CT angiography in acute ischemic stroke: the right tool for the job?

The edifice of peer-reviewed literature, particularly in neuroradiology, has new technology as one of its major building blocks. Indeed, one of the easiest paths to publication is the application of a new technological “toy” to a neurodiagnostic problem, or a refinement of that toy to allow a new application (take if from one with shameless experience). For example, with each new iteration of technology, a large number of articles dealing with the pituitary gland has been republished, sometimes even by the same authors. Thus, literature on pluridirectional tomography of the pituitary fossa was followed by articles on computed tomography (CT) evaluation of the pituitary gland, which in turn were supplemented by reports on “dynamic” CT evaluation of the pituitary fossa with reformations, only to be replaced by magnetic resonance (MR) imaging of the pituitary, and (not surprisingly) more recently by “dynamic” MR imaging of the pituitary gland. It is also the fate of literature dealing with evaluation of the cerebrovascular system, particularly in the setting of ischemic stroke. The advent of the digital age brought the first alternative to conventional angiography, with intravenous digital subtracted angiography receiving a great deal of attention in the early 1980s, a promise unfulfilled. At that time, when dynamic table incrementation was developed for CT scanners, the concept of CT angiography was proposed (1). However, only duplex ultrasound and MR angiography have seen widespread use alongside conventional angiography for the evaluation of the cervicocranial vessels. Now comes the most recent wrinkle in the technology continuum. Slip-ring capabilities and higher-heat loading capacity of X-ray tubes have been combined in the CT instrument to allow continuous table translation in respect to the CT gantry while the tube spins. An intravenously injected contrast bolus can thus be “chased” in its early transit, while still concentrated, through the cerebrovascular arterial system. Two articles in this issue of the AJNR (2, 3) demonstrate beautifully the resulting image quality and potential of this technique in evaluating the intracranial vasculature when assessing patients with acute stroke. The authors are to be congratulated for providing level 2 data for CT angiography in Thornbury’s model of diagnostic efficacy (4). However, as with other promising techniques that provided high-quality images, the question must be asked whether the typical early enthusiasm of pioneering workers will hold up in the long run. To answer this question, another question needs to be asked: What do we need to know in the earliest stages of acute ischemia in order to direct patient treatment? After excluding hemorrhage, the first and most important piece of information needed for proper treatment is whether the ischemic insult is still reversible or already permanent. Next, the cause of the ischemic insult needs to be determined if possible. If vascular, is the lesion proximal (a high-grade preocclusive or occlusive cervical carotid lesion), a dissection at the skull base, an occlusive lesion in one of the major vessels leading to the circle of Willis, or a lesion in one of the major vessels distal to the circle of Willis, such as the M1 segment of the middle cerebral artery? Based on the answers, the next management steps might include investigation of the heart as a possible source of embolus, conventional angiography (when the noninvasive examinations fail to reveal a cause of the symptoms), or perhaps even an attempt at endovascular intervention.