Previous treatment options pertaining to myopic CNV revealed unsatisfactory long-term outcomes.23,25-27
The RADIANCE study proved that treatment using ranibizumab resulted in rapid improvements in visual acuity in patients with myopic CNV that were statistically superior to the results pertaining to verteporfin PDT after three months.28 The BRILLIANCE study demonstrated the superiority of the efficacy of ranibizumab over verteporfin PDT after three months and reported that the beneficial effects persisted for twelve months.29 The MYRROR study revealed that the patients with myopic CNV who underwent intravitreal aflibercept injection displayed significant improvements in visual and anatomic parameters over a time period of 48 weeks.
In the current study all the patients displayed good improvement in vision. The injection burden observed in the present study was concurrent with the MYRROR study.
After intravitreal aflibercept injections, the subjects in the younger group displayed a greater reduction in CFT at twelve months compared to the order group. The aforementioned observation was concurrent with the clinical findings reported by Yoshida et al and Bruè et al.7,30 Moreover, at three and twelve months, the younger group displayed a trend towards better improvement in vision after intravitreal aflibercept injections, although the results were not statistically significant.
The inferior treatment outcomes associated with myopic CNV in the older age group could be attributed to the decreased integrity and function of the myopic retinal pigment epithelium that downgrades the inhibition of angiogenesis.31 Indeed, myopic CNV in older patients could manifest simultaneous AMD and high myopia, resulting in poor natural outcomes. Moreover, older patients tend to develop chorioretinal atrophy or degeneration that negatively influences the final visual acuity.
Previous studies reported that the role of spherical equivalent refraction and axial length as predictive factors associated with the final visual acuity after anti-VEGF treatment in myopic CNV remained controversial. Bruè et al reported that higher myopia is associated with decreased visual acuity after intravitreal aflibercept injections in young patients with myopic CNV. Some studies identified a significant positive correlation between BCVA and macular choroidal thickness after anti-VEGF treatment.32 Greater axial length was often considered to be major factor associated with macular choroidal thinning. The current study did not observe any difference in the treatment outcomes between the groups with long and short axial length (Table 4). This could be attributed to the relatively small sample size and short duration of follow-up. Further investigations should be performed to elucidate the significance of axial length in predicting treatment outcomes.
Recently, OCTA is being widely used in the diagnosis of various macular diseases and provides detailed, layered images of enface retinal and choroidal vasculature.33-36 Cheng et al used OCT B-scan and OCTA to perform quantitative analysis and monitor the therapeutic effects of intravitreal ranibizumab injection (0.5 mg/0.05 mL) (Lucentis; Genentech, Inc., South San Francisco, CA) in myopic CNV.23 In the aforementioned study, OCTA revealed significant attenuation of the capillaries and small caliber feeder vessels after intravitreal ranibizumab injection. Moreover, the study reported a simultaneous reduction in the size of the selected CNV area and flow area.
To the best of our knowledge, this is the first study that used OCTA to detect the changes in the shape, selected area, and flow area of CNV after intravitreal aflibercept injection at three and twelve months. In the current study, all the 12 patients who underwent OCTA displayed improvement in BCVA, decrease in the selected CNV area and flow area at three months and the treatment effects persisted until the twelfth month. This result is consistent with the results reported by Cheng et al, wherein myopic CNV was treated using intravitreal ranibizumab injection.23
According to a previous study, it was often considered that active CNV presented as retinal hemorrhage, intra-retinal fluid, sub-retinal fluid, or pigment epithelial detachment, as shown by the fundus photography and OCT B-scan images. These active CNV lesions were also considered to be more sensitive to anti-VEGF therapy. However, some cases of myopic CNV that presented with scars or fibrosis on fundoscopic examinations after anti-VEGF treatment could also show residual, high reflective neovascularization lesions on OCT B-scan images. The aforementioned condition often confused clinicians regarding the necessity for further treatment and the presence of CNV activity. Hence, OCTA is expected to provide additional biomarkers and parameters to guide further management.
Bruyère et al used OCTA to identify two types of neovascular membranes in high myopia. The first involved small, disorganized vascular loops, suggesting an immature neovascular network. The second is a larger, highly structured, interlacing network, suggesting a mature lesion.26 Querques et al reported that on OCTA images, active myopic CNVs could be mainly interlacing, whereas the abnormal vascular network with tangled pattern could be inactive myopic CNV.27 Cheng et al reported that sea-fan and lacy-wheel types tend to be clinically active CNVs, whereas poorly defined lesions, such as filamentous CNV vessels, could be either active or inactive.
In the present study, the interlacing group displayed a trend towards better improvement in CFT reduction and decrease in the selected CNV area and flow area (Table 5). Additionally, previous studies have reported consistent results and stated that the capillaries and small caliber feeder vessels were significantly attenuated after intravitreal ranibizumab injection, whereas the main central trunk vessel and large caliber feeder vessels remained unchanged. 25,37
However, owing to the limited sample size, the scenario warrants further investigation to clarify the significance of CNV patterns in predicting the response to anti-VEGF therapy. OCTA could be considered as a useful tool that can be employed to identify the different CNV patterns and detect CNV activity, predict treatment response, and monitor the need for repeated treatments in patients with myopic CNV.
The current study has certain limitations. The series was relatively small, and the duration of follow-up was short. The current study is retrospective in nature and a prospective analysis is necessary to study the current observations. The present study observed some poor image quality with significant motion artifacts, due to unstable fixation, which could make the analysis more challenging. Additionally, the accuracy of the layers automatically divided by OCTA could be affected by the long axial length of the pathologic myopia eyeballs. Moreover, the accuracy of measurement could be affected, owing to the manual selection of the CNV area using the software provided by OCTA.