Our next speaker is Dr. Alex Witek. He's going to be presenting reduction in radiation and contrast dose using MR-Trix protocol prior to spinal angiography. Good afternoon. I'm Alex Witek. I'm a sixth-year neurosurgery resident at the Cleveland Clinic and this is a project I did under the guidance of Dr. Bain and Dr. Masaryk. These are our disclosures. So for patients with a dural arteriovenous fistula, the diagnosis typically follows this sequence. They have symptoms that lead to the ordering of an MRI. The MRI will often show signs that are suggestive of a fistula such as abnormal flow voids, abnormal enhancement, or edema within the spinal cord. But ultimately, these patients undergo a catheter angiogram which is the gold standard for diagnosis. Drawbacks to catheter angiography include that it is a procedure that is invasive, that can be time-intensive, and that often involves large doses of radiation and contrast. So with these drawbacks in mind, we recently implemented a protocol of time-resolved MR angiography prior to catheter angiography in these patients. This MR protocol utilizes a sequence of rapidly acquired images. And when combined with contrast enhancement and digital subtraction, the result is a series of 3D volumes that allow us to follow the contrast from the arteriole all the way through to the venous phase. And our initial experience using this protocol led to the hypothesis that time-resolved MR angiography can help localize a dural AVF and make the angiographic diagnosis more efficient. So this is an example of this MRA protocol. The first thing we typically look for is arteriovenous shunting. If you look here on the panel on the left, this is a sequence of sagittal MIPS, and you can see that there are some perimedullary veins that fill during the arterial phase and appear to originate from the thoracolumbar junction. So then if we pick the optimal time point here, we can take that volume and apply a volume rendering technique. And when we do that, we can see that these abnormal veins actually have a direct connection to this artery, which is the left L1 segmental artery. So this patient then underwent a normal spinal catheter angiogram and was in fact found to have a dural AVF that originated from the left L1 segmental artery. And because we had this MRA information beforehand, we were able to make this diagnosis with only a single injection. So this experience led to this study, which was a single center retrospective study and included patients with a spinal dural AVF diagnosed on catheter angiography. We looked at a more recent group that had time result MRA prior to catheter angiography, and we compared them to a historical control group of patients diagnosed with angiography alone. The primary outcome was radiation dose and secondary outcomes included volume of contrast, number of injections, and procedural time. These are the patient characteristics. We had 38 patients total with eight in the MRA group, 30 in the control group. You can see they were well matched with regard to age and BMI. Most of the patients were male and almost all of them had the typical MRI findings of serpiginous phleboids and cord edema. These are the results of the MRA studies. So of the eight MRA patients, AV shunting was positively identified in all eight of the studies. And the fistula was localized in six out of the eight, including five patients in whom we could precisely localize the fistula to a specific segment and side, and one additional patient in which it was less precisely localized to an area spanning two to three segments. In two of the patients, we could not localize the fistula, and this was because the fistula ended up being outside of our selected field of view on the MR. Okay? So these are the main study results here. With regard to the primary outcome, the radiation dose was lower in the MRA group with a median of 2.2 grays compared to 3.9 in the control group. Number of injections and contrast volume were also significantly lower in the MRA group. The difference in procedural time was not significant. So limitations of this study is that it was small and retrospective and included only patients with a dural AVF diagnosed on angiography. One technical nuance to consider is the limited field of view, which is about 10 segments in our experience. And some future directions for research with regard to this protocol involve investigating how it performs in the whole population of patients presenting with a possible fistula. So in conclusion, time-resolved MRA is a useful adjunct to catheter angiography for diagnosis of these lesions. It can result in lower radiation dose, lower contrast dose, and fewer injections. And more data is needed to fully define its role in the diagnosis and treatment of these lesions. Thank you.

Dr. Alex Witek

time-resolved MR angiography

dural arteriovenous fistulas

spinal angio

catheter angiography