Intracranial Aneurysm Treatment

Abstract

This chapter presents endovascular strategies for the treatment of intracranial aneurysms, including embolization (e.g., coiling or Onyx infusion), flow diversion, and parent vessel sacrifice. Techniques are discussed in detail, and complications are covered. A primer on imaging of intracranial aneurysms is also presented.

Appendix: Primer on Imaging of Intracranial Aneurysms by Joel K. Curé, M.D.

Imaging techniques for intracranial aneurysms include CTA, MRA, catheter angiography and indocyanine green videoangiography. Evaluation of suspected aneurysmal subarachnoid hemorrhage typically begins with non-contrast computed tomography.

1.1 Computed Tomography

Non-contrast computed tomography (CT) is the first-line imaging modality for patients suspected of having an aneurysmal subarachnoid hemorrhage. The most common diagnostic error leading to failure to correctly diagnose SAH is failure to obtain a non-contrast brain CT.89 Using xanthochromia on lumbar puncture as the gold standard, CT performed on a third generation scanner was 98% sensitive for detecting SAH within 12 h of symptom onset.90 The sensitivity of CT for SAH decreases with time after the hemorrhage and with small-volume hemorrhages.91 The most common cause of rebleeding soon after aneurysm surgery is failure to obliterate the offending aneurysm, and this in turn is most commonly due to failure to identify the aneurysm at angiography.92 The thickness and distribution of the intracranial hemorrhage and aneurysm contour are useful in predicting the site of the offending aneurysm on subsequent angiography.93,94 The amount of subarachnoid blood on CT (Fisher scale) correlates with the risk of vasospasm.95 While most aneurysm ruptures produce subarachnoid hemorrhage, aneurysms that have become adherent to the brain surface may produce both subarachnoid and intraparenchymal hemorrhage,96 entirely intraparenchymal hemorrhage,97 or more rarely primarily intraventricular hemorrhage.98 CT in a substantial number of patients with non-aneurysmal SAH demonstrates a characteristic pattern of SAH surrounding the midbrain, the so-called “perimesencephalic subarachnoid hemorrhage” pattern.

1.2 Catheter Angiography

Conventional catheter digital subtraction angiography, more recently supplemented with 3D rotational angiography, is still regarded as the gold standard for imaging of intracranial aneurysms.89 It is considered by many to be the initial study of choice for evaluation of the patient with SAH, particularly in centers with 24/7 availability of a dedicated neurointerventional team.99,100 DSA provides the highest spatial resolution (0.124 mm pixel size)91 and is the optimal imaging technique for preoperative assessment of aneurysm anatomy and for evaluating morphologic features that have a direct bearing on endovascular or open surgical treatment.99 The complication rate for catheter angiography by neurointerventionalists in an academic setting has been estimated at 0.3%.101 No source of hemorrhage is identified in up to 25% of initial catheter angiograms performed for SAH. Repeat catheter angiography discovers an initially unidentified aneurysm in an additional 1–2% of cases.89

1.3 CT Angiography

CT angiography (CTA) employs a thin-slice high resolution rapidly acquired helical CT scan of the brain during the rapid intravenous infusion (“bolus”) of 80–120 cc of iodinated contrast material. Imaging is performed during peak arterial opacification, ideally before significant venous opacification occurs. Sub-millimeter thickness slices can be achieved on modern multi-detector CT scanners with a maximum pixel size of 0.35 mm.91 Scan time depends on the number of available detectors, but is less than 1 min for a 64 detector scanner with coverage from the foramen magnum to the vertex. Scanners with up to 320 detectors are currently available (Toshiba Aquilion One), the latter enabling whole-brain coverage with a single rotation of the CT gantry. Finally, a CTA evaluation can initiated and completed immediately upon detection of SAH on a non-contrast CT study. This can be accomplished far more quickly than a Neurointerventional team can be mobilized to complete a conventional angiogram under the best of circumstances.

The acquired images are reconstructed and available for multiplanar and 3D viewing at the CT scanner console within approximately 5 min, although network transmission via a PACS system for enterprise-wide review requires a variably longer time interval. The resulting axial source images are reviewed on a PACS viewing station. This is followed by a review of 2D maximum intensity projection (MIP) “sliding slabs” in the coronal, axial, and sagittal planes and of color 3D volume rendered images on an integrated thin client 3D workstation. Review of the source images and 2D reconstructions is particularly helpful for detecting aneurysms that are near or within bone (i.e. in or near the skull base and anterior clinoid processes). 2D reformatted and 3D volume rendered images can be generated to optimally demonstrate the relationship of the aneurysm to adjacent vascular (e.g. parent artery, adjacent perforating vessels) or osseous structures (e.g. the anterior clinoid process), but review of the “source” images is essential.

Several recent studies have supported the use of CTA as the initial imaging modality for suspected intracranial aneurysms. In a recent study of 179 patients (with 239 DSA-documented aneurysms) presenting with SAH to a single institution who underwent CTA and DSA, sensitivity of CTA was 99.6% and specificity was 100% 102 Notably, 19% of the aneurysms in this study were ≤2.9 mm. A recent meta-analysis of 45 studies comparing CTA to DSA for detection of suspected intracranial aneurysms found an overall sensitivity of 97.2% for CTA (95% CI, 95.8–98.2%). CTA specificity was 97.9% (95.7–99.0). Subgroup analysis demonstrated significantly greater sensitivity for 16 and 64 detector CT scanners than for single or 4 detector scanners, especially for smaller (≤4 mm) aneurysms. This was most likely linked to the availability of thinner (sub-millimeter) slices on the 16 and 64 detector CT scanners.103 Factors affecting the sensitivity and specificity of CTA for aneurysm detection include: aneurysm size and location, vascular tortuosity, radiologist experience, mode of image acquisition and presentation.89 CTA advantages compared to DSA include ability to demonstrate mural calcification, intraluminal thrombus, orientation of the aneurysm with respect to intraparenchymal hemorrhage, and relationship of the aneurysm to adjacent bony structures. Disadvantages include concealment of aneurysms by bony structures or aneurysm clips and decreased ability to demonstrate small vessels.89

1.4 MR Angiography

3D time of flight (3D TOF) MR angiography (MRA) is the most commonly used MR angiographic technique for imaging intracranial aneurysms. This technique is based on a T1 weighted 3D Fourier transform spoiled gradient echo MRI sequence (3D-SPGR). Here, the combined use of a short TR and small flip angle produces location-specific signal suppression in stationary tissues due to saturation. Inflowing blood that has not experienced these repeated RF pulses appears “bright” in contrast to the stationary background tissue. The individual imaging “slices” obtained in this fashion are subjected to a maximum intensity projection algorithm that creates a three dimensional angiogram. The data can also be subjected to 3D volume rendering similar to that employed in CT angiographic post-processing.

Advantages of MRA include no requirement for injected contrast (beneficial in patients with renal failure or pregnancy). Use of MR angiography may be limited by patient stability, inability to remain motionless for the study, or by contraindications to MRI, in particular implanted ferromagnetic surgical devices or pacemakers. While modern non-ferromagnetic aneurysm clips and endovascular coils are not contraindications to MRI/MRA, local field distortions and susceptibility effects produced by aneurysm clips compromise vascular analysis in the immediate region of the device as well as downstream from the device. Sensitivity of 3D TOF MRA for cerebral aneurysms ranges from 55–93%. Sensitivity is greatest with larger aneurysms, with a sensitivity of 85–100% reported for aneurysms ≥5 mm.89 Considering the many logistical barriers to MR scanning of unstable patients in the acute setting and generally higher sensitivity and spatial resolution of CTA and conventional angiography, CTA has become the primary non-invasive modality for aneurysm diagnosis at most centers.

1.5 Follow-up of Treated Aneurysms

Follow-up imaging of treated aneurysms is standard for both clipping and coiling. Imaging is necessary to ensure adequate treatment of the aneurysm, and long-term follow-up imaging has the additional advantage of identifying de novo aneurysm formation, which occurs in patients with a history of aneurysmal SAH at a rate of 1–2% per year.89,104 DSA remains the gold standard for follow-up imaging of both clipped and coiled aneurysms.105 CTA can provide adequate imaging of some previously clipped aneurysms, but beam-hardening and streak artifacts preclude effective post-coiling evaluation. MRA can provide excellent imaging of coiled aneurysms but is useless with clipped aneurysms because of artifact. The authors of this handbook use DSA for early post-operative assessment of clipped aneurysms and MRA for routine surveillance imaging of coiled aneurysms.

1.6 Follow-up of Clipped Aneurysms: CTA

Early post-treatment goals for post-clipping angiography include assessing completeness of aneurysm occlusion, ruling out arterial narrowing or occlusion by the aneurysm clip, and evaluating for possible vasospasm. Late goals of imaging in clipped aneurysms include assessing the stability of the clipped aneurysm and ruling out de novo aneurysms in other intracranial arteries.105 Susceptibility artifacts generated by aneurysm clips limit the use of MRA for evaluation of clipped aneurysms. Anatomy in the region of the clipped aneurysm and flow within arteries located “downstream” from the clip-related artifact are often impossible to visualize due to these effects. DSA or CTA is therefore required for follow-up of these patients. Recently, CTA has been assuming a larger role in post-clipping evaluation. Reports comparing CTA to DSA for detection of post-clip aneurysm residuals have yielded variable results that may reflect differences in the types of aneurysm clips employed and different MDCT systems and scanning and post-processing techniques. CTA is less useful in patients in whom multiple surgical clips have been used for reconstruction, or in patients with cobalt alloy-containing clips.106,107 Reported CTA sensitivities for aneurysm remnants have tended to be highest (100%) in studies where only titanium clips were used.108,109 A recent series of 31 consecutive patients undergoing both DSA and 64-detector CTA after aneurysm clipping with a variety of aneurysm clips (including cobalt-alloy clips) found an overall CTA sensitivity and specificity of only 50% and 100% respectively for aneurysm remnants. When only considering remnants measuring >2 mm on DSA, the sensitivity and specificity of CTA improved to 67% and 100%, respectively. The authors noted: “Conventional DSA remains the most accurate postoperative radiological examination to evaluate the quality of the clipping in every circumstance.”110 The sensitivity of CTA for evaluating vessel patency adjacent to the clipped aneurysm is also lower than that of DSA.108,110 Post-processed CTA data in which clips have been digitally “removed” with a bone removal software package improves the sensitivity of CTA for detection of postoperative aneurysm residual.111

1.7 Follow-up of Coiled Aneurysms: MRA

MRA is well-suited to for non-invasive follow up of coiled aneurysms since coil-related artifacts on MR are mild and can be minimized with use short echo time (TE) 3DTOF MRA acquisitions and/or contrast enhanced MRA.112,113 Contrast enhanced MRA is performed with a very short TE that minimizes susceptibility effects produced by aneurysm coils and signal loss due to turbulent flow (intravoxel phase dispersion). Intravascular contrast also reduces signal loss due to these factors and reduces the loss of signal in slowly flowing blood that occurs due to spin saturation.114 Finally, use of elliptic centric view-order sampling of κ-space enables data filling of the central portions of K-space responsible for image contrast during early arterial phase of the gadolinium injection, helping to reduce the effects of venous contamination. While some studies have found no added benefit of contrast administration for MRA of coiled aneurysms.22, others have found contrast enhanced centric phase encoded MRA more sensitive for residual flow in coiled aneurysms.114, and especially aneurysms treated with stent-assisted coiling.115 Residual aneurysm filling may be difficult to detect within coiled aneurysms with DSA (“the gold standard”) against the opaque coil mass or due to subtraction artifacts. Therefore, it is difficult to compare sensitivity of MRA versus DSA. Some have observed that contrast enhanced MRA.23 or 3DTOF MRA at 3T may actually demonstrate contrast filling within the coil mass more clearly than DSA.116 Finally, use of MRA (including contrast enhanced MRA) has been found to be more cost-effective than follow-up with intra-arterial DSA.24

1.7.1 MRA Protocol for Imaging of Coiled Aneurysms:117

• NVPA Coil

• Axial plane

• Pulse sequence: Vascular TOF SPGR

• Imaging options: Variable bandwidth, Fast, 2ip512, Zip2, Smart Prep

• TE: Minimum

• Flip angle: 45

• Bandwidth: 41.67

• Freq: 320

• Phase: 224

• Nex: 1

• Phase FOV: 0.75

• Scan time: 1:01

• FOV: 22

• Slice thickness: 1.4

• LOCS per slab: 60

• Frequency direction: AP

• User CVs screen

• Maximum monitor period: 30

• Image acquisition delay: 5

• Turbo mode (1) Faster

• Elliptical Centric (1) on

1.8 Indocyanine Green Videoangiography

Microscope-integrated indocyanine green (ICG) videoangiography is a useful imaging technique during aneurysm surgery118,119 and other kinds of neurosurgical procedures. ICG is a near-infrared fluorescent dye that binds tightly to plasma globulins and remains intravascular with normal vascular permeability. It has a half-life of 3–4 min and is eliminated exclusively by the liver. Following an IV injection of ICG, fluorescence is induced with a microscope-integrated light source with a wavelength of 700–850 nm and is imaged with a video camera. It is useful in aneurysm surgery to (1) check for complete exclusion of the aneurysm after clipping and (2) make sure the adjacent parent vessels are still patent. ICG is only visible within exposed vessels in the surgical field; it cannot be visualized through tissue.

• 1) Additional applications. Aside from aneurysm surgery, ICG has been used in surgery for:

  • a) Dural arteriovenous fistulas.120

  • b) EC-IC bypass.121

  • c) Decompressive craniectomy for stroke.122

  • d) Brain tumors.123,124

  • e) Spinal vascular lesions.125–127

• 2) Devices and drug

  • a) The microscope (Zeiss Pentero) must be outfitted with the ICG videoangiography module (FLOW 800, Carl Zeiss, Oberkochen, Germany)

  • b) IC-Green™ (Akorn, Inc., Buffalo Grove, IL) comes in 25 mg vials. The drug contains 5% sodium iodine. It should not be given to patients with a history of adverse reactions to iodine or iodinated contrast media.

    • i) Dose: 25 mg per dose, one-size-fits-all.

    • ii) Alternative dose: 0.2–0.5 mg/kg; daily dose should not exceed 5 mg/kg.128

• 3) Technique

  • a) Consent. Informed consent should include the risk of anaphylaxis (1 in 500)129

  • b) Just prior to completing the surgical exposure, have the anesthesiologist prepare the ICG.

  • c) Activate video recording on the microscope.

  • d) Inject the ICG.

  • e) Continue recording until the bolus of dye passes through the area of interest.

• 4) Tips:

  • a) Repeat doses can be given 20 min or less after the most recent dose without significant residual fluorescence interference from the previous injection.

  • b) Oxygen saturation measurements may show falsely low values during the first pass of the drug.130