We thank Drs Calugaru M. and Calugaru D. for their interest in our article,1 and we welcome this opportunity to address their concerns. The purpose of our study was to investigate the outcomes of intravitreal antivascular endothelial growth factor (VEGF) therapy in eyes with both neovascular age-related macular degeneration (AMD) and diabetic retinopathy (DR) as higher levels of VEGF due to concomitant DR in eyes with active neovascular AMD may lead to a higher consumption of anti-VEGF molecules, thus impairing the efficacy of treatment.1 As pointed out, in our series we included patients that had undergone other treatments for DR, either before or during administration of anti- VEGF drugs for neovascular AMD. This is an obvious limitation of our retrospective study, even though all the patients included were treatment na?ve for anti-VEGF agents. We recorded that best corrected visual acuity, after a significant improvement at 1 year, returned to baseline values at the last follow-up visit, while mean central macular thickness (CMT) significantly decreased from 408 ?m to 335 ?m. As pointed out, these results may suggest that disease process could be still active and progressive requiring further treatment. 2 In fact, at the end of follow up, CNV was still active in 39% eyes, while 61% eyes developed an atrophic/fibrotic scar with no signs of activities. On the other hand, CMT could have been influenced by the two concomitant diseases, the diabetic retinopathy and the choroidal neovascularization. In particular, we cannot exclude that some patients underwent anti-VEGF treatment, DR-related macular edema. We agree that, a lot of cytokines, chemokines, and growth factors may be associated with DR pathophysiology.3,4 Further prospective studies should consider the effects of these molecules and the use of non-specific anti- VEGF substances which inhibits the up-regulation of the VEGF and suppresses the expression of the whole panoply of the proinflammatory and proangiogenic factors.

References 1. Bandello F, Corvi F, La Spina C, et al. Outcomes of intravitreal anti- VEGF therapy in eyes with both neovascular age-related macular degeneration and diabetic retinopathy. Br J Ophthalmol 2016; http:/dx.doi.org/10.1136/bjophthalmol-2016- 308400. 2. Gover S, Murthy RK, Brar VS, et al. Normative data for macular thickness by high-definition spectral-domain optical coherence tomography (spectralis). Am J Ophthalmol 2009;148:266-271. 3. Sohn HJ, Han DH, Kim IT, et al. Changes in aqueous concentrations of various cytokines after intravitreal triamcinolone versus bevacizumab for diabetic macular edema. Am J Ophthalmol 2011;152:686-694. 4. Shah SU, Harless A, Bleau L, et al. Prospective randomized subject- masked study of intravitreal bevacizumab monotherapy versus dexamethasone implant monotherapy in the treatment of persistent diabetic macular edema. Retina. 2016 Apr 27. [Epub ahead of print]

Conflict of Interest:

None declared

Response to e-letter to the editor

We thank Dr. Mourits and co-workers for their interest and comments regarding our manuscript on impression-free three-dimensional (3D) printed anophthalmic socket, and appreciate their recognition of its value in the manufacturing of an ocular prosthesis.(1)

Dr. Mourits and co-workers propose an alternative technique, impression mould through silicon injection in the anophthalmic socket, which is subsequently scanned with magnetic resonance imaging (MRI), and further 3D processed. This technique, however, renews the manual step of impression moulding, which reintroduces a non-digital procedure in the digital domain.

We confirm the benefit of MRI over cone-beam computed tomography (CBCT) as radiation-free option.(2) However, we would like to highlight our concerns on the current use of MRI for the purpose of imaging the anophthalmic socket. First, spatial resolution of MRI is lower than CBCT, since MRI is characterized by geometric distortion produced by magnetic susceptibility, particularly at the air-tissue interface.(3,4) Second, longer acquisition times in MRI potentially result in more variable delineation.(5) Manual delineation in MRI, in contrast to computer- assisted delineation in CT, requires expertise and is time-consuming. Third, adequate MRI protocols and preferred coils for orbital scanning with high resolution and signal-to-noise ratio within reasonable measurement times are as yet not established.(6) By contrast, in CT, the process of delineation of soft tissue for calculation of orbital soft tissue volume is validated, reliable and accurate.(7) Finally, MRI in children requires general anaesthesia, which is associated with increased morbidity.(8)

Continuous improvement in MRI imaging will help to further develop this innovative concept. We encourage Dr. Mourits et al to publish the data they shared on the use of a mould and special MRI orbital scanning.

Bibliography

1. Ruiters S, Sun Y, de Jong S, et al. Computer-aided design and three-dimensional printing in the manufacturing of an ocular prosthesis. Br J Ophthalmol 2016 ;100:879- 881.

2. Brenner DJ, Hall EJ. Computed tomography--an increasing source of radiation exposure. N Engl J Med 2007;357(22):2277-84.

3. Ludwig U, Eisenbeiss AK, Scheifele C, et al. Dental MRI using wireless intraoral coils. Sci Rep 2016;6:23301.

4. Sumanaweera TS, Glover GH, Binford TO, et al. MR susceptibility misregistration correction. IEEE Trans Med Imaging 1993;12(2):251-9.

5. Karlo C, Reiner CS, Stolzmann P, et al. CT- and MRI-based volumetry of resected liver specimen: comparison to intraoperative volume and weight measurements and calculation of conversion factors. Eur J Radiol 2010;75(1):e107-11.

6. Georgouli T, James T, Tanner S, et al. High-resolution microscopy coil MR-Eye. Eye (Lond) 2008;22(8):994-6.

7. Regensburg NI, Kok PH, Zonneveld FW, et al. A new and validated CT -based method for the calculation of orbital soft tissue volumes. Invest Ophthalmol Vis Sci 2008;49(5):1758-62.

8. Cot? CJ, Wilson S. Guidelines for Monitoring and Management of Pediatric Patients Before, During, and After Sedation for Diagnostic and Therapeutic Procedures: Update 2016. Pediatr Dent 2016;38(4):13-39.

Conflict of Interest:

None declared

I read with great interest the article by Romero- Aroca et al. titled "Changes observed in diabetic retinopathy: eight-year follow-up of a Spanish population" [1]. Insulin usage has found to be a risk factor for progression of DRP. Authors conclude this result as relaxation of patients in metabolic control. Beside this conclusion, insulin analogues may directly cause progression of DRP. Insulin and its analogues stimulate insulin like growth hormone-1 (IGF-1) receptors, like growth hormone. There is a strong relationship between growth hormone and progression of DRP. Diabetic retinopathy regresses after surgical ablation or spontaneous infarction of pituitary gland [2]. Growth hormone deficiency is a protective factor for development of diabetic retinopathy in dwarfs [3]. Development of diabetic retinopathy is significantly higher in pubertal subjects than pre -pubertal subjects, despite the same glycemic control [4]. Thus, insulin analogues may cause progression of DRP through growth hormone-like effect. But I hypothesize that insulin analogues may cause progression of DRP only after deterioration of inner blood retinal barrier [5]. Hyperglycemia is a major risk factor for development of DRP. High blood glucose levels cause inner blood-retinal barrier deterioration by polyol pathway, non-enzymatic protein glycation and oxidative stress. Insulin analogues decrease blood glucose level and protect pericytes and inner blood-retinal barrier. When inner blood-retinal barrier is intact, insulin analogues may pass into the retinal tissue in very small amounts. After impairment of inner blood retinal barrier, insulin analogues may pass into the retinal tissue in much more amounts. Thus, insulin analogues may pass retinal tissue after impairment of inner blood retinal barrier and cause progression of DRP by growth hormone like effect. But before impairment, analogues can not pass through inner blood retinal barrier to cause progression of DRP, even delay onset of DRP by decreasing high blood glucose that impair pericytes.

References 1. Romero-Aroca P, de la Riva-Fernandez S, Valls-Mateu A, Sagarra-Alamo R, Moreno-Ribas A, Soler N. Changes observed in diabetic retinopathy: eight- year follow-up of a Spanish population. Br J Ophthalmol. 2016 Jan 14. 2. Adams, D.A., Rand, R.W., Roth, N.H., Dashe, A.M., Gipstein, R.M., Heuser, G. Hypophysectomy in diabetic retinopathy. The relationship between the degree of pituitary ablation and ocular response. Diabetes. 1974;23:698???707. 3. Merimee, T.J. A follow-up study of vascular disease in growth-hormone-deficient dwarfs with diabetes. N Engl J Med. 1978;298:1217???1222. 4. Murphy, R.P., Nanda, M., Plotnick, L., Enger, C., Vitale, S., Patz, A. The relationship of puberty to diabetic retinopathy. Arch Ophthalmol. 1990;108:215???218. 5. Kaya A, Kar T, Aksoy Y, Ozalper V, Basbug B. Insulin analogues may accelerate progression of diabetic retinopathy after impairment of inner blood-retinal barrier. Med Hypotheses. 2013 Dec;81(6):1012-4. 4.

Conflict of Interest:

None declared

Changing from pro re nata treatment regimen to a treat and extend regimen with ranibizumab in neovascular age-related macular degeneration Dan Calugaru, Mihai Calugaru Department of Ophthalmology, Univ of Medicine Cluj-Napoca/Romania

Re: Changing from pro re nata treatment regimen to a treat and extend regimen with ranibizumab in neovascular age-related macular degeneration. Hatz and Prunte. Br J Ophthalmol 2016; http:/dx. doi.org/10.1136/bjophthalmol-2015-307299.

Dear Editor We would like to address several limitations arisen from the interesting study by Hatz and Prunte (1) and which can be specifically summarized as follows:

1. The article was retrospectively conducted with the existence of a selection bias due to the inclusion in the final analysis only those patients who completed treat and extend (TE) follow-up of 12 months. Of note, only patients with active disease during the last 3 months of the pro re nata (PRN) phase were transitioned to TE treatment.

2. With the exception of data concerning the baseline choroidal neovascularization (CNV), there were no details on the anatomical types of neovascular maculopathy (CNV/serous and/or hemorrhagic detachment of the neurosensory retina or retinal pigment epithlium [RPE]/retinal hard exudates/subretinal and sub-RPE fibrovascular proliferation/disciform scar [subretinal fibrosis]), at baseline visit, as well as before and after switching from a PRN to a TE treatment regimen.

3. There were no data referring to the proportion of eyes considered "dry" on optical coherence tomography as per criterion of central retinal thickness (CRT) < 320 microns (2), before and after switching to a TE algorithm.

4. The comparative analysis of the visual and morphologic outcomes of the two treatment regimens was fairly inconclusive. Thus, there was a mean visual acuity (VA) gain of approximately 5 Early Treatment Diabetic Retinopathy Study (ETDRS) letters in comparison with the baseline VA at the end of the PRN phase; this value increased subsequently by an average of approximately 5 letters until the month 12 of the TE phase. The CRT decreased by a mean of 86 microns in relation to the baseline value, up to the end of the PRN phase; this CRT reduction increased thereafter by a mean of 35 microns until the month 12 of the TE phase. Importantly, during the TE period of this study, patients received approximately two more injections over a 12-month period than during 12 months of PRN treatment.

5. Implementation of the 2-sequence, 2-period, 2-treatment algorithm design in which each patient was assigned to a sequence of treatment, did not provide the answer to the question which of the 2 treatment approaches was more efficiently. On the contrary, the disadvantages of such a study could not be avoided. Thus, the washout period, which is essential between periods of such a study in terms of aliased effects, was not precisely delimited and the impact of the significant carryover effects may be confounded with direct treatment effects, in the sense that these effects could not be estimated separately being able to bias the interpretation of data analysis.

Altogether, regardless of the treatment approaches chosen (TE/PRN algorithm), the efficacy of therapy depends primarily on the promptness of the therapy after neovascular age-related degeneration diagnosis (3,4).

References 1. Hatz K, Prunte C. Changing from pro re nata treatment regimen to a treat and extend regimen with ranibizumab in neovascular age-related degeneration. Br J Ophthalmol 2016; http:/dx. doi.org/10.1136/bjophthalmol -2015-307299. 2. Grover S, Murthy RK, Brar VS, and Chalam KV. Normative data for macular thickness by high-definition spectral-domain optical coherence tomography (spectralis). Arch Ophthalmol 2009;148-271. 3. Calugaru D, Calugaru M. Treat-and-extend intravitreal bevacizumab for branch retinal vein occlusion. Ophthalmic Surg Lasers Imaging Retina 2015;46:994. 4. Calugaru D, Calugaru M. Comment on:"Central retinal vein occlusion:modyfing current treatment protocols." Eye 2016; http:/dx.doi.org/10.1038/eye.2016.83.

Conflict of interest: None declared

Conflict of Interest:

None declared