Interocular Difference in Retinal Nerve Fiber Layer Thickness Predicts Optic Neuritis in Pediatric-Onset Multiple Sclerosis

Optical coherence tomography (OCT) is capa- ble of quantifying retinal damage. Defining the extent of anterior visual pathway injury is important in multiple sclerosis (MS) as a way to document evidence of prior disease, including subclinical injury, and setting a baseline for patients early in the course of disease. Retinal nerve fiber layer (RNFL) thickness is typically classified as low if values fall outside of a predefined range for a healthy population. In adults, an interocular difference (IOD) in RNFL thickness greater than 5 mm identified a history of unilateral optic neuritis (ON).

Original Contribution Section Editors: Clare Fraser, MD Susan Mollan, MD Interocular Difference in Retinal Nerve Fiber Layer Thickness Predicts Optic Neuritis in Pediatric-Onset Multiple Sclerosis Amy T. Waldman, MD, MSCE, Leslie Benson, MD, John R. Sollee, BS, Amy M. Lavery, PhD, Geraldine W. Liu, ALM, Ari J. Green, MD, MCR, Emmanuelle Waubant, MD, Gena Heidary, MD, PhD, Darrel Conger, CRA, Jennifer Graves, MD, PhD, Benjamin Greenberg, MD, MHS Background: Optical coherence tomography (OCT) is capable of quantifying retinal damage. Defining the extent of anterior visual pathway injury is important in multiple sclerosis (MS) as a way to document evidence of prior disease, including subclinical injury, and setting a baseline for patients early in the course of disease. Retinal nerve fiber layer (RNFL) thickness is typically classified as low if values fall outside of a predefined range for a healthy population. In adults, an interocular difference (IOD) in RNFL thickness greater than 5 mm identified a history of unilateral optic neuritis (ON). Through our PERCEPTION (PEdiatric Research Collaboration ExPloring Tests in Ocular Neuroimmunology) study, we explored whether RNFL IOD informs on remote ON in a multicenter pediatric-onset MS (POMS) cohort. Methods: POMS (defined using consensus criteria and first attack ,18 years) patients were recruited from 4 academic centers. A clinical history of ON (.6 months prior to an OCT scan) was confirmed by medical record review. RNFL thickness was measured on Spectralis machines (Heidelberg, Germany). Using a cohort of healthy controls from our centers tested on the same machines, RNFL thickness ,86 mm (,2 SDs below the mean) was defined as abnormal. Based on previously published findings in adults, an RNFL IOD .5 mm was defined as abnormal. The proportions of POMS participants with RNFL thinning (,86 mm) and abnormal IOD (.5 mm) were calcu- Division of Neurology (ATW, JRS, AML, GWL), Children’s Hospital of Philadelphia and Departments of Neurology and Pediatrics (ATW), Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania; Departments of Neurology (LB) and Ophthalmology (GH), Boston Children’s Hospital and Harvard Medical School, Boston, Massachusetts; Division of Neuroimmunology and Glial Biology (AJG, EW), Department of Neurology, Weill Institute of Neurosciences, University of California San Francisco, San Francisco, California; Department of Neurology and Neurotherapeutics (DC, BG), University of Texas Southwestern Medical Center, Dallas, Texas; Department of Neurology, University of California San Diego, San Diego, California; and Department of Ophthalmology (AJG), University of California San Francisco, San Francisco, California. Supported by NIH (NINDS K23NS069806, A.T.W., PI). A. T. Waldman has received research support from the NIH (NINDS K23NS069806, PI); she has received research support from the NIH (R01NS071463, site investigator; U54NS115052, project PI), Biogen Idec (PI), IONIS Pharmaceuticals (PI), United Leukodystrophy Foundation, and the Children’s Hospital of Philadelphia (Foerderer Award, PI), royalties from UpToDate, and served as a consultant to Optum. L. Benson does not have disclosures directly related to the content of this article; Biogen sponsored clinical trial; paid consultant for the national vaccine injury compensation program. G. W. Liu discloses her spouse’s royalties for Liu, Volpe, Galetta: Neuro-Ophthalmology, Diagnosis and Management. 2010, Elsevier. A. J. Green does not have disclosures directly related to the content of this article; he has received research support from NINDS, NMSS, NINDS SBIR, Adelson Medical Research Foundation, Hilton Foundation, Hellman Family Foundation/That Many May See, and Inception Sciences (Prior); he has received fees or other compensation from Bionure, Pipeline Therapeutics, Inception Sciences and Mylan Pharma; he is an Associate Editor at JAMA Neurology and has previously served on an endpoint adjudication committee for Medimmune/Viela Bio and has intellectual property and patents related to the UCSF Small Molecule Remyelination Program. E. Waubant is the site principal investigator for ongoing trials with Genentech and Biogen; she has research funding from NIH, PCORI, NMSS, and Race to Erase MS. She has received honoraria for lectures from Medscape, The Corpus, and AAN and for consulting work from Jazz Pharmaceuticals, Emerald, and DBV; she is the cochief editor for MSARD. G. Heidary has no disclosures relevant to the content of the article; she has grant support from the Children’s Tumor Foundation and NIH. J. Graves does not have disclosures directly related to the content of this article; she has received honoraria from Genzyme for nonpromotional trainee education events; she has received personal fees from Novartis and Celgene; she has received recent grant and clinical trial support from the National MS Society, Race to Erase MS, Biogen, Genentech, and Octave. B. Greenberg has received grant support from the NIH, NMSS, Transverse Myelitis Association, PCORI, Guthy Jackson Charitable Foundation, Chugai, Medimmune, Medday, and Genentech. He has received consulting fees from Alexion, Novartis, EMD Serono, Genentech, and Celgene; he is an unpaid board member of the Transverse Myelitis Association. The other authors report no conflicts of interest. Address correspondence to Amy T. Waldman, MD, MSCE, Division of Neurology, Children’s Hospital of Philadelphia, 3401 Civic Center Boulevard, Philadelphia, PA 19104; E-mail: [e-mail] Waldman et al: J Neuro-Ophthalmol 2021; 41: 469-475 469 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution lated. Logistic regression was used to determine whether IOD was associated with remote ON. Results: A total of 157 participants with POMS (mean age 15.2 years, SD 3.2; 67 [43%] with remote ON) were enrolled. RNFL thinning occurred in 45 of 90 (50%) ON eyes and 24 of 224 (11%) non-ON eyes. An IOD .5 mm was associated with a history of remote ON (P , 0.001). An IOD .5 mm occurred in 62 participants, 40 (65%) with remote ON. Among 33 participants with remote ON but normal RNFL values ($86 mm in both eyes), 14 (42%) were confirmed to have ON by IOD criteria (.5 mm). Conclusions: In POMS, the diagnostic yield of OCT in confirming remote ON is enhanced by considering RNFL IOD, especially for those patients with RNFL thickness for each eye in the normal range. An IOD .5 mm in patients with previous visual symptoms suggests a history of remote ON. Journal of Neuro-Ophthalmology 2021;41:469–475 doi: 10.1097/WNO.0000000000001070 © 2020 by North American Neuro-Ophthalmology Society O ptical coherence tomography (OCT) can be used to quantifiably estimate retinal thickness in ophthalmologic and neurologic conditions including multiple sclerosis (MS). In particular, quantification of the retinal nerve fiber layer (RNFL), which consists of the unmyelinated axonal fibers of the optic nerve, provides insight into active or previous injury to the optic nerve. OCT-derived measurements of RNFL thickness are primarily compared with normative values supplied by the manufacturer’s reference database to make inferences about the presence of underlying pathology (1). Using this approach, RNFL thickness values are usually considered abnormal if they fall below the lower fifth or first percentiles. RNFL measurements demonstrate excellent reproducibility (1). The presence of RNFL thinning has shown good sensitivity and specificity for confirming remote optic neuritis (ON) in the context of MS (2). Documenting a previous ON event can be useful in the diagnostic evaluation of a child with possible MS. Yet, although ON is a common symptom in POMS, studies have demonstrated that only 50% of pediatric ON eyes may show RNFL thinning after a clinical ON attack (2–4). Thus, ON pathology may be missed by OCT if RNFL thickness is in the normal range. Regarding the prevalence of measurable retinal insult after ON, pediatric ON and POMS are rare disorders; therefore, OCT studies have generally been limited by small sample sizes. For example, among 5 published OCT studies on POMS, each reported less than 28 participants, of whom only a portion had ON (2–6). The largest study included 53 POMS participants (27 ON eyes) (7). There is also discordance among these studies regarding the prevalence of RNFL thinning in non-ON eyes. Beyond relying on age-related normative values for RNFL thickness in individual eyes, the field would benefit from data that documents what amount of interocular difference (IOD) 470 in RNFL thickness would be indicative of a previous ON attack in a POMS patient. We previously demonstrated in a single-center cohort of 24 POMS participants using Cirrus OCT that considering the RNFL IOD helped detect an additional 12.5% of POMS-ON participants who had failed to demonstrate RNFL thinning in the affected eye (2). In an international adult MS cohort, an RNFL IOD greater than 5 mm was shown to predict remote unilateral ON (8). To improve the diagnostic yield of RNFL measurements, we first determined the frequency of RNFL thinning in ON and nonON eyes in a large multicenter POMS cohort. We then explored whether considering the RNFL IOD improves detection of remote ON. METHODS PEdiatric Research Collaboration ExPloring Tests in Ocular Neuroimmunology Group We developed the PERCEPTION (PEdiatric Research Collaboration ExPloring Tests in Ocular Neuroimmunology) Study to address the gaps in knowledge regarding visual outcome metrics and their interpretation in pediatric neuroinflammatory diseases. For this study, children with POMS had been enrolled at 4 academic centers (Boston Children’s Hospital, Children’s Hospital of Philadelphia [CHOP], University of California San Francisco [UCSF], and University of Texas Southwestern/Children’s Dallas). The centers were chosen based on their respective visual sciences research interests with expertise in both pediatric neurology and neuro-ophthalmology. The study is approved by the respective institutional review boards at each center. All participants gave informed written consent, and child assent was obtained. Participants Youth with POMS whose first attack occurred ,18 years of age were enrolled. POMS diagnosis was confirmed by the 2017 McDonald criteria (9). Children with other neuroinflammatory diseases or radiographic isolated syndrome (abnormal MRI suggestive of MS with no clinical attacks) were excluded from the current study. Clinical ON was defined as having visual impairment lasting .24 hours, accompanied by pain with eye movement, abnormal color vision, and/or a central scotoma; a history of ON was confirmed by medical records. POMS-ON participants were excluded if their most recent ON attack occurred ,6 months prior the OCT scan. POMS non-ON participants were defined as those that did not have a clinical history of ON in either eye. For participants with a history of unilateral ON, the unaffected eye is defined as the fellow eye. In children, RNFL thickness is greater than in adults; thus, pediatric values cannot be abstracted from adult cohorts (10). Therefore, to obtain the appropriate normative data, a healthy control cohort with no history of known ocular disease Waldman et al: J Neuro-Ophthalmol 2021; 41: 469-475 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution affecting the visual pathway was also enrolled. Healthy controls were recruited at 3 of the 4 centers (all centers except Boston Children’s Hospital) by local advertisement. RESULTS Optical Coherence Tomography Across the 4 sites, 157 participants with POMS (mean age 15.2 years, SD 3.2) and 33 healthy controls (mean age 13.6 years, SD 5.0) were enrolled (Table 1). The difference in age between the groups was significant (P = 0.016), although there was no relationship between age and RNFL thickness in the healthy controls (P = 0.584). There was no difference in sex between the POMS and control cohorts (P = 0.139). A clinical history of ON was reported in 67 POMS participants (90 eyes) (Table 1). Additional clinical information for the POMS group is provided in Table 1. In the healthy control cohort, the mean RNFL thickness was 104.0 mm (SD 9.0, range 86–130; Table 2); accordingly, abnormal RNFL thickness was defined as ,86 mm (,2 SD from the control mean). Spectral domain OCT was obtained by trained technicians at each site using Spectralis (Heidelberg, Germany, software version 6.12) at 40,000 A scans per second. Participants underwent an RNFL scan using the Nsite Analytics RNFL protocol for each undilated eye. The technician verified that the images were focused and centered with uniform illumination and assessed for artifacts. All Spectralis OCT scans were reviewed to ensure acceptable quality, as defined by the OSCAR-IB guidelines (11). OCT data from UCSF has been previously published (7). All OCT data are reported as recommended by the APOSTEL guidelines (12). Database Data were managed and stored using the research-focused electronic web-based data capture system REDCap (13), hosted at CHOP under an agreement with the software’s development consortium, led by Vanderbilt University. Statistical Analyses Demographic features were compared using the Student t test for age and the test of proportions for sex. The mean, median, SD, and range of RNFL thickness and RNFL IOD values were calculated. RNFL thinning was defined as ,2 SD below the mean for the control participants. Based on a recently published international collaboration in adults, an abnormal RNFL IOD was defined as .5 mm (8). RNFL thickness values were compared between groups using a generalized estimating equation (using an independent covariance matrix to account for intrasubject intereye relationships). RNFL IOD values were compared between groups using multivariate linear regression. Linear regression was also used to investigate the relationship between RNFL IOD and a history of remote ON within the POMS cohort. In a sensitivity analysis, we explored other cutoffs for abnormal RNFL IOD values to determine whether our results are concordant with previously published adult studies. We calculated the sensitivity and specificity of different RNFL IOD cutoffs to predict remote ON and created a receiver operating characteristic (ROC) curve to determine the cutoff that maximized both specificity and sensitivity. The area under the curve (AUC) was calculated to determine the capacity of the test (the consideration of RNFL IOD values) to distinguish POMS-ON subjects from POMS subjects with no history of remote ON. Statistical analyses were performed using Stata Statistical Software (STATA, version 12.1; StataCorp LP, College Station, TX), and the ROC curve and AUC calculation were performed using GraphPad Prism (Version 8.3.0; GraphPad Software Inc, La Jolla, CA). The statistical significance was defined as P , 0.05. Waldman et al: J Neuro-Ophthalmol 2021; 41: 469-475 Participants Retinal Nerve Fiber Layer Thickness Of the 90 eyes with remote ON, 45 (50%) demonstrated RNFL thickness ,86 mm (Fig. 1). In the 224 non-ON eyes, 24 (11%) had RNFL thickness ,86 mm (Fig. 1). As a group, mean RNFL thickness was reduced in the pooled eyes of POMS participants compared with eyes from controls (P , 0.001; Table 2). As expected, POMS eyes with a history of ON (both eyes of participants with bilateral ON and the affected eyes of unilateral ON participants) had thinner RNFL values compared with healthy controls (P , 0.001; Table 2). Compared with healthy controls, POMS participants with no history of remote ON (neither unilateral nor bilateral) demonstrated reduced RNFL thickness (P , 0.001; Table 2), suggesting the presence of subclinical injury. Likewise, the fellow eyes of POMS participants with a history of unilateral ON also demonstrated RNFL thinning compared with control eyes (P = 0.001; Table 2). There was no difference in RNFL thickness between fellow unaffected eyes and the eyes of participants with no history of remote ON (neither unilateral nor bilateral) (P = 0.735). Within the POMS group, as anticipated, the RNFL thickness of eyes affected by ON was thinner than those not affected (P , 0.001). We acknowledge that a proportion of the participants were previously published (7). When removed, the mean RNFL thickness among non-ON eyes (96.8 mm, SD 11.6) and clinical ON eyes (83.0 mm, SD 21.4) was unchanged. Retinal Nerve Fiber Layer Interocular Difference RNFL IOD was greater in POMS participants than in controls (P = 0.006; Table 2). Compared with controls, RNFL IOD was greater for POMS participants with a clinical history of both unilateral (P , 0.001) and bilateral (P = 0.003) ON. 471 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 1. Demographic, clinical, and OCT data for healthy controls and POMS participants. Age, yr, mean (SD) Sex, N (% female) Time since clinical onset, yr, median (range) History of ON (bilateral or unilateral), N (%) Unilateral ON, N (% ON participants) Bilateral ON, N (% ON participants) Healthy Controls (N = 33) POMS (N = 157) P value 13.6 (5.0) 18 (55) N/A N/A N/A N/A 15.2 101 1.0 67 44 23 0.016* 0.139† N/A N/A N/A N/A (3.2) (64) (0.02–14.2) (43%) (66%) (34%) *Age compared using Students t test. † Sex compared using the test of proportions. N, number of subjects; N/A, not applicable; OCT, optical coherence tomography; ON, optic neuritis; POMS, pediatric-onset multiple sclerosis. However, there was no difference in RNFL IOD between POMS participants with no history of ON and controls (P = 0.087, Table 2). Retinal Nerve Fiber Layer Interocular Difference as an Indicator of Remote Optic Neuritis RNFL IOD was abnormal (.5 mm) in 62 (39%) POMS participants (Fig. 1). Of these individuals, 40 (65%) had a history of remote ON. Among participants with a history of clinical ON but RNFL thickness values $86 mm in both eyes (N = 33), 14 (42%) individuals were identified as having ON by IOD criteria (having an RNFL IOD .5 mm; Fig. 1). Of the 14 additional ON participants identified by IOD criteria, 10 had unilateral ON and 4 had bilateral ON. Using both RNFL thickness and IOD as criteria for remote ON, 48 (72%) participants with a clinical history of ON had unilateral or bilateral RNFL thinning ,86 mm and/or an RNFL IOD .5 mm. We also explored RNFL thickness and IOD in those POMS participants without a clinical history of ON (N = 90) (Fig. 1). Bilateral or unilateral RNFL thinning ,86 mm occurred in 12 (13%) non-ON POMS participants or 20 of 180 (11%) eyes. An RNFL IOD .5 mm occurred in 22 (24%) non-ON POMS participants. Together, 27 (30%) POMS participants with no clinical history of ON had unilateral or bilateral RNFL thinning ,86 mm and/or an abnormal RNFL IOD. By contrast, only 3 (9%) healthy control subjects had RNFL IOD .5 mm; of those subjects, 2 had an IOD of 6 and 1 had an IOD of 7 mm. The area under the ROC curve, the capacity of the test to distinguish patients with a history of remote ON vs those without, was 0.76 (Fig. 2). In predicting the ON history, an RNFL IOD cutoff of 5 mm had a sensitivity of 60% and specificity of 76%. An IOD cutoff of 6 mm had a slightly lower sensitivity (55%) and higher specificity (86%), but the difference between the 2 (specificity 2 sensitivity = 31%) was nearly double that for a cutoff of 5 mm TABLE 2. Absolute RNFL thickness and IOD in healthy controls and POMS participants. Healthy Controls N RNFL thickness, mm, 66 eyes mean (SD, range) RNFL IOD, mm, mean (SD, range) Result POMS Group 104.0 All (9.0, 86–130) Non-ON eyes* ON eyes Fellow eyes 33 subjects 2.4 (2.0, 0–7) All Non-ON subjects† Unilateral ON Bilateral ON N Result P value 314 eyes 180 eyes 92 eyes 42 eyes 157 subjects 90 subjects 44 subjects 23 subjects 92.9 (17.5, 36–127) 97.7 (13.4, 40–127) 81.9 (21.1, 36–119) 96.9 (13.2, 67–124) 6.9 (9.4, 0–65) 4.0 (5.2, 0–28) 12.3 (12.4, 1–65) 8.5 (10.9, 0–42) ,0.001 ,0.001 ,0.001 0.001 0.006 0.087 ,0.001 0.003 P values for group comparisons between healthy controls and POMS were calculated using univariate generalized estimating equations (using an independent covariance matrix to account for intrasubject intereye relationships) for RNFL thickness values and linear regression for RNFL IOD. *The subject did not have a history of ON in either eye; fellow eyes (the unaffected eye in a subject with unilateral ON) are listed separately in the table. † The subject did not have a history of ON in either eye. IOD, interocular difference; N, number of eyes or subjects; ON, optic neuritis; POMS, pediatric-onset multiple sclerosis; RNFL, retinal nerve fiber layer. 472 Waldman et al: J Neuro-Ophthalmol 2021; 41: 469-475 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 1. RNFL IOD as an indicator of remote ON in POMS. Diagram demonstrating the number of POMS patients (with and without optic neuritis) with RNFL thinning (,86 mm) and an IOD greater than 5 mm. In patients with a clinical history of ON, RNFL thinning was not present in all affected eyes. The IOD can help the clinician identify optic nerve pathology. The presence of RNFL thinning or an abnormal IOD in those without a clinical ON history suggests a subclinical insult. IOD, interocular difference; the numbers in parestheses represent the number of participants in each category; ON, optic neuritis; POMS, pediatric-onset multiple sclerosis; RNFL, retinal nerve fiber layer. (specificity 2 sensitivity = 16%), suggesting that an IOD cutoff of 5 mm is preferable. In comparison, the sensitivity for an IOD cutoff of 4 mm was 69% with a specificity of 74%; a cutoff of 4 mm therefore achieved the smallest difference between sensitivity and specificity for all IOD cutoffs (specificity 2 sensitivity = 5%). Thus, an IOD cutoff of 4 mm had the greatest capacity to distinguish POMS patients with remote ON vs those without an ON history; however, both cutoffs of 4 and 5 mm yield similar specificities. For consistency with previous adult studies and ease of implementation across the age span, we recommend a cutoff of 5 mm. Within the POMS group, RNFL IOD was associated with a history of ON (either bilateral or unilateral) and unilateral ON (both P , 0.001). The relationships remained significant when age and sex were included in the model. FIG. 2. Receiver operating characteristic (ROC) curve for retinal nerve fiber layer interocular difference (RNFL IOD). The sensitivity and specificity of each potential RNFL IOD cutoff for identifying remote optic neuritis (ON) was calculated. For each IOD cutoff, the sensitivity was plotted against (1 – specificity) to generate a ROC curve. The area under the curve, the capacity of the test to distinguish patients with a history of remote ON vs those without, was 0.76. The optimal IOD cutoff at which sensitivity and specificity are closest is 4 mm; however, both cutoffs of 4 and 5 mm yield similar specificities. Waldman et al: J Neuro-Ophthalmol 2021; 41: 469-475 473 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution DISCUSSION RNFL thinning below normative ranges in POMS is not an obligate finding after an ON attack. This study confirms our observation, as only 50% of POMS-ON eyes POMSON eyes demonstrated RNFL thinning of mean values (averaged across quadrants or sectors) below the normal range for age-matched controls (,86 microns) (2). The value of OCT in identifying a history of clinical ON in POMS is enhanced when RNFL IOD is considered in addition to RNFL thickness values in individual eyes. Our findings have clinical implications for improved diagnostic yield when using OCT to confirm the ON history in POMS. We defined an RNFL IOD .5 mm as indicative of disease based on a large adult cohort (8) that included 368 healthy and 1,530 MS participants. Nolan-Kenney and colleagues determined that an RNFL IOD threshold of 5 mm maximized sensitivity and specificity for identifying a history of unilateral ON. The narrow IOD in our control population (mean 2.4 mm, SD 2.0) further supports the use of 5 mm as a cutoff. The adult study also explored IOD cutoffs for ganglion cell and inner plexiform layer thickness; however, our multicenter cohort did not uniformly collect such data for analysis. Although the presence of RNFL thinning is not a universal finding in POMS, the magnitude of average RNFL thinning in POMS-ON eyes compared with control eyes (mean loss of 22.1 mm) was similar to adults with MS. A meta-analysis using time-domain OCT demonstrated a loss, on average, of 20.4 mm in RNFL thickness (95% confidence interval [CI] 17.9–22.9 mm) among adult MS-ON eyes (14). The same study reported an average loss of RNFL thickness in adult MS non-ON eyes of 7.1 mm (95% CI 5.5–8.7 mm) compared with controls, which is similar to our POMS cohort (mean loss of 6.3 mm). We further explored the OCT parameters in POMS participants with no history of clinical ON. Although these participants did not have a clinical attack suggestive of ON, their RNFL thickness values were significantly thinner than controls; 30% had RNFL thinning (,86 mm) and/or an abnormal IOD (.5 mm). These participants may have experienced subclinical ON, as demonstrated in a previous POMS study (7). There are several caveats to our work. First, we defined a history of ON by review of medical records for clinical symptoms. For this study, we did not require a dedicated MRI of the orbits at the time of the clinical symptoms to confirm the diagnosis of ON. It is possible that a higher proportion of ON participants would have RNFL thinning if a stricter definition (such as the presence of an enhancing ON lesion) was used. The Pediatric Optic Neuritis Prospective Outcomes Study (15) defined their cohort based on the presence of an enhancing optic nerve lesion using a dedicated MRI of the orbits with fat-saturated sequences 474 and will be able to specifically address the proportion of participants demonstrating RNFL thinning after an MRI-confirmed attack in the future. Although NolanKenney et al only considered the use of RNFL IOD to detect monocular ON (8), we chose to also include POMS participants with a history of bilateral involvement to mimic “real world” experiences. Most ON occurrences in POMS are unilateral, although both eyes can be affected at different times; therefore, although an individual attack may be considered unilateral, a patient who experienced unilateral attacks in each eye, regardless of whether the attacks temporally coincided, would be considered to have a clinical history of bilateral ON. Our study is limited by the inclusion of retrospective data, including a previously published cohort. Graves et al reported subclinical RNFL thinning in POMS non-ON eyes (7), which could have influenced the proportion of abnormal eyes in the current study; however, when these participants were removed from the analyses, there was no change in the mean RNFL values. Our findings suggest that in POMS, the value of OCT as a tool to detect a history of clinical ON is enhanced when RNFL IOD is considered in addition to monocular RNFL thickness values. In clinical practice, RNFL IOD should be considered when interpreting OCT results, especially for POMS patients who may have experienced clinical symptoms but have RNFL thickness values for each eye within the normal range. STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: A. T. Waldman and J. R. Sollee; b. Acquisition of data: A. T. Waldman, L. Benson, J. R. Sollee, A. M. Lavery, G. W. Liu, A. J. Green, E. Waubant, G. Heidary, D. Conger, J. Graves, and B. Greenberg; c. Analysis and interpretation of data: A. T. Waldman, L. Benson, J. R. Sollee, A. M. Lavery, G. W. Liu, A. J. Green, E. Waubant, G. Heidary, D. Conger, J. Graves, and B. Greenberg. Category 2: a. Drafting the manuscript: A. T. Waldman and J. R. Sollee; b. Revising it for intellectual content: A. T. Waldman, L. Benson, J. R. Sollee, A. M. Lavery, G. W. Liu, A. J. Green, E. Waubant, G. Heidary, D. Conger, J. Graves, and B. Greenberg. Category 3: a. Final approval of the completed manuscript: A. T. Waldman, L. Benson, J. R. Sollee, A. M. Lavery, G. W. Liu, A. J. Green, E. Waubant, G. Heidary, D. Conger, J. Graves, and B. Greenberg. REFERENCES 1. Avery RA, Rajjoub RD, Trimboli-Heidler C, Waldman AT. Applications of optical coherence tomography in pediatric clinical neuroscience optical coherence tomography background. Neuropediatrics. 2015;46:88–97. 2. Waldman AT, Liu GT, Lavery AM, Liu G, Gaetz W, Aleman T, Banwell B. Optical coherence tomography and visual evoked potentials in pediatric MS. J Neurol Neuroimmunol Neuroinflamm. 2017;4:e356. 3. Yeh EA, Weinstock-Guttman B, Lincoff N, Reynolds J, Weinstock A, Madurai A, Agarwal N, Buch P, Karpinski M, Ramanathan M. Retinal nerve fiber thickness in inflammatory Waldman et al: J Neuro-Ophthalmol 2021; 41: 469-475 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution 4. 5. 6. 7. 8. demyelinating diseases of childhood onset. Mult Scler. 2009;15:802–810. Waldman AT, Hiremath G, Avery RA, Conger A, Pineles SL, Loguidice MJ, Talman LS, Galetta KM, Shumski MJ, Wilson J, Ford E, Lavery AM, Conger D, Greenberg BM, Ellenberg JH, Frohman EM, Balcer LJ, Calabresi PA. Monocular and binocular low-contrast visual acuity and optical coherence tomography in pediatric multiple sclerosis. Mult Scler Relat Disord. 2014;3:326–334. Yilmaz U, Gucuyener K, Erin DM, Yazar Z, Gurkas E, Serdaroglu A, Tepe N, Demir E. Reduced retinal nerve fiber layer thickness and macular volume in pediatric multiple sclerosis. J Child Neurol. 2012;27:1517–1523. Yeh EA, Marrie RA, Reginald YA, Buncic JR, Noguera AE, O’Mahoney J, Mah JK, Banwell B, Costello F; Canadian Pediatric Demyelinating Disease Network. Functional-structural correlations in the afferent visual pathway in pediatric demyelination. Neurology. 2014;83:2147–2152. Graves JS, Chohan H, Cedars B, Arnow S, Yiu H, Waubant E, Green A. Sex differences and subclinical retinal injury in pediatric-onset MS. Mult Scler. 2017;23:447–455. Nolan-Kenney RC, Liu M, Akhand O, Akhand O, Calabresi PA, Paul F, Petzold A, Balk L, Brandt AU, Martinez-Lapiscina EH, Saidha S, Villoslada P, Al-Hassan AA, Behbehani R, Frohman EM, Frohman T, Havla J, Hemmer B, Jiang H, Knier B, Korn T, Leocani L, Papadopoulou A, Pisa M, Zimmermann H, Galetta SL, Balcer LJ; International Multiple Sclerosis Visual System Consortium. Optimal intereye difference thresholds by optical coherence tomography in multiple sclerosis: an international study. Ann Neurol. 2019;85:618–629. Waldman et al: J Neuro-Ophthalmol 2021; 41: 469-475 9. McNicholas N, Hutchinson M, McGuigan C, Chataway J. 2017 McDonald diagnostic criteria: a review of the evidence. Mult Scler Relat Disord. 2018;24:48–54. 10. 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.e1. 11. Tewarie P, Balk L, Costello F, Green A, Martin R, Schippling S, Petzold A. The OSCAR-IB consensus criteria for retinal OCT quality assessment. PLoS One. 2012;7:1–7. 12. Cruz-Herranz A, Balk LJ, Oberwahrenbrock T, Saidha S, Martinez-Lapiscina E, Lagreze W, Schuman J, Villoslada P, Calabresi P, Balcer L, Petzold A, Green A, Paul F, Brandt A, Albrecht P; IMSVISUAL consortium. The APOSTEL recommendations for reporting quantitative optical coherence tomography studies. Neurology. 2016;86:2303– 2309. 13. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap): a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2008;42:377– 381. 14. Petzold A, de Boer JF, Schippling S, Vermersch P, Kardon R, Green A, Calabresi PA, Polman C. Optical coherence tomography in multiple sclerosis: a systematic review and meta-analysis. Lancet Neurol. 2010;9:921–932. 15. Pineles SL, Liu GT, Waldman AT, Lazar E, Kupersmith MJ, Repka MX; Pediatric Eye Disease Investigator Group and the Neuro-Ophthalmology Research Disease Investigator Consortium. Pediatric optic neuritis prospective outcomes study. J Neuroophthalmol. 2016;36:115–117. 475 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.