16 Chapter 1 were outside the area of coverage. The authors stated that these studies had been performed with four- detector-row scanners, but in fact in only one of these five a four-row scanner was used.178 The other four were studies in which single-row scanners were used. This nicely illustrates one of the main drawbacks of the older, single-row scanners, which is the limited coverage within a certain time window. Larger coverage would require longer scan times which would lead to more motion artefacts, as well as to more undesired enhancement of venous structures. Another improvement of modern multi-detector scanners is the higher spatial resolution. Even though spatial resolution depends on the collimation of the scanner and not necessarily on the number of detector rows, modern scanners with more detector rows use smaller collimation and therefore have better spatial resolution. In the meta analysis of Menke et al156 a subgroup analysis based on scanner type was carried out. They reviewed 45 studies with 3,643 patients, 26 of which have previously been included in the meta-analysis by Westerlaan et al. and 9 of which were used in older meta-analyses.135- 137 Sixteen of the studies reviewed by Menke et al. utilized single-detector-row scanners; the other 29 utilized four to 64 detector row scanners. Pooled sensitivity and specificity were found to be 95.0 and 96.2% respectively. When stratified to the number of detector rows however, the pooled sensitivities ranged from 91.2% for single-row scanners to 97.8% for 64-row scanners and specificity ranged from 90.8% for 4-row-scanners to 97.7% for 64-row scanners (pooled specificity for single-row scanner studies was 94.0%). Sub-millimeter slice thickness was only achieved in 16- and 64-detector-row scanners: the 16 studies in Menke’s meta-analyses using sub-millimeter slice thickness had significantly better results in detecting aneurysms than the studies using a slice thickness of 1 millimeter or more: sensitivity 98.8% vs. 95.9% (p=0.006) and specificity 99.4% vs. 96.4% (p=0.029). A few more studies have been published since these meta-analyses appeared, using 16-,179-181 64-182,183 and even 320 detector-row scanners.184,185 One of these studies reported detection of aneurysms smaller than 3mm with a sensitivity of more than 95%.183 Another new CT scanning technique currently under investigation is dual energy CT186 which allows for automatic bone subtraction.187-190 The results so far look promising, but further evaluation of this technique with larger study population is necessary. The studies mentioned above consistently use DSA as the standard of reference, sometimes in combination with perioperative findings. A few studies report aneurysms missed by DSA that were discovered with CTA and confirmed during treatment,179,191,192 and one study reports a false-positive DSA finding with true-negative CTA, confirmed at surgery.193 Some claim that, for the detection of small aneurysms, 3D CTA has a superior sensitivity than conventional DSA192 or even rotational DSA.183 Considering these reports we are probably justified in stating that the accuracy of CTA with modern multi-slice scanners at least approaches that of DSA in the detection of intracranial aneurysms. The development of new techniques to treat aneurysms has influenced the requirements of aneurysmal detection and evaluation tools. In the last decade, and especially since the publication of the results of the International Subarachnoid Aneurysm Trial (ISAT),194 the new endovascular treatment technique of intracranial aneurysms using platinum Guglielmi detachable coils (GDC)195,196 has gained acceptance and become the primary treatment mode in many centers. As a consequence of this new treatment, not only must a diagnostic tool detect the aneurysm but it must also be able to assess the feasibility of endovascular treatment.197,198 Thus, studies assessing the performance of non-invasive diagnostic tools in patients suspected of harboring an intracranial aneurysm now focus on both its detection accuracy as
proefschrit Van Zwam inhoud met kaft en stellingen.indd
To see the actual publication please follow the link above