We warmly thank Lee HJ and Kim JY for their interest in our article entitled “Effect of Intravitreal Ranibizumab on Ganglion Cell Complex and Peripapillary Retinal Nerve Fiber Layer in Neovascular Age-Related Macular Degeneration Using Spectral Domain Optical Coherence Tomography.”1
We agree with them regarding some observations. In particular, we recognize that automatic segmentation errors in retinal nerve fiber layer and ganglion cell complex calculation along with lack of fixation stability may lead to considerable artifacts in patients with maculopathy.
We are glad to clarify in this letter some further details regarding our study.
As stated in the Methods section, patients have been included only if they had sufficiently clear ocular media, adequate pupillary dilation, and stable fixation to permit good-quality optical coherence tomography imaging. A signal strength of 7/10 was used as the threshold for quality assessment, so that images below 7/10 were not taken into consideration for the study.
In addition, included patients suffered from naive choroidal neovascularization, with relatively early-onset exudation and low presence of large fibrotic/atrophic scars. Indeed, mean best-corrected visual acuity of the included patients was 0.5 ± 0.8 logMAR (20/63) at baseline, with 14 patients (58%) presenting with a best-corrected visual acuity of 0.5 logMAR or better. Thus, the good baseline visual acuity could allow a satisfying fixation for the optical coherence tomography scans, as well as the limited retinal disorganization because of the early-onset lesions ensured a good visualization of the inner retinal layer boundaries.
We did not use any manual correction of retinal nerve fiber layer and ganglion cell complex boundaries. However, each acquisition was reviewed by two retina specialists (I.Z. and M.B.P.) separately, and images unacceptable for segmentation and/or motion artifacts were excluded from the final analyses. As far as it regards the central macular thickness, the device used in the study (SD-OCT Cirrus HD; Carl Zeiss Meditec Inc, Jena, Germany) has demonstrated high accuracy and test–retest reliability in its calculation.2–4 Moreover, the average error score severity has been reported to improve with restoring of normal macular morphology with intravitreal anti–vascular endothelial growth factor therapy.
Regarding the macular ganglion cell inner plexiform layer, we agree with our colleagues about the risk that the results presented in our study could have been influenced, at least in part, by the presence of autosegmentation errors caused by macular distortion. Recently, Alshareef et al have recently published a review of the artifacts in ganglion cell inner plexiform layer thickness measurement in eyes with neovascular age-related macular degeneration. According to this analysis, scan line artifacts were observed in 64.47% of B scans being misidentification of the ganglion cell inner plexiform layer boundaries the most common error observed. However, this deviation was rated as mild (<10 μm) in the 94.9% of cases.5
In conclusion, we can assure that our analysis represents a reliable description of changes in macular shape and ganglion cell inner plexiform layer thickness in patients with wet age-related macular degeneration treated with anti–vascular endothelial growth factor. However, segmentation artifacts and motion errors might always be taken into consideration in the decision-making strategies and the clinical approach.