Jump to comment:

• Re: Optical coherence tomography angiography identifies peripapillary microvascular dilation and focal non-perfusion in giant cell arteritis
  Eric D Gaier, Aubrey L Gilbert, Dean M Cestari and John B Miller
  Published on: 31 August 2018
• Re: Optical coherence tomography angiography identifies peripapillary microvascular dilation and focal non-perfusion in giant cell arteritis
  Nicole Balducci, Corrado Gizzi, Vincenzo Parisi and Piero Barboni
  Published on: 23 April 2018

• Published on: 31 August 2018

  Re: Optical coherence tomography angiography identifies peripapillary microvascular dilation and focal non-perfusion in giant cell arteritis

  • Eric D Gaier, Ophthalmologist Massachusetts Eye and Ear Infirmary, Harvard Medical School, Boston, MA
  • Other Contributors:
    • Aubrey L Gilbert, Ophthalmologist
    • Dean M Cestari, Ophthalmologist
    • John B Miller, Ophthalmologist

  Dear Editor,
  We thank Dr. Balducci and her colleagues for their interest in our paper [1]. They raise several important points regarding optic nerve angiography, and we are thankful to have the opportunity to discuss these items further.
  In preparation of our manuscript, we felt that diffuse changes in the peripapillary capillary network were best appreciated at lower magnification. Balancing this objective in presentation with a sufficient resolution to appreciate the focal deficits we highlighted, the image sizes published represent what we felt was the best compromise. For those who feel that higher magnification images are needed, we have included in this letter Figure 1, which includes the same 6x6 mm OCT-A images in the acute phase for all cases in our study. Quantitation of OCT-A signal can be a powerful way to objectively assess regional as well as between eye and patient differences. We have recently performed a quantitative assessment of angiographic signal in non-arteritic anterior ischemic optic neuropathy using a different device, the Optovue Avanti (Fremont, CA) [1]. However, in the current study, the small number of affected eyes did not allow for meaningful statistical analysis of quantitative data. In addition, quantitative analyses can be misleading when confounding artifacts or segmentation errors are present as discussed below.
  Jia and colleagues [2] showed a strong non-linear correlation between RNFL thickness and radial peripapillary...

  Show More

  Dear Editor,
  We thank Dr. Balducci and her colleagues for their interest in our paper [1]. They raise several important points regarding optic nerve angiography, and we are thankful to have the opportunity to discuss these items further.
  In preparation of our manuscript, we felt that diffuse changes in the peripapillary capillary network were best appreciated at lower magnification. Balancing this objective in presentation with a sufficient resolution to appreciate the focal deficits we highlighted, the image sizes published represent what we felt was the best compromise. For those who feel that higher magnification images are needed, we have included in this letter Figure 1, which includes the same 6x6 mm OCT-A images in the acute phase for all cases in our study. Quantitation of OCT-A signal can be a powerful way to objectively assess regional as well as between eye and patient differences. We have recently performed a quantitative assessment of angiographic signal in non-arteritic anterior ischemic optic neuropathy using a different device, the Optovue Avanti (Fremont, CA) [1]. However, in the current study, the small number of affected eyes did not allow for meaningful statistical analysis of quantitative data. In addition, quantitative analyses can be misleading when confounding artifacts or segmentation errors are present as discussed below.
  Jia and colleagues [2] showed a strong non-linear correlation between RNFL thickness and radial peripapillary capillary density in 10 healthy subjects. That healthy subjects with normal vision and no (ischemic) posterior pole disease were analyzed is an important distinction. Indeed, there is an increase in RNFL thickness associated with ischemic optic neuropathies, but caution should be used in extrapolating trends based on normative data to the pathologic states of ischemic optic nerve disorders. In addition, arteritic anterior ischemic optic neuropathy (AAION) produces edema and is commonly associated with peripapillary cotton-wool spots that produce blockage effects, as demonstrated in cases 1-3 in our series (denoted with blue arrows) [1]. These secondary effects can alter normal anatomic boundaries and thus segmentation, which can lead to erroneous results and inaccurate conclusions in a small cohort.
  With respect to the correspondence of choroidal filling defects seen on FA and OCT-A, it is important to first acknowledge that spectral domain OCT is suboptimal for imaging sub-RPE (vascular) structures due to limitations in resolution at deeper penetrance lamina [3]. Of note, we are currently collecting a series of patients with ischemic optic neuropathies imaged using swept-source OCT-A with the specific focus on choroidal vascular perfusion (Tieger et al, in preparation), though there are limitations with this approach as well. We agree that wide-field imaging may been better suited for capturing choroidal perfusion defects and have incorporated this into our imaging strategy moving forward. In the current study, we performed macular scans for some patients in addition to optic disc-centered scans in a subset of patients, including the initial presentation of case 2 (Figure 2). Fluorescein angiography revealed a clear choroidal filling defect affecting the macular region temporal to the disc and extending inferiorly (red arrow). Counter-intuitively, OCT-A showed a corresponding increase in angiographic signal in the superficial lamina and choroidal lamina (Figure 2B,D). We did not feel that the correspondence between the apparent increased choriocapillary OCT-A signal and the reduced fluorescence on FA in the area inferior to the disc (red arrows) was clear enough for publication especially in the context of other imaging limitations [3].
  Balducci and colleagues [4] reported a single case of acute AAION imaged with the Optovue device and described corresponding perfusion defects on FA and OCT-A that were continuous with the optic disc. However, there are several limitations with the image presented. First, the image is confounded by motion artifact. Second, the image is too highly magnified such that the region of interest is at the edge of the sampling area, where segmentation errors can occur. Given that the retinal arterioles visible on FA are not visible in the superficial lamina on OCT-A, segmentation error is likely present, similar to case 2 in our series (Figure 3). Comparison with composite images can reveal such errors. Third, no fundus photograph is presented to exclude edema, which can impart a blockage effect as also demonstrated in case 2 in our series (Figure 2; blue arrow) [1].
  In conclusion, OCT-A is a new and powerful tool for imaging microvascular anatomy in normal and pathologic conditions of the posterior pole, but OCT-A image acquisition and segmentation can be prone to artifact and misrepresentation. Balducci and colleagues highlight a number of these pitfalls in the early application of this new technology in the analysis of ischemic optic neuropathies. As the field continues to explore the application of OCT-A, it will be critical to remember that the study of the pathologic state teaches us about the technology (and limitations therein) just as the technology provides new insights into the pathophysiology.
  Figures can be downloaded at the following link and will be available through 8/31/2019: https://www.dropbox.com/sh/kf1fp9gsuas4zm1/AACLiIA2cKLF_J9nkWGdymPza?dl=0
  Conflicts of Interest: None
   
  FIGURE LEGENDS
  Figure 1. Enlarged Optical Coherence Tomographic Angiography (OCT-A) images (6 mmx 6 mm; Zeiss) of all controls and cases. Superficial, choriocapillary and choroidal laminar segmentations are included. The right (top) and left (bottom) eyes are included for each case, which are demarcated by red lines.
  Figure 2. Enlarged and expanded OCT-A sampling regions for case 2. (A) Corresponding fluorescein angiography image (early) depicting blockage by an overlying cotton wool spot along the superior arcade (blue arrow) and choroidal hypoperfusion temporal to the optic disc most prominent inferiorly (red arrow) (represented as Figure 3H in initial report). (B-D) Superficial, choriocappilary and choroidal laminar segmented OCT-A images are included with montaged samplings centered on the optic disc and macula. Blue and red arrow represent the same regions corresponding to those in the fluorescein angiogram; the blue arrows indicate signal blockage extending through all lamina, and the red arrows indicate the region with increased OCT-A signal in the superficial and choroidal laminae. The yellow arrow denotes attenuated OCT-A signal in a region superior to the optic disc in the superficial lamina. Images are each 6 mm x 6 mm and were acquired using the Zeiss device.
  Figure 3. Enlarged OCT-A images for re-presentation of case with optic disc edema and AAION. (A) Fundus photograph of the optic disc depicting pallid optic disc edema that is obscuring the peripapillary retinal arterials (blue arrow). (B,C) Composite and superficial laminar OCT-A images (3 mm x 3 mm; Optovue) for the corresponding region depicted in the fundus photograph. There is apparent retinal arterial signal attenuation in the superficial lamina (C) as denoted by the blue arrow that is partially secondary to segmentation error as it is better preserved in the composite segmentation (B; blue arrow). There is some component of blockage due to overlying edema as evident in the fundus photograph, but some regions of peripapillary OCT-A signal attenuation likely represent true hypoperfusion as they do not correspond to overlying blockage elements (for example: yellow arrow).
   
  REFERENCES

  1. Gaier ED, Gilbert AL, Cestari DM, Miller JB. Optical coherence tomographic angiography identifies peripapillary microvascular dilation and focal non-perfusion in giant cell arteritis. Br J Ophthalmol 2018;102(8):1141-46.
  2. Jia Y, Simonett JM, Wang J, et al. Wide-Field OCT Angiography Investigation of the Relationship Between Radial Peripapillary Capillary Plexus Density and Nerve Fiber Layer Thickness. Invest Ophthalmol Vis Sci 2017;58(12):5188-94.
  3. Diaz JD, Wang JC, Oellers P, et al. Imaging the Deep Choroidal Vasculature Using Spectral Domain and Swept Source Optical Coherence Tomography Angiography. Journal of vitreoretinal diseases 2018;2(3):146-54.
  4. Balducci N, Morara M, Veronese C, et al. Optical coherence tomography angiography in acute arteritic and non-arteritic anterior ischemic optic neuropathy. Graefes Arch Clin Exp Ophthalmol 2017;255(11):2255-61.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 23 April 2018

  Re: Optical coherence tomography angiography identifies peripapillary microvascular dilation and focal non-perfusion in giant cell arteritis

  • Nicole Balducci, Ophthalmic doctor Studio Oculistico d'Azeglio, Bologna, Italy
  • Other Contributors:
    • Corrado Gizzi, Ophthalmic Doctor
    • Vincenzo Parisi, Ophthalmic Doctor
    • Piero Barboni, Ophthalmic doctor

  Dear Editor,
  We read with interest the paper by Gaier et al.1 The collection of 5 eyes affected by acute A-AION and analyzed with OCT-A is remarkable, due to the rare disease presentation. The main finding of the paper was that during the acute phase of A-AION, diffusely dilated superficial peripapillary capillaries were detectable on OCT-A. Interestingly, peripapillary capillary dilatation was also noted in the fellow eye. Unfortunately, the figures presented by the authors are too small to allow the reader to qualitatively appreciate the capillaries dilatation. More detailed images and a quantitative vessels analysis would have helped to document the microvascular changes.
  They hypothesized that the capillary dilatation may represent a form of luxury perfusion in the setting of short ciliary arterial compromise or a centrally mediated autoregolatory mechanisms in the setting of reduced perfusion of the optic nerve. These hypotheses are interesting, but it is important to differentiate the RNFL thickness increase from the capillary dilatation, as peripapillary capillary plexus density and RNFL thickness are highly correlated and fit well with a nonlinear stacked-layer model.2
  Moreover, the authors stated that OCT-A laminar analysis did not highlight the choroidal/choriocapillaris perfusion defects seen on FA. However, a recent study3 showed a tight correspondence between the choroidal perfusion defects visible on FA (and even better on indocyanine green...

  Show More

  Dear Editor,
  We read with interest the paper by Gaier et al.1 The collection of 5 eyes affected by acute A-AION and analyzed with OCT-A is remarkable, due to the rare disease presentation. The main finding of the paper was that during the acute phase of A-AION, diffusely dilated superficial peripapillary capillaries were detectable on OCT-A. Interestingly, peripapillary capillary dilatation was also noted in the fellow eye. Unfortunately, the figures presented by the authors are too small to allow the reader to qualitatively appreciate the capillaries dilatation. More detailed images and a quantitative vessels analysis would have helped to document the microvascular changes.
  They hypothesized that the capillary dilatation may represent a form of luxury perfusion in the setting of short ciliary arterial compromise or a centrally mediated autoregolatory mechanisms in the setting of reduced perfusion of the optic nerve. These hypotheses are interesting, but it is important to differentiate the RNFL thickness increase from the capillary dilatation, as peripapillary capillary plexus density and RNFL thickness are highly correlated and fit well with a nonlinear stacked-layer model.2
  Moreover, the authors stated that OCT-A laminar analysis did not highlight the choroidal/choriocapillaris perfusion defects seen on FA. However, a recent study3 showed a tight correspondence between the choroidal perfusion defects visible on FA (and even better on indocyanine green angiography) and on OCT-A at the level of the choriocapillary in a single case of acute A-AION.
  The choroidal defects noted on FA by Gaier et al.1 were slightly distant from optic disc border and they did not fit inside the small scan area of the OCT-A exam. Larger OCT-A scan size could help to highlight choriocapillaris perfusion defect in those cases.

  References
  1 Gaier ED, Gilbert AL, Cestari DM, Miller JB. Optical coherence tomography angiography identifies peripapillary microvascular dilation and focal non-perfusion in giant cell arteritis. Br J Ophthalmol Published Online First: [2017 Nov 9] DOI: 10.1136/bjophthalmol-2017-310718

  2 Jia Y, Simonett JM, Wang J1, et al. Wide-Field OCT Angiography Investigation of the Relationship Between Radial Peripapillary Capillary Plexus Density and Nerve Fiber Layer Thickness. Invest Ophthalmol Vis Sci. 2017;58:5188-5194.

  3Balducci N, Morara M, Veronese C, et al. Optical coherence tomography angiography in acute arteritic and non-arteritic anterior ischemic optic neuropathy. Graefes Arch Clin Exp Ophthalmol 2017;255:2255-2261.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top