Purpose
To describe the unique diagnostic fundus autofluorescence (FAF) patterns in patients with achromatopsia and the associated findings on optical coherence tomography (OCT).
Design
Observational case series.
Methods
We evaluated 10 patients with achromatopsia by means of best-corrected visual acuity (BCVA), ophthalmoscopy, Goldmann visual field, full-field electroretinography (ffERG), OCT, and FAF photography. FAF patterns were compared with patient age and foveal changes on OCT.
Results
Patients fell into two dichotomous age groups at the time of evaluation: six patients ranged from 11 to 23 years of age, and 3 patients ranged from 52 to 63 years of age. All patients had severely reduced photopic ffERG responses, including those exhibiting preserved foveal structure on OCT. The younger patients had absent to mild foveal atrophy on OCT, and four of the six demonstrated foveal and parafoveal hyperfluorescence on FAF. In addition, a 7-month-old child with compound heterozygous mutations in CNGA3 demonstrated similar foveal hyperfluorescence. The older patients demonstrated advanced foveal atrophy and punched-out foveal hypofluorescence with discrete borders on FAF imaging corresponding to the area of outer retinal cavitation on OCT.
Conclusions
Foveal hyperfluorescence is an early sign of achromatopsia that can aid in clinical diagnosis. In our cohort, patients with achromatopsia demonstrated age-dependent changes in FAF, which are likely to be progressive and to correlate with foveal atrophy and cavitation on OCT. This finding may be useful in charting the natural course of the disease and in defining a therapeutic window for treatment.
Achromatopsia is an autosomal recessive inherited disorder of cone photoreceptors. Affected individuals present in early childhood with deficient color discrimination, reduced visual acuity, photophobia, nystagmus, and absent or reduced cone function on electroretinography (ERG). Approximately 70% of achromatopsia cases can be accounted for by mutations in either CNGA3 or CNGB3 , which encode the alpha and beta subunits, respectively, of the cyclic nucleotide-gated cation channel in cone photoreceptors. Gene therapy has successfully improved vision and rescued electroretinographic cone function in mouse models of achromatopsia with both CNGA3 and CNGB3 mutations. In addition, application of gene therapy in two canine models of achromatopsia with CNGB3 mutations demonstrated successful restoration of vision. With gene therapy trials in humans imminent, determining the ideal candidates for gene therapy and defining a therapeutic window for treatment is of paramount importance.
Traditionally, achromatopsia has been characterized as a congenital and stationary disease. However, evidence is emerging that at least some patients demonstrate a progressive loss of vision, cone photoreceptor function, and foveal structural integrity. A subset of patients demonstrated progressive decline in visual acuity from childhood into adulthood. Khan and associates observed that some patients retain low-level cone function into adulthood and then experience progressive loss of residual function on ERG. In addition, progressive changes in foveal morphology on optical coherence tomography (OCT) have been described both longitudinally within patients and in an age-dependent manner among patients. Adaptive optics imaging has been used to further characterize photoreceptor structure in achromatopsia patients, revealing residual intact cones even in older patients exhibiting foveal atrophy on OCT.
Fundus autofluorescence (FAF) imaging detects pathology at the level of the retinal pigment epithelium and serves as an important diagnostic tool in macular disease. In particular, cone dystrophies are characterized by a ring of hyperfluorescence surrounding the central macula and, less commonly, by patchy hyperfluorescence of the central macula. FAF patterns in achromatopsia have not been well characterized in the literature. This study describes and quantifies a unique pattern of fundus autofluorescence in achromatopsia and correlates these findings with changes on OCT.
Methods
This is an observational case series. Subjects with achromatopsia with mutations in CNGA3 or CNGB3 were ascertained from the Retinal Dystrophy Clinic at the Kellogg Eye Center. The cohort included all patients with achromatopsia as determined by both OCT and FAF imaging, including two patients from one family with CNGA3 mutations and six patients from five families with CNGB3 mutations. An additional patient with mutations in CNGA3 had no available FAF photos but was included for assessment by serial OCTs. A 7-month-old child with confirmed compound heterozygous mutations in CNGB3 was included for FAF photos. Seven patients had more than one visit. The study complied with the tenets of the Declaration of Helsinki, and we obtained informed consent from all participating subjects, in accordance with the University of Michigan institutional review board.
DNA was extracted from peripheral blood lymphocytes and screened for mutations in CNGA3 and CNGB3 , as previously described, by polymerase chain reaction (PCR) amplification of all coding exons and flanking introns and untranslated sequences, followed by direct Sanger sequencing of PCR products.
Clinical data included best-corrected visual acuity; ophthalmoscopy; macular spectral-domain OCT (SD-OCT) (Cirrus; Carl Zeiss Meditec, Jena, Germany; Spectralis, Heidelberg Engineering, Heidelberg, Germany); full-field electroretinography (ffERG); fundus autofluorescence photography (FAF) (Spectralis); color vision testing (Farnsworth D-15 test); and Goldmann kinetic visual field perimetry.
Quantification of FAF was performed using a modification of a previously described technique. FAF quantification curves were produced using pixel intensity values (grayscale values) that were extracted from a 50 (vertical) × 400 (horizontal) pixel area centered at the foveola. Variability in grayscale values was reduced by vertical averaging, which generated an average pixel value at each horizontal location across this area. The resultant 1 × 400 matrix of averaged grayscale values was represented graphically. A control curve from a healthy control subject without retinal disease was included.
Results
Patients and Mutations
Patients fell into dichotomous age groups, with six patients between 11 and 23 years of age and three patients between 52 and 63 years of age. One additional patient was 7 months old at evaluation. Genetic testing confirmed mutations in CNGA3 in four patients and in CNGB3 in six patients. The clinical data and genotypes are summarized in the Table . The four patients with CNGA3 mutations were from three families and had a total of five different mutations, including four missense mutations and one single amino acid deletion. Patients A1 and A2 are siblings. The six patients with CNGB3 mutations were from five families and had a total of three different mutations, including one splice defect and two frameshift mutations. Patients B4 and B5 were siblings. All CNGB3 patients carried at least one p.Thr383fsX mutant allele, which accounts for 70% of disease-causing CNGB3 mutations and approximately 40% of all achromatopsia-related disease alleles in Caucasians. Three of these patients were homozygous, while three were compound heterozygous with other CNGB3 mutations ( Table ).
Clinical Data
BCVA ranged from 20/125 to 20/400 for CNGA3 patients and from 20/100 to 20/300 for CNGB3 patients. Color-vision testing using the Farnsworth D-15 test most commonly followed a scotopic axis, although there was some variability, and one patient (A1) had nearly normal ordering of hues. Common findings on fundus examination included foveal atrophy, macular granularity, and retinal vascular attenuation ( Figure 1 ). A few patients exhibited granularity of the peripheral retina but no peripheral pigmentary changes were seen. All patients demonstrated severely reduced photopic ffERG responses ( Table ). Patients with CNGA3 mutations had largely nonrecordable (NR) photopic a-waves and b-waves and flicker responses. Patient A3 had barely detectable photopic b-waves, and patient A4 had barely detectable flicker responses. Photopic ffERG parameters were more variable in the CNGB3 patients. Two of six had NR a-waves, all had recordable b-waves, and three of six had NR flicker responses. Patient B6 progressed from barely detectable photopic b-waves (<5 μV) to NR responses over 5 years, and patient B3 progressed from reduced photopic a- and b-waves to NR responses over 6 years.
Fundus Autofluorescence and Optical Coherence Tomography Findings
Figure 2 depicts FAF photos with corresponding OCT images for eight patients. One patient had serial OCT images but did not have FAF imaging, and the 7-month-old infant had FAF imaging but not OCT. Pixel intensity across the macula in each FAF photo is represented graphically as an autofluorescence curve to demonstrate the areas of hyperfluorescence in each image.
In four of six younger patients, FAF photos revealed a unique foveal hyperfluorescence ( Figure 2 ). Patients B1 and A2 demonstrated focal foveal hyperfluorescence, which corresponds with a positive deflection in the FAF quantification curve instead of the typical foveal depression seen in the control curve. Patients B2 and B3 demonstrated foveal and parafoveal hyperfluorescence sparing the foveola, which is more robust in patient B3 and more subtle in patient B2. FAF photos of the 7-month-old child also revealed a characteristic foveal and parafoveal hyperfluorescence ( Figure 3 ). The OCT findings in these patients had little to no foveal atrophy.
The FAF findings appeared to correlate with age and advancing atrophy on OCT. In contrast to the foveal hyperfluorescence observed in the younger patients, FAF revealed a punched-out foveal hypofluorescence with discrete borders and varying degrees of surrounding hyperfluorescence in all three older patients, which similarly corresponded to a punched-out depression on the FAF quantification curve ( Figure 2 ). All three of these patients also demonstrated advanced foveal atrophy and cavitation on OCT. When FAF photos were aligned with OCTs, the area of foveal cavitation correlated with the area of hypofluorescence in all six eyes ( Figure 4 ).
Serial Optical Coherence Tomography Findings
In addition to the age-dependent progression of foveal atrophy seen by comparison between the younger and older patient groups, progressive foveal atrophy was also demonstrated longitudinally within an individual patient with compound heterozygous mutations in CNGA3 , p.Arg277His and p.Glu376del (patient A4 in Table ). Three serial OCTs showed progressive foveal atrophy and cavitation in both eyes from ages 13 to 18 ( Figure 5 ). No FAF photos were available for this patient.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
