Retinitis pigmentosa (RP) is a group of inherited retinal diseases characterized by progressive degeneration of photoreceptor cells.1 Impairment of night vision is typically the first clinical manifestation of RP, followed by peripheral vision loss and gradual central vision deterioration.2 The pathogenesis of RP is considerably complex and multifactorial. Past literature implicates inner retina disorganization, retina pigment epithelium (RPE) degeneration, retinal ganglion cell death, and vascular changes as factors that may contribute to the disease.1,2
The ring of hyperautofluorescence is a well-recognized feature in RP. Geographic atrophy of the RPE appears hypoautofluorescent on fundus autofluorescence (FAF) imaging due to decreased metabolic activity.3 As the disease progresses, the hyperautofluorescent ring corresponds to the junction between the hypoautofluorescent atrophic peripheral retina and the active posterior pole, where increased oxidative stress results in the accumulation of hyperautofluorescent lipofuscin.4,5 This hypoautofluorescence in FAF imaging corresponds to scotomas on Goldman perimetry,4 and constriction of the hyperautofluorescent ring area may be a sign of RP disease progression.5
In addition, it has been reported that RP is associated with vascular changes such as decreased macular blood flow, arterial narrowing, and vessel attenuation.2,6 Previous studies have shown that RP eyes are associated with a significantly reduced macular vessel density and increased foveal avascular zone (FAZ) size compared to healthy controls.2 However, little is known about the relationship between the area of hyperautofluorescence and vascular perfusion parameters as obtained by optical coherence tomography angiography (OCTA). In this retrospective, cross-sectional study, we measured the area of the hyperautofluorescent ring in patients with RP to investigate whether a correlation exists with best-corrected visual acuity (BCVA), FAZ size, and retinal capillary perfusion density (CPD).
Patients and Methods
We identified 18 eyes of nine patients with genetically confirmed RP. Two patients had USH2A variant RP, one patient had TULP1 variant, one patient had VPS13B variant, and one patient had GUCY2D/ROM1/RP1L variant. Genotype data were not available for four patients. Institutional review board (IRB) approval was obtained through IRB Services. Eyes with macular edema were excluded from this study. Patients were imaged using ultra-widefield FAF (California; Optos, Dunfermline, Scotland) at the Vitreous Retina Macula Specialists of Toronto. The Optos California uses a 532-nm wavelength green laser for its autofluorescence mode.
Medical records were reviewed to obtain demographic and visual acuity (VA) data. ImageJ software (version 1.51k; NIH, Bethesda, MD) was used to measure the area of the hyperautofluorescent ring in each RP FAF image (Figure 1). Spectral-domain OCTA (SD-OCTA) (AngioVue; Optovue, Fremont, CA) was used to assess retinal vascular perfusion in the superficial capillary plexus (SCP) and deep capillary plexus (DCP) at the macula using a 3 mm × 3 mm strategy as well as a 6 mm × 6 mm strategy. The SCP and DCP were auto-segmented using the device's software, which removed artifacts. CPD in the parafoveal SCP (Figure 2) and DCP (Figure 3) was measured in each eye. In addition, the device's software was used to measure and calculate the size of the foveal avascular zone (FAZ) at the SCP (Figure 4).
Fundus autofluorescence image of the left eye in a patient with retinitis pigmentosa. The yellow circle indicates the area of hyperautofluorescence as measured with ImageJ (version 1.51k).
(A) 3 mm × 3 mm optical coherence tomography angiography image of the left macular superficial capillary plexus in a patient with retinitis pigmentosa. (B) Vessel density map indicating percentage density of blood vessels. Parafoveal capillary perfusion density was measured at 40.74%.
(A) 3 mm × 3 mm optical coherence tomography angiography image of the left macular deep capillary plexus in a patient with retinitis pigmentosa. (B) Vessel density map indicating percentage density of blood vessels. Parafoveal capillary perfusion density was measured at 51.86%.
Optical coherence tomography angiography image of the left foveal avascular zone (FAZ) in a patient with retinitis pigmentosa. The area of the FAZ was measured as 0.322 mm2.
Statistical analysis was conducted using Excel 2016 (version 1806; Microsoft, Redmond, WA). BCVA values were converted to the logarithm of the minimal angle of resolution (logMAR). Pearson correlation coefficients were calculated to determine the strength of linear relationship between hyperautofluorescent ring size and the following variables: VA, SCP parafoveal CPD, DCP parafoveal CPD, and FAZ area. Regression analysis was performed to determine if the correlation between two measured variables was statistically significant (P < .05).
Nine subjects with bilateral RP were identified for a total of 18 eyes. The mean age of the subjects was 48.6 years old (range: 29 years to 80 years), and one subject was male (11.1%). The mean BCVA was 0.31 ± 0.21 logMAR units in the right eye, and 0.31 ± 0.20 logMAR units in the left eye.
Hyperautofluorescent ring area was highly concordant between the right and left eyes in each patient (R = 0.987; P < .001). As expected, there was a statistically significant positive correlation between CPD measurements in the SCP and DCP at both 3 mm × 3 mm and 6 mm × 6 mm capture strategies (Table 1). The P value was less than .001 for the following variable comparisons: 3 mm × 3 mm SCP CPD vs. 3 mm × 3 mm DCP CPD, 3 mm × 3 mm SCP CPD vs. 6 mm × 6 mm SCP CPD, 3 mm × 3 mm SCP CPD vs. 6 mm × 6 mm DCP CPD, 3 mm × 3 mm DCP CPD vs. 6 mm × 6 mm SCP CPD, 3 mm × 3 mm DCP CPD vs. 6 mm × 6 mm DCP CPD, and 6 mm × 6 mm DCP CPD vs. 6 mm × 6 mm SCP CPD.
Pearson Correlation Coefficients Between Hyperautofluorescent Ring Area, SCP CPD, DCP CPD, FAZ Area, and logMAR VA
There was no significant correlation between hyperautofluorescent ring size and parafoveal CPD in the SCP, or the DCP for both the 3 mm × 3 mm and 6 mm × 6 mm imaging strategies. VA was not correlated with hyperautofluorescent ring size (R = 0.242; P = .33). FAZ area showed a negative trend with hypoautofluorescent ring size, but the correlation was not statistically significant (R = —0.362; P = .14). Overall, none of the primary outcome variables (CPD, VA, FAZ area) were statistically correlated with hyperautofluorescent ring area in this study.
There was a significant negative relationship between FAZ area and 3 mm × 3 mm DCP CPD (P = .037), and a significant negative relationship between 3 mm × 3 mm SCP CPD and VA (P = .046).
Assessing the hyperautofluorescent ring and retinal vasculature density are important clinical tools for evaluating disease severity and progression in RP. Previous studies have indicated that disease progression in RP is associated with constriction in hyperautofluorescent ring size and decrease in visual function.4,7 Sugahara et al.6 identified that parafoveal DCP perfusion density was significantly negatively correlated with BCVA (R = −0.72; P < .001), whereas Alnawaiseh et al.8 found a significant negative correlation between SCP perfusion density and BCVA (R = −0.77; P < .001). Both studies reported that RP eyes have significantly lower SCP and DCP parafoveal flow densities compared to age-matched healthy controls.6,8 The literature also describes a positive correlation between SCP FAZ area and BCVA in RP eyes.6,9 This study expands on those findings, revealing that 3 mm × 3 mm SCP CPD has a significant negative correlation with VA in RP eyes. Since the significant correlation between DCP CPD and FAZ area and SCP CPD and VA was only found in the 3 mm × 3 mm imaging strategy, these results suggest that the 3 mm × 3 mm strategy may be preferable to the 6 mm × 6 mm strategy for research analysis. As the image capture area increases, poor fixation, and motion artifacts become more of an issue due to increased imaging time, and accurate auto-segmentation is more of a challenge.
Although constriction of the hyperautofluorescent ring and reduced parafoveal CPD have both been associated with progression of RP, the two factors were not found to have a statistically significant linear relationship. Central VA may be preserved despite progression of the disease, whereas foveal blood flow has been shown to strongly correlate with VA.8 Therefore, our finding that increased SCP FAZ area was correlated with worse BCVA, whereas hyperautofluorescent ring size showed no correlation conforms well with existing knowledge. Since RP encompasses a heterogenous group of disorders with variable inheritance patterns and more than 50 identified causal genes,5 it is difficult to predict how the pathogenesis of photoreceptor degeneration and microvascular impairment affect each other. Nonetheless, FAF and OCTA are both valuable imaging modalities for monitoring the progression of RP and will play increasingly vital roles in the assessment of future therapeutic trials.10
This study was limited by several factors. As the disease has a low prevalence of 1:3,000,4 our study has a small sample size of 18 eyes. Therefore, the statistical power was low, and it is possible that a larger sample size with more RP genotypes included may yield a statistically significant correlation between our study outcomes and hyperautofluorescent ring size. Additionally, since RP represents a heterogenous disease with dozens of genotype mutations identified,11 the small cohort of genotypes included in this study may not be representative of all presentations of RP. Furthermore, visual field data were not routinely available in all cases, and not all patients were able to reliably perform the visual field test, therefore we could not analyze the relationship between hyperautofluorescent ring size and visual field impairment. With a larger sample size and longitudinal evaluation of patients, we may be able to better understand the how the morphology of the hyperautofluorescent ring can affect the pathogenesis and progression of RP.
In conclusion, OCTA and FAF are valuable imaging modalities for assessing RP. Evaluation of parafoveal microvasculature metrics such as CPD and FAZ area may be useful for correlation with VA, but the area of hyperautofluorescence as detected by FAF is not a reliable predictor of CPD, FAZ area, or VA.
- Toto L, Borrelli E, Mastropasqua R, et al. Macular features in retinitis pigmentosa: Correlations among ganglion cell complex thickness, capillary density, and macular function. Invest Ophthalmol Vis Sci. 2016;57(14):6360–6366. doi:10.1167/iovs.16-20544 [CrossRef]
- Parodi MB, Cicinelli MV, Rabiolo A, et al. Vessel density analysis in patients with retinitis pigmentosa by means of optical coherence tomography angiography. Br J Ophthalmol. 2017;101(4):428–432. doi:10.1136/bjophthalmol-2016-308925 [CrossRef]
- Ogura S, Yasukawa T, Kato A, et al. Wide-field fundus autofluorescence imaging to evaluate retinal function in patients with retinitis pigmentosa. Am J Ophthalmol. 2014;158(5):1093–1098. doi:10.1016/j.ajo.2014.07.021 [CrossRef]
- Tee JJL, Kalitzeos A, Webster AR, Peto T, Michaelides M. Quantitative analysis of hyperautofluorescent rings to characterize the natural history and progression in RPGR-associated retinopathy. Retina. 2018;38(12):2401–2414.
- Trichonas G, Traboulsi EI, Ehlers JP. Correlation of ultra-widefield fundus autofluorescence patterns with the underlying genotype in retinal dystrophies and retinitis pigmentosa. Ophthalmic Genet. 2017;38(4):320–324. doi:10.1080/13816810.2016.1227450 [CrossRef]
- Sugahara M, Miyata M, Ishihara K, et al. Optical coherence tomography angiography to estimate retinal blood flow in eyes with retinitis pigmentosa. Sci Rep. 2017;7:46396. doi:10.1038/srep46396 [CrossRef]
- Dysli C, Schürch K, Pascal E, Wolf S, Zinkernagel MS. Fundus autofluorescence lifetime patterns in retinitis pigmentosa. Invest Ophthalmol Vis Sci. 2018;59(5):1769–1778. doi:10.1167/iovs.17-23336 [CrossRef]
- Alnawaiseh M, Schubert F, Heiduschka P, Eter N. Optical coherence tomography angiography in patients with retinitis pigmentosa. Retina. 2019;39(1):210–217. doi:10.1097/IAE.000000000000190 [CrossRef]
- Koyanagi Y, Murakami Y, Funatsu J, et al. Optical coherence tomography angiography of the macular microvasculature changes in retinitis pigmentosa. Acta Ophthalmol. 2018;96(1):e59–e67. doi:10.1111/aos.13475 [CrossRef]
- Sujirakul T, Lin MK, Duong J, Wei Y, Lopez-Pintado S, Tsang SH. Multimodal imaging of central retinal disease progression in a 2-year mean follow-up of retinitis pigmentosa. Am J Ophthalmol. 2015;160(4):786–798.e4. doi:10.1016/j.ajo.2015.06.032 [CrossRef]
- Chang S, Vaccarella L, Olatunji S, Cebulla C, Christoforidis J. Diagnostic challenges in retinitis pigmentosa: Genotypic multiplicity and phenotypic variability. Curr Genomics. 2011;12(4):267–275. doi:10.2174/138920211795860116 [CrossRef]
Pearson Correlation Coefficients Between Hyperautofluorescent Ring Area, SCP CPD, DCP CPD, FAZ Area, and logMAR VA
|Ring Area||3 mm × 3 mm SCP||3 mm × 3 mm DCP||6 mm × 6 mm SCP||6 mm × 6 mm DCP||FAZ Area||VA|
|3 mm × 3 mm SCP||0.084||1.000|
|3 mm × 3 mm DCP||0.285||0.912*||1.000|
|6 mm × 6 mm SCP||0.189||0.834*||0.754*||1.000|
|6 mm × 6 mm DCP||0.269||0.763*||0.725*||0.822*||1.000|