In the present study, the key role of photoreceptors in visual impairment with PM-CNV was determined by using OCT to observe microstructural changes in the fundus and predict factors of EZ tear. The significant relationship between photoreceptor morphology and density degradation and BCVA deterioration has been extensively documented in previous studies. 6,13,14 Therefore, early intervention and protection of the EZ will be important to maintain visual function in PM-CNV.
While the pathological changes with PM-CNV progress slowly and vision gradually loses over 10 years without treatment.15 However, since the PM eye axis gradually lengthens, the EZ in the outer layer of the retina become thinner and the BCVA worse.5 Therefore, the present study had the advantage of retrospectively studying the effect of EZ tear with PM-CNV patients on their vision, to predict the factors contributing to EZ tear and to be able to provide a biological indicator to monitor whether the EZ was torn or not. This study showed that higher HF and larger area, bigger HF/CNV area ratio, greater CNV area and thinner MEZ increased the risk of EZ tear, as well as type 2 PM-CNV patients had all EZ tear, whereas type 1 had almost continuous and intact EZ. There was no correlation between age and the factors within groups and between groups. It suggests that EZ tear may be related to the disease duration, type and size of PM-CNV. Therefore, it is helpful to identify the above risk factors before starting treatment for PM-CNV to warn patients and provide prognostic information.
In previous studies, EZ tear was not only significantly associated with BCVA, but also predicted visual acuity recovery after anti-VEGF treatment.16 Similar findings were found in the present study. It was found that the BCVA in the EZ tear group was about 1/2 of normal BCVA in the non-tear group, and after anti-VEGF treatment the BCVA in the EZ tear group was only 0.65±0.22 lower than the 0.45±0.12 in the non-tear group before anti-VEGF treatment. Indicating that the EZ tear severely affected the BCVA of the PM-CNV, leading to even worse visual function. However, Milani17 expressed the opposite view, who found that an increase in BCVA was observed even in the absence of EZ at baseline and final follow-up, hence he believed that the integrity of the EZ may not be the major factor for BCVA, and that it was the integrity of the external limiting membrane and EZ of the bleeding and lesion attachment that may be the main factor influencing BCVA. His findings were similar to those of the 10/12 eyes in tear group of present study, where the EZ integrity was not recovered but the BCVA was increased after anti-VEGF treatment. Thus, we hypothesize that it is possible that the cone cells are not destroyed after EZ tear and are distributed above the lesion to maintain normal visual function, but this hypothesis will need to be confirmed in the future by large-scale and long-term studies.
Integrity of regional photoreceptors was a powerful predictor of whole retinal vision function.18 The photoreceptor cone cells are predominantly located in the fovea, with a gradual decrease in the periphery.19 Likewise, PM-CNV also occurs mostly below the macular fovea, or in the parafovea.20 In this study, the EZ was torn and disappeared in the tear group, and the BCVA deteriorated severely, which was not consistent with Milani's17 study either. Therefore, we speculated that the EZ tear may have led to changes in the density and arrangement rules of the cone cells as well leading to severe functional deterioration. Previous studies have also found that the density of the cone cells was also an important factor in visual dysfunction, and were able to confirm our conjecture, however they examined high myopia.6 The density and arrangement patterns of PM-CNV cone cells need to be studied in an efficient way in the future.
Among the photoreceptor morphological parameters, MEZ thickness was one of the most important predictors of photosensitivity, especially in the inner region.6 In this study MEZ thickness also played a role in the impact of EZ tear. The MEZ zone contains the mitochondria, dictyosome, endoplasmic reticulum and can produce ATP, light of protease and chemicals necessary to maintain photosensitivity.21,22 Previous studies have found a significant thinning of the MEZ in PM with a high correlation with BCVA deterioration. This could also suggest that thinner MEZ is an important factor in the deterioration of PM-CNV visual acuity.
This study has some limitations. To begin with, the effect of different refraction and axial length on MEZ thickness and EZ tear in PM-CNV patients was not studied. Secondly, because of the limitations of the OCTA technique itself, OCTA displays the blood flow signal by detecting the red blood cell flow rate. when the red blood cell flow rate was too slow or too fast, OCTA was unable to detect the blood flow signal (neovascularization) and the image appeared as a non-perfused or non-vascular area.23 This may lead to errors during our CNV area calculations, resulting in errors in the HF/CNV area ratio. Thirdly, all measurements of HF height, HF area, CNV area and MEZ thickness in this study were made manually and it was difficult to guarantee the accuracy of this data. Finally, the present study was a retrospective study with a small sample size. Follow-up was short and irregular, and it was not possible to study the long-term visual effects of EZ for PM-CNV patients.
In conclusion, PM-CNV with higher HF, larger HF area, and greater HF area relative to CNV area are highly likely to have a relatively high risk of EZ tear. It is beneficial for both the clinician and the patient to assess these risk factors prior to treatment for PM-CNV progression. Patients with non-tear EZ will have better visual quality.