Abstract
Objective
Familial Dysautonomia (FD) disease, lacks a useful biomarker for clinical monitoring. In this longitudinal study we characterized the structural changes in the macula, peripapillary and the optic nerve head (ONH) regions in subjects with FD.
Methods
Data was consecutively collected from subjects attending the FD clinic between 2012–2019. All subjects were imaged with spectral-domain Optical Coherence Tomography (OCT). Global and sectoral measurements of mean retinal nerve fiber layer (RNFL) and macular ganglion cell and inner plexiform layer (GCIPL) thickness, and ONH parameters of rim area, average cup-to-disc (C:D) ratio, and cup volume were used for the analysis. The best fit models (linear, quadratic and broken stick linear model) were used to describe the longitudinal change in each of the parameters.
Results
91 subjects (149 eyes) with FD of ages 5–56 years were included in the analysis. The rate of change for average RNFL and average GCIPL thicknesses were significant before reaching a plateau at the age of 26.2 for RNFL and 24.8 for GCIPL (−0.861 μm/year (95% CI: −1.026, −0.693) and −0.553 μm/year (95% CI: −0.645, −0.461), respectively). Significant linear rate of progression was noted for all ONH parameters, except for a subset of subjects (24%), with no cupping that did not show progression in any of the ONH parameters.
Conclusions
The rapidly declining RNFL and GCIPL can explain the progressive visual impairment previously reported in these subjects. Among all structural parameters, ONH parameters might be most suitable for longitudinal follow-up, in eyes with a measurable cup.
Introduction
Familial Dysautonomia (FD) is an incurable, recessively inherited, autonomic and sensory neuropathy, characterized by low levels of the ELP1 protein (previously known as IKAP), affecting the development and survival of specific neuronal tissue, including retinal ganglion cells (RGC) [1–3]. This ELP1 protein deficiency is caused by single T to C change in base pair 6 of intron 20, located in chromosome 9, which appears almost exclusively in European Jews (Ashkenazi) [2]. Subjects with FD have a complex neurological phenotype with reduced pain and temperature perception, absent deep tendon reflexes, afferent baroreflex failure causing orthostatic hypotension and paroxysmal hypertension, gait ataxia and visual impairment [3–9].
Assessment of ocular disease progression in FD patients is essential. Gradually but consistently, vision deteriorates over time, and can progress to legal blindness, usually after the third decade of life.[3,10] Nowadays, as FD patients’ life expectancy is increasing with average age of 25 years old, the debilitating effect of the vision loss is more prominent [1]. Subjects with FD lack muscle sensation that usually fine-tunes motor coordination and therefore heavily depend on their sight to correct muscle movements [11]. Without vision, patients have higher prevalence of falls and bone fractures, and as a result, may lose the ability to walk independently.
The retina presents a peculiar histologic structure for neurodegeneration study as it consists of segregated nuclei in the ganglion cell layer (GCL) and their unmyelinated axons in the retinal nerve fiber layer (RNFL), which subsequently form the optic nerve [12]. Ocular manifestation of neurodegenerative disorder can be examined using, optical coherence tomography (OCT), a clinic-based device that provides a non-invasive, high-resolution, repeatable cross-sectional imaging of the retina [13].
Previous cross sectional study, using OCT, demonstrated gradual thinning of peripapillary RNFL in all subjects with FD [3], but longitudinal research assessing the trajectory of a given individual was never done. The OCT device enables not only the monitoring of the RNFL but also other retinal components in the macula and optic nerve head (ONH) [14,15]. Previous studies assessing RGCs degeneration showed advantage in using disparate retina areas, as the different characteristics of each of these areas allow tracking progression in different disease severity stages [16,17].
In this study, we investigated the hypothesis that the well-documented worsening visual loss in FD patients is caused by ongoing retinal neuronal degeneration. We used spectral-domain OCT system to quantify longitudinal changes in macular, ONH and peripapillary regions in subjects with FD.
Methods
A longitudinal, observational study was carried out at Dysautonomia Center in the Department of Neurology at NYU School of Medicine to assess ocular structure changes over time in subjects with FD.
Participants
Subjects with FD were recruited to the study if they were 4 years of age or older. FD diagnosis was based on genetic confirmation of homozygous ELP1 mutation. All subjects were scanned with OCT during annual visits in the FD clinic, unless otherwise medically indicated. All scans acquired between 2012–2019 were considered for this project.
Study Protocol
All subjects had their medical records reviewed and underwent, slit lamp biomicroscopy and fundus examination. Subjects were imaged with spectral-domain OCT (Cirrus HD-OCT; Zeiss, Dublin, CA) using the macular cube 512 × 128 and the optic disc cube 200 × 200 scans. Scans were disqualified if they had a signal strength of less than 6, decentration of the image (foveola for macular scans and disc center for ONH scans), segmentation errors, or motion artifacts defined as a discontinuity of the blood vessels that exceeded the width of 1 major vessel diameter. Average circumpapillary RNFL (global and quadrants), vertical cup-to-disc (C:D) ratio, average C:D ratio, rim area, cup volume, macular GCIPL thickness (global and sectorial), total retina and outer retina thickness were used for the analysis.
Statistical Analysis
A linear spline model was used with one breaking point and two models without breaking point - a linear model and a model with a quadratic term for age - to evaluate the longitudinal change for each OCT outcome measure. The model with the lowest Akaike Information Criterion (AIC) out of the three was chosen as the preferred model. We used a mixed effects model, which is known as well suited to deal with unbalanced designs [18,19], for analyzing a sample with different number of observations per eye, including eyes with one observation. All models included fixed effects of age at each visit and OCT signal strength, and random intercepts to account for the between-visit correlation for each eye and the inter-eye correlation within a subject. The models for ONH outcomes were also adjusted for disc area as a fixed effect. Moreover, the rates of change for the average and sectorial RNFLs and GCIPLs were compared between eyes with and without cupping using linear mixed effects models. Furthermore, the predicted RNFLs (average and sectoral) at age 6 for each FD subject were generated based on the preferred models via bootstrapping. Percentage loss was calculated by comparing the predicted RNFL to the normal reference value [20]. The overall comparison of percentage loss across all RNFLs was conducted using one-way repeated ANOVA, while the difference in percentage loss between temporal RNFL and the rest was tested by pair-wise t-test. Statistical analysis was performed using R software version 3.5.2. p < 0.05 was considered statistically significant.
Results
Two hundred and five eyes (122 subjects) were eligible for the study, out of which 149 eyes (91 subjects) had at least one qualified scan (Figure 1). Subjects’ age range at baseline was 5–56 years with 54% women (Table 1). The average follow-up duration for subjects with 2 or more qualified tests (85 eyes) was 45.7±21.2 months. Marked tissue loss was recorded by OCT at baseline in conventional parameters (Table 2). Of note is the small disc diameter in this cohort.
Fig. 1.
OCT scans selection process. n = subjects(eyes)
Table 1.
Demographic data by age group
| Male | Female | Total | ||||
|---|---|---|---|---|---|---|
| Subjects | Eyes | Subjects | Eyes | Subjects | Eyes | |
| 5–15 | 6 | 10 | 10 | 16 | 16 | 26 |
| 15–25 | 20 | 32 | 21 | 39 | 41 | 71 |
| 25–35 | 12 | 21 | 9 | 14 | 21 | 35 |
| 35–45 | 3 | 4 | 7 | 9 | 10 | 13 |
| 45- | 1 | 2 | 2 | 2 | 3 | 4 |
| Total | 42 | 69 | 49 | 80 | 91 | 149 |
Table 2.
Baseline OCT measurement of the study population
| Mean (Standard deviation) | |
|---|---|
| Average RNFL (μm) | 54.55 (3.42) |
| Temporal RNFL (μm) | 36.81 (9.16) |
| Superior RNFL (μm) | 72.99 (13.16) |
| Nasal RNFL (μm) | 52.05 (8.48) |
| Inferior RNFL (μm) | 75.83 (14.59) |
| Average GCIPL (μm) | 56.08 (4.73) |
| Superior GCIPL (μm) | 56.70 (4.84) |
| Inferior GCIPL (μm) | 55.10 (4.98) |
| Rim area (mm2) | 1.25 (0.26) |
| Cup volume (mm2) | 0.05 (0.08) |
| Average C:D ratio | 0.34 (0.19) |
| Vertical C:D ratio | 0.36 (0.21) |
| Disc diameter (mm) | 1.32 (0.18) |
RNFL retinal nerve fiber layer, GCIPL ganglion cell-inner plexiform layer, C:D cup-to-disc
The linear spline model with one breaking point was the best fitted model for the rate of change of average RNFL and GCIPL thicknesses with a statistically significant rate of decline before the tipping point (−0.861 μm/year (95% CI: −1.026, −0.693) and −0.553 μm/year (95% CI: −0.645, −0.461), respectively) and a non-significant slope after the tipping point (Table 3). The normal aging rate of change in average RNFL thickness was previously reported to be substantially slower (−0.365 μm/year)[21] than the rate we reported in the accelerated phase in subjects with FD (−0.861 μm/year).
Table 3.
Rate of change in OCT parameters of the peripapillary and macula
| Rate of Change per year | |||||
|---|---|---|---|---|---|
| Before the Tipping Point | After the Tipping Point | ||||
| Age at Tipping Point (Years) | Slope (95% CI) | P Value | Slope (95% CI) | P Value | |
| Average RNFL (μm) | 26.2 | −0.861 (−1.026, −0.693) | <0.001 | −0.030 (−0.212, 0.151) | 0.740 |
| Temporal RNFL (μm) | 20.8 | −0.635 (−0.978, −0.291) | <0.001 | 0.257 (0.060, 0.455) | 0.010 |
| Superior RNFL (μm) | 31.4 | −1.104 (−1.315, −0.891) | <0.001 | −0.053 (−0.407, 0.300) | 0.770 |
| Nasal RNFL (μm) | 32.2 | −0.439 (−0.634, −0.245) | <0.001 | 0.362 (0.019, 0.706) | 0.039 |
| Inferior RNFL (μm) | 27.1 | −1.485 (−1.750, −1.212) | <0.001 | −0.298 (−0.644, 0.048) | 0.090 |
| Average GCIPL (μm) | 24.8 | −0.553 (−0.645, −0.461) | <0.001 | −0.080 (−0.183, 0.021) | 0.120 |
| Superior GCIPL (μm) | 23.3 | −0.551 (−0.708, −0.394) | <0.001 | −0.104 (−0.229, 0.021) | 0.100 |
| Inferior GCIPL (μm) | 23.8 | −0.623 (−0.771, −0.473) | <0.001 | −0.045 (−0.173, 0.083) | 0.490 |
CI confidence interval, RNFL retinal nerve fiber layer, GCIPL ganglion cell-inner plexiform layer
Statistically significant slopes are shown in bold
While the rate of decline in all RNFL sectors was significant before the tipping point, there were significant and positive (thickening) slopes only in the temporal and nasal quadrants after the tipping point. The tipping point, after which the subjects reach what was previously described as floor effect - where no further meaningful change occurs to neuronal tissue [22], appears in our cohort at the age of 26.2 years for average RNFL and 24.8 years for average GCIPL. Of the four RNFL quadrants, subjects reached the tipping point in the temporal quadrant at the youngest age (Table 3, Fig. 2). The predicted RNFL thickness at the age of 6 was significantly lower compared to healthy 6 years’ old subjects in average and all quadrants (all P<0.001) (Table 4). At this age, the percent loss in the temporal quadrant was significantly higher compared to other sectors (P<0.001).
Fig. 2.
RNFL rate of change per year. The red marks represent the age at the tipping point of each parameter. In the temporal area, subjects reached the tipping point at the youngest age (~21)
Table 4.
Retinal nerve fiber layer (RNFL) thickness (and percentage loss) at the age of six years old in FD subjects and healthy individuals. Values for FD are predicted based on our models while values in healthy are measurements were previously reported20
| Predicted | Healthy | Percentage Loss | P value | |
|---|---|---|---|---|
| Average RNFL (μm) | 71.78 | 96.29 | 25.5 | <0.001 |
| Temporal RNFL (μm) | 43.93 | 68.85 | 36.2a | <0.001 |
| Superior RNFL (μm) | 89.28 | 120.12 | 25.7 | <0.001 |
| Nasal RNFL (μm) | 58.34 | 67.46 | 13.5 | <0.001 |
| Inferior RNFL (μm) | 98.03 | 129.02 | 24.0 | <0.001 |
RNFL retinal nerve fiber layer
Statistically significant Differences are shown in bold
The percent loss in the temporal RNFL is significant higher compared to other sectors <0.001
The best fitted model for all ONH parameters (rim area, average C:D ratio and cup volume) was the linear model with a statistically significant rate of progression for all tested parameters (Table 5). A subset of subjects (24%) had no cupping and did not show progression in any of the ONH parameters (Figure 3), but demonstrated a significant rate of decline in RNFL and GCIPL parameters at similar rates as those with cupping, except for the temporal RNFL. Subjects with no cup had a significantly thicker temporal RNFL at young age compared with those with cup and a significantly faster RNFL thinning rate thereafter (p=0.002; Table 6). This subset had a small disc size compared to the rest of the cohort (disc diameter = 1.2mm), and in some eyes tortuosity of some retinal veins were noted. There was no evidence to retinal hemorrhages or optic disc drusen.
Table 5.
Rate of change in OCT ONH parameters
| Rate of Change per year | ||
|---|---|---|
| Slope (95% CI) | P Value | |
| Rim area (mm2) | −0.009 (−0.013, −0.006) | <0.001 |
| Cup volume (mm2) | 0.002 (0.001, 0.003) | <0.001 |
| Average C:D ratio | 0.007 (0.005, 0.009) | <0.001 |
| Vertical C:D ratio | 0.007 (0.005, 0.010) | <0.001 |
CI confidence interval, C:D cup-to-disc
Statistically significant slopes are shown in bold
Fig. 3.
a Cup volume rate of change per year. A subset of subjects showed no cupping (red square). b A subject with small ONH (disc diameter of 1.1 mm), presenting with slightly elevated ONH, small cup and tortuous veins. c Longitudinal structural change in ONH as reflected in OCT scan of a subject with no cupping
Table 6.
Temporal Retinal nerve fiber layer (RNFL) thickness (and percent reduction) at the age of six years old in FD subjects (with and without cupping) and healthy individuals. Values for FD are predicted based on our models while values in healthy are measurements were previously reported20
| Predicted | Healthy | Percentage Loss | P value (against healthy) | P for cupping vs. no cupping | ||
|---|---|---|---|---|---|---|
| Temporal RNFL (μm) | C:D ≤ 0.2 | 44.26 | 68.85 | 35.7 | <0.001 | <0.001 |
| C:D > 0.2 | 33.43 | 68.85 | 51.4 | <0.001 |
RNFL retinal nerve fiber layer
Discussion
This study is the first to characterize the longitudinal structural changes in the macula, peripapillary, and ONH, as measured in-vivo by OCT, in subjects with familial dysautonomia. Our results showed an accelerated and statistically significant RNFL and GCIPL loss until they reach the floor effect around the age of 26 years. Concurrently, ONH parameters showed a continuous and significant linear change throughout the age span of our cohort. However, approximately one-fourth of the subjects had no or minimal ONH cupping with no change in ONH measurements over time and therefore other locations should be used for monitoring the disease.
Our findings of accelerated retinal damage that reach the floor effect at the third decade of life explain the severe deterioration in vision reported in the second and third decades of life of these subjects, which was originally attributed to corneal opacification [3,10,23–25]. We showed statistically significant structural damage already present at the age of 6, which corroborate with the findings of a previous cross sectional study that showed considerably thin RNFL in all subjects with FD [3]. The mean RNFL thickness at the plateau in our cohort (53 μm) is similar to the threshold of minimum practical measurement that was previously reported in subjects with glaucoma (57 μm) [26].
As noted in a previous study, the measurement variability in average RNFL is smaller than in sectoral measurements [27]. Although we assessed each scan for qualification using strict criteria, RNFL sectors might be affected by a small degree of torsion of the eye causing the location of boundaries between sectors to varies with subsequent differences in the measurements [28]. The measurement variability is most notable in thin sectors adjacent to thick sectors, namely the temporal and nasal sectors, where aberrant thickening over time was previously reported. Contrary to this feature, measurements along the entire circle are not affected by torsion and thus showing lower variability. As such, this parameter is the most reliable for follow-up purposes.
Our results might indicate a preferential loss of RGCs in the macula, as was previously suggested and confirmed pathologically by Mendoza et al. [3,29]. In our cohort, the percent RNFL loss at age 6 was significantly higher in the temporal area compared to the average and other quadrant RNFL thickness. This preferential damage in the papillomacular bundle closely mimics mitochondrial optic neuropathies, such as autosomal dominant optic atrophy (ADOA) and Leber’s hereditary optic neuropathy (LHON) that demonstrates preferential loss of small axons that originate from parvocellular RGCs in the macula [30,31].
ONH parameters, rather than RNFL and GCIPL, might be better suited for a longer follow-up period in eyes with a measurable cup, as they showed linear decline throughout the duration of our cohort. Similarly, a previous study has shown the ability of OCT’s ONH parameters to detect progression in subjects with advanced glaucoma, when no further structural changes are detectable with RNFL and GCIPL [17]. The reason for this discrepancy is unclear and might be due to non-neural tissue in the ONH that is more slowly affected than the neural tissue in the retina, origination of the damage at the level of the retina that then propagate toward the ONH, or other reasons.
A subset of subjects had minimal to no-cupping, sometimes accompanied with tortuous blood vessels, as was previously observed in other mitochondrial optic neuropathies [32]. In most of these eyes the ONH was slightly elevated and completely stable even through years of follow-up. Of note is the small ONH size in the entire cohort (disc diameter = 1.3 mm) that was further smaller in this subgroup (diameter = 1.2 mm), which might indicate crowded discs [33,34]. Clinical and OCT evaluation of these eyes did not disclose evidence of elevated intracranial pressure, optic disc drusen or other main causes of elevated ONH.
The disease manifestation in the temporal quadrant of subjects with no cupping has some unique characteristics compared to subjects with cupping, as it is thicker at a younger age and the rate of decline is faster. It is unclear why this subset of subjects has relative protection to their RNFL at younger ages. Future histo-pathological studies are warranted.
Limitations
The fragile neurological status of FD subjects brings particular challenges. We could not compare the retinal structural change with a reliable longitudinal functional data such as visual acuity and visual field tests as they require cooperation that is not obvious for those patients. Furthermore, despite previous attempts to develop validate rating scales, there is a lack of systemic scales quantifying signs and symptoms of the disease. Of note, about a quarter of the subjects scanned with OCT were disqualified for the study mostly due to motion artifacts. Using an OCT device with a retinal tracking system might minimize this problem. An additional limitation is that none of the parameters were suitable for all patients at all ages. This finding should be taken into consideration when planning future studies or tracking progression as part of periodic clinical visits.
Conclusions
Under the described characteristics, the OCT device can detect a significant progressive retinal structural change in subjects with FD. Considering the paucity of biomarkers associated with FD severity, having an in-vivo reproducible indicator of neural damage in the eye might be useful for clinical monitoring of these subjects and the basis for future clinical trials on potential disease-modifying drugs.
Funding
NIH R01-EY013178, a grant from the Familial Dysautonomia Foundation, and unrestricted grant from Research to Prevent Blindness. The sponsor or funding organization had no role in the design or conduct of this research.
Footnotes
Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.
Conflict of interest Joel S Schuman: Royalties from intellectual property licensed by Massachusetts institute of Technology and Massachusetts Eye and Ear infirmary to Zeiss. None of the other authors have any conflict of interest to declare.
Ethics approval The institutional review board and ethics committee at NYU Langone Health approved the study. The study has been performed in accordance with the ethical standards laid down in the Declaration of Helsinki. Enrollment was voluntary and a written informed consent was obtained from all subjects or their legal proxy.
Availability of data and material The data that support the findings of this study are available from the corresponding author upon reasonable request.
Code availability N/A
References
- 1.Norcliffe-Kaufmann L, Slaugenhaupt SA, Kaufmann H. Familial dysautonomia: History, genotype, phenotype and translational research. Prog Neurobiol. 2017;152:131–148. [DOI] [PubMed] [Google Scholar]
- 2.Dong J, Edelmann L, Bajwa AM, Kornreich R, Desnick RJ. Familial dysautonomia: Detection of the IKBKAP IVS20+6T → C and R696P mutations and frequencies among Ashkenazi Jews. Am J Med Genet. 2002;110(3):253–257. [DOI] [PubMed] [Google Scholar]
- 3.Mendoza-Santiesteban CE, Hedges TR, Norcliffe-Kaufmann L, Axelrod F, Kaufmann H. Selective retinal ganglion cell loss in familial dysautonomia. J Neurol. 2014;261(4):702–709. [DOI] [PubMed] [Google Scholar]
- 4.Axelrod FB, Lyer K, Fish I. Progressive sensory loss in familial dysautonomia. Pediatrics. 1981;67(4):517–522. [PubMed] [Google Scholar]
- 5.Norcliffe-Kaufmann L, Axelrod F, Kaufmann H. Afferent baroreflex failure in familial dysautonomia. Neurology. 2010;75(21):1904–1911. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Norcliffe-Kaufmann L, Kaufmann H. Familial dysautonomia (Riley-Day syndrome): When baroreceptor feedback fails. Auton Neurosci Basic Clin. 2012;172(1–2):26–30. [DOI] [PubMed] [Google Scholar]
- 7.Macefield VG, Norcliffe-Kaufmann L, Axelrod FB, Kaufmann H. Cardiac-locked bursts of muscle sympathetic nerve activity are absent in familial dysautonomia. J Physiol. 2013;591(3):689–700. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Macefield VG, Norcliffe-Kaufmann L, Gutiérrez J, Axelrod FB, Kaufmann H. Can loss of muscle spindle afferents explain the ataxic gait in Riley-Day syndrome? Brain. 2011;134(11):3198–3208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Josaitis CA, Matisoff M. Familial dysautonomia in review: Diagnosis and treatment of ocular manifestations. In: Advances in Experimental Medicine and Biology. 2002;506:71–80. [DOI] [PubMed] [Google Scholar]
- 10.Mendoza-Santiesteban CE, Hedges TR, Norcliffe-Kaufmann L, et al. Clinical neuro-ophthalmic findings in familial dysautonomia. J Neuro-Ophthalmology. 2012;32(1):23–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Macefield VG, Norcliffe-Kaufmann LJ, Axelrod FB, Kaufmann H. Relationship between proprioception at the knee joint and gait ataxia in HSAN III. Mov Disord. 2013;28(6):823–827. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Frohman EM, Fujimoto JG, Frohman TC, Calabresi PA, Cutter G, Balcer LJ. Optical coherence tomography: A window into the mechanisms of multiple sclerosis. Nat Clin Pract Neurol. 2008;4(12):664–675. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Huang D, Swanson EA, Lin CP, et al. Optical coherence tomography. Science 1991;254(5035):1178–1181. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Leung CKS. Diagnosing glaucoma progression with optical coherence tomography. Curr Opin Ophthalmol. 2014;25(2):104–111. [DOI] [PubMed] [Google Scholar]
- 15.Dong ZM, Wollstein G, Schuman JS. Clinical utility of optical coherence tomography in glaucoma. Investig Ophthalmol Vis Sci. 2016;57:556–567. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Sung KR, Sun JH, Na JH, Lee JY, Lee Y. Progression detection capability of macular thickness in advanced glaucomatous eyes. Ophthalmology. 2012;119(2):308–313. [DOI] [PubMed] [Google Scholar]
- 17.Lavinsky F, Wu M, Schuman JS, et al. Can Macula and Optic Nerve Head Parameters Detect Glaucoma Progression in Eyes with Advanced Circumpapillary Retinal Nerve Fiber Layer Damage? Ophthalmology. 2018;125(12):1907–1912. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Jennrich RI, Schluchter MD. Unbalanced Repeated-Measures Models with Structured Covariance Matrices. Biometrics. 1986;42(4):805–820. [PubMed] [Google Scholar]
- 19.Schluchter MD. Analysis of incomplete multivariate data using linear models with structured covariance matrices. Stat Med. 1988;7:317–324. [DOI] [PubMed] [Google Scholar]
- 20.Elía N, Pueyo V, Altemir I, Oros D, Pablo LE. Normal reference ranges of optical coherence tomography parameters in childhood. Br J Ophthalmol. 2012:96(5):665–670. [DOI] [PubMed] [Google Scholar]
- 21.Celebi ARC, Mirza GE. Age-related change in retinal nerve fiber layer thickness measured with spectral domain optical coherence tomography. Investig Ophthalmol Vis Sci. 2013;54:8095–8103. [DOI] [PubMed] [Google Scholar]
- 22.Mwanza JC, Budenz DL, Warren JL, et al. Retinal nerve fibre layer thickness floor and corresponding functional loss in glaucoma. Br J Ophthalmol. 2015;99(6):732–737 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.RILEY CM, DAY RL. Central autonomic dysfunction with defective lacrimation; report of five cases. Pediatrics. 1949;3(4):468–478. [PubMed] [Google Scholar]
- 24.LIEBMAN SD. Ocular Manifestations of Riley-Day Syndrome. AMA Arch Ophthalmol. 1956;56(5):719. [DOI] [PubMed] [Google Scholar]
- 25.Kroop IG. The production of tears in familial dysautonomia. Preliminary report. J Pediatr. 1956;48(3):328–329. [DOI] [PubMed] [Google Scholar]
- 26.Mwanza JC, Kim HY, Budenz DL, et al. Residual and dynamic range of retinal nerve fiber layer thickness in glaucoma: Comparison of three OCT platforms. Investig Ophthalmol Vis Sci. 2015;56(11):6344–6351. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Wu H, De Boer JF, Chen TC. Reproducibility of retinal nerve fiber layer thickness measurements using spectral domain optical coherence tomography. J Glaucoma. 2011:20(8):470–476. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Tewarie P, Balk L, Costello F, et al. The OSCAR-IB consensus criteria for retinal OCT quality assessment. PLoS One. 2012;7(4):e34823. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Mendoza-Santiesteban CE, Palma JA, Hedges TR, et al. Pathological confirmation of optic neuropathy in familial dysautonomia. J Neuropathol Exp Neurol. 2017;76(3):238–244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Sadun AA, Win PH, Ross-Cisneros FN, et al. Leber’s hereditary optic neuropathy differentially affects smaller axons in the optic nerve. In: Transactions of the American Ophthalmological Society. 2000;98:223–235. [PMC free article] [PubMed] [Google Scholar]
- 31.Yu-Wai-Man P, Griffiths PG, Chinnery PF. Mitochondrial optic neuropathies - Disease mechanisms and therapeutic strategies. Prog Retin Eye Res. 2011;30(2):81–114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Fraser JA, Biousse V, Newman NJ. The neuro-ophthalmology of mitochondrial disease. Surv Ophthalmol. 2010;55(4):299–334. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Jonas JB, Gusek GC, Naumann GOH. Optic disc, cup and neuroretinal rim size, configuration and correlations in normal eyes. Investig Ophthalmol Vis Sci. 1988;29(7):1151–1158. [PubMed] [Google Scholar]
- 34.Borchert M, Garcia-Filion P. The syndrome of optic nerve hypoplasia. Curr Neurol Neurosci Rep. 2008;8(5):395–403. [DOI] [PubMed] [Google Scholar]