Optical Gap Biomarker in Cone-Dominant Retinal Dystrophy


To characterize the progression of optical gaps and expand the known etiologies of this phenotype.


Retrospective cohort study.


Thirty-six patients were selected based on the identification of an optical gap on spectral-domain optical coherence tomography (OCT) from a large cohort of patients (N = 746) with confirmed diagnoses of inherited retinal dystrophy. The width and height of the gaps in 70 eyes of 36 patients were measured by 2 independent graders using the caliper tool on Heidelberg Explorer. Measurements of outer and central retinal thickness were also evaluated and correlated with gap dimensions.


Longitudinal analysis confirmed the progressive nature of optical gaps in patients with Stargardt disease, achromatopsia, occult macular dystrophy, and cone dystrophies ( P < .003). Larger changes in gap width were noted in patients with Stargardt disease (78.1 μm/year) and cone dystrophies (31.9 μm/year) compared with patients with achromatopsia (16.2 μm/year) and occult macular dystrophy (15.4 μm/year). Gap height decreased in patients with Stargardt disease (6.5 μm/year; P = .02) but increased in patients with achromatopsia (3.3 μm/year) and occult macular dystrophy (1.2 μm/year). Gap height correlated with measurements of central retinal thickness at the fovea ( r = 0.782, P = .00012). Interocular discordance of the gap was observed in 7 patients. Finally, a review of all currently described etiologies of optical gap was summarized.


The optical gap is a progressive phenotype seen in an increasing number of etiologies. This progressive nature suggests a use as a biomarker in the understanding of disease progression. Interocular discordance of the phenotype may be a feature of Stargardt disease and cone dystrophies.


  • The causes of the optical gap phenotype are continually increasing and need to be characterized.

  • The rate of optical gap widening is progressive and differs across etiologies of the phenotype.

  • Changes in optical gap size do not necessarily correlate with changes in visual acuity.

  • Changes in optical gap height correlate well with changes in central foveal retinal thickness.

  • Interocular discordance is a feature of optical gap caused by Stargardt disease or cone dystrophy.

Optical gap is an occult macular phenotype that is most discernible on spectral-domain optical coherence tomography (OCT) in a number of inherited retinal dystrophies. It is characterized by a focal loss of the photoreceptor-attributable ellipsoid zone (EZ) band, previously known as the inner segment (IS) and outer segment (OS) junction, in the fovea and parafoveal region. This phenotype was first described in 2006 by Barthelemes and asosciates in patients with achromatopsia. Soon afterward, Leng and associates expanded the differential associated with optical gaps to include Stargardt disease and dominant cone dystrophies caused by mutations in cyclase proteins. Since then, individual case reports detailing novel etiologies of both hereditary and nonhereditary optical gap have been slowly growing in the literature.

The pathogenesis of this finding has previously been a subject of interest in several etiologies of the phenotype. Nõupuu and associates developed a 3-part staging system using cross-sectional data to suggest progression of the phenotype and noted longitudinal progression from one stage to another in several patients with Stargardt disease. Greenberg and associates similarly created a 5-step staging system in patients with achromatopsia, but that study was also limited by the use of cross sectional data. The clinical significance of the optical gap has been suggested as a potential outcome measure in clinical trials for achromatopsia. Longitudinal data of phenotypic progression in individual patients has been attempted in the pediatric achromatopsia population. However, it has otherwise not been well characterized and has implications to not only improve our understanding of the optical gap but also determine the necessary length of clinical trials that may use optical gap as an outcome measure.

The present study describes the analysis of 36 genetically heterogeneous patients with the optical gap phenotype, including 2 patients harboring mutations in novel candidate genes associated with this phenotype, RAB28 and PITPNM3 . Moreover, this study details the longitudinal analysis of 19 individual patients to better understand phenotype progression and natural history over time. The authors additionally perform a review of the current literature of this phenotype. This information provides an updated differential diagnosis of optical gap since it was first described in 2006 to better guide clinicians who may encounter this phenotype in practice.


Subject Selection

A retrospective review was performed of 746 patients with both a clinical and molecular genetic diagnosis of inherited retinal dystrophy who were seen and evaluated at the Edward S. Harkness Eye Institute at Columbia University Irving Medical Center between 2009 and 2019. Retinal images from previous visits were evaluated for the presence of an identifiable loss of reflectance and disruption of the EZ line in the fovea or the parafoveal regions on spectral-domain OCT. Patients with disruptions of the outer retinal layers secondary to macular telangiectasias, vitelliform macular dystrophies, vitreomacular traction, secondary macular neovascular disease, and macular holes were excluded. A total of 36 patients were identified who fit the inclusion criteria. This study offered minimal risk to the patients, and because of its retrospective design, patient consent was waived as described in Columbia University Irving Medical Center Institutional Review Board–approved protocol AAAR8743. All procedures were reviewed and deemed to be in accordance with the tenets of the Declaration of Helsinki.

Ophthalmic Examination and Imaging

Patients underwent initial ophthalmic examination including measurement of best-corrected visual acuity (BCVA), dilation with topical tropicamide (1%) and phenylephrine (2.5%), and fundus examination by a retinal specialist (S.H.T). In addition, patients underwent multimodal imaging including spectral-domain OCT, short wavelength autofluorescence (SW-AF), and wide field color fundus photography. Full-field electroretinograms (ffERGs) were obtained from patients using Dawson, Trick, and Litzkow electrodes and Ganzfeld stimulation using a Diagnosys Espion Electrophysiology System (Diagnosys LLC, Littleton, Massachusetts, USA) according to International Society for Clinical Electrophysiology of Vision standards. Spectral-domain OCT and SW-AF were acquired using a Spectralis HRA+OCT (Heidelberg Engineering, Heidelberg, Germany). Color fundus photography was obtained using an Optos 200 Tx (Optos, PLC, Dunfermline, United Kingdom).

Optical Gap Progression and Statistical Analysis

Disease progression was assessed between the initial and follow-up visits using the change in the horizontal (nasal-temporal axis) and vertical (anterior-posterior axis) lengths of the optical gap on corresponding fovea-aligned spectral-domain OCT scans as measured by 2 independent graders (J.O., J.R.). Optical gap width was defined as the longest contiguous length of the disruption in the EZ band with the presence of a vacant space in its place ( Supplemental Figure 1, A through D , green lines; Supplemental Material at AJO.com ). In cases where residual EZ was observed at the fovea but a distinct optical gap was identified on either side of the residual EZ, the residual EZ was included within the measurements. Optical gap height was defined as the distance between the external limiting membrane and the retinal pigment epithelium (RPE) and Bruch membrane complex at the fovea ( Supplemental Figure 1, A , yellow lines). In cases with multiple follow-up visits, measurements were taken between the initial and most recent follow-up visit in which an optical gap was seen in both eyes. In cases with only 1 follow-up visit in which one eye demonstrated the presence of a gap and the other eye had a gap that had become atrophic, measurements were taken from the single eye with the gap. Areas of peripheral collapse of the retina into the optical gap were not included as part of the optical gap measurements. Measurements were performed with the caliper tool on the Heidelberg Explorer (HEYEX). Intraclass correlation coefficients (ICCs) were calculated for measurements to determine the reliability of intergrader variability.

Measurements of central retinal thickness at the fovea (CRTF) and outer retinal thickness (ORT) across the retina were also evaluated between initial and follow-up visits. CRTF was defined as the distance between the RPE–Bruch membrane complex and the internal limiting membrane. ORT was defined as the distance between the RPE–Bruch membrane complex and the boundary between the outer nuclear and outer plexiform layers. Each scan was manually segmented on HEYEX and then exported. The thicknesses were calculated with MATLAB software (The Mathworks, Inc, Natick, Massachusetts, USA) as developed and described by Hood and associates.

When available, all measurements were taken using follow-up scans that used automatic real-time tracking to align with baseline images. When matched scans were unavailable for the initial and the most recent visits, measurements were taken at the fovea from both visits. A paired sample t test was performed to compare longitudinal measurements of optical gap dimensions and retinal thickness. Correlation of changes in optical gap dimensions with logarithm of minimal angle of resolution visual acuity and retinal thickness measurements were also performed. Analysis was performed using R statistical software (v 3.6.1; R Foundation for Statistical Computing, Vienna, Austria).

Review of the Literature

A review of the current literature was performed using the search terms “foveal cavitation,” “optical gap,” “disruption of the EZ line,” and “disruption of the IS/OS junction” on PubMed to identify other etiologies of optical gap and clarify the differential diagnosis of the phenotype. Review of all other known etiologies of cone and cone-rod retinal dystrophies was also performed. All reviews, original articles, case series, and case reports were included. Articles with evidence of OCT findings consistent with the aforementioned definition of foveal cavitation or optical gap were examined and included in the review.


Clinical Summary

The clinical, genetic, and demographic information of these patients are summarized in Table 1 . A total of 76 eyes from 38 patients were evaluated for optical gaps. The patients had a mean and median age of 34.6 and 27.5 years (range 11-77 years), respectively, at the time of the initial evaluation. Twelve patients (P1-P12) presented with a diagnosis of electrophysiologic group I Stargardt disease caused by mutations in ABCA4 . Twelve patients (P13-P24) presented with a diagnosis of achromatopsia—6 caused by mutations in CNGA3 , 2 in CNGB3 , 3 in AFT6 , and 1 in PDE6C . Five patients (P25-P29) presented with occult macular dystrophy caused by mutations in RP1L1 , and 3 patients (P30-P32) were diagnosed with cone dystrophy caused by guanylate cyclase mutations in the GUCY2D and GUCA1A genes. Two cases of pattern macular dystrophy caused by PRPH2 mutation were identified (P33 and P34). Two novel candidate etiologies of optical gap were also evaluated: cone dystrophy caused by 2 mutations in RAB28 (P35) and cone dystrophy caused by a single mutation in PITPNM3 (P36). Example spectral-domain OCT images of each etiology of optical gap are shown in Figure 1 .

Table 1

Clinical Characteristics of 36 Patients with Optical Gaps

Patient No. Age/Sex Diagnosis Gene Variant Age of Onset BCVA (OD, OS) Symmetry
P1 26/M Stargardt ABCA4 c.4139C>T:p.Pro1380Leu, c.5882G>A:p.Gly1961Glu 20 20/60, 20/60 Y
P2 26/F Stargardt ABCA4 c.1622T>C:p.Leu541Pro, c.5882G>A:p.Gly1961Glu 24 N/A Y
P3 a , b 26/F Stargardt ABCA4 c.1622T>C:p.Leu541Pro, c.5882G>A:p.Gly1961Glu 18 20/250, 20/250 N
P4 a , b 23/F Stargardt ABCA4 c.1622T>C:p.Leu541Pro, c.5882G>A:p.Gly1961Glu 15 20/80, 20/80 N
P5 b 25/F Stargardt ABCA4 c.286A>G:p.Asn96Asp, c.5882G>A:p.Gly1961Glu 24 20/40, 20/30 N
P6 b 27/F Stargardt ABCA4 c.5882G>A:p.Gly1961Glu, c.6448T>C:p.Cys2150Arg 15 20/150, 20/150 Y
P7 b 23/M Stargardt ABCA4 c.5882G>A:p.Gly1961Glu, c.5318C>T:p.Ala1773Val 22 20/40, 20/30 Y
P8 b 23/F Stargardt ABCA4 c.4139C>T:p.Pro1380Leu, c.5882G>A:p.Gly1961Glu 18 20/40, 20/30 Y
P9 23/M Stargardt ABCA4 c.1622T>C:p.Leu541Pro, c.5882G>A:p.Gly1961Glu 23 N/A N
P10 25/F Stargardt ABCA4 c.5882G>A:p.Gly1961Glu, c.5196+1056A>G 21 20/50, 20/100 Y
P11 b 13/M Stargardt ABCA4 c.2461T>A:p.Trp821Arg, c.6448T>C:p.Cys2150Arg 11 20/150, 20/200 Y
P12 26/M Stargardt ABCA4 c.3065A>G:p.Glu1022Gly, c.5882G>A:p.Gly1961Glu 26 20/30, 20/30 3
P13 c 30/M Achromatopsia CNGA3 c.1391T>G:p.Leu464Arg, c.1621C>A:p.Leu541Phe 25 20/125, 20/125 Y
P14 c 63/M Achromatopsia CNGA3 c.829C>T p.Arg277Cys, c.847C>T:p.Arg283Trp N/A 20/150, 20/150 Y
P15 45/F Achromatopsia CNGA3 c.1702G>A:p.Gly568Arg, c.1823T>A: p.Leu608Gln Childhood 20/100, 20/100 Y
P16 c 34/M Achromatopsia CNGA3 c.830G>A:p.Arg277His, c.1070A>G:p.Tyr357Cys N/A 20/100, 20/150 Y
P17 c 23/M Achromatopsia CNGA3 c.1391T>G:p.Leu464Arg, c.1641C>A:p.Phe547Leu 1.5 20/50, 20/50 Y
P18 47/M Achromatopsia CNGA3 c.1669G>A:p.Gly557Arg, c.667C>G:p.Arg223Gly 5 20/150, 20/125 Y
P19 c 32/M Achromatopsia CNGB3 c.1432C>T p.Arg478Ter, c.1432C>T p.Arg478Ter 23 20/160, 20/200 Y
P20 c 46/F Achromatopsia CNGB3 c.1056-3C>G het Childhood 20/80, 20/80 Y
P21 a , d 12/F Achromatopsia ATF6 c.970C>T:p.Arg324Cys, c.970C>T:p.Arg324Cys 6 20/200, 20/200 Y
P22 a , d 23/M Achromatopsia ATF6 c.970C>T:p.Arg324Cys, c.970C>T:p.Arg324Cys 18 20/63, 20/100 Y
P23 a , d 18/F Achromatopsia ATF6 c.970C>T:p.Arg324Cys, c.970C>T:p.Arg324Cys 12 20/100, 20/63 Y
P24 32/M Achromatopsia PDE6C c.1759T>C:p.Tyr587His, c.1759T>C:p.Tyr587His Childhood 20/100, 20/150 Y
P25 56/F Occult macular dystrophy RP1L1 c.133C>T:p.Arg45Trp, c.449C>T:p.Thr150Ile 51 20/100, 20/125 Y
P26 65/M Occult macular dystrophy RP1L1 c.133C>T:p.Arg45Trp 49 20/70, 20/60 Y
P27 a 25/M Occult macular dystrophy RP1L1 c.133C>T:p.Arg45Trp N/A N/A Y
P28 a 23/M Occult macular dystrophy RP1L1 c.133C>T:p.Arg45Trp 12 20/80, 20/80 Y
P29 43/F Occult macular dystrophy RP1L1 c.133C>T:p.Arg45Trp 33 20/60, 20/60 Y
P30 77/F Cone dystrophy GUCA1A c.526C>T:p.Leu176Phe 45 20/125, 20/200 N
P31 66/M Macular dystrophy GUCY2D c.2516C>T:Thr839Met 64 20/100, 20/80 Y
P32 24/F Cone dystrophy GUCY2D c.2513G>A:p.Arg838His 24 N/A N
P33 49/F Macular dystrophy PRPH2 c.424C<T:p.Arg142Trp Childhood 20/50, 20/50 Y
P34 56/F Macular dystrophy PRPH2 c.514C>T p.Arg172Trp 56 20/60, 20/60 Y
P35 41/M Cone dystrophy RAB28 c.136G>C:p.Gly46Arg, c.70G>C: p.Gly24Arg 42 20/70, 20/70 N
P36 51/F Cone dystrophy PITPNM3 c.227_229del:p.Gly76del 45 20/80, 20/CF N

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Aug 17, 2020 | Posted by in OPHTHALMOLOGY | Comments Off on Optical Gap Biomarker in Cone-Dominant Retinal Dystrophy

Full access? Get Clinical Tree

Get Clinical Tree app for offline access