Fig 1.
Manual segmentation and morphometric analysis of retinal layers.
(A) Three horizontal OCT scans passing through the fovea were used for morphometric analyses. (B) The boundaries of the inner limiting membrane (straight blue line), the outer border of the retinal nerve fiber layer (red straight line), the mid-outer plexiform layer (orange straight line), and the inner border of the RPE-Bruch’s membrane complex (straight brown line) were traced manually. “Retinal Nerve Fiber Layer Area” (RNFLA) was described as the area between the inner border of the internal limiting membrane and the outer border of the retinal nerve fiber layer. The area between the internal limiting membrane and the outer plexiform layer was defined as the “Inner Retinal Area” (IRA), including the RNFLA and the cysts. Inner retinal neuronal area (IRNA) was defined as the area occupied by inner retinal neurons and estimated by subtracting the total cyst area and RNFLA from the IRA. The area between the outer plexiform layer and RPE was regarded as the “Outer Retinal Area” (ORA). In the presence of cysts in the outer retinal zone, the cyst area was excluded from the outer retinal area calculation. The lengths of the intact OLM (interrupted yellow line) and EZ (blue interrupted line) were also traced and calculated.
Table 1.
Causative genes and inheritance patterns of the patients.
Fig 2.
Distribution of intraretinal cyst sizes among underlying mutations causing retinitis pigmentosa.
Comparison of cyst sizes in subjects with different underlying genetic mutations did not reveal any differences (p = 0.08).
Table 2.
Correlations of anatomical and functional parameters.
Fig 3.
Best-fit curves defining the relationship between cyst size and other retinal anatomical parameters.
(A) A significant linear correlation exists between the size of the intraretinal cysts and RNFLA (r = 0.27, p = 0.001), indicating that RNFL becomes thickened at times when intraretinal cysts are at the peak of their size. (B) The change in retinal cyst size and the IRNA can be best expressed with a 3-parameter Sigmoidal curve in the form of f = a/(1+exp(-(x-x0)/b)). Retinal cysts resolve as the IRNA decreases in the late phases of the degenerative process. (r = 0.20, p = 0.02). (C) Intraretinal cysts diminish with the thinning of the inner retina (r = 0.74, p < 0.001). The high correlation is mostly due to the presence of cysts in the inner nuclear layer. (D) Intraretinal cysts are the biggest in the early phases of the disease. As outer retinal retina degenerates, the size of the cysts gets smaller (r = 0.26, p = 0.003). (E) Resolution of intraretinal cysts coincides with decrease in total retinal area (r = 0.77, p < 0.001). (F) There is a positive linear correlation between cyst size and remaining OLM, indicating that cysts develop early in retinal degeneration and fade out as the outer retina degenerates (r = 0.21; p = 0.005). (G) Like iOLML, iEZL positively correlates with the cyst area, indicating that degeneration of the photoreceptors is coupled with the resolution of the cysts (r = 0.16; p = 0.03).
Fig 4.
Impact of change in anatomical features on visual acuity.
(A) A 2-parameter single rectangular hyperbola best-fits to explain the impact of retinal cysts on visual acuity (r = 0.14, p = 0.05). Cysts cause a 2-line drop in visual acuity early during development. Although cysts may enlarge in time, their impact on visual acuity remains the same. (B) RNFLA has the highest linear correlation with logMAR visual acuity among other studied parameters (r = 0.325, p < 0.001). (c) A weak linear correlation existed between the IRNA and visual acuity (r = 0.160, p = 0.028). (D) An increase in IRA with the development of cysts decreases the patients’ visual acuity (r = 0.238, p = 0.002). (E) No significant correlation was detected between ORA and visual acuity (r = - 0.072, p = 0.197). (F) Similar to the relationship of visual acuity with IRA, the size of cysts affects both the TRA and visual acuity (r = 0.159, p = 0.029). Patients have better visual acuity if the cysts are small; as they get bigger, their visual acuity worsens, and TRA increases. (G) A significant correlation exists between the OLM and visual acuity, which fits-best to a 2-parameter hyperbolic curve in the form f = (a*b)/(b+x) (r = 0.19, p = 0.01). The 50th percentile of this effect corresponds to 8115 mm of intact OLM and visual acuity of 20/34. (H) iEZL significantly correlates with visual acuity. Distribution of the data best-fits to a 2-parameter hyperbolic curve in the form f = (a*b)/(b+x) (r = 0.30, p<0.001). The 50th percentile of this effect corresponds to 1210 mm of intact EZ and visual acuity of 20/55.