The flat detector technology that is now available in the angiography suite on most commercial C-arm systems provides a platform for both CT and angiographic imaging, and with it, the potential for high spatial resolution anatomic and physiologic imaging for intraprocedural diagnosis and evaluation.

“Time is Brain” has now become the mantra for stroke management and workflow in the neurologic and endovascular neurosurgical communities. Efforts to identify the cause of acute ischemic stroke quickly, streamline the triage of patients with stroke, and institute appropriate attempts at revascularization, eg, IV thrombolytics or endovascular revascularization (when appropriate) have become a priority in modern stroke centers. In parallel, the latest generation of endovascular mechanical thrombectomy devices shows promising results in revascularizing patients with large vessel occlusions who do not respond to or are poor candidates for intravenous thrombolytic therapy.

As imaging researchers, the goal is to understand clinical needs and workflows and then develop imaging tools that have the potential to add value to these procedures. Perfusion imaging (using CTP or MR) is a method that is commonly used in an attempt to distinguish ischemic core from penumbral tissue. Many physicians would agree that this information is potentially valuable, but choose not to perform these advanced imaging studies because of the additional time they require. The goal of our research was to assess the potential to perform cerebral perfusion imaging directly in the angiography suite using C-arm CT.

With dynamic perfusion imaging, we know that there are two important factors required to adequately sample the time attenuation properties of the contrast bolus: 1) adequate temporal sampling and 2) adequate signal-to-noise ratio for parenchymal voxels, particularly voxels located in the ischemic regions that may have significantly lower enhancement relative to that of healthy parenchyma. Relative to a commercial MCCT scanner, C-arm CT systems have lower temporal resolution (4–6 seconds vs 1 second) and lower detector sensitivity. The initial work in the field of

dynamic C-arm CT perfusion imaging was performed by Ganguly et al,¹ who used a serial series of aortic arch injections combined with a series of acquisitions (offset in time) to overcome the temporal resolution limits of the C-arm CT. In our study, we modified a commercial robotic angiography system to rotate at its maximum angular velocity to understand how this might improve our ability to obtain dynamic perfusion parameters with high-speed C-arm CT when compared with standard CTP. Our results show that the current C-arm CT systems have adequate SNR to measure cerebral perfusion; however, the systems still lack the required temporal resolution needed to adequately characterize the arterial input function associated with a standard injection protocol. This limitation resulted in a systematic overestimation of measured CBF and CBV values in our experiments.

New flat panel detectors (recently commercialized) provide improved SNR and readout rates and should further improve the image quality of the C-arm CT scans. Further engineering efforts are also likely to result in small increases in temporal resolution for C-arm CT. Finally, new algorithms that include compressed sensing^2,3 and iterative model-based reconstruction techniques⁴ have provided promising preliminary results in obtaining the temporal resolution necessary to provide quantitative cerebral perfusion measurements equivalent to MDCT perfusion with existing C-arm CT systems.

References

1. Ganguly A, Fieselmann A, Marks M, et al. Cerebral CT perfusion using an interventional C-arm imaging system: cerebral blood flow measurements. AJNR Am J Neuroradiol 2011;32:1525–31. doi: 10.3174/ajnr.A2518
2. Tang J, Xu M, Niu K, et al. A novel temporal recovery technique to enable cone beam CT perfusion imaging using an interventional C-arm system. Proc SPIE 8668, Medical Imaging 2013: 86681A. doi: 10.1117/12.2007620
3. Nett BE, Brauweiler R, Kalender W, et al. Perfusion measurements by micro-CT using prior image constrained compressed sensing (PICCS): initial phantom results. Phys Med Biol 2010;55:2333–50. doi: 10.1088/0031-9155/55/8/014
4. Wagner M, Deuerling-Zheng Y, Möhlenbruch M, et al. A model based algorithm for perfusion estimation in interventional C-arm CT systems. Med Phys 2013;40:31916. doi: 10.1118/1.4790467

 

Read this article at AJNR.org . . .