, and released. Techniques involving stents and liquid embolic agents have also been described for the treatment of cavernous sinus fistulas, but are not discussed further here.
Fig. 1.3–14a-d Endovascular treatment of a spontaneous fistula in the left cavernous sinus. Angiographic analysis of the fistula using an injection into potentially afferent vessels. The fistula in the left cavernous sinus and intercavernous sinus is primarily supplied by the left external carotid artery. From the right, vessels from the external and internal carotid arteries are also involved (Barrow type D fistula). The vertical images (a) demonstrate the supply of the fistula from the carotid flow area from the right (right column). The image at the top left shows the afferent supply to the fistula from the right external carotid artery. Gaps in the cavernous sinus, which mark the course of the internal carotid artery, are clearly recognizable. The next six images (b) demonstrate the supply to the fistula from the external carotid artery, with the first four in a lateral projection with contrasting of the right superior ophthalmic artery in the first image in the lower row. The two last images in the lower row again show the cavernous sinus with gaps in the internal carotid artery (arrow). The following images (c) show access of the ophthalmic artery with a microcatheter, from top left to bottom right. The last five images (d) illustrate treatment using platinum coils; top row: filling of the ophthalmic vein and of the cavernous sinus; middle left: filling of the intercavernous sinus. The last two images show complete occlusion of the fistula.
Transverse sinus and superior sagittal sinus fistulas
As mentioned above, these are usually treated endovascularly, but there are various techniques (see above) for closing the shunts in the sinus wall (see section 1.3.2.6). In addition to the methods mentioned, there is also a technique in which the fistulas in the sinus wall are probed directly from a transvenous or transarterial access route. Techniques involving releasing stents into the sinus have also been described.
1.3.3 Occult (“low-flow”) cerebral vascular malformations
These are vascular malformations involving slow flow. There is no arteriovenous shunt. It is often not possible to demonstrate these malformations on angiography.
1.3.3.1 Cavernous hemangioma (cavernoma)
Cavernomas are congenital or acquired intracerebral vascular malformations with a cavernous venous structure that lie encapsulated in the cerebral parenchyma. Cavernomas are found as cerebral manifestations in approximately 10% of cases in patients with hereditary hemorrhagic telangiectasia (Rendu–Osler–Weber syndrome). In the familial form, there are usually multiple lesions. Seizures are the most frequent clinical manifestation, followed by neurological deficits due to spontaneous bleeding or space-occupying lesions. The risk of bleeding is approximately 2–3% per year, but it is assessed variously depending on whether or not clinically unremarkable blood deposits in the cavernoma are regarded as bleeding. Cavernomas may occur in combination with other vascular malformations, and are most often associated with venous malformation.
MRI is the examination method of choice, and it shows circumscribed areas with signal changes, which may resemble acute bleeding, fresh thrombosis, older bleeding, or only small punctate hemosiderin deposits, depending on the stage. A typical finding is a “popcorn-like” center with areas of bright and dark signal intensity, surrounded by a dark hemosiderin border (Fig. 1.3-15). CT only reveals larger cavernomas. Angiography is usually negative (cryptic or occult malformation).
There are no options for conservative treatment. Clinically silent cavernomas usually only receive observation. There is no endovascular treatment procedure. There is an indication for neurosurgical removal if the cavernoma increases in size or when there are focal neurological deficits, symptomatic bleeding, or seizures. Depending on the location, surgical removal may be considered in younger patients for prophylaxis to prevent bleeding. The lesion is removed in toto, and this can be safely accomplished even in critical locations such as the brainstem.
Fig. 1.3–15a, b Cavernoma. (a) The axial T2-weighted MRI sequences show a focal lesion in a deep parietal location on the left alongside the lateral ventricle, with a center formed by bright and dark points (“popcorn”), surrounded by a dark ring (hemosiderin deposits). (b) The hemosiderin-sensitive T2* gradient echo sequences, now with a coronal section, confirm the presence of blood breakdown products as dark signal losses.
1.3.3.2 Capillary telangiectasia
Capillary telangiectasia is another intracerebral vascular malformation with a cavernous venous structure. In contrast to cavernomas, however, the vessels are diffusely embedded in the cerebral parenchyma. These malformations are usually observed incidentally in the brainstem, thalamus, or basal ganglia during MRI examinations. There are no clinical symptoms associated with them. In individual patients with intracerebral bleeding, it has been debated whether this finding is coincidental or whether it occurs in combination with another vascular malformation that is not visible on imaging, such as a cavernoma or AVM.
On MRI, T1-weighted and T2-weighted images are usually normal, or only show discrete signal changes. Hemosiderin-sensitive sequences show slight signal fading, but not as marked as in cavernoma. After contrast administration, there is signal enhancement of small, striate vessels or of an extensive area, with no perifocal edema and no space-occupying characteristics (Fig. 1.3-16). The enhancement already declines again in late images—in contrast to most focal inflammatory or tumor findings. There is no indication for treatment.
Fig. 1.3–16a, b Telangiectasia. (a) In a young woman with formication in the left arm, the hemosiderin-sensitive T2* gradient echo sequence on MRI shows flat signal fading in the left paramedian area in the pons, much weaker than in a cavernoma (see Fig. 1.3-12c). The T1-weighted and T2-weighted images were normal. (b) On the T1-weighted sequences after contrast administration, there is circumscribed enhancement of small vascular structures in this area, compatible with capillary telangiectasia as an incidental finding in the examination.
Fig. 1.3–17a, b Venous angioma or developmental venous anomaly (DVA). (a) Digital subtraction angiography identifies a venous malformation in the cerebellum solely in the venous phase, with small afferent veins (“caput medusae”) and an efferent collecting vein (arrow). (b) Magnetic resonance imaging (T1-weighted sequence) after contrast administration shows a similar finding, with the small afferent veins contrasted (arrow). The collecting vein was imaged on adjacent sections (not shown).
1.3.3.3 Venous malformations (venous angioma)
Venous malformations are very often seen on MRI images. These represent a circumscribed persistent embryonic venous system consisting of small, spider-burst veins that flow into a dilated collecting vein (“caput medusae”). The term “developmental venous anomaly” (DVA) has therefore also been introduced to describe this type of vascular malformation. They often occur in combination with cavernomas, a radiographic search for which should therefore be carried out. Venous malformations usually have no clinical significance. An association with seizures or intracranial bleeding is more likely to be explained by small, radiographically undiagnosed cavernomas.
Venous malformations can be demonstrated angiographically. However, the venous structures are only first filled with contrast in the venous phase, as compared to arteriovenous malformations with early arteriovenous shunts. The MRI appearance is also typical, with small, confluent spider-burst veins and larger, efferent collecting veins becoming visible after contrast administration (Fig. 1.3-17). Treatment is not necessary and might be dangerous, as DVAs drain normal brain tissue.
References
Barrow DL, Spector RH, Braun IF et al. Classification and treatment of spontaneous carotid-cavernous fistulas. J Neurosurg 1985; 62: 248–56.
Barth M, Capelle HH, Weidauer S, et al. Effect of nicardipine prolonged-release implants on cerebral vasospasm and clinical outcome after severe aneurysmal subarachnoid hemorrhage: a prospective, randomized, double-blind phase IIa study. Stroke 2007; 38: 330–6.
Cognard C, Gobin Y, Pierot L et al. Cerebral dural arteriovenous fistulas: clinical and angiographic correlation with a revised classification of venous drainage. Radiology 1995; 194: 671–80.
Fisher C, Kistler J, Davis J. Relation of cerebral vasospasm to subarachnoid hemorrhage visualized by computerized tomographic scanning. Neurosurgery 1980; 6 (1): 1–9.
Forsting M, Wanke I. Intracranial vascular malformations and aneurysms. 2nd revised edition. Springer, Heidelberg, 2008.
Guglielmi G, Vinuela F, Dion J et al. Electrothrombosis of saccular aneurysms via endovascular approach. Part 2: preliminary clinical experience. J Neurosurg 1991; 75: 8–14.
Hunt WE, Hess RM. Surgical risk as related to time of intervention in the repair of intracranial aneurysms. J Neurosurg 1968; 28: 14–20.
Kassel NF, Torner JC, Haley C et al. The International Cooperative Study on the timing of aneurysmal surgery. Part 2: surgical results. J Neurosurg 1990; 73: 37–47.
Kis B, Weber W. Treatment of unruptured cerebral aneurysms. Clinical Neuroradiology 2007; 17 (3): 159–66.
Lasjaunas P, Burrows P, Planet C. Developmental venous anomalies (DVA): the so-called venous angioma. Neurosurg Rev 1986; 9: 233–44.
Lobato RD, Perez C, Rivas JJ et al. Clinical, radiological, and pathological spectrum of angiographically occult intracranial vascular malformations. J Neurosurg 1988; 68: 518–31.
Molyneux A, Kerr R, Stratton I et al. International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2134 patients with ruptured intracranial aneurysms: a randomised trial. Lancet 2002; 360: 1267–74.
Pelz DM, Fox AJ, Vinuela et al. Preoperative embolization of brain AVMs with isobutyl-2 cyanoacrylate. AJNR Am J Neuroradiol 1988; 9: 757–64.
Porter PJ, Willinsky RA, Harper W, Wallace MC. Cerebral cavernous malformations: natural history and prognosis after clinical deterioration with or without hemorrhage. J Neurosurg 1997; 87: 190–7.
Puchner MJA, Freckmann N, Herrmann HD. Familial subarachnoid hemorrhage in east Finland, 1977–1990 (letter). Neurosurgery 1994; 35: 173–4.
Raabe A, Beck J, Gerlach R, Zimmermann M, Seifert V. Near-infrared indocyanine green video angiography: a new method for intraoperative assessment of vascular flow. Neurosurgery 2003; 52: 132–9.
Spetzler RF, Martin NA. A proposed grading system for arteriovenous malformations. J Neurosurg 1986; 65: 476–83.
Stefani MA, Porter PJ, ter Brugge KG et al. Large and deep brain arteriovenous malformations are associated with risk of future hemorrhage. Stroke 2002; 33: 1220–4.
Tseng MY, Czosnyka M, Richards H, et al. Effects of acute treatment with pravastatin on cerebral vasospasm, autoregulation, and delayed ischemic deficits after aneurysmal subarachnoid hemorrhage: a phase II randomized placebo-controlled trial. Stroke 2005; 36: 1627–32.
Vajkoczy P, Meyer B, Weidauer S, et al. Clazosentan (AXV-034343), a selective endothelin A receptor antagonist, in the prevention of cerebral vasospasm following severe aneurysmal subarachnoid hemorrhage: results of a randomized, double-blind, placebo-controlled, multicenter phase IIa study. J Neurosurg 2005; 103: 9–17.
Vajkoczy P. Revival of extra-intracranial bypass surgery. Curr Opin Neurol 2009; 22: 90–5.
Wanke I, Doerfler A, Dietrich U et al. Endovascular treatment of unruptured intracranial aneurysms. AJNR Am J Neuroradiol 2002; 23: 756–61.
Westphal M, Cristante L, Grzyska U et al. Treatment of cerebral arteriovenous malformations by neuroradiological intervention and surgical resection. Acta Neurochir (Wien) 1994; 130: 20–7.
Wiebers D, Whisnant JP, Huston J et al. Unruptured intracranial aneurysms: natural history, clinical outcome, and risks of surgical and endovascular treatment. Lancet 2003; 362: 103–10.
Zimmerman RS, Spetzler RF, Lee KS et al. Cavernous malformations of the brain stem. J Neurosurg 1991; 75: 32–9.
1.4 Treatment for acute stroke
Gerhard Schroth, Christoph Ozdoba, Marwan El-Koussy
1.4.1 Clinical picture
According to the definition established by the World Health Organization, stroke is characterized by the acute onset of focal or global disturbances of brain function due to vascular causes. Synonyms for acute stroke include apoplexy, cerebrovascular accident, cerebral infarction, and ictus.
In Europe, cerebral infarction is the most frequent cause of disability among adults. After cardiac diseases and cancer, stroke is the third most frequent cause of death and it is one of the main causes of epilepsy and dementia in the elderly.
The incidence of stroke increases with age. Approximately two cases of stroke per year can be expected per 1000 population. In the over-50s age group, the incidence increases approximately two- to threefold in each decade of life.
Approximately 15–20% of stroke cases are caused by intracerebral or subarachnoid bleeding. The remaining 80% have ischemic causes, with occlusion of the cerebral arterial vessels. These two entities cannot be distinguished clinically; therefore every patient with stroke needs to undergo computed tomography (CT) or magnetic resonance imaging (MRI) as quickly as possible. If bleeding is excluded by the imaging examinations, then intravenous thrombolysis can be started on the CT/MRI table even while the location and effects of the vascular occlusion are demonstrated using CT angiography (CTA) or MR angiography (MRA).
The causes of acute vascular occlusion differ fundamentally between myocardial infarction and cerebral infarction, and require different treatment approaches. Heart attacks are due to local, and usually longstanding, arteriosclerotic changes in the vascular walls in over 90% of cases. This leads directly to acute occlusion of the coronary artery due to rupture of the plaque and secondary in situ thrombosis. In addition to rapid recanalization with thrombectomy and/or angioplasty, coronary stents are also used to stabilize the local vascular process.
By contrast, occlusions in the cerebral vessels are caused by arterio-arterial embolism in approximately 75% of cases. The aim of treatment for cerebral stroke is to remove the thrombus from an otherwise healthy cerebral vessel. Thrombi that are transported to the cerebral circulation often originate in the heart, but may also have formed on plaques in the aortic arch or cervical vessels. Thrombi that form in the pelvic and crural veins can be transported to the cerebral circulation via a right-to-left circulation defect such as a patent oval foramen. This phenomenon is termed “paradoxical” embolism.
The brain weighs approximately 1200 g, equivalent to around 2% of body weight, but it requires roughly 20% of cardiac output and oxygen consumption. The brain initially responds to a reduction in regional cerebral blood flow (CBF) with reversible cerebral dysfunction, which may progress to an irreversible defect—infarction—depending on the duration and extent of the drop in CBF (Fig. 1.4-1).
1.4.2 Clinical findings
A global interruption of blood supply to the brain (global ischemia)—e.g., due to cardiovascular arrest—leads to unconsciousness after approximately 10 seconds, and the cerebral tissue is already irreversibly damaged after only a few minutes. These intervals can be prolonged by hypothermia which acts as in a neuroprotective fashion.
In focal cerebral ischemia, cerebral blood flow is only reduced in the area supplied by the occluded cerebral vessel. The associated neurological deficits depend on the size, location, and collateral supply in the underperfused areas of the brain.
Fig. 1.4–1 Diagram showing threshold values for cerebral blood flow (CBF).
Occlusion of the main trunk of the middle cerebral artery (M1 occlusion) or its proximal (M2) or peripheral (M3) branches is by far the most frequent form of cerebral infarction. The main symptoms are contralateral predominantly arm and facial hemiparesis along with speech disorders if the dominant hemisphere is affected (Fig. 1.4-2).
Circulatory disturbances in the basilar artery lead to infarction of the brainstem, cerebellum, and thalamus. The main symptoms are vertigo, ataxia, and gaze paralyses, which can quickly progress to coma, quadriplegia, or “locked-in syndrome,” in which there is complete paralysis of the body and cranial nerves, but with preservation of cerebral cognitive function, and communication is only possibly using eye movements (Fig. 1.4-3).
Fig. 1.4–2a-f Acute stroke with left-sided hemiparesis in a 13-year-old boy. The diffusion-weighted image (DWI) (a) and apparent diffusion coefficient (ADC) map (b) show cytotoxic edema in the anterior territory of the middle cerebral artery (MCA). (c) On the mean transit time (MTT) map, the entire MCA area and parts of the posterior territory show a perfusion deficit. (d) The T2-weighted spin-echo image only shows discrete signal enhancement in the right insula. (e) First-pass contrast magnetic resonance angiography reveals the cause as an occlusion of the MCA. (f) On susceptibility-weighted imaging (SWI), the thrombus in the MCA leads to a flow void that demarcates the occlusion in an enlarged form (a round frontal artefact caused by the adjoining sphenoid sinus).
Fig. 1.4–3a-e A 64-year-old man with somnolence. The CT 85 minutes after the onset of symptoms shows an occlusion in the basilar artery. Intravenous thrombolysis was started 105 minutes after the onset of symptoms. The noncontrast CT (a) shows a hyperdense basilar artery that does not enhance after contrast administration (b). Magnetic resonance imaging 4 h 15 min after the onset of symptoms shows areas of acute ischemia in the brainstem and cerebellum on the diffusion-weighted image (c), which are already demarcated on the T2w image. Contrast-enhanced first-pass MRA of the cerebral arteries (e) shows the occlusion in the basilar artery.
Fig. 1.4–4a-c Sudden paralysis of the right leg by a thromboembolic occlusion in the left anterior cerebral artery. (a) The diffusion-weighted MRI shows cytotoxic edema in the anterior cerebral artery territory. Occlusion of the left anterior artery was confirmed on MRA (b) and digital subtraction angiography (c) before endovascular mechanical removal of the thrombus.
In occlusions of the anterior cerebral artery (Fig. 1.4-4), the main symptom is central paralysis of the legs, while with posterior infarction (Fig. 1.4-5) the main symptom is narrowing of the visual field (homonymous hemianopia).
The symptoms of acute occlusion of the internal carotid artery, which is the cause of stroke in approximately 10% of cases, depend on the location of the occlusion and on the collateral vessels that are available to supply the downstream hemisphere. If the collateral supply is good, the carotid occlusion may be asymptomatic, whereas simultaneous occlusion of the middle and anterior cerebral artery (T occlusion) leads to death or extremely severe deficits in around 70% of cases.
Circulatory disturbances in the ophthalmic artery lead to monocular hemianopia due to retinal infarction that can be directly visualized on ophthalmoscopy. Acute occlusion of the central retinal artery may occur in isolation or as a partial symptom of carotid occlusion (Fig. 1.4-6).
The same applies to infarction of the anterior choroidal artery, which—after the ophthalmic artery and posterior communicating branch—originates as the third branch from the internal carotid artery, before the latter divides into the anterior and middle cerebral arteries. The main symptoms are sensory or, more rarely, motor hemisyndromes and homonymous hemianopia if the optic tract is affected after the optic chiasm.
The severity of stroke in the acute stage is classified using the National Institutes of Health Stroke Scale (NIHSS), in which points are assigned to neurological deficits and added, leading to maximum score of 42. These are summed up briefly in Table 1.4-1.
As a rule of thumb, thrombolysis is indicated starting from NIHSS 4. Up to NIHSS 9, the condition is described as minor stroke. From NIHSS 10, an occluded cerebral vessel is identified angiographically in over 95% of cases. From NIHSS 12, a large cerebral vessel is affected in over 90% of cases—such as the internal carotid artery (diameter 4–6 mm), the M1 segment of the middle cerebral artery (approx. 3 mm), or the basilar artery (3–4 mm). The recanalization rate with intravenous thrombolysis after occlusions of large cerebral vessels is relatively low.
Fig. 1.4–5a, b The CT (a) shows a subacute, already partly demarcated left-sided posterior infarction due to occlusion of the posterior cerebral artery (b), displayed as a secondary 3D reconstruction from the CT angiogram.
Fig. 1.4–6 Occlusion of the central retinal artery by a cholesterol crystal that is clearly visible on ophthalmoscopy, as is the occluded vessel and retinal edema due to an “ocular stroke.” (Image kindly provided by Dr. Wolf, Dept. of Ophthalmology, University of Berne, Switzerland.)
Table 1.4–1 National Institutes of Health Stroke Scale (NIHSS).
1.4.3 Differential diagnosis
Approximately 20% of stroke cases are caused by bleeding, while the remainder (80–85%) are due to an acute onset of circumscribed hypoperfusion or ischemia. Approximately 60% of the cases of bleeding involve spontaneous intracerebral hematoma as a result of arterial or venous hemorrhage into the cerebral tissue. The remainder consist of subarachnoid bleeding, usually resulting from aneurysmal rupture (Figs. 1.4-7 and 1.4-8).
Approximately 3–5% of stroke cases are caused by bland or septic thrombosis of the cerebral veins or dural sinuses (sinus thrombosis). The symptoms have a wide range of severity and acuteness. The symptoms occur acutely in around 30% of cases, not rarely in the form of epileptic seizures—e.g., when the inferior anastomotic vein (Labbé vein) is affected, which opens into the transverse sinus and drains the ipsilateral temporal lobe. Treatment consists of heparin administration, even if signs of typical venous congestion and hemorrhagic infarction are already visible. Extensive sinus thromboses that progress during systemic anticoagulation treatment represent an indication for local endovascular treatment. Large-lumen catheters can be used that allow aspiration (Fig. 1.4-9).
Fig. 1.4–7a-c (a) Intracerebral bleeding with ventricular penetration on CT. Subarachnoid bleeding is seen on CT (b) and MRI (c). In the fluid-attenuated inversion recovery (FLAIR) sequence, the MRI shows subarachnoid bleeding with a high level of sensitivity as a signal enhancement in the subarachnoid space in the right insular cistern.
Fig. 1.4–8a-e Intracerebral bleeding, right basal ganglia with diffusion restriction on the b1000 diffusion-weighted image (a) and apparent diffusion coefficient (ADC) image (b). The bleeding has a hyperintense appearance on T2-weighted imaging (c) and shows a flow void on susceptibility-weighted imaging (SWI) (d). There was no evidence of a bleeding source on intracranial time-of-flight (TOF) magnetic resonance angiography (e).
Fig. 1.4–9a-f Thrombosis in the superior sagittal sinus. (a) The CT shows a “negative triangle sign” (absence of contrast in the lumen, while the sinus wall takes up contrast). (b) The sagittal T1-weighted MRI shows the thrombus in the sinus. (c) The late venous catheter angiogram shows the congested cerebral veins and an absence of contrast in the superior sagittal sinus. (d) A 5F aspiration catheter (VASCO-ASP) in the transverse sinus. (e) Venogram showing the tip of the aspiration catheter in the anterior third of the partly recanalized superior sagittal sinus. (f) Thrombus material aspirated from the sinus.
Dissections of the carotid artery (approximately 75% of all dissections) and vertebral artery (approximately 20%), which may also occur multiply and in intracranial locations (approximately 5%), are rare causes of infarction in children and adolescents. The resulting stenoses can lead to infarction directly by reducing blood flow, or indirectly due to arterioarterial transport of thrombi out the false lumen into the cerebral circulation. If the dissection expands intradurally, it can also lead to severe intracranial bleeding. When there are progressive symptoms in spite of anticoagulation and/or the dissection is expanding intracranially, stenting with several overlapping stents and/or a flow diverter may be able to stabilize the situation (Figs. 1.4-10 and 1.4-11).
Fig. 1.4–10 Follow-up after a bilateral carotid dissection. Starting on day 4, the wall hematoma with the bright signal on the fat-saturated T1-weighted axial MRI can be easily distinguished from the dark residual lumen (flow void).
Fig. 1.4–11a-c Symptomatic dissection in the distal cervical segment of the internal carotid artery. (a) Digital subtraction angiography (DSA), (b) angiography without subtraction, (c) DSA after implantation of two carotid Wallstents.
Extremely rare causes of ischemic infarction include cerebral auto-somal-dominant arteriopathy with subcortical infarcts and leuko-encephalopathy (CADASIL), vasculitides such as temporal arteritis, Takayasu arteritis, and in particular central nervous system angiitis, which is difficult to diagnose.
1.4.4 Diagnosis
It is not possible to determine clinically whether a sudden neurological deficit is due to bleeding in the cerebral parenchyma or to a circulatory disturbance. Rapid imaging diagnosis is therefore key to decision-making.
If bleeding has been excluded using CT or MRI, it can be assumed that the cause of the acute neurological deficit is an ischemic cerebral infarction. In the second step, the imaging task is then to locate the vascular occlusion. Either CT angiography (CTA) or magnetic resonance angiography (MRA) methods can be used. Imaging of the cerebral vessels from the aortic arch to the peripheral branches of the cerebral arteries is obligatory, and with modern CT and MRI systems it takes less than a minute after contrast administration. Measurement of cerebral blood flow using perfusion CT or perfusion MR then follows, which also takes less than a minute.
When clinical and imaging diagnostic procedures have been completed, taking a maximum of 15–20 minutes, the following information must be available:
That no cerebral bleeding is present
Which cerebral vessel is occluded and where
Whether the occlusion explains the clinical symptoms
To what extent the cerebral tissue primarily supplied by the occluded vessel is already necrotic
How extensive the ischemic penumbra is—i.e., the area in which cerebral blood flow is reduced but the brain tissue is not yet necrotic (Figs. 1.4-12 and 1.4-13).
For the purposes of targeted treatment planning, an attempt is also made to use multimodal datasets from the initial imaging procedures to obtain information about the chemical composition and biomechanics of the thrombus, as well as its length.
Fig. 1.4–12a-c Multimodal MRI with penumbra. (a) The cytotoxic edema in the diffusion-weighted MRI in the anterior middle cerebral artery (MCA) territory is outlined in blue. (b) In the perfusion image, almost the entire MCA territory shows delayed perfusion (outlined in red). (c) The hypoperfused but still uninfarcted area corresponds to the penumbra (“tissue at risk”) and remains when area A is subtracted from area B.
Fig. 1.4–13 On multimodal perfusion CT, the penumbra is defined as an area of reduced cerebral blood flow (CBF) or a delayed mean transit time (MTT), but still with a normal cerebral blood volume (CBV). The CBV is also reduced in the cerebral tissue that has already undergone irreversible infarction.
1.4.4.1 Computed tomography
Although the cerebral cortex (gray matter) has a higher water content at around 82% than the medullary layer (white matter, water content approximately 70%), it has greater X-ray absorption and is therefore displayed with greater hyperdensity. The reason for this is that there is a lower lipid concentration in the cerebral cortex (33% in comparison with 55% of the dry weight) and higher concentrations of protein (55% vs. 39%) and oxygen. The difference is approximately 8 Hounsfield units (HU). Good, neuro-optimized CT devices, technically accurate examinations and well-windowed images make visual differentiation of as little as 4 HU possible. The infarct leads to a continuous increase in water content, which can be recognized on CT as a decline in density. The infarct’s hypodensity distinguishes it from the normal brain, but usually only after 2–4 hours.
At the same time, the density difference between the medulla and cortex declines. In the infarcted area, the basal ganglia are no longer distinguishable from the surrounding tracts (obscuration of the lentiform nucleus) (Fig. 1.4-14) and the contrast between the insula and the neighboring extreme capsule and external capsule disappears (insular ribbon sign; Fig. 1.4-15).
Slightly later, increasing water retention leads to local cerebral swelling, which becomes visible through compression of the adjacent sulci, with flattening of the relief of the cerebral gyri.
Large, compact thrombi are more dense and can be directly demonstrated on CT using the “hyperdense artery sign” in the absence of iodinated contrast (Fig. 1.4-16).
Fig. 1.4–14a-e (a) On the noncontrast CT, there is obscuration of the lentiform nucleus on the right side. (b) The hypodensity (ischemia) is better visualized with contrast enhancement. (c) Clear hypoperfusion, with the territory of the middle cerebral artery on the right. (d, e) Occlusion of the right internal carotid artery and middle and anterior cerebral artery (T occlusion) on the angio-CT.
Fig. 1.4–15a-c Early signs of middle cerebral artery (MCA) infarction on a noncontrast CT 2 hours after thromboembolic occlusion of the MCA. (a, b) There is hypodensity in the insula on the right, which is no longer distinguishable from the neighboring tracts of the external capsule and extreme capsule (insular ribbon sign). (c) The corresponding perfusion CT shows delayed perfusion in the entire circulation area of the middle cerebral artery on the right.
Fig. 1.4–16 Direct imaging of a large, hyperdense thrombus as the cause of acute middle cerebral artery infarction (dense artery sign).
1.4.4.2 Magnetic resonance imaging (MRI)
A diagnosis of cerebral infarction can be made within the first few minutes using multimodal MR techniques:
On conventional spin echo imaging, there are no flow-related signal losses (flow voids).
Time-of-flight (TOF) angiography displays the vascular occlusion directly, with no need for contrast administration.
Diffusion-weighted imaging (DWI) even only a few minutes after the vascular occlusion can detect not only increased water retention in the hypoperfused cerebral tissue (vasogenic edema), but also redistribution of the water from the extracellular space to the intracellular space. This rapidly occurring cytotoxic edema results from failure of the cellular sodium-potassium pump due to oxygen and glucose deficiency in the territory. This leads to inflow of sodium and water into the neuroglia and neurons and thus to a redistribution of the water component from the extracellular to the intracellular space. The extracellular space, through which water can flow relatively unobstructed and which represents approximately 15% of the brain’s volume, decreases in size. The intracellular space, in which water diffusion through the cell organelles and membranes is inhibited, expands. Calculating the apparent diffusion coefficient (ADC) value allows semiquantitative assessment of the extent of the cytotoxic edema and makes it easier to distinguish it from artifacts (known as “shining through”).
Susceptibility-weighted imaging (SWI) has a high level of sensitivity for detecting the thrombus in the vessel and hemostasis in the downstream arteries, as well as deoxygenated slowly flowing blood or thrombosed blood in the veins.
Comparison of conventional MR images with diffusion-weighted images makes it possible to narrow the time point of the infarction a bit better, which may be important if the stroke symptoms appear on waking (“wake-up stroke”) or the patient is found unconscious or with global aphasia. If the DWI-marked infarct is not yet visible on T2 and/or FLAIR images, it is relatively fresh and there is more of an indication for acute thrombolytic or mechanical thrombectomy treatment.
1.4.4.3 CT angiography (CTA) and MR angiography (MRA)
Imaging of the cerebral arteries from the aortic arch up to the peripheral branches of the middle cerebral, anterior and vertebrobasilar territories is obligatory in cases of ischemic infarction, and with modern equipment it takes only a few minutes. Intravenous contrast medium is injected to demonstrate the vessels, and imaging is started on the first passage of contrast through the aortic arch and cervical vessels. Standardized postprocessing programs allow selective three-dimensional display of the vessels (lumenography).
1.4.4.4 Perfusion CT imaging (PCT) and perfusion magnetic resonance imaging (PMRT)
Magnetic resonance and computed-tomographic perfusion imaging are procedures used for diagnostic demonstration and quantification of organ perfusion. They allow at least semiquantitative measurement of cerebral perfusion, can be carried out within a few minutes, and display hypoperfused areas of the brain immediately after vascular occlusion has occurred. Positron-emission tomography (PET) and single-photon emission computed tomography (SPECT), like ultrasound and Doppler ultrasonography, have no role in the modern diagnostic work-up for acute stroke.
Contrast-enhanced perfusion imaging is based on the indicator dilution theory: the passage of an intravenously administered contrast bolus, as compact as possible, through the cerebral circulation is displayed at a frame rate of no less than 1 image per second, if possible.
On the CT, the radiographic density of the normally perfused brain increases transiently during passage of the contrast (Fig. 1.4-17). On perfusion MRI, either T1-weighted imaging is used to determine the signal increase, or T2-weighted gradient imaging (T2*) is used to measure the signal decrease that occurs when the MR contrast flows through the capillaries, leading to local magnetic field changes (susceptibility disturbances) (Fig. 1.4-18).
Arterial spin labeling (ASL) is another elegant MR perfusion technique, and it does not require any contrast administration. Blood flowing into the brain is marked, and the blood itself serves as an endogenous marker during its passage through the brain. Due to the longer measurement time of approximately 5 minutes, in comparison with 1 minute for contrast-enhanced measurements, this technique is only used in special cases (contrast intolerance, renal problems) in patients with acute stroke (Fig. 1.4-19).
Functional parameters for cerebral perfusion are calculated from the signal curves using various mathematical models and algorithms and are presented in the form of parameter images. Changes in the mean transit time (MTT) and time to peak (TTP) parameters are the easiest to interpret and detect, and they allow perfusion to be described with a high level of sensitivity.
Cerebral blood flow (CBF) describes how much blood per unit of time is flowing through the cerebral tissue. Normal findings are 50–70 mL per 100 g tissue per minute. Neurological deficits occur starting from 20 mL/100 g/min. Irreversible cell damage occurs below a threshold of approximately 15 mL/100 g/min. This applies especially to the core of the infarct, although it is usually surrounded by tissue that is still temporarily receiving sufficient blood from collaterals. In a model that is not uncontroversial and which is of little assistance in treatment planning, this tissue is described as the penumbra, or “tissue at risk.”
Cerebral blood volume (CBV) is the percentage proportion of the blood (arterial, capillary, and venous) within a defined quantity of brain (usually also 100 g).
In cases of acute stroke, autoregulation can initially keep the cerebral perfusion constant. A decline in perfusion leads to dilation of the cerebral vessels. This has the effect that during a stroke, the CBV increases as long as the affected cerebral tissue is still receiving blood via collaterals. Hemodynamically, the penumbra is characterized by a reduction in CBF but with normal or increased CBV, whereas in cerebral tissue that has already suffered infarction or in the core of the infarct, the CBV and CBF are both reduced—the latter to values below 15 mL/100 g cerebral tissue/min.
Fig. 1.4–17a–c CT perfusion. (a) Noncontrast CT in a 54-year-old patient with acute left-sided hemiparesis, with no clear pathological findings. (b) Interactive measurement of the arterial contrast increase in the anterior cerebral artery and of venous outflow in the superior sagittal sinus. (c) The CT perfusion image calculated from the data shows that hypoperfusion in the right central region is the cause of the left-sided hemiparesis.
Fig. 1.4–18a-e Perfusion MRI. A signal intensity–time curve (a) is measured for each pixel, and on the basis of the indicator dilution theory a concentration–time curve (b) is adapted to it. The perfusion parameters are calculated using the concentration–time curve: the relative cerebral blood volume (RCBV) corresponds to the area under the concentration–time curve (c); the mean transit time (MTT) corresponds to the first moment of the concentration–time curve (d). The regional cerebral blood flow is calculated from the CBV and MTT (RCBF = RCBV / MTT) (e). The corresponding parameter maps are calculated for each of these parameters. The perfusion parameters (c-e) show hypoperfusion in the circulation area of the middle cerebral artery.
Fig. 1.4–19 The principle of arterial spin labeling (ASL).
1.4.4.5 Catheter digital subtraction angiography (DSA)
Catheter angiography is only used as a primary diagnostic tool in exceptional cases—e.g., to discover whether an occlusion or pseudo-occlusion is present, or to examine vessels using high spatial and temporal resolution—e.g., when vasculitis is the suspected cause of cerebral perfusion disturbances. On the other hand, imaging of the entire cerebral circulation is obligatory in the context of endovascular therapy in order to detect the morphology of the vascular occlusion and the extent of the collateral supply, particularly now that 3D images are available using rotation angiography and dyna-CT can be used on the angiography table as well to depict the brain parenchyma and measure CBV.
1.4.5 Treatment
Acute ischemic stroke is treatable. Rapid reopening of the occluded cerebral vessel leads to a reduction in the mortality rate and can help avoid disability in one in three patients. Spontaneous recanalization occurs in approximately 25% of cases, but mainly in occlusions of small cerebral vessels and usually too late to salvage the downstream cerebral tissue.
Many treatment approaches for revascularizing cerebral vessels have been borrowed from the field of cardiology. The main difference between infarction in the brain and in the heart is that cerebral infarction is usually caused by an embolism that occludes an otherwise healthy cerebral vessel. In contrast, occlusion in the coronary vessels usually takes place against the background of a local atherosclerotic vascular process. This difference has meant that interventional neuroradiology has had to find new and innovative ways of treating cerebral infarction different from that used to treat myocardial infarction.
Targeted, expert neurological care for stroke patients in specialized stroke units and protocol-driven basic measures beforehand have led to a significant reduction in the late sequelae of stroke. Optimizing respiration and blood-pressure management are decisive elements in this chain of treatment. After cerebral hemorrhage has been excluded as the cause of the symptoms, and when there is definitive evidence of vascular occlusion, blood pressure values above the normal level of a maximum of 200–220 mmHg systolic are targeted. The patient’s temperature and blood glucose level have to be monitored and kept normal. Any form of excitement or agitation for the patient must be avoided, as the brain’s oxygen consumption increases by up to 50% due to stress-related neuronal activation. The primary goal of treatment for acute stroke is recanalization of the occluded cerebral vessel. Methods available included intravenous thrombolysis and endovascular, imaging-guided recanalization by interventional neuroradiologic methods. Treatment success is assessed at a clinical examination 3 months after the event, and the modified Rankin scale has become the internationally accepted standard for this (Table 1.4-2).
Table 1.4–2 Modified Rankin scale (mRS).
| 0 | No symptoms, no disability in daily living |
| 1 | No significant disability, despite some symptoms: able to carry out all usual tasks and activities. Slight neurological impairment possible |
| 2 | Slight disability: unable to carry out all previous activities, but able to look after own affairs without assistance. Clear neurological deficit |
| 3 | Moderate disability: requires some help, but able to walk unassisted. Clear neurological deficit |
| 4 | Moderately severe disability: unable to walk unassisted, unable to attend to own bodily needs without assistance. Limited mobility, limited communication |
| 5 | Severe disability: bedridden, incontinent, requires constant nursing care and attention. Barely any communication |
In some studies, death is scored as mRS 6.
1.4.5.1 Intravenous thrombolysis (IVT)
Thrombolytic agents were introduced for the treatment of stroke as early as the 1960s and 1970s. Patients were selected without any imaging procedures; the treatment was usually started too late, leading to very high rates of bleeding and mortality.
In the meantime, large studies totaling nearly 5000 patients have been carried out that have confirmed the efficacy of IVT within a time window of up to 4.5 hours after the onset of stroke. The National Institute of Neurological Disorders and Stroke (NINDS) study, published in 1995, included 624 patients with severe acute stroke (NIHSS 14). Within a 3-hour time window, half of the patients received a placebo and the other half were treated with intravenous administration of 0.9 mg rt-PA/kg body weight. A prior noncontrast CT examination excluded bleeding, but imaging evidence of a vascular occlusion was not required. No significant difference between the treatment group and the placebo group was observed within 24 hours. After 3 months, however, the patients treated with rt-PA showed a significantly better result than the placebo group (OR 1.7; 95% CI, 1.2 to 2.6; P = 0.008; number needed to treat: 7). Symptomatic intracerebral bleeding was observed in 6.4% of the treated patients, in comparison with 0.6% of the placebo patients. The publication of this study in 1995 led to approval in the United States for intravenous rt-PA administration at the above dosage within the 3-hour time window in the treatment of ischemic stroke.
Additional prospective, multicenter, and placebo-controlled studies in which the time window was extended to 6 hours achieved a significant improvement in the clinical outcome—the European Cooperative Acute Stroke Study (ECASS I and II) and the Alteplase Thrombolysis for Acute Noninterventional Therapy in Ischemic Stroke (ATLANTIS) trial. Patients treated with IVT within up to 4.5 hours (ECASS III) were found after 3 months to have significantly better clinical findings—defined as mRS scores of 0 and 1— in comparison with the placebo group (52.4% vs. 45.2%, P = 0.04; n = 821). However, if mRS 2 is also included as a “good outcome,” the result is no longer significant, and the ECASS III study treated relatively mild cases of stroke (average NIHSS 9). The mortality rate also did not differ between the two groups, while symptomatic intracranial bleeding at 2.4% in the rt-PA group was very low, but significantly higher than in the control group (0.2%).
The disadvantage in the above studies is that no imaging documentation was available regarding the site of occlusion and the recanalization results. It is therefore clear that the studies must also have included patients in whom a vascular occlusion was not present or had already recanalized. The patients were therefore exposed to an unnecessary and indefensible risk of cerebral bleeding. Administering thrombolytic agents without positive CT, MRI, or angiographic evidence of vascular occlusion explaining the clinical symptoms is no longer acceptable given the currently available information.
When the initial findings and effect of intravenous treatment are documented with imaging or ultrasound methods, it is found that the efficacy of IVT declines with increasing vascular calibers. Adequate recanalization with IVT has been demonstrated using transcranial Doppler ultrasound 1 hour later in cases of occlusion of the internal carotid artery, middle cerebral artery, and anterior cerebral artery (carotid T occlusion) in 6% of cases, while recanalization of the main trunk of the middle cerebral artery (M1 occlusion) occurs in approximately 30% of cases and recanalization of branch occlusions (M2 occlusion) occurs in approximately 44% (Clotbust study). IVT-treated M1 occlusions that were reexamined using MRI after approximately 24 hours showed a persistent M1 occlusion in 30% of cases; minimal capillary recanalization was seen in 30% (TIMI 1); partial recanalization was found in 21% and complete recanalization (TIMI 3) in only 17%.
The advantage that IVT is rapidly and widely available stands in distinct contrast to the disadvantages of the low recanalization rate in large vessels, the associated bleeding complications, and the narrow time window of a maximum of 4.5 hours.
1.4.5.2 Endovascular stroke treatment
Local intra-arterial thrombolysis
Following pioneering studies by Zeumer, Mori, and Theron, it was the Prolyse in Acute Cerebral Thromboembolism (PROACT) I and II studies that subsequently confirmed the efficacy of intra-arterially administered prourokinase in the treatment of severe stroke (NIHSS 17). In both studies, angiographically documented occlusions of the middle cerebral artery were treated if it was possible to initiate the treatment within 6 hours of the start of symptoms. After documentation of the vascular occlusion, the treatment consisted of navigating a microcatheter into the M1 segment up to just in front of the occlusion. In 26 patients, 6 mg of prourokinase was injected locally for a period of up to 90 minutes, and in 14 patients only saline was injected (PROACT I). In the PROACT II trial, a maximum of 9 mg of prourokinase was injected for a maximum of 2 hours in front of the thrombus in 121 patients and the results were compared with 59 patients in whom only saline was injected as a placebo. Recanalization of the middle cerebral artery was observed in 66% of the patients who received intra-arterial thrombolysis, in comparison with 18% in the placebo group. After 3 months, only 25% of the patients in the placebo group had good clinical results (defined and measured as mRS 0–2), in comparison with 40% of those who were treated with prourokinase. The rate of symptomatic intracerebral bleeding among the patients who underwent thrombolysis was higher, at 10%, compared with the control group (2%).
Two other prospective studies have also confirmed the efficacy of local thrombolysis, although they were both ended prematurely. In the Middle Cerebral Artery Embolism Local Fibrinolytic Intervention Trial (MELT), urokinase was injected into the thrombus rather than in front of it, and mechanical fragmentation of the thrombus was also allowed. A recanalization rate of 74% was achieved in this way. A clear positive trend for thrombolysis therapy was also noted clinically: 49.1% of the patients treated with urokinase had a good clinical outcome (mRS 0–2), in comparison with 38.6% of those in the placebo group. The study, conducted in Japan, was prematurely stopped after the introduction of IVT in the country, and significance could therefore not be achieved due to the small numbers of patients included. A third prospective and randomized study, the Australian Urokinase Trial (AUST), compared local thrombolysis with a placebo in patients with basilar artery occlusion, in the same way as in the PROACT study. The trial was stopped prematurely because the placebo group had an outcome (mRS 0–2) that was 38% poorer.
Table 1.4–3 Scales used to grade recanalization of an occluded cerebral vessel.
| Thrombolysis in Myocardial Infarction (TIMI): assesses the local vascular findings | |
| TIMI 0 | No recanalization |
| TIMI 1 | Minimal, capillary recanalization |
| TIMI 2 | Partial recanalization |
| TIMI 3 | Vessel is completely patient |
| The Mori scale assesses cerebral perfusion after recanalization of the cerebral vessel | |
| Mori 0 | No perfusion |
| Mori 1 | Minimal reperfusion |
| Mori 2 | Reperfusion area less than 50% |
| Mori 3 | More than 50% reperfusion |
| Mori 4 | Restoration of normal perfusion |
| The Thrombolysis in Cerebral Infarction (TICI) classification represents a combination of TIMI and Mori | |
| TICI 0 | No perfusion, no anterograde flow distal to the occlusion |
| TICI 1 | Capillary flow through the occlusion site, with minimal perfusion and with no contrast in the distal vascular tree |
| TICI 2 | Partial perfusion; the arterial vascular tree distal to the occlusion shows contrast on angiography. However, inflow and/or wash-out are clearly delayed |
| TICI 2a | Only a maximum of two-thirds of the vascular territory after the occlusion is contrasted |
| TICI 2b | The entire vascular territory is contrasted, but with a marked delay as described under 2 |
| TICI 3 | Complete restoration of perfusion with no time delay in the arterial, capillary/parenchymal, and venous phases |
In summary, these three studies show that patients with severe cerebral infarction benefit from local thrombolysis within a time window of 6 hours. In any comparison between the published results on intravenous thrombolysis and those for intra-arterial local thrombolysis, it needs to be taken into account that more severe cases of stroke (NIHSS around 17) were treated with the latter method, while in the intravenous studies strokes with severity grades of 14 (NINDS), 11 (ECASS II) and nine (ECASS III) were treated. The intra-arterial recanalization rates of 66–75% are notable, particularly as they were documented angiographically, with the advantage that the administration of the thrombolytic agent can be stopped once the vessel has become patent, and can checked approximately every 15 minutes by injecting contrast through the guide catheter. Local thrombolysis with a microcatheter, in front of or into the occlusion in the cerebral vessel, thus still represents a simple, relatively low-complication alternative to mechanical recanalization techniques when endovascular access to the occlusion site is difficult due to ectasia, kinking, or stenoses in the cervical vessels, expert interventional experience is lacking, or the diameter of the carotid artery or vertebral artery is too small for devices. The disadvantage is the relatively long time required for recanalization, and for this reason the available dosage (1 million units of urokinase, 0.6 mg rt-PA/kg body weight) is administered using a perfusion system for a period of 90–120 minutes. As the collateral supply to the brain tissue is not impaired by the microcatheter, however, this need not lead to enlargement of the infarction and a decrease in the penumbra provided that blood pressure is kept slightly elevated and stable.
A retrospective meta-analysis including 53 thrombolysis studies reporting recanalization rates within the first 24 hours after the start of symptoms also confirmed the good efficacy of endovascular thrombolysis. According to the data, spontaneous recanalization of an occluded middle cerebral artery occurs in approximately 22% of cases. After intravenous thrombolysis, this percentage can be raised to approximately 50%, while local intra-arterial thrombolysis led to recanalization in 67% of cases (PROACT) and 74% of cases (MELT). If additional mechanical recanalization techniques are used, the occluded cerebral vessel can be reopened in 80–90% of cases. A retrospective comparison of two groups with clinically severe M1 occlusions (NIHSS 17), one of which received intravenous treatment while the other was treated intra-arterially, also showed significantly better clinical results in the group with endovascular treatment. Only 23% of the patients in the IVT arm had good results, in comparison with 53% of those who received endovascular treatment, with “good results” being defined as mRS 0–2 after 3 months. This finding is all the more remarkable in that the IVT was only allowed within a 3-hour time window, so that many stroke patients were excluded from the treatment who could still be treated with intra-arterial thrombolysis within the 6-hour time window.
Sonothrombolysis
The application of ultrasound during thrombolysis is intended to increase its effectiveness, although the precise mechanism involved is not known. In the Clotbust study, 126 patients with MCA occlusions received IVT within the 3-hour time window. The 63 patients in the treatment arm also received continuous transcranial ultrasound at 2 MHz during the infusion of the thrombolytic agent. Complete recanalization (TIMI 3) was significantly more frequent in the treatment arm (46% vs. 18%; P < 0.001). By contrast, the rate of symptomatic intracranial hemorrhage, the mortality rate, and the final clinical results did not differ significantly.
Another study, in which transcranial ultrasound (300 kHz) was used to supplement IVT within a time window of up to 6 hours, had to be stopped prematurely because symptomatic intracranial bleeding occurred in five of the 14 patients (36%) in the treatment arm. In the Interventional Management of Stroke (IMS II) study, a combination of IVT/IAT and endovascular ultrasound applied using a 3.3F catheter led to complete recanalization in 69% of cases after a treatment period of 2 hours. However, six of the 33 patients (18%) suffered symptomatic intracranial hemorrhage. The catheter is only suitable for the treatment of large arteries (up to the M2 segment) and could not be advanced intracranially through an extremely tortuous internal carotid artery (ICA) in 9% of the patients.
Aspiration (Figs. 1.4-20, 1.4-21)
Recanalization by aspirating the thromboembolic foreign material for endovascular treatment of acute stroke is a particularly attractive method for several reasons:
It is not necessary to pass the occlusion site and navigate the microwire and microcatheter in the occluded downstream vascular segment, which is not angiographically visible. This reduces the risk of vascular perforation, spasm, and dissection.
In cerebral infarction, the vascular wall at the occlusion site has not undergone any arteriosclerotic changes, in contrast to myocardial infarction. The occlusion is caused by embolic transport of thrombi into the cerebral circulation, and the material is therefore less adherent to the vascular wall than in an arterio-sclerotic occlusion of a peripheral or coronary vessel.
Sophisticated neuroangiographic techniques allow a precise, imaging-guided procedure: the occlusion site is identified using 3D rotation angiography, or at least biplanar angiography with a resolution of less than 200 μm. The access route and occlusion site are “frozen” on the imaging display and the aspiration catheter is navigated to a point in front of the occlusion with imaging guidance (biplanar road map).
Stabilization of the cerebral vessels at the skull base and meninges allows a precise approach with endovascular navigation, as the images are only slightly disturbed by pulse and respiratory artifacts.
In contrast to the peripheral vessels, in which the effect of aspiratory negative pressure is limited due to collapse of the vessel, the cerebral vessels are fixed to the bone and hard meninges after they enter the skull base.
Aspiration techniques are mainly used to recanalize occlusions in the large vessels supplying the brain. Aspiration catheters with gauges of 5F and 6F are now available with a highly flexible distal third and with a curve at the tip that is preformed or can be shaped using steam or hot air. As they have a hydrophilic coating and reinforced proximal catheter shaft, they can be advanced using a telescoping technique through a 7F or preferably 8F catheter, sometimes without wire guidance, to occlusion sites in the M1 segment of the middle cerebral artery, or into the basilar artery. The siphon in the internal carotid artery usually offers the greatest resistance, as it is attached to bone and hard meninges and cannot be stretched. The tip of the aspiration catheter is advanced to the proximal end of the thrombus under imaging guidance (road map). As soon as the occlusion has been reached and blood stops flowing back through the aspiration catheter, a lockable 50-mL aspiration syringe is used to attach the thrombus to the tip of the aspiration catheter. The negative pressure that can be created with this technique at the tip of a 5F aspiration catheter is approximately 10 times greater than the suction pressure created by the Penumbra pump (see below), and it usually leads to deformation and fixation of the thrombus at the tip of the aspiration catheter. After approximately 1–2 minutes, with negative pressure being maintained and proximal flow arrest, the aspiration catheter is withdrawn. If spontaneous return flow from the guide catheter is not observed during this withdrawal maneuver, then it also has to be carefully aspirated; irrigation of the guide catheter must cease during this maneuver.
Fig. 1.4–20a-g Technique for revascularization of an acutely occluded internal carotid artery (T occlusion) in a 70-year-old patient with aphasia and right-sided hemiplegia (NIHSS 19). (a) The first-pass contrast magnetic resonance angiogram shows the occluded internal carotid artery on the left. (b) Common carotid angiogram on the left, with the stump of the left internal carotid artery. (c) The occlusion is passed with a long (3 m) 0.038” wire using a biplanar road map. (d) Passage with a 5F aspiration catheter and the 8F guide catheter during continuous aspiration. (e) Stenting, with distal protection provided by a filter-wire protection system. (f) Checking with 3D rotation angiography after aspiration and stent placement. (g) The thrombus aspirated using the 8F catheter.
The disadvantage of this technique is that the withdrawal of the aspiration catheter causes loss of access, and the segment from the tip of the guide catheter up to the occlusion has to be traversed again if the thrombus cannot be removed on the first attempt.
The Penumbra system (Fig. 1.4-22) consists of an aspiration catheter that is navigated to the front of the vascular occlusion, with an aspiration pump that ensures continuous aspiration of the thrombus, which is simultaneously fragmented inside and outside the catheter tip by a wire with an olive-shaped tip. This means that the thrombus can be suctioned out piece by piece, with the aspiration catheter being kept clear. Despite higher recanalization rates of over 80%—which have usually been achieved in combination with administration of thrombolytic agents in bridging therapy—the clinical results are not satisfactory in all cases. This is because on the one hand, the negative pressure created by the pump is relatively low, while on the other the fragmentation movement of the separator takes place outside of the aspiration catheter and thus in an area that is not visualized angiographically and/or by the road map. Although the separator wire is relatively soft, vascular injury, dissection, or spasm can occur quite rapidly if the tip of the aspiration catheter is directed at the vascular wall while passing a curve, so that the separator has no space available, or if the occlusion (as is often the case) is located in a vascular bifurcation towards which the tip of the aspiration catheter is directed. In addition, if there are small perforating branches that originate immediately in front of or inside the occlusion, thrombus fragments may be pressed by the separator itself into still-patent branches beyond the bifurcation, or may be washed back into reopened side branches during the recanalization procedure. The aspiration catheters supplied by Penumbra also had relatively small lumens, although there is now a 5F system recently added.
Fig. 1.4–21a-c Acute occlusion of the internal carotid artery after heart surgery. (a) Lateral digital subtraction angiography image of the internal carotid artery. (b) The recanalized internal carotid artery after aspiration. (c) Deformation of the thrombus due to forced aspiration through the 5F aspiration catheter.
Fig. 1.4–22a, b The Penumbra aspiration system, with microcatheters in various sizes, corresponding separators, and the aspiration system. The thrombus is broken up by advancing and withdrawing the separator, and the fragments are continuously aspirated.
Mechanical retriever systems
Since 2000, there have been increasing numbers of publications describing small series and case reports of successful recanalization of cerebral vessels using microinstruments that were initially developed to manage complications during endovascular procedures. In individual cases, the grasping instruments developed for the purpose (such as microsnares and micro-alligator clips) can be used to grasp compact thrombi and remove them from the vessel. Numerous mechanical thrombectomy systems have been developed on the basis of this experience.
The retriever systems act at the distal end of the occlusion. It is therefore necessary to pass the occlusion site with a microwire and microcatheter, so that on the one hand there is a risk of vascular perforation and on the other a possibility of further distal migration of the thrombus. As a result these systems are increasingly being used in combination with proximal protection devices, and the following procedure is recommended for M1 occlusions:
Imaging of the occlusion and collateral supply with a diagnostic 5F catheter.
Placement of an 8F or 9F balloon catheter in the internal carotid artery, if necessary exchanging the catheter over a long wire.
Preparation of a biplane road map, with the lateral projection also displaying the guide catheter balloon while the anteroposterior projection shows an enlarged image of the occlusion site.
Navigation of a microcatheter and microwire through the occlusion, which should be achievable without difficulty in over 90% of cases.
Careful injection of contrast through the microcatheter in order to ensure that the tip is positioned distal to the occlusion and that during the blind transit a vessel has been entered that is large enough for the retriever system to be opened inside it; positioning of the retriever through the microcatheter distal to the occlusion.
Irrigation of the guide catheter must cease at this point, if not before.
Arrest of flow by inflating the guide catheter balloon.
Careful removal of the retriever system, with simultaneous aspiration of the guide catheter once the system is inside it.
Removal of the retriever through the widely opened Tuohy valve and repeat aspiration if spontaneous backflow of blood from the guide catheter is not observed.
Restoration of cerebral flow as quickly as possible by deflating the balloon.
Careful inspection of both the retriever system and of spontaneous back-bleeding, or as a result of aspiration, from the guide catheter, in which fragments of thrombus can often be found.
Follow-up angiography and if necessary a repeat procedure, or a move to a different technique.
The Catch system (Balt, Montmorency, France; Fig. 1.4-23) was one of the first to receive certification in Europe in animal trials and to be introduced into everyday clinical practice. The self-expanding basket, which is opened distal to the thrombus, is available in various diameters.
The Phenox Clot Retriever (Phenox Ltd., Bochum, Germany) consists of a soft wire with outward-pointing microfilaments woven into it, which are intended to grasp the whole length of the thrombus like a brush or pipe cleaner and mobilize it. Two systems can be used simultaneously for thrombi in the area of the middle cerebral artery bifurcation. The withdrawal maneuver has to be carried out with flow arrest and aspiration through the balloon-equipped guide catheter. The Merci retriever (Concentric Medical Inc., Fremont, California, USA; Fig. 1.4–24) is a nitinol wire, which after being extended from the microcatheter is designed to open like a corkscrew in order to catch and grasp the thrombus. It is delivered with an 8F or 9F balloon guide catheter, which blocks blood flow while the retriever is being withdrawn. The retriever is available in various diameters and with or without microfilaments (type L), which are intended to additionally attach the thrombus to the system. In a prospective registry study, recanalization rates of 43% were achieved with the initial system, which were increased to 64% when it was combined with local thrombolysis using rt-PA. In the Multi Merci Trial, including 131 patients with severe stroke (NIHSS 19), a recanalization rate (TIMI 2 and 3) of just under 70% was achieved with simultaneous intra-arterial rt-PA administration. Recanalization was achieved using the new Merci retriever alone in 57% of the cases.
Fig. 1.4–23 The Catch retriever system.
In general, we regard these mechanical systems as providing an opportunity to force recanalization in rare, exceptional situations when the clinical and imaging findings show it to be necessary. Use of the systems requires experience not only with the instruments themselves, but also with the cerebral circulatory system, cerebral function, and neurophysiology. A careful review of the literature confirms the impression that the clinical improvement achieved cannot fully match the reported recanalization results, for several reasons:
The results of recanalization are practically all classified using the Thrombolysis in Myocardial Infarction (TIMI) classification. It remains an open question whether the local reopening of the vessel also led to restoration of perfusion in the downstream cerebral tissue (TICI/Mori classification).
Even brief flow arrest can completely destabilize the precarious hemodynamic situation in the penumbra, particularly when flow remains reduced by dissection or spasm after deflation of the balloon.
We have observed in animal experiments that the thrombus is compressed by the mobilization procedure and parts of it can be pressed into side branches. When the thrombus is removed, these parts are sheared off and the side branches are occluded. This mechanism appears to play a role particularly during mechanical recanalization of the basilar artery, with its multiple small branches to the brain stem, and in the lenticulostriate branches to the middle cerebral artery.
Fig. 1.4–24a-c The Merci retriever. (a) Type A. (b) Type L with filaments. (c) Recanalization of an M1 occlusion using a Merci retriever, with the type X Merci in the M1 segment of the middle cerebral artery.
1.4.5.3 Percutaneous transluminal angioplasty (PTA) and stenting (Figs. 1.4-25, 1.4-26)
The principle of PTA and stent treatment involves pressing the thrombus into the wall and fixing it there with the stent if appropriate. It is known from animal experiments that the microcatheter does not penetrate the thrombus, but passes the occlusion site between the vascular wall and the thrombus. The thrombus is thus not attached to the vascular wall circumferentially, but attached to the part of the vessel opposite the point at which it has been passed with the wire or catheter. Attention therefore needs to be given to ensure that the microwire over which the stent will later be introduced passes the occluded basilar artery in the dorsal part of the vascular lumen, and this is only technically possible with the microcatheter when an enlarged lateral road map projection is used. Then, when the stent opens, the thrombus will be shifted ventrally and thus away from the branches to the pons that arise from the dorsal circumference of the basilar artery. Passage of an occluded M1 segment between the thrombus and the upper circumference of the middle cerebral artery is achieved analogously, so that the upward-directed origins of the lenticulostriate branches are kept open.
Animal studies and our own experience have confirmed that PTA can lead to rapid recanalization, but that the lumen closes again relatively rapidly if the thrombus is not fixed and compressed against the wall by a stent. The risk of fragmentation and embolization of the thrombus into distal branches is also greater with PTA alone. Rapid recanalization rates of up to 90% have been achieved with both balloon-mounted and self-expanding stents (Wingspan, Enterprise).
The disadvantage of stent recanalization is the relatively high rate of stent thrombosis, which requires a relatively aggressive form of management with double aggregation (i.v. aspirin and rapid titration of clopidogrel). The cause of these early stent occlusions is that the thrombus attached to the wall slowly expands back into the lumen through the mesh of the stent, and the lumen is then occluded again by thrombus. This observation has led to the development of a “removable stent”, also known as “stentrievers”.
Fig. 1.4–25a-e A 61-year-old man with NIHSS 15 due to a left-sided M1 occlusion. (a, b) The wire and microcatheter are passed between the upper wall and the thromboembolic blockage. (c) PTA thus compresses the thrombus caudally. (d) The lenticulostriate end arteries arising from the upper wall, which supply the basal ganglia, are thus kept free or reopened. (e) Remodeling of the M1 lumen by secondary placement of a self-expanding stent.
Fig. 1.4–26a-i A 62-year-old patient with mild motor aphasia. The diffusion-weighted image shows ischemic areas in the territory of the left middle cerebral artery (a), with delayed perfusion in the mean transit time perfusion image (b) and evidence of stenosis in the distal M1 segment of the left middle cerebral artery on time-of-flight magnetic resonance angiography (MRA) (c). (d) The first-pass contrast-enhanced MRA of the cervical vessels does not show any hemodynamically relevant stenoses in the cervical arteries. The acute, symptomatic stenosis was treated by placing a balloon-mounted neurostent. (e, f) Digital subtraction angiography (DSA) and 3D rotation angiogram of the stenosis. (g) Inflation of the balloon-mounted stent using a triple-catheter technique. (h, i) DSA and 3D rotation angiogram after stent placement.
1.4.5.4 Stent retriever device (Figs. 1.4-27, 1.4-28)
Initial attempts to achieve rapid, albeit transient, recanalization without definitive placement of a stent was accomplished by not fully releasing the self-expanding stent, so it could be withdrawn again into its sheath and retrieved when desired. The Solitaire stent (Covidien, ev3 Endovascular Inc., Plymouth, Minnesota, USA), which was originally developed for stent protection in endovascular treatment of broad-based aneurysms, proved to be particularly suitable. This is a laser-cut self-expanding stent that is attached to a guide wire with which it can be introduced and removed again through a 0.0021-inch microcatheter (Rebar, Prowler). As with coils for aneurysm treatment, the stent can be released from the wire electrolytically and can remain in place. The stents are available with various lengths and diameters, from 4 × 15 mm to 6 × 30 mm. When the stent is withdrawn in its opened state after approximately 5 minutes, animal experiments and small case series have shown that the thrombus is also removed in the stent mesh and that this leads to rapid recanalization of the vessel in around 90% of cases. Although the stent is withdrawn in its expanded state, spasm occurs rarely and dissections very rarely.
If possible, withdrawal of the stent retriever should be carried out in flow arrest through a large-lumen (8 or 9F) balloon guide catheter that is “parked” in the distal internal carotid artery and briefly opened during withdrawal of the stent.
Prospective single-center and multicenter registry studies have confirmed that using the stent retriever significantly reduces the time required for the intervention and achieves recanalization rates of 80–90%. This is paralleled by a marked improvement in the clinical results: 121 patients with acute stroke (NIHSS 18) were treated within the first 6 hours using the Solitaire retriever system in five centers in Europe. The recanalization rate was 90%, and 55% of the patients had good clinical results after 3 months, defined as mRS 0–2. This is much better than the results in the intravenous thrombolysis studies, although the latter only treated patients within a time window of 3–4.5 hours and only included patients with significantly milder cases of stroke with NIHSS scores of between 9 and 16.
Stent retriever systems are now being supplied by various companies, each with a slightly different design.
1.4.5.5 Combined intravenous thrombolysis and endovascular treatment (bridging approach)
This combined treatment approach is increasingly being used in stroke networks. In smaller hospitals, cerebral bleeding can be excluded as the cause of stroke using computed tomography, which is widely available, and intravenous thrombolysis can be started without delay. While this treatment is already taking effect, the patient can be moved to a center in which the technical facilities are then available for carrying out endovascular treatment if there is no improvement with the intravenous thrombolysis. If the patient’s condition has changed on admission to the center, and/or more than 1–2 hours has passed, the center should carry out further imaging diagnosis in order to exclude bleeding into the infarct, or in any event to document successful recanalization using the intravenous bridging thrombolysis.
Fig. 1.4–27a, b (a) A retrievable stent. (b) Documentation of the thrombus in the stent after recanalization of a thromboembolic occlusion of the middle cerebral artery.
Fig. 1.4–28a-d Acute stroke in a self-sufficient 89-year-old who called the emergency physician herself (NIHSS 19). (a) Occlusion of the right middle cerebral artery. (b) Immediate capillary recanalization after initial deployment of the retrievable stent. (c) Detailed angiography (spatial resolution 0.1 mm) of the reopened capillary lumen after compression of the thrombus onto the wall by the stent, which is still in place. (d) Complete recanalization after removal of the stent. The patient wanted to go home again the next day to look after her husband and cat.
Studies on combined intravenous/intra-arterial thrombolysis have used lower intravenous dosages (0.6 mg rt-PA/kg body weight) in order to complete the intra-arterial thrombolysis with the remaining dose of rt-PA or urokinase. It also appears acceptable to administer the full dosage of rt-PA, 0.9 mg/kg body weight, followed by endovascular treatment if recanalization does not take place, as endovascular treatment is increasingly limited to the use of mechanical recanalization techniques.
1.4.5.6 Multimodal endovascular therapy
Modern approaches to the treatment of acute stroke take into account both the location of the occlusion and also the time between the start of symptoms and the therapeutic intervention. The treatment of vertebrobasilar infarctions begins with IVT within a 4.5-hour time window. The patients should be transferred without delay to a center in which mechanical recanalization can follow if needed. If the patient has in the meantime become comatose and is showing more extensive clinical symptoms of a brainstem lesion, multimodal magnetic resonance diagnosis should be carried out first in order to allow interdisciplinary assessment of the prognosis—e.g., imminent “locked-in” syndrome—and discussion of it with relatives.
Patients with an infarct in the carotid territory should receive IVT if they are within the 4.5-hour time window and there is no occlusion of the ICA or M1/M2 segment. For patients in whom the 4.5-hour window has already been exceeded, or who present with a larger vascular occlusion (ICA, M1, and M2), endovascular therapy is prepared, with or without bridging therapy, the time window for which is 6 hours for the middle cerebral artery. The endovascular procedure is carried out with anesthesia facilities on hand, through a wide-lumen port (8F), which keeps all options open for the subsequent procedure. After diagnostic imaging of both carotid territories and the posterior circulation, a 7F or 8F guide catheter is positioned in the relevant cervical artery. If there is a combined ICA/ MCA occlusion, the ICA is initially recanalized using aspiration via the 8F catheter, with stent placement if needed. This improves perfusion of the collaterals and prevents re-occlusion. If the MCA is not already recanalized by the aspiration at this point, we follow this with IAT (< 6 h) or thromboembolectomy (< 8 h).
Depending on how much time has passed and upstream ICA conditions, occlusions of the MCA can initially be treated with IAT or aspiration (with a 4–5F catheter). IAT is carried out via a 2.4F microcatheter, the tip of which is placed in the thrombus. We infuse no more than 1 million IU urokinase via an infusion pump that distributes the dosage over a period of 90–120 minutes, with angiography being carried out via the guide catheter after 30 minutes and then every 15 minutes. The intra-arterial thrombolysis is then stopped once recanalization occurs. This makes it possible to reduce the rate of symptomatic intracerebral bleeding to less than 5%. Thrombolysis can be speeded up with careful passage and manipulation of the thrombus with the microwire and microcatheter. However, the risk of vascular perforation needs to be considered in all forms of manipulation. With persistent occlusions, or when IAT is contraindicated, mechanical procedures can be used for thrombectomy, PTA, and/or implantation of a stent as the last resort. The introduction of retrievable stents has changed the situation such that this technique can now be used as a primary form of endovascular therapy, with or without intravenous bridging treatment, since recanalization can be achieved rapidly and safely in a very high percentage of cases.
References
Alexandrov AV, Molina CA, Grotta JC, et al. Ultrasound-enhanced systemic thrombolysis for acute ischemic stroke. N Engl J Med 2004; 351: 2170–8.
Arnold M, Schroth G, Nedeltchev K, et al. Intra-arterial thrombolysis in 100 patients with acute stroke due to middle cerebral artery occlusion. Stroke 2002; 33: 1828–33.
Brekenfeld C, Remonda L, Nedeltchev K, et al. Symptomatic intracranial haemorrhage after intra-arterial thrombolysis in acute ischaemic stroke: assessment of 294 patients treated with urokinase. J Neurol Neurosurg Psychiatry 2007; 78: 280–5.
Brekenfeld C, Schroth G, Mattle HP, et al. Stent placement in acute cerebral artery occlusion: use of a self-expandable intracranial stent for acute stroke treatment. Stroke 2009; 40: 847–52.
del Zoppo GJ, Poeck K, Pessin MS, et al. Recombinant tissue plasminogen activator in acute thrombotic and embolic stroke. Ann Neurol 1992; 32: 78–86.
Furlan A, Higashida R, Wechsler L, et al. Intra-arterial proUK for acute ischemic stroke. The PROACT II study: a randomized controlled trial. Prolyse in Acute Cerebral Thromboembolism. JAMA 1999; 282: 2003–11.
Gralla J, Brekenfeld C, Mordasini P, Schroth G. Mechanical thrombolysis and stenting in acute ischemic stroke. Stroke 2012; 43: 280–5.
Hacke W, Kaste M, Bluhmki E, et al. Thrombolysis with alteplase 3 to 4.5 hours after acute ischemic stroke. N Engl J Med 2008; 359: 1317–29.
IMS Study Investigators. Combined intravenous and intra-arterial recanalization for acute ischemic stroke: the Interventional Management of Stroke Study. Stroke 2004; 35: 904–11.
Lewandowski CA, Frankel M, Tomsick TA, et al. Combined intravenous and intra-arterial r-TPA versus intra-arterial therapy of acute ischemic stroke: Emergency Management of Stroke (EMS) Bridging Trial. Stroke 1999; 30: 2598–605.
Mattle HP, Arnold M, Georgiadis D, et al. Comparison of intra-arterial and intravenous thrombolysis for ischemic stroke with hyperdense middle cerebral artery sign. Stroke 2008; 39: 379–83.
Mattle HP, Arnold M, Lindsberg PJ, Schonewille WJ, Schroth G. Basilar artery occlusion (Review). Lancet Neurol 2011; 11: 1002–14.
Nedeltchev K, Arnold M, Brekenfeld C, et al. Pre- and in-hospital delays from stroke onset to intra-arterial thrombolysis. Stroke 2003; 34: 1230–4.
Nedeltchev K, Brekenfeld C, Remonda L, et al. Internal carotid artery stent implantation in 25 patients with acute stroke: preliminary results. Radiology 2005; 237: 1029–37.
Neumann-Haefelin T, du Mesnil de Rochemont R, Fiebach JB, et al. Effect of incomplete (spontaneous and postthrombolytic) recanalization after middle cerebral artery occlusion: a magnetic resonance imaging study. Stroke 2004; 35: 109–14.
Ogawa A, Mori E, Minematsu K, et al. Randomized trial of intra-arterial infusion of urokinase within 6 hours of middle cerebral artery stroke: the middle cerebral artery embolism local fibrinolytic intervention trial (MELT) Japan. Stroke 2007; 38: 2633–9.
Rha JH, Saver JL. The impact of recanalization on ischemic stroke outcome: a meta-analysis. Stroke 2007; 38: 967–73.
Shaltoni HM, Albright KC, Gonzales NR, et al. Is intra-arterial thrombolysis safe after full-dose intravenous recombinant tissue plasminogen activator for acute ischemic stroke? Stroke 2007; 38: 80–4.
The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med 1995; 333: 1581–7.
Theron J, Courtheoux P, Casasco A, et al. Local intra-arterial fibrinolysis in the carotid territory. AJNR Am J Neuroradiol 1989; 10: 753–65.
Tomsick T, Broderick J, Carrozella J, et al. Revascularization results in the Interventional Management of Stroke II trial. AJNR Am J Neuroradiol 2008; 29: 582–7.
Zeumer H, Hacke W, Ringelstein EB. Local intra-arterial thrombolysis in vertebrobasilar thromboembolic disease. AJNR Am J Neuroradiol 1983; 4: 401–4.
Zeumer H, Hündgen R, Ferbert A, et al. Local intra-arterial fibrinolytic therapy in inaccessible internal carotid occlusion. Neuroradiology 1984; 26: 315–7.
