Retinal Disease Expression in Bardet-Biedl Syndrome-1 (BBS1) Is a Spectrum from Maculopathy to Retina-Wide Degeneration

Abstract

purpose. To define the retinal phenotype in patients with the Bardet-Biedl syndrome and mutations in the BBS1 gene.

methods. Ten patients (age range, 16–48 years), representing eight pedigrees, with BBS1 gene mutations were studied clinically and with kinetic perimetry, chromatic static perimetry, electroretinography (ERG), and optical coherence tomography.

results. Of the 10 patients, 8 were M390R homozygotes and 2 were compound heterozygotes with one allele also M390R. A spectrum of retinal disease expression was present. The mildest disease was a subtle maculopathy with relatively limited peripheral retinal dysfunction. Moderate disease showed retina-wide rod > cone dysfunction, and often there was a negative ERG waveform. More severe disease expression had different patterns: either loss of central function but retained abnormal peripheral function or a retained small central island of impaired function only. Moderate and severe disease showed loss of retinal and photoreceptor layer thickness across wide expanses of retina. Severity differed in family members and was independent of age. In addition, severity was not explained by genotype at a recently reported BBS epistatic gene, MGC1203.

conclusions. The cardinal feature of retinal degeneration in BBS1 can show a wide spectrum of disease expression.

The Bardet-Biedl Syndrome (BBS) is an autosomal recessive disorder characterized by retinal degeneration accompanied variably by systemic manifestations such as obesity, polydactyly, renal and cardiac abnormalities, hypogonadism, and cognitive impairment. ^{1 2} To date, 11 BBS genes have been identified. ^{3 4 5 6 7} The most common mutation causing BBS is in the BBS1 gene, which resides on chromosome 11q13. ⁸ BBS1 is composed of 17 exons and spans 23 kb. Although numerous BBS1 mutations have been identified, approximately 80% of patients with BBS1 have the M390R mutation. ⁹ Experimental studies have demonstrated that BBS1 expression is strictly limited to ciliated cells, including photoreceptors which are the primary ciliated cells in the retina. ^{3 10} Bbs2-, Bbs4-, and Bbs6-null mice have shown a retinal degeneration phenotype, ^{11 12 13} but there are currently no published data on the retinal disease expression in Bbs1-null mice. ¹⁴ A very recent report, however, shows retinal degeneration in a mouse knock-in model of Bbs1 M390R (Philp AR et al. IOVS 2006;47:ARVO E-Abstract 5777). 

The rapid increase in known BBS genotypes has not yet stimulated a parallel increase in detailed studies of BBS phenotype. The ocular phenotypes in BBS2, BBS3, and BBS4 were recently studied, and the retinal disease in these BBS genotypes was reported to be early and severe. ¹⁵ Details of the retinal disease expression have not been reported in patients with the BBS1 M390R mutation, the most common allele of the most common form of BBS. Progress in understanding the mechanism of BBS and the promise of future treatment options for retinal degenerations prompted us to study the human BBS1 retinopathy with the ultimate goal of determining the candidacy of this genotype for therapeutic intervention. 

Methods

Subjects

Patients with the Bardet-Biedl syndrome (n = 10, Table 1 ), representing eight families, underwent clinical ocular examinations and visual function studies. Nonocular abnormalities were determined by history and/or available medical records; these cardinal features were not specifically assessed in the patients. All subjects gave informed consent in compliance with the Declaration of Helsinki, and institutional review board approval was obtained. 

DNA Sequencing

Genotyping of patients for both the BBS1 gene and the MGC1203 gene was performed by amplifying patient genomic DNA with PCR and purifying the amplified band from a 2.0% agarose gel. ^{8 9} The purified PCR products were sequenced by using dye-terminator chemistry (3730XL DNA Analyzer; Applied Biosystems, Inc., Foster City, CA). 

Visual Function Studies

Psychophysics.

Kinetic perimetry was performed with two targets (V-4e, I-4e). Dark- and light-adapted static threshold chromatic perimetry was performed with a modified automated perimeter. Techniques, methods of data analysis, and normal results have been described. ^{17 18 19 20 21}  

Electroretinography.

Full-field electroretinograms (ERGs) were performed with a standard protocol. Details of the methods and normal data have been published. ^{18 19 21}  

Optical Coherence Tomography.

Cross-sectional images of the central retina were obtained with commercially available optical coherence tomography (OCT) instruments (OCT1 and OCT3; Carl Zeiss Meditec, Dublin, CA) using described methods. ^{22 23 24 25 26 27} Scans were 4.5 mm long. Each scan, formed by a series of longitudinal reflectivity profiles (LRPs), was analyzed with custom-developed computer programs. Measurements were based on the distance between previously defined features on averaged LRPs. ^{21 22 23 24 25 26 27} Retinal thickness was defined as the distance between the signal transition at the vitreoretinal interface (labeled T1 in Ref. ²² ) and the major signal peak corresponding to the RPE. ²⁵ In normal subjects, the RPE peak was assumed to be the deepest peak within the deep retinal multipeaked scattering signal complex (labeled ORCC [outer retinal–choroidal complex] in Ref. ²² ). In patients, the multipeaked ORCC was identifiable or there was a single deep peak in the retina, which was presumed to be the RPE or Bruch’s membrane based on previous work in retinal degenerations. ²² Spatial maps of retinal thickness were derived from groups of raster scans, and abnormal thinning was determined as published. ^{25 26 27} Outer nuclear layer (ONL) thickness was defined as the major intraretinal signal trough delimited by the signal slope maxima and measured as previously reported. ²⁷  

Results

Variation in Severity of Retinal Disease Expression

Presenting visual symptoms of the patients varied widely (Table 1 ; mean age at first visit, 28.7 years; age range, 16–48 years), as did the diagnoses given before our evaluation. Nine of 10 patients had not been diagnosed with BBS. Among the diagnoses at the time of referral were nonsyndromic retinitis pigmentosa (RP) or macular degeneration; other patients were asymptomatic and their referral was prompted by a routine eye examination in the patient or even in a relative with overt retinal disease. Visual acuity ranged from normal to severely impaired in the 10 patients at the earliest ages examined, and there was no clear relationship between age and acuity (Table 1) . No specific refractive error was associated with these patients, as opposed to reports in other genotypes. ¹⁵ No correlation between age and kinetic perimetry results was observed in this cohort. Kinetic visual fields from six patients with BBS1, ordered by increasing severity and not by age (Figs. 1A 1B 1C 1D 1E 1F) , illustrate that patients of widely different ages could have a similarly mild disease expression, whereas those of similar ages could have expression that differs dramatically in severity. For example, P3, at age 17, and P7, at age 44, both had normal kinetic fields (Figs. 1A 1B ; Table 1 ). P10, at age 48, had a full field with the larger target but generalized limitation in response to the smaller target (Fig. 1C ; Table 1 ). P1, P5, and P4 were all between ages 16 and 21 but showed various degrees of peripheral field loss (Figs 1D 1E 1F ; Table 1 ) and contrasted with the normal function of P3 who was in the same age group. Serial kinetic perimetry results in two further patients illustrated progressive central field loss. P2 showed loss of central field over an interval of 4 years between ages 17 and 21 (Fig. 1G) . P9 lost the island of central function between ages 37 and 45 (Fig. 1H) . 

Dark-adapted static perimetric profiles across the central 60° of visual field are shown for the less severely affected patients with BBS1 (Fig. 1I) , for patients from the same family (Fig. 1J) , and for a severe phenotype (Fig. 1K) . Rod-mediated sensitivity could be normal in the central field, as in P3. Peripheral rod function in P3 was measurable throughout; 19% of the loci (mid- and far-peripheral field) were abnormal by 1 log unit, on average. P5 and P6 showed similar rod-mediated abnormalities in the central profiles; sensitivities were slightly reduced below normal in the central 30° of field, but there was diminished sensitivity with eccentricity. P5 showed measurable rod-mediated sensitivity in islands of the far periphery; the 24% of loci with detectable function were reduced by, on average, 3.5 log units. P6 had no measurable function in the peripheral field. P1 had more paracentral depression of sensitivity than did P5 or P6 but greater sensitivity at 20° to 30° from fixation. Peripheral rod-mediated function was measurable at 94% of loci, but was reduced on average by 2.7 log units. 

Within families, there was major variation in severity, and this was independent of age (Fig. 1J) . For example, P7 had central dark-adapted sensitivities that were within normal limits at age 40, whereas her 44-year-old sibling, P8, showed only scattered islands of severely impaired sensitivity (Fig. 1J , family 1). Peripheral rod-mediated function in P7 was measurable and most loci fell within normal limits; approximately 20% of loci were reduced by, on average, 0.8 log units. P8 had no measurable peripheral function. In another family, a 48-year old patient (P10) had central sensitivities reduced by less than a log unit whereas a 37-year-old cousin (P9) retained only small islands of severely reduced function (Fig. 1J , family 2). P10 had measurable peripheral rod-mediated function; 90% of loci were abnormally reduced by, on average, 2 log units. P9 had approximately 15% of loci (peripheral nasal field loci) with measurable rod function, and these were reduced by ≥3.5 log units. Illustrating an even more extreme loss of function are the data of P4; there is no rod function and only a small central island of reduced cone function; no peripheral function was measurable (Fig. 1K) . No clear relationship between genotype and severity of phenotype was thus discernible in BBS1-related disease. 

Negative ERG Waveform at Certain Disease Stages

All eight patients studied with full-field ERGs had abnormalities. Figure 2Aillustrates how the four detectable ERGs compared with normal waveforms. The ERGs are ordered by severity from reduced to not detectable; there was no relationship between age and severity. Of note, three of the four patients with detectable waveforms showed b-/a-wave ratios to the maximum white flash (normal mean, 1.6; SD, 0.2; n = 96, age range, 9–50 years) that were abnormally reduced (P7, 0.8; P10, 1.1; P5, 0.7), indicating greater inner than outer retinal dysfunction. P3 had a normal b-/a-wave ratio of 1.8. 

Figure 2Bsummarizes the ERG parameters from the patients with BBS1. Among the amplitude parameters, only the cone ERG amplitudes of P7 were within normal limits (represented by rectangles, ±2 SD from the mean). Cone flicker timing was delayed in all patients but P3. Using standard rod b-wave and cone flicker ERG amplitude measurements to compare the degree of retina-wide rod and cone dysfunction, we found approximately equal losses in the higher-amplitude signals of P3 and P7; greater rod than cone dysfunction was present in patients with lower signals, which mainly showed retained measurable flicker (P10, P5, and P2). 

Central Retinal Micropathology of BBS1

Retinal imaging studies in the patients with BBS1 revealed a spectrum of abnormalities from relatively subtle macular changes to the severe retinal degeneration that has been reported in patients with BBS. ^{15 16} P3 at age 17 showed a focal foveal lesion that was vitelliform-like, ²⁸ but limited in extent. The ophthalmoscopic appearance was identical with that which we previously illustrated in P7, when her genotype was unknown (Fig. 6 in Ref. ¹⁶ ). By ophthalmoscopy, P1 had a granular appearance to the macula. P9 showed retinal thinning with a maculopathy characterized by areas of depigmentation near and around the fovea. 

Cross-sectional retinal images by OCT in these patients increased understanding of the ophthalmoscopic abnormalities. P3 has normal foveal thickness but a focal lesion that disturbs the deep layer we have termed the ORCC or outer retinal–choroidal complex ^{22 25 26 27 29 30 31} (Fig. 3B) . The depth scan in P1 indicates a wider area of disturbance with thinning and specific loss of the more vitreal component of the ORCC (Fig. 3C) . More extreme thinning is evident in the cross-sectional image of P9 (Fig. 3D) . Longitudinal reflectivity profiles (LRPs) through the fovea further illustrate differences between patients with BBS1 and normal subjects (Fig. 3E) . P3 shows normal total retinal thickness at the fovea (179 μm; ±2 SD from the mean = 172–245 μm; n = 25) and a photoreceptor nuclear layer (highlighted) that is near the normal lower limit (65 μm; ±2 SD from the normal mean = 66–130 μm; n = 10). The abnormality in the LRP of P3 was the multipeaked backscattering signal deep to the photoreceptor nuclear layer, presumably representing disruption of photoreceptor inner and outer segment (IS/OS) layers and involving the RPE. P1 had reduced foveal retinal thickness (126 μm) and a reduced photoreceptor layer (52 μm). There was no multipeaked backscattering signal as in P3; there was also reduced complexity of ORCC peaks, which has been associated with loss of photoreceptor IS/OS. ^{22 25 26 27} P9 had a thinned fovea (52 μm), with a single-peaked ORCC and little or no measurable photoreceptor layer (≤15 μm). 

Topographical maps of retinal thickness over a wide central retinal area are shown in three patients with BBS1 and in a normal subject (Figs. 4A 4B 4C 4D) . The normal retina showed a central depression at the foveal pit, a surrounding ring of increased thickness attributable to displaced inner retinal layers from foveal development, and a decline in thickness with eccentricity. The crescent of thickening extending superior and inferior from the optic nerve represents the converging axons from ganglion cells. All three patients had thinning at and around the fovea and various degrees of paracentral thinning (Figs. 4B 4C 4D , insets). There was retained thickness near the optic nerve, and it appeared no different from normal. Quantitation of retinal thickness and ONL thickness were performed in two younger patients (P1, P3) who were asymptomatic but had definite retinal disease expression when examined in detail. P3, who had shown the unusual foveal backscattering feature (Fig. 3E) , had retinal thickness that was at the lower limit of normal across 9 mm of temporal retina and 2 mm of nasal retina (Fig. 4E) . ONL thickness, however, was abnormally reduced across most of the central retina and only approached normal limits at approximately 8 mm in the temporal retina. P1 had definite retinal thinning in and around the fovea but was within or at the lower limit of normal at 4 mm eccentricity in the temporal retina. Across the entire central retina, there was thinning of the ONL. These data are in contrast to the severe expression in P9 whose retinal thickness and ONL were abnormal spanning the entire measured region. 

For comparison with retinal reflectivity waveforms at the foveal locus (Fig. 3E) , LRPs are shown for the same patients (P3, P1, P9) at a temporal retinal locus 7.8 mm from the fovea (Fig. 4H) . P3 had a near-normal LRP, with the exception of ONL thinning; inner lamination was normal in appearance and the IS/OS components of the ORCC were intact. P1, in contrast, had far greater ONL thinning and a single-peaked ORCC, suggesting loss of IS/OS. Inner retinal waveform features between the ONL and inner plexiform layer (IPL) were less discrete than normal. The LRP waveform in P9 at this locus retained a single-peaked ORCC and inner retinal hyperreflectivity with intervening hyporeflectivity. Normal lamination is not discernible at this disease stage. To confirm and extend the observation of normal-appearing thickness near the optic nerve head in the three maps of retinal topography in some patients with BBS1 (Figs. 4A 4B 4C 4D) , an analysis of nerve fiber layer (NFL) thickness was performed in an annular region surrounding the optic nerve. A band of high reflectivity at the retinal surface has been related to histologic measurements of the NFL. ²⁵ In six patients with BBS1 (P1, P3, P4, P5, P9, and P10), NFL thickness was normal when measured at a ring of 3.4 mm diameter around the optic nerve (Fig. 4I) . 

Putative BBS Epistatic Gene and Phenotypic Variability in BBS1-Associated Retinal Disease

A recent report has suggested that a variant allele (C430T) of the MGC1203 gene which encodes a pericentriolar protein results in a more severe retinal phenotype in BBS patients who are heterozygous for the 430T allele. ³² To determine whether the variation in retinal phenotype observed in the patients in the present study can be explained by an epistatic effect from the MGC1203 gene, we genotyped by direct sequencing the MGC1203 gene in all 10 patients. The genotype results showed that all 10 patients were homozygous for the common 430C allele. These results indicate that the variation in phenotype in the patients with BBS1 in our study is not explained by genotype at the MGC1203 locus. 

Discussion

Phenotypic differences in retinal disease among patients with BBS of unknown genotype have been documented (for example, Refs. ¹⁶ , ^{33 34 35 36 37 38} ), but the phenotypes of the new molecular subtypes are not understood in detail. Hypotheses about the effect of complex inheritance on disease expression, ^{39 40} for example, may be better tested given some increased understanding of phenotype. In the present study, it is notable that many patients were not initially diagnosed as having BBS and may not have entered BBS genotype analyses. Some subjects were asymptomatic and, if screened for BBS genes, could have been classified as having mutations without evidence of eye disease; yet, they all showed definite retinal abnormalities. Certainly in BBS1, there is a spectrum of retinopathy and the relatively mild expression in some patients with BBS1, even into the fifth decade of life, contrasts with the early and severe ocular disease reported in other genotypes. ^{15 41} That patients with BBS1 are not always legally blind by their teens can be taken as an encouraging counseling note for their families. 

Negative ERGs are a feature of many retinopathies, ^{42 43} but, to our knowledge, they have not been described in BBS. We hypothesize from the present data that a negative waveform is a disease stage rather than a primary disease feature. The phenomenon may not have been noted previously because only patients at earlier stages (without negative waveforms) or later stages (with no detectable rod waveforms and only reduced cone signals) were examined. Our data suggest that the retinal disease initially affects rod and cone photoreceptor function but that there is a subsequent effect on inner retinal function, mainly evident in the rod retinal pathway. Remodeling has been found in the retina of many animal models of retinal degeneration, and there are definite morphologic effects in the inner retina (involving rod and cone bipolar cells) and some evidence of physiological correlates. ^{44 45 46} A proposed sequence from reduced photoreceptor and bipolar ERG components to a negative waveform in our patients with BBS1 may represent one of the few examples in the literature of remodeling of the human retina. Of additional interest is a recent report in which serum antibodies to BBS1 were found in patients with metastatic melanoma, ⁴⁷ which can be associated with a paraneoplastic autoimmune retinopathy characterized by a negative ERG. ⁴⁸ Whether mutant BBS1 protein can provoke retinopathy (or a stage thereof) with a negative ERG is, in theory, testable by experiment. It is also possibly relevant that b-wave but not a-wave abnormalities have been detected in obligate carriers of BBS. ⁴⁹  

Because of the variability of retinal phenotype in patients with BBS1, we genotyped the 10 patients examined in this study at the MGC1203 locus. A rare variant at this locus has been reported to result in severe retinal disease. ³² All 10 patients had the common 430C allele, despite the fact that some of them had a severe retinal phenotype. Although it is possible that the rare 430T allele contributes to a severe phenotype in some cases, the rarity of this allele and the fact that patients homozygous for the common 430C allele can have mild to severe retinal phenotypes, indicates that MGC1203 genotype is not a general explanation for the variation in retinal phenotype in patients with BBS. 

Two findings in our present BBS1 studies have immediate clinical relevance. First, there was the identification of a macular disease phenotype in many patients; in some patients with mild disease expression, maculopathy was the presenting or most clinically detectable finding. Questions about systemic features of BBS should be asked of patients with modest visual acuity reductions and OCT abnormalities similar to those we have illustrated. At this stage of our understanding, the OCT findings are nonspecific, but with higher-resolution scanning in the future, more details of the localization of early defects may be forthcoming. The main clinical value of recognition of a subtle ocular phenotype in BBS1 remains in the diagnosis of BBS and the prevention of undetected systemic disease in such patients. 

Second, unlike some developmental retinal abnormalities, ²⁵ there was definable lamination and no extreme retinal thickening. Nerve fiber layer estimates with OCT were normal even at stages of extreme retinal dysfunction and retinal thinning. This bodes well for future consideration of treatment. 

 Supported by the National Institutes of Health, the Macula Vision Research Foundation, Foundation Fighting Blindness, Inc., The Macular Disease Foundation, The Ruth and Milton Steinbach Fund, Alcon Research Institute, The Paul and Evanina Bell Mackall Foundation Trust, and the F. M. Kirby Foundation.

 Submitted for publication May 9, 2006; revised June 13, 2006; accepted September 6, 2006.

 Disclosure: A.A. Azari, None; T.S. Aleman, None; A.V. Cideciyan, None; S.B. Schwartz, None; E.A.M. Windsor, None; A. Sumaroka, None; A.Y. Cheung, None; J.D. Steinberg, None; A.J. Roman, None; E.M. Stone, None; V.C. Sheffield, None; S. G. Jacobson, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Clinical Characteristics of the BBS1 Patients

Patient	Age at First Visit (y)/Gender	Presenting Symptoms/Diagnosis	BBS1 Mutations	Nonocular Abnormalities*	Visual Acuity (RE-LE), †	Refraction, ‡	Kinetic Visual Field Extent (V-4e/I-4e), §
P1	16/F	None/rule out RP	M390R	Polydactyly, renal cardiac, cognitive	20/30	−1.50	94/ND
			M390R
P2	16/F, ∥	VA loss/BBS	M390R	Polydactyly, obesity	20/100	+1.50	53/ND
			E549X
P3	17/M	None/rule out Stargardt disease	M390R	Polydactyly, cardiac, hepatic	20/20	−1.50	99/79
					M390R
P4	19/F	VF loss/RP	M390R	Polydactyly	20/40–20/50	−1.75	<1/ND
			L518P
P5	20/M	VA loss/maculopathy	M390R	Epilepsy, cognitive	20/60–20/100	−4.00	61/29
			M390R
P6	30/M	VA, VF, NV loss/RP	M390R	Polydactyly	20/100	−2.00	<1/ND
			M390R
P7, ¶	40/F	None/presbyopia	M390R	Obesity	20/25–20/30	−0.50	95/91
			M390R
P8, ¶	44/M	VA, VF, NV loss/RP	M390R	Polydactyly, obesity, renal	20/400–20/800	−3.5	<1/ND
			M390R
P9, #	37/F	VA loss/RP	M390R	Polydactyly, renal	20/200	+0.75	43/ND
			M390R
P10, #	48/M	None/presbyopia	M390R	Polydactyly, cognitive	20/25–20/30	+1.75	87/54
			M390R

 Some patient information was provided previously for P2, P4, P6, P7, and P8 (Jacobson et al. ¹⁶ ). RP, retinitis pigmentosa; VA, visual acuity; VF, visual field; NV, night vision; ND, not detectable.

*  By history and/or available medical records.

†  Best corrected visual acuity; similar in the two eyes unless specified.

‡  Spherical equivalent; average of both eyes.

§  Average of both eyes; expressed as a percentage of the normal mean for the V-4e or I-4e target; 2 SD below normal, or 90%.

∥  Patient examined multiple times from ages 15 to 17; data shown are those obtained at times of greatest cooperation.

¶  From the same pedigree.

#  From the same pedigree.

Figure 1.

 

Perimetry in patients with BBS1 indicated a spectrum of severity that was unrelated to age. (A–H) Kinetic perimetry in patients from six pedigrees showed different degrees of severity: mild (A–C), moderate (D, E, G), and severe (F, H). Serial results for two patients (P2, P9) are shown (G, H). In P2, serial results span a 4-year interval and in P9, an 8-year interval. (I–K) Dark-adapted sensitivity profiles across the horizontal meridian measured with (I, J) 500- and (K) 650-nm stimuli in nine patients with BBS1. All data are depicted as from the right eye. Shaded areas: limits (mean ± 2 SD) of normal sensitivity obtained under dark-adapted conditions with the (I, J) 500- or the (K) 650-nm stimulus at the cone plateau. Hatched area: location of the physiological blind spot. N, nasal field; T, temporal field; DA, dark-adapted.

Figure 2.

 

Abnormal electroretinograms with negative waveforms in BBS1. (A) Rod b-wave, mixed (maximum white) a- and b-wave, and cone ERG amplitudes can be abnormal in the patients with BBS1. Among the four patients with detectable responses, three (P7, P10, and P5) showed greater b-wave than a-wave amplitude reduction in the mixed response. (B) ERG results from one eye of eight patients with BBS1 compared with normal limits (rectangle: ±2 SD from the mean). ND, not detectable.

Figure 3.

 

Spectrum of macular disease in BBS1. (A) Normal OCT centered on the fovea and across 1.9 mm of the central retina. (B, C) P3 showed an abnormality in the ORCC at the fovea, whereas there was thinning of the ORCC across the scan length in P1. (D) Image in P9 illustrates extreme central retinal thickening; the scan was taken in the vertical meridian. (E) Longitudinal reflectivity profiles (LRPs) at the anatomic fovea in a normal subject and the three patients with BBS1. Highlighted in the LRP is the cone photoreceptor nuclear layer, indicating diminished depth of this layer and retinal thickness in the patients. Of note is multipeaked backscattering signal in the deep retina of P3.

Figure 4.

 

Regional variation in retinal thickness in BBS1. (A–D) Topographical maps from depth images of the central retina in a normal subject and three patients with BBS1. Map insets: difference maps showing regions of significant thinning (gray) compared with normal. (E–G) Retinal and ONL thickness across the horizontal meridian in three patients compared with normal (gray band, ±2 SD from the mean). (H) Reflectivity waveforms at a locus 9 mm into the temporal retina, comparing laminar waveform components of the same three patients. Green-highlighted region: presumed location of the ONL. (I). Polar plot of NFL thickness at a circle (diameter, 3.4 mm) around the optic nerve. Black traces: patient data; gray band: normal data (±2 SD from the mean). S, superior; I, inferior; T, temporal; N, nasal.

The authors thank Elaine Smilko, Michelle Doobrajh, Alexandra Windsor, Margorzata Swider, and Marisa Roman for critical help with the studies.