18 Chapter 1 In 1985 Wedeen et al first described the possibility of using magnetic resonance to image arteries noninvasively.215 Although a variety of early techniques were investigated, the most promising techniques were based on the phase change experienced by moving spins in the presence of a bipolar magnetic field gradient (phase contrast, PC) or on the saturation of stationary tissue and the bright signal obtained from the inflow of fresh unsaturated magnetization in blood. The latter technique is most commonly called time-of-flight (TOF) MRA but has been more recently referred to as inflow- enhancement MRA.216 The same period that saw the development of flow dependent MRA techniques, also saw the publication of the first reports on a contrast medium for MRI.217-219 Contrast-enhanced MRA allows for better background suppression and shorter acquisition times as the Gadolinium-based contrast agents shorten the T1 value of inflowing blood. This was initially combined with TOF MRA sequences.220,221 Due to the presence of a blood-brain barrier, contrast media do not enter the normal brain tissue, thus enhancing the contrast between brain tissue and blood vessels. More recently, first-pass dynamic subtraction angiography was developed, further reducing acquisition times.222-224 First-pass dynamic subtraction angiography requires a gradient system with high slew rates for ultrafast acquisitions and additional tools for the proper timing of the scan sequence initiation to coincide with the contrast bolus arrival,225 feasible with modern scanners. Due to the short arteriovenous transit time in the cerebral circulation, early enhancement of venous structures limits the time window for scan acquisition, and a trade-off is therefore necessary between scan time, volume coverage and spatial resolution.226 Subtraction of an unenhanced data set from the contrast enhanced data set is performed to delete residual high signal of nonvascular structures.227,228 Further improvements in contrast-enhanced MRA were accomplished through optimized k-space filling:229-231 the principle that the center of k-space determines the contrast and the periphery resolution of the image is used to manipulate the order in which k-space is filled with the acquired signal. The technique which uses automatic triggered first-pass contrast acquisitions with centric ordered k-space filling (called ‘ATECO’ by Farb et al232,233) is employed in this thesis, and is henceforth referred to as contrast-enhanced MRA (CEMRA). In the first report of MRI of intracranial aneurysms, the aneurysms were visualized by using flow voids on standard T1- and T2-weighted spin echo images,234 and it was concluded that ‘MRI was able to show aneurysms of 5 x 5 mm or larger’. The use of specific flow-dependent MRA sequences for imaging of the cerebral vessels was described a few years later235-237 and the first reports on the application of this technique to the detection of intracranial aneurysms appeared in the early 1990s.238-240 A multitude of publications followed, evaluating the different MRA techniques and comparing them with each other,240- 244 with CTA,132,133 or with DSA,245-253 which in all cases was considered the standard of reference. PC MRA proved to be inferior to TOF MRA in most studies242,243,254 and was soon abandoned as a technique for aneurysm detection.228 All authors concluded that MRA could not replace DSA in the detection of cerebral aneurysms due to insufficient sensitivity, especially for smaller aneurysms, but that it could be useful as a screening tool in patients at higher risk of harboring intracranial aneurysms.239,245,251 The first, and to date the only, meta-analysis on evaluating MRA in detecting intracranial aneurysms is the previously mentioned review of White and Wardlaw.135 In this analysis, besides the 16 CTA studies, 20 studies using MRA were evaluated, comprising 926 patients. All these studies used TOF MRA acquisition sequences and three of them247,253,255 used both TOF and PC sequences. Pooled MRA
proefschrit Van Zwam inhoud met kaft en stellingen.indd
To see the actual publication please follow the link above