Dear Editor,

We read the article published by McKenna, et al (1) with great interest and laud them on the quality and design of their study. Screening for diabetic retinopathy in rural, low resource settings is the need of the hour, however models which are cost effective, yet provide intensive screening and continuum of care are limited. Keeping this in mind, we feel that there are a few points requiring further clarity in this article.

The odds-ratio calculated in table 3 displays the significant effect of didactic training on correct diagnosis by rural doctors. However, for the odds-ratio to be calculated, there would have been a comparison group of rural doctors who were not provided didactic training. The numbers of these doctors have not been mentioned, and no details have been provided as to whether they were given any basic level of training related to the program. In the results provided for comparison between rural doctors and the non-medical graders, it has not been made clear whether doctors who had not been provided didactic training were included. In that case, results presented in the study may have been biased towards the non-medical graders.

In the study, the arbitrator changed the grade for a high percentage of the cases, moreover, 33% of the images were not found to be of adequate quality. Hiring an arbitrator, re-checking the grading and assuring high quality images (2) through standard equipment and trained personnel would drive up the cost of the programme. The telemedicine model by itself has already been proved as cost effective for diabetic retinopathy screening. (3) Thus, as mentioned by the authors, calculation of cost-effectiveness of the model mentioned in the study, should be carried out assessing its’ scalability.

We appreciate the opportunity to be able to discuss our views on the subject and the article in question.

References

1. McKenna M, Chen T, McAneney H, et al.
Br J Ophthalmol 2018;102:1471–1476.
2. Taylor DJ, et al. Image-quality standardization for diabetic retinopathy screening. Expert Rev Ophthalmol. 2009;4(5):469–76.
3. Sharafeldin N, Kawaguchi A, Sundaram A, et al. Br J Ophthalmol 2018;102:1485–1491.

In their report, entitled “Intravitreal chemotherapy in retinoblastoma: expanded use beyond intravitreal seeds“, Abramson and corkers report on the successful use of intravitreal chemotherapy in 52 patients for subretinal seeds and recurrent retinal tumours [1]. They state that, prior to their experience, intravitreal chemotherapy had been used exclusively to control persistent or recurrent vitreous seeding in retinoblastoma that had been refractory to systemic intravenous or intra-arterial chemotherapy.

In fact, intravitreal chemotherapy as an adjuvant treatment for both subretinal seeds and recurrent retinal tumours, including its use instead of systemic chemotherapy in the setting of chemothermotherapy for small unresponsive primary retinoblastomas, has been in regular use already for a decade at the Ocular Oncology Service, Helsinki University Eye Hospital. Indeed, three of the first four patients that we reported during the congress of the International Society of Ocular Oncology in 2009 [2], and published in 2011 [3], received intravitreal methotrexate for reasons other than vitreous seeds. Subsequent experience with intravitreal chemotherapy with methotrexate and, later, with melphalan has strengthened our initial findings, as does the comprehensive report of Abramson and coworkers.

1. Abramson DH, Ji X, Francis JH, et al. Intravitreal chemotherapy in retinoblastoma: expanded use beyond intravitreal seeds. Br J Ophthalmol 2018 Jun 6. pii: bjophthalmol-2018-312037. doi: 10.1136/bjophthalmol-2018-312037.

2. Kivelä T, Eskelin S, Lindahl P, Majander A. Intravitreal methotrexate monotherapy as salvage treatment for recurrent retinoblastoma after standard chemoreduction. In: ISOO Metting 2009: Programme and Abstracts, p. 279. http://www.xcdsystem.com//isoo/webimages/pdf%20%236%20isoo%202009%20camb... (accessed on November 1, 2018)

3. Kivelä T, Eskelin S, Paloheimo M. Intravitreal methotrexate for retinoblastoma. Ophthalmology 2011;118(8):1689, 1689.e1-6. doi: 10.1016/j.ophtha.2011.02.005.

Dear Editor
We read the above paper with much interest and welcome the review and analysis of trends in acanthamoeba keratitis – a very important complication from contact lens wear. The paper discusses the incidence of acanthamoeba keratitis at Moorfields Eye Hospital, a large tertiary referral centre.
We note an incidence of 18.5 cases per annum in 1997-1999, rising to a mean of 50.3 per annum in 2011-2016 and hence has been quite rightly quoted as almost a 3 fold increase in cases.
We would however suggest some caution when using those figures to state that there is an epidemic at present.
When one attempts to take into account the fluctuations in numbers of contact lens wearers with the United Kingdom per year and relate that to incidence of acanthamoeba keratitis one has a slightly modified view.1 There has been a steady increase in contact lens wear with figures from the ACLM estimating 4.2 million CL wearers in 2016. A figure has been created showing this relative incidence in a chart format.2
The figure represents the number of cases diagnosed at Moorfields divided by the number of contact lens users (rising from 1.6 million in 1992 to 4.2 million in 2016). Therefore the mean number of cases when adjusted for CL wearers is 8.5 per year with a standard deviation of 5.8, with 11.8 in 2015 and 14 in 2016.
Whilst there is still a significant rise in cases, compared to the mainly stable period of 1996-2010, the rates are still lower than those experienced in 1993-4 once numbers of wearers is taken into account.
In addition, one should consider other causes for increased numbers of cases diagnosed.
Firstly one should acknowledge the increased awareness by optometrists and ophthalmologists as to the possibility of acanthamoeba keratitis and improvements in the early diagnosis.
There are also possible changes in referral patterns to Moorfields Eye Hospital from other hospitals (i.e. a high index of suspicion may lead to early transfer of suspected cases). This is certainly what we have seen in the Midlands.
The results from treatment can also be relatively good with prompt modern treatment. In a recent review of 13 confirmed cases at UHCW, Coventry over the last 12 months, none of the patients required surgical intervention and all recovered vision (with over 85% complete improvement in acuity). 45% of patients were correctly diagnosed with acanthamoeba keratitis on the first eye casualty visit and treatment was dual topical therapy.
Whilst we applaud the sentiment in this paper and the recommendations made, perhaps one needs to consider that is what was seen in one location, all be it a very large hospital – the hospitals of the 2 senior authors have seen more than 20 cases in the last year alone and hence the presentation of about 50 cases per annum cannot be easily extrapolated without taking into account a number of other factors.
1The Association for Contact Lens Manufacturers. www.aclm.org.uk. Technical summary. Contact lens year book 2016.
2Separate figure comparing CL wearers to acanthamoeba numbers from figures given by Carnt et al.

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.