OCT Angiography

[Dusn]

---

Members don't see this ad.

A clinic I work at has gotten funding to get a new OCT. I'd like to get one with OCT Angiography but unfortunately there seem to be few options at this point. I know Zeiss and Optovue both have one but it sounds like optovue's Angiovue is not available in the US yet? So does this mean there's only one option at this point, which is Zeiss?

I wanted to get experiences from those of you who may have used it in clinical practice. Is there a lot of artifact which makes interpretation impossible? Does it take a long time to get a scan? Is it helpful in differentiating CNV from a non-CNV cause of sub-retinal fluid (such as atypical CSR).

Thanks.

 

[MullerCell]

I've had a Zeiss demo in clinic and did not find that it impacted my clinical decision making in any way. The images are easy to acquire and fast (few seconds per eye). The cost of the OCT unit with angiography is about $20,000 extra and you are not able to bill anything for the angio study, so doesn't really make economic sense. I don't think it replaces doing an FA, and I found that sometimes the images aren't easy to interpret (likely in part due to my lack of experience).

 

[macula87]

Currently in the US, the Angiovue module is FDA approved for Zeiss only. I think the approval for the Optovue is going to be soon.

The scan quality depends on many factors including visual acuity and the ability of the patient to fixate. Artifacts are seen frequently but these are well described and easy to identify. http://www.ncbi.nlm.nih.gov/pubmed/26428607

OCT angiography can very helpful in detecting CNV cases and might spare you the need to do an FA. http://www.ncbi.nlm.nih.gov/pubmed/25795476

 

[100D]

OCTA is very cool and certainly will play a large role in clinical practice in the future, but IMO, it's probably not prime time yet. It's an amazing research tool, but the difficult interpretation for non-experts and limited scanning area would probably not make OCTA beneficial for most busy retina practices. It certainly should not replace a conventional FA; most OCTA studies get both OCTA and FA. I think the OCTA studies have provided great insights into pathophysiology and can add information for difficult cases, but I doubt that it would change clinical management in 99% of our patients currently.

 

[MstaKing10]

I heard that there are patent issues that preclude octa becoming available from other companies such as Heidelberg etc. the Zeiss machine is high quality. Scans add an extra 6-10 seconds to aquire. Getting good scans is highly operator dependent. Motion artifact is the biggest issue but we've gotten much better at eliminating this. Just like any new technology, the clinical interpretation and use of this modality is evolving but it's fascinating to see the degree of capillary drop out in vascular disease, microaneurysms in diabetics and the anatomy of CNV in Amd especially after treatment or in chronic/mature membranes. I think eventually it will supplant angiography especially as software advances to incorporate larger scan areas (only 6x6 now) and flow dynamics.

 

[speyeder]

OCT-A is a fascinating new technology but its clinical relevance is still not understood yet. In addition to motion artifacts, lesions that are slow-flow may not show up on OCT-A. It appears to be helpful in identifying CNV and capillary non-perfusion in ischemic vascular diseases. But how much will OCT-A alter treatment decisions? It may aid us in initiating therapy in some ambiguous cases but in the overwhelming majority of our patients, our treatment decisions will be made on OCT alone. In addition it may not be as useful in following response to treatment, as vascular complexes may or may not regress depending on CNV type. How reliable is it in differentiating active from inactive CNV? Furthermore, should we inject lesions we see on OCT-A that don't show fluid on OCT? Like any new imaging modality, it will take time before we have answers to these and many other questions.

 

[elementals]

FWIW, I work with OCT-A in my research and it's not a replacement for FA, it's a complement. FA gives you timecourse info and shows you extravasation, but misses some of the finer vessels. OCT-A shows incredible but static vascular detail deep into the retina. I don't think we really know how to use OCT-A yet, but that's what the research is for 🙂

 

[Visionary]

Agree with a lot of what has been said here. We demoed the Zeiss software for several months. The acquisition takes significantly longer than standard OCT, and, as such, it's prone to artifacts. The images provide a lot of information, perhaps too much for day-to-day clinic flow; however, I did not see a single case in which it would have changed my management approach. With a $25k price tag and no ability to bill beyond the traditional OCT, I don't see this being widely adopted. Seems like more of a research tool, at this point. Should help us learn more about vascular pathology in the retina and choroid. Software needs refinement, based on additional normative data. Acquisition speed needs improvement, as well. It can show flow abnormalities, sort of, but cannot show leakage. For that reason alone, it will not replace traditional angiography, at least in its current form. I do foresee OCTa being part of regular OCT packages, in the future. Having the ability to perform simultaneous OCT, OCTa, and FA/ICGA on a machine such as Spectralis would be very nice.

 

[Dusn]

Thanks for your comments.

Was the Zeiss OCTA able to reliably replicate the kind of images of the CNV complex that you can get with a Spectralis video ICG and a great photographer who knows how to get the net in proper focus? I often have to use techs who are not the most experienced photographers and my angiography is of pretty poor quality, was hoping an OCTA might help make up for this.

These are examples of the types of video ICGs I'm talking about:
http://www.heidelbergengineering.com/international/products/imaging-modes/icg-angiography/

 

[Visionary]

Sort of. Problem is the CNV actually pop out a lot easier with ICGA. In the first 30-40 seconds, you can usually spot them readily. With OCTa, you're seeing all the vasculature at once. You can segregate image planes, which helps, but you're reliant on the software, which isn't always going to give you the optimal segmentation lines. ICGA is still superior, IMO. If you have Spectralis, you might have your photographer shoot simultaneous FA/ICGA. Helps with quality, since they can focus easier with the FA filter.

 

[Slide]

I did a lot of my research in this field as a resident, more on the clinical side. As the software goes, it is really nice but it seems more practical for research and academic uses right now. We were able to get an early version of the Optovue and the machine is pretty fantastic for regular OCT (high res and wide angle images). The great thing about it is if your photographer/tech can get a good quality OCT, they can get a good quality OCTA too. The pictures are really pretty in that you can see a lot of anatomical detail of blood flow and you can map out nonperfusion as well.

There's still a lot of work clinically that needs to be done to show it does have clinically utility, and it won't replace FA - as other posters have said, it'll likely serve as an adjunct. Some areas my PI and I looked into for possible clinical application included:
-Assessing capillary nonperfusion in higher detail. FA can give you that information but if there are things that will block capillary perfusion like heme, OCTA can get you that info in questionable areas. It also gives you a higher detailed version of the FAZ, but I'm not sure if it will necessary translate to correlating macular ischemia with vision when you can't find another cause for vision to be less than expected.
-Picking up flow in supposedly controlled or quiescent CNVs. On OCT there may be no fluid but you'll see some faster flow sometimes. However, it's still questionable if you would still treat it or not.

Just based on my work with it, the current information we can get from it is nice but I don't know if it affects clinical judgement. It's nice to show to patients if they can get it, but you don't need to get OCTA just for that. I personally think the key to making OCTA not only useful but game changing is having it give you a quantifiable way to measure blood flow velocity in a certain area. Last I saw, there was a way to highlight areas on the macula that was above or below a perfusion/flow index, but unfortunately it was a unitless index. They've also worked on cutting down the acquisition speed but it still takes longer than standard OCT. It took our photographer anywhere from 5 to 10 minutes to get an OCTA along with standard OCT.

Ironically there's probably more potential for OCTA with glaucoma. There's work being done by some of its founders to see if the microvasculature around the optic nerve changes before arcuate defects/NFL loss presents itself.

Edit: one other thing that may make it much more palatable than angiography is getting wider scans and software packages that let you stitch together multiple 6x6 or 9x9 slices that will give you basically an en face OCTA image of the macula and mid-periphery. It would be a great substitute if you had those patients whom you can't get venous access for dye injection.