Optical Coherence Tomography in Optic Nerve Hypoplasia: Correlation With Optic Disc Diameter, Nerve Fiber Layer Thickness, and Visual Function

The correlation between optic disc diameters (DDs) with average retinal nerve fiber layer thickness (RNFLT) and visual function in children with optic nerve hypoplasia (ONH) having nystagmus is unknown. Data were obtained from a retrospective review of 28 children (mean age: 9.4 years; ±5.1). Optic DD was defined as the maximal horizontal opening of Bruch membrane with spectral optical coherence tomography combined with a confocal laser ophthalmoscope. Average RNFLT was obtained from circumpapillary b-scans. RNFLT was also remeasured at eccentricities that were proportionate with DD to rule out potential sampling artifacts. Visual function was assessed by visual acuity at last follow-up and by visual evoked potentials (VEP) in 11 patients. The eye with the larger DD, which had better visual acuity, was analyzed to exclude potential effects of amblyopia. DD was correlated with average RNFLT (r = 0.61), visual acuity (r = 0.32), and VEPs (r = 0.66). The relationship between RNFLT and DD was as follows: average RNFLT (μm) = 0.074 * DD (μm) - 18.8. RNFLT also correlated with the ratio of horizontal optic DD to macula-disc-margin distance (DD:DM; r = 0.59). RNFLT measured at eccentricities proportionate with DD showed progressive decrease in thickness only for DDs <1,100 μm. All patients with DD <1,000 μm had subnormal visual acuity, whereas those with DD <1,200 μm had subnormal VEPs. DD correlates with average RNFLT and with visual function in children with ONH. Using OCT imaging, DD can be obtained in children with nystagmus and provides objective information.

Original Contribution Optical Coherence Tomography in Optic Nerve Hypoplasia: Correlation With Optic Disc Diameter, Nerve Fiber Layer Thickness, and Visual Function John P. Kelly, PhD, Francine Baran, MD, James O. Phillips, PhD, Avery H. Weiss, MD Background: The correlation between optic disc diameters (DDs) with average retinal nerve fiber layer thickness (RNFLT) and visual function in children with optic nerve hypoplasia (ONH) having nystagmus is unknown. Methods: Data were obtained from a retrospective review of 28 children (mean age: 9.4 years; ±5.1). Optic DD was defined as the maximal horizontal opening of Bruch membrane with spectral optical coherence tomography combined with a confocal laser ophthalmoscope. Average RNFLT was obtained from circumpapillary b-scans. RNFLT was also remeasured at eccentricities that were proportionate with DD to rule out potential sampling artifacts. Visual function was assessed by visual acuity at last follow-up and by visual evoked potentials (VEP) in 11 patients. The eye with the larger DD, which had better visual acuity, was analyzed to exclude potential effects of amblyopia. Results: DD was correlated with average RNFLT (r2 = 0.61), visual acuity (r2 = 0.32), and VEPs (r2 = 0.66). The relationship between RNFLT and DD was as follows: average RNFLT (mm) = 0.074 * DD (mm) 2 18.8. RNFLT also correlated with the ratio of horizontal optic DD to macula-disc-margin distance (DD:DM; r2 = 0.59). RNFLT measured at eccentricities proportionate with DD showed progressive decrease in thickness only for DDs ,1,100 mm. All patients with DD ,1,000 mm had subnormal visual acuity, whereas those with DD ,1,200 mm had subnormal VEPs. Conclusions: DD correlates with average RNFLT and with visual function in children with ONH. Using OCT imaging, DD Roger H. Johnson Vision Lab (JPK, FB, JOP, AHW), Division of Ophthalmology, Seattle Children's Hospital, Seattle, Washington; Department of Ophthalmology (JPK, FB, AHW), University of Washington Medical Center, Seattle, Washington; and Department of Otolaryngology (JOP), University of Washington School of Medicine, Seattle, Washington. Supported by an unrestricted grant from the Peter LeHaye, Barbara Anderson, and William O. Rogers Endowment Funds. None of the authors have any proprietary or financial interest in this manuscript, its software or devices stated wherein, or the funding agencies. Address correspondence to John P. Kelly, PhD, Division of Ophthalmology, OA.5.342, Seattle Children's Hospital, 4800 Sand Point Way NE, Seattle, WA 98105; E-mail: [e-mail] 312 can be obtained in children with nystagmus and provides objective information. Journal of Neuro-Ophthalmology 2018;38:312-319 doi: 10.1097/WNO.0000000000000596 © 2017 by North American Neuro-Ophthalmology Society O ptic disc anatomy and electrophysiology help predict visual function in children with optic nerve hypoplasia (ONH) (1-4). However, the relationship between optic disc size, retinal nerve fiber layer thickness (RNFLT), and visual function has not been fully characterized in ONH. In controls, optic nerve fiber count is strongly correlated with the retrobulbar optic nerve cross-sectional area (5). Also, average RNFLT is associated with optic disc area on optical coherence tomography (OCT) (6,7). Estimation of optic disc diameter (DD) by clinical ophthalmoscopy (2,4) requires skill and can be imprecise in the presence of nystagmus. Alternatively, ONH can be assessed by objective methods, such as fundus photography (1,3,8). ONH severity is assessed by the ratio of horizontal optic DD to the distance from the temporal disc margin to the center of the macula (DD:DM), whereas DD:DM ratio can approximate visual acuity outcomes in the fixating eye (1,3). Additional factors such as macular hypoplasia, amblyopia, and central scotoma are thought to contribute to visual outcome (8). By contrast, Pilat et al (9) found that RNFLT parameters obtained with OCT did not correlate with visual acuity. The purpose of our study was to examine several factors in children with bilateral ONH using OCT. First, we measured horizontal DD on OCT to provide a rapid and objective assessment of ONH in the presence of nystagmus. Ideally, optic disc area and RNFLT are acquired by a 2-dimensional volume scan, which requires several seconds of stable gaze. When nystagmus is present, volumetric analysis of optic nerve area and circumpapillary RNFLT cannot be obtained reliably. We defined DD as the maximum horizontal distance between the edges of the Kelly et al: J Neuro-Ophthalmol 2018; 38: 312-319 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution retinal pigment epithelium/Bruch membrane complex across multiple single-line OCT scans of the optic nerve (10). We postulated that there would be a relationship between DD and average RNFLT based on data in controls (6,7). Second, we compared DD:DM ratios on confocal imaging using objective DD. Third, we examined whether RNFLT from OCT may contain an artifact due to disproportionate scan size with respect to the hypoplastic optic disc (7). Fourth, we investigate the presence of foveal hypoplasia in patients with bilateral ONH (8,11). Fifth, we compared DD with visual function as measured by visual acuity and visual evoked potentials (VEP) in the eye with the larger DD to exclude effects of amblyopia. METHODS This retrospective study was approved by the Institutional Review Board of Seattle Children's Hospital and conformed to the requirements of the United States Health Insurance Portability and Privacy Act. Inclusion criteria were children with clinically detected bilateral ONH who subsequently had OCT imaging. All had to have full eye examinations (cycloplegic refraction, assessment of gaze holding, ocular motility, pupillary responses, and slit-lamp evaluation). The diagnosis of ONH was based on the clinical appearance of an abnormally small optic disc (with or without pallor, a double-ring sign, or segmental irregularities of the disc margins) detected by direct ophthalmoscopy, or by indirect ophthalmoscopy when nystagmus was severe. Patients were excluded if they were born premature, had major brain malformations (e.g., periventricular leukomalacia, schizencephaly, encephalocele, generalized cortical migration defects), or had other ocular defects (e.g., aniridia, albinism, cataracts, persistent fetal vasculature etc.), or a known history of maternal drug abuse. From our records, we identified 28 children meeting the inclusion criteria. Patients had assessments of growth velocity to screen for growth hormone deficiency or endocrine abnormalities. A brain MRI was available in 20 subjects (71%) to exclude major brain malformations or hypoxic-ischemic events that could lead to optic nerve loss because of trans-synaptic retrograde degeneration from cortical damage (12,13). MRIs deferred in the remaining 8 patients because of findings of normal growth velocity and minimal developmental delays. The optic disc and fovea were examined by simultaneous spectral domain OCT and scanning laser ophthalmoscope (Spectralis OCT-SLO; Heidelberg Engineering, Heidelberg, Germany). The SLO was used to check the alignment of each OCT with the center of the optic disc. Images were acquired in high speed mode (40,000 A-scans/second 768 · 768 pixels). Image acquisition was faster along the horizontal scan (0.125 ms) compared with the vertical scan direction (96 ms). Therefore, we acquired horizontal OCT line scans of the optic disc and the fovea in children with nystagmus. Kelly et al: J Neuro-Ophthalmol 2018; 38: 312-319 When possible, average RNFLT measurements were obtained from circumpapillary b-scans (3.5 mm diameter, 768 · 497 pixels). Circumpapillary scans had to be aligned to within 400 mm of the optic disc center and also required the average RNFLT to be within 10 mm on a repeat scan. DD and RNFLT segmentation were measured using Heidelberg HEYEX software (version 5.4.6.0). RNFLT segmentation errors were manually corrected if needed. All line scans were also inspected for adequate representation of the fovea (longest outer segments with the thickest outer nuclear layer). It has been shown in healthy eyes that the decrease in RNFLT with smaller optic discs can be eliminated if the diameter of circumpapillary b-scan is scaled proportionate to the optic disc size (7). This finding has important implications for our study because Spectralis OCT software uses a fixed radius of 1,770 mm for circumpapillary b-scans (a factor of 27.5 mm per diopter is used to correct for refractive error). To address this concern, we resampled RNFLT at a distance of 850 mm from the disc edge. Resampling was possible because multiple OCTs were acquired in patients with nystagmus or unsteady fixation, resulting in random alignment with the center of the optic disc. Therefore, a proportion of these scans included the desired location of 850 mm from the disc edge. Software was developed that allowed manual piecemeal resampling of RNFLT from misaligned scans at the desired distance from the optic disc center. Each RNFLT measurement was then converted to a percentage of the Spectralis OCT normative thickness after correcting for the appropriate circumpapillary scan angle [Normative values for RNFLT are not available with Spectralis OCT for children. However, global RNFLT in children is approximately 12 mm thicker than the Spectralis normative value (6,14), which is likely due to a decline in RNFLT with older age (15). A tutorial of the analysis and software are available at http://faculty.washington.edu/jokelly/ manualoct.] The software also allowed scoring of the fovea to be used for DD:DM ratios on the corresponding confocal SLO image. The experimenter (JPK) performing the OCT analyses was masked to the clinical findings of the patients. We assessed visual acuity using recognition optotypes (Snellen, Allen, or HOTV) from the last follow-up examination. If the patient was unable to read the largest optotype (20/200), the viewing distance was reduced until a correct response was obtained using the 20/200 optotype. The acuity value was then converted to logarithm of the minimum angle of resolution (logMAR) after correcting for distance if needed (4). Subjects with low vision (#20/1,600) had visual acuity confirmed by Teller Acuity Cards. VEP recording and analysis followed previously published methods (4,16,17). Stimuli were brief onset-offset of horizontal 0.5 cycle/degree gratings, which reduced artifacts associated with nystagmus. The response from the eye with the larger DD was analyzed to exclude secondary effects of amblyopia. The signal from the Oz electrode was analyzed using custom software (16). The software automatically 313 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 1. A. Representative confocal scanning laser ophthalmoscope images (above) placed in registration with the corresponding optical coherence tomography (OCT) (below). The horizontal white dotted line shows the location of the OCT section through the optic nerve. The edges of Bruch membrane opening with respect to the optic disc were determined in all patients by vertical registration between the upper and lower images (as shown in A). The horizontal disc diameters, (mm), are 830, 400, 767, 904, 1,126, and 389 for images A-F, respectively. The right and left eyes from the same patient are shown in C and D, and for another patient in E and F. 314 Kelly et al: J Neuro-Ophthalmol 2018; 38: 312-319 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution generated an index of signal-to-noise ratio (SNR) based on consistent timing and amplitude of all individual VEP epochs. In this manner, the VEP metric was objective in the context of abnormal waveforms. An SNR .1.3 indicated a statistically significant response compared with subjects with no light perception, or to spontaneous electroencephalographic activity in subjects with nystagmus (16). Note that the SNR can have a value ,1.0 because the SNR is the ratio of averaged magnitude of the response divided by the underlying variance in the EEG signal in the Fourier domain. The experimenter performing the VEP analysis was masked to the clinical status of the patient. RESULTS Subject ages ranged from 2.4 to 20 years (mean: 9.4 years ±5.4 years) at the time of OCT imaging. Visual acuity was available from 26 patients (mean age: 9.9 years; ± 5.1). Pattern VEP recordings were available from 11 subjects at a younger age (mean age = 2.3 years; SD = 1.7). Age was not correlated with DD or with VEP SNR (r2 # 0.16 for both). Cycloplegic refractive error (in spherical equivalent) ranged from 24.75 to +5.00 diopters (mean 0.25 diopters; ±2.25). For the eye with the larger optic disc, there was no significant relationship between ametropia (in spherical equivalent) with age-corrected logMAR, VEP SNR, or DD (r2 # 0.06 for all). Seven patients had findings consistent with "septo-optic dysplasia" (partial or complete agenesis of the corpus callosum and septum pellucidum, with or without pituitary involvement). All 11 patients showing MRI abnormality in the region of the pituitary gland had evidence of hypothalamic pituitary dysfunction. One had focal gray matter heterotopia in parietal and temporal lobes, another had a small focus of gray matter heterotopia in the left lateral ventricle, and a third had a Chiari I malformation without hydrocephalus. Representative images from 4 patients (6 eyes) are shown in Figure 1 and demonstrate a range of horizontal DD that were found in this study. Right and left eyes are placed in the left column and right columns, respectively. All images were cropped to highlight the optic nerve. The edges of Bruch membrane opening were registered by aligning a vertical line from the OCT image (below) to the corresponding confocal SLO image of the optic nerve (above). Nystagmus was present for all scans, so the automatic real-time eye tracking feature of the Spectralis instrument could not be used. The maximal horizontal DD was determined for each subject by comparing across several repeat scans. RNFLT from circumpapillary b-scans was available from 15 of 28 patients. Figure 2A shows that average RNFLT and DD were correlated (r2 = 0.61; P , 0.0001). The equation defining the relationship was average RNFLT = 0.074 * DD (mm) 2 18.8. Each eye's slope was within ±1 SE of the slope for combined eyes. Figure 2B shows that RNFLT and DD:DM ratio were also correlated (r2 = 0.59; Kelly et al: J Neuro-Ophthalmol 2018; 38: 312-319 FIG. 2. A. The relationship between average retinal nerve fiber layer thickness (RNFLT) and optic disc diameter (DD) as measured by the horizontal diameter of Bruch membrane opening using optical coherence tomography. B. The relationship between average RNFLT and the ratio of horizontal optic DD to macula-disc-margin distance (DD:DM). Right and left eyes are plotted as open and closed circles, respectively. P , 0.0001). The equation defining the relationship was average RNFLT = 223.1 * DD:DM 2 2.46. Note that DD was highly correlated with DD:DM ratio (r = 0.97) because DD is used in both metrics. For reference, a DD of 1395 mm corresponded to a DD:DM ratio of 0.35. Resampling of RNFLT was performed at scaled eccentricities from the center of the optic disc. Figure 3A shows samples from 6 patients with DD ,600 mm (after correcting for angular rotation about the optic disc). The data were overlaid on the Spectralis instrument's normative curve to show the relative reduction in RNFLT, despite scaling for each patient's optic DD. The average RNFLT was approximately 40% of the instrument's normative value. Figure 3B summarizes all the data by averaging RNFLT from all subjects and then binning the averages into discrete DD sizes. 315 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 3. Retinal nerve fiber layer thickness (RNFLT) after scaling for horizontal disc diameter (DD). The sampled locations are 850 mm from the disc edge and are scaled to each subject's optic DD. A. Resampled RNFLT is plotted circumferentially around the optic disc. Data from 6 subjects with horizontal DDs ,600 mm are overlaid on the adult normative template for the instrument (normative values are not available for children). B. RNFLT scaled as a percentage of the adult normative thickness across DDs for all patients. The data were fit by a third-order polynomial (r2 = 0.74). Error bars are 99% confidence interval of the mean. RNFLT was relatively constant for DD .1,100 mm. RNFLT decreased for DD ,1,100 mm and had a plateau near 40%. Therefore, RNFLT showed a progressive decline with smaller horizontal optic DDs, despite scaling the circumpapillary b-scan with the diameter of the optic disc. Figure 4A shows that DD correlated with visual acuity when using the eye with the larger DD. After removing the subject with visual acuity measured by Teller Acuity Cards, the correlation was statistically significant (r2 = 0.32; P , 0.005). The equation defining this relationship was logMAR = 20.001 * DD (microns) + 1.28. All subjects with DD 316 #1,000 mm were worse than 0.3 logMAR. Visual acuity in the eye with the smaller DD ranged from no light perception to 20/20. Using the eye with the smaller DD visual acuity did not correlate with DD (r2 = 0.14; P = 0.06). Figure 4B shows that VEP SNRs were strongly correlated with DD. After removing the 2 data points with SNR #1.0, the correlation was statistically significant (r2 = 0.66; P , 0.002). VEP data were compared with 20 age-matched controls (2.9 vs 2.3 years age; SD = 1.7 for both groups; P = 0.4). Only 2 subjects with DD of larger than 1,300 mm had a VEP SNR in the 95% confidence range of controls. Kelly et al: J Neuro-Ophthalmol 2018; 38: 312-319 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 4. A. The relationship between horizontal disc diameter (DD) on optical coherence tomography and visual acuity in children with optic nerve hypoplasia. One patient with a 470-mm horizontal DD and visual acuity of 20/1,600 on Teller Acuity Cards was excluded. B. The relationship between horizontal DD and visual evoked potential (VEP) signal strength. Open circles represent data from 2 patients in which VEP signal was not statistically different from noise. These 2 data points were excluded from the regression. The shaded area represents the ±95% mean confidence interval of age-matched controls. LogMAR, logarithm of the minimum angle of resolution. The area encompassing the fovea could be visualized in one or both eyes of 24 patients, of which 17 had evidence of mild foveal hypoplasia based on multiple OCT through the macula. Figure 5 shows a patient with a DD of 760 mm in which multiple sections through the fovea revealed persistence of the inner retinal layers. The fovea showed thickening of the outer nuclear layer and the elongation of the outer segments. Foveal hypoplasia was more common in the eye with the smaller DD (17/23 eyes) compared with the eye with larger DD (10/24 eyes). In the eye with the larger DD, logMAR acuity was lower in subjects with foveal hypoplasia compared with those with a normal fovea (0.71 vs 0.32 logMAR; P = 0.003). In the eye with the smaller DD, Kelly et al: J Neuro-Ophthalmol 2018; 38: 312-319 logMAR acuity was not statistically lower in those with foveal hypoplasia (1.45 vs 0.78 logMAR; P = 0.08). DISCUSSION Our results show that average RNFLT and visual function are correlated with DD when measured by the maximum horizontal diameter between Bruch membrane on OCT. Our data are consistent with the notion that amblyopia is an important factor because the correlation of visual function with optic DD was not significant in eyes with the smaller optic DD. DD can be measured with OCT when nystagmus limits the assessment by ophthalmoscopy or by 317 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 5. Imaging of a patient with foveal hypoplasia. A. Confocal scanning laser ophthalmoscope (SLO) image of the right eye from a 7-year-old with bilateral optic nerve hypoplasia. A foveal reflex was visible on many of the confocal images. B. The optic nerve disc has a horizontal disc diameter of 767 mm. C. Six optical coherence tomography sections through the central macula are shown stacked on top of each other. The arrows to the left show their corresponding sections through the confocal SLO image. The third image from the top (highlighted with scale bar) shows the scan with the best match to the foveal center. volumetric imaging of the optic nerve by OCT. Measurement of the circumpapillary RNFLT requires alignment of the OCT onto the center of the optic disc, which can be very difficult in the presence of nystagmus. Circumferential RNFLT measurements were obtained only in any 53% of the patients, whereas horizontal line scans across the optic nerve were readily obtained. In comparison, success rates are approximately 70% for obtaining optic nerve b-scans (single line) in healthy children older than 1 week of age using a hand-held OCT (10). DD is an objective measure that helped classify visual function in our patient cohort. Those with DD ,1,000 mm had reduced VEPs and significant loss of RNFLT after adjusting for the size of the optic disc. By contrast, subjects with a DD .1,200 mm had 20/80 or better visual acuity and a normal VEP response. Visual acuity in ONH is likely determined by the localized area with the highest density of the remaining retinal ganglion cells but does provide information regarding the central visual field. Because children with ONH cannot provide a reliable visual field test, we use the VEP to objectively assess the central 10-18° of the visual field (17,18). Therefore, the correlation between VEP and DD likely reflects the corresponding reduction of cortical inputs with smaller optic nerves. One significant advantage of our VEP analysis is that it does not require subjective scoring of waveform peaks in the presence of low amplitude and/or large background noise. Our study supports and strengthens previous reports that vision can be predicted with morphometric studies 318 of the optic nerve and that the novel techniques used in this study make it easier to perform on awake children. Furthermore, our data and other studies (8,9,11) report that foveal development and inner retinal layering can be abnormal in ONH. However, our study has limitations. First, high-quality imaging is not likely to be successful in children with significant cognitive delays. Second, MRIs were not available in 8 patients. Excluding some patients with MRIs while including other patients without MRIs may have introduced significant bias. Of note, this study excluded patients having major brain malformations or hypoxic ischemic events that could lead to reduced optic nerve size resulting from transsynaptic retrograde degeneration because of cortical damage (13). Third, we attempted to remove the affects of amblyopia using the eye with the larger DD and then also measure the VEP under binocular viewing. This is a technical consideration because at a young age, many of these patients are intolerant to monocular testing and that nystagmus intensity often increases with monocular viewing. We cannot exclude the possibility that children with ONH might have better visual acuity in the eye with the smaller DD. Fourth, this is a relatively small sample size and further studies will be needed to confirm our findings. STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: John P. Kelly, James O. Phillips, and Avery H. Weiss; b. Acquisition of data: John P. Kelly, Avery H. Weiss, and Francine Baran; c. Analysis and interpretation of data: John P. Kelly and Avery H. Weiss. Category 2: a. Drafting the Kelly et al: J Neuro-Ophthalmol 2018; 38: 312-319 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution manuscript: John P. Kelly, James O. Phillips, Avery H. Weiss, and Francine Baran; b. Revising it for intellectual content: John P. Kelly, James O. Phillips, Avery H. Weiss, and Francine Baran. Category 3: a. Final approval of the completed manuscript: John P. Kelly, James O. Phillips, Avery H. Weiss, and Francine Baran. REFERENCES 1. Borchert M, McCulloch D, Rother C, Stout AU. Clinical assessment, optic disk measurements, and visual-evoked potential in optic nerve hypoplasia. Am J Ophthalmol. 1995;120:605-612. 2. Weiss AH, Kelly JP. Acuity, Ophthalmoscopy, and visually evoked potentials in the prediction of visual outcome in infants with bilateral optic nerve hypoplasia. J AAPOS. 2003;7:108-115. 3. McCulloch DL, Garcia-Filion P, Fink C, Chaplin CA, Borchert MS. Clinical electrophysiology and visual outcome in optic nerve hypoplasia (ONH). Br J Ophthalmol. 2010;94:1017- 1023. 4. Kelly JP, Phillips JO, Weiss AH. VEP analysis methods in children with optic nerve hypoplasia: relationship to visual acuity and optic disc diameter. Doc Ophthalmol. 2010;94:1017-1023. 5. Jonas JB, Schmidt AM, Miller-Bergh JA, Schlotzer-Schrehardt UM, Naumann GO. Human optic nerve fiber count and optic disc size. Invest Ophthalmol Vis Sci. 1992;33:2012-2018. 6. Turk A, Ceylan OM, Arici C, Keskin S, Erdurman C, Durukan AH, Mutlu FM, Altinsoy HI. Evaluation of the nerve fiber layer and macula in the eyes of healthy children using spectral domain optical coherence tomography. Am J Ophthalmol. 2012;153:552-559. 7. Carpineto P, Cianaglini M, Aharrh-Gnama A, Cirone D, Mastropasqua L. Custom measurement of retinal nerve fiber layer thickness using STRATUS OCT in normal eyes. Eur J Ophthalmol. 2005;15:360-366. Kelly et al: J Neuro-Ophthalmol 2018; 38: 312-319 8. Zeki SM, Dudgeon J, Dutton GN. Reappraisal of the ratio of disc to macula/disc diameter in optic nerve hypoplasia. Br J Ophthalmol. 1991;75:538-541. 9. Pilat S, Sibley D, McLean RJ, Proudlock FA, Gottlob I. Highresolution imaging of the optic nerve and retina in optic nerve hypoplasia. Ophthalmology. 2015;122:1330-1339. 10. Patel A, Purohit R, Lee H, Sheth V, Maconachie G, Papageorgiou E, McLean RJ, Gottlob I, Proudlock FA. Optic nerve head development in Health infants and children using handheld spectral-domain optical coherence tomography. Ophthalmology. 2016;123:2147-2157. 11. Abbasian J, Blair N, Shahidi M, Ying GS, Huang J, Kaufman L, Blair M. Macular optical coherence tomography in patients with unilateral optic nerve hypoplasia. J AAPOS. 2015;19:57-61. 12. Hoyt CS, Good WV. Do we really understand the difference between optic nerve hypoplasia and atrophy? Eye. 1992;6:201-204. 13. Taylor D. Development abnormalities of the optic nerve and chiasm. Eye (Lond). 2007;21:1271-1284. 14. Yanni SE, Wang J, Cheng CS, Locke KI, Wen Y, Birch DG, Birch EE. Normative reference ranges for the retinal nerve fiber layer, macula, and retinal layer thicknesses in children. Am J Ophthalmol. 2013;155:354-360. 15. Bendschneider D, Tornow RP, Horn FK, Laemmer R, Roessler CW, Juenemann AG, Kruse FE, Mardin CY. Retinal nerve fiber layer thickness in normal measured by spectral domain OCT. J Glaucoma. 2010;19:475-482. 16. Kelly JP, Darvas F, Weiss AH. Waveform variance and latency jitter of the visual evoked potential in childhood. Doc Ophthalmol. 2014;128:1-12. 17. Odom JV, Bach M, Brigell M, Holder GE, McCulloch DI, Mizota A, Tormene AP. ISCEV standard for clinical visual evoked potentials: (2016 update). Doc Ophthalmol. 2016;133:1-9. 18. Kelly JP, Weiss AH. Comparison of pattern visual-evoked potentials to perimetry in the detection of visual loss in children with optic pathway gliomas. J AAPOS. 2006;10:298-306. 319 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.