18.1 Digital subtraction angiography

Digital subtraction angiography (DSA) is usually performed via femoral access with a biplane unit with rotational/three-dimensional (3D) capabilities. Intra-arterial injection is performed using iodinated contrast after guiding the catheter to each of the four main vessels of the neck. Using X-rays, 2D views are taken, with rotational radiography used to create 3D models useful for treatment planning. DSA has the highest spatial resolution of all imaging modalities and is considered the gold standard for evaluating intracranial aneurysms. The high spatial resolution allows for aneurysm neck characterization to guide endovascular versus open surgical treatment and better visualization of associated branch vessels. Furthermore, DSA has temporal resolution for evaluation of arterial, capillary, and venous phases. In addition to conventional DSA, 3D rotational angiography (3DRA) further increases the sensitivity of small aneurysms less than 3 mm in size and can identify aneurysms missed by computed tomography angiography (CTA)/magnetic resonance angiography (MRA) or DSA alone. ¹ , ² These factors make DSA the gold standard in evaluating treatment response after flow diversion. While initial results showed a small but significant mortality of 0.06% and a more worrisome stroke rate of 4%, ³ more recent reports show DSA complication rates below 0.3%. ⁴ Because DSA is an invasive procedure, the information acquired from this study must be weighed against the small procedural risk associated with it. Certainly, DSA and 3DRA are essential for treatment planning ( Fig. 18.1) and immediate evaluation of successful flow diversion ( Fig. 18.2).

18.2 Computed Tomography Angiography)

On CTA, contrast fills the aneurysm lumen and allows easy detection of pre- and posttreatment status. For treatment planning, CTA can accurately determine aneurysm lumen location and geometry, with measurements of size, shape, neck size, presence of daughter sacs, and other concerning features including wall irregularities or, rarely, active contrast extravasation diagnostic of aneurysm rupture ( Fig. 18.3). CTA is performed on multi-row detector helical scanners in the axial plane from which multiplanar reformats, maximum intensity projections, and 3D reconstructions are produced (see Table 18.1 for our institutional protocol). Early CT technology lacked sensitivity for detecting aneurysms 3 mm and smaller, with sensitivity as low as 84% using four-channel multi-row detector CT scanners. ⁵ Several investigators and a recent meta-analysis have demonstrated that current CTA protocols with multidetector scanners have a spatial resolution that can reliably diagnose aneurysms compared to DSA, with sensitivity approaching 100%. ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰

CTA can also be used to monitor aneurysms following treatment, though one must be aware of potential artifacts related to metallic hardware, including some stents, coils, and clips. These can result in extensive beam hardening from the metal. This is due to the use of multiple X-ray energies and the “hardening” of the beam after it passes through any high attenuation material leaving only the high energy X-rays. While this can limit the evaluation of aneurysms adjacent to metallic hardware, current flow-diverting stents produce little, if any, beam-hardening artifact ( Fig. 18.4). Following flow diversion, the lumen internal to the stent can be evaluated with appropriate windowing and leveling to maximize differences between the contrast within the lumen and the stent. In addition, post-treatment CTA can also show progressive thrombosis of the aneurysm which can occasionally be detected as a high attenuation crescent on noncontrast CT. However, in the subacute setting, from 6 hours to 1 week after clot formation, the ability to detect blood products drops dramatically.

One drawback to CTA imaging is the radiation dose imparted to the patient. This results in a small, but significant, risk of carcinogenesis as extrapolated from certain studies. Over the past 30 years, medical radiation sources have increased the average radiation dose per person in the United States from an average of 3.6 mSv in the 1980s to 6.2 mSv in 2006. ¹¹ Extrapolating data from radiation exposure at Hiroshima, Three Mile Island, and Chernobyl and radiation workers to current CT use, it has been estimated that about 1.5 to 2.0% of all cancers in the United Stated may be attributable to CT radiation. ¹² A noncontrast CT and CTA of the head results in a mean effective dose of approximately 3.6 versus 4.3 mSv between two studies (1.7 vs. 2.7 mSv for noncontrast head CT and 1.9 vs. 1.6 mSv for head CTA) using estimations from CT phantoms, dosimeters during CT scans on patients, and dose-length products (DLPs) to estimate the effective dose. ¹³ , ¹⁴ This imparts an overall low cancer risk to patients undergoing noncontrast CT and CTA of the brain. As a comparison, the radiation dosages for flow diversion follow-up are significantly less than that during workup of acute ischemic stroke with additional CTA neck and CT perfusion studies (which is estimated at 1 in 1,200 patients). ¹⁴

The effective dose is calculated from the DLP (mGy–cm) and takes into account the radiosensitivity of imaged organs, 80% of which is attributed to the thyroid. ¹⁵ One reason for this low risk is that most patients undergoing scanning for aneurysms are often within the sixth to ninth decades, and exposure in this age group carries a lower risk of excess cancer mortality due to decreased DLP conversion factors. Although the individual risk of cancer with CT/CTA aneurysm workup is likely small, databases have been developed for future large-scale epidemiologic studies addressing CT radiation risk. ¹⁶