In this study, we attempted to investigate the effect of dexamethasone implant and ranibizumab on the regression of HRDs in patients with ME secondary to BRVO. We demonstrated that the mean number of HRDs in inner and outer retinal layers was significantly decreased in patients who administered IVD at 1 year after treatment. Furthermore, the reduction in HRDs and better BCVA after IVD intimates that the HRDs should be considered as inflammatory markers in the follow-up of ME in BRVO. Thus, IVD injection could be more appropriate for patients with higher HRDs after BRVO.
Previous studies explain the possible constitution of HRDs detected on OCT, but these dots are still unknown. Bolz et al. reported that the isolated HRDs were found on OCT, but they could not be found on fundus photographs taken simultaneously with OCT. However, they demonstrated that the confluent accumulated HRDs on OCT were detected as hard exudates in the corresponding fundus photograph. Therefore they supported that these isolated dots characterized by the same hyperreflectivity was accumulated dots might be small intraretinal protein and/or lipid deposits as precursors of hard exudates.(10) In contrast, several other studies have asserted that HRDs are associated with inflammatory responses in the retina.(11–13) Coscas et al. suggested that these HRDs were most likely microglia cells activated by inflammation, which subsequently swell and spread to outer retinal layers.(7) Furthermore, in another study, authors reported that, in patients with BRVO, HRD disappeared immediately after intravitreal bevacizumab treatment, suggesting that HRF could represent inflammatory cells, particularly activated microglia, rather than lipid extravasation.(14)
The RIDE and RISE phase III clinical trials also showed that resolution of hard exudates (deposition of lipoproteins) after IVR was not evident before 6 months of treatment. By contrast, these studies recognized diminished numbers of HRD within 6 months of treatment, especially in eyes that received dexamethasone implants.(15) Similar to their results, in our study, the decrement of HRDs was properly early after IVD treatment. (Within 3 months) Vogel et al. suggested that the HRDs on the OCT are cellular, rather than accumulations of proteins or lipids, given their discrete, granular, and demarcated presentation and consistent size in a study with patients with central serous chorioretinopathy via adaptive optics scanning light ophthalmoscopy findings. They support the hypothesis that HRF could be activated microglia.(16)
Lee et al. reported a strong correlation between sCD14 and HRDs in the inner retina and also they suggested that the HRDs observed on OCT may be due to activated microglia in diabetic ME.(17) These studies suggested that HRDs could reflect the presence of activated microglia due to retinal inflammation. Zeng et al showed that, as DR in human donor eyes progresses, activated microglia infiltrate and migrate to the outer retinal layer. Singhal et al demonstrated that the number of activated microglia decreases after intravitreal triamcinolone injection.(18) Retinal glial cells, contribute to the development of ME.(19) In a healthy retina, resting microglia are essentially located in the inner retina, but with inflammation, the activated glial cells migrate to the site of injury.(20) The activated microglia also discharge proinflammatory and proangiogenic mediators.(21) Vujosevic et al stated the migration of HRDs from the inner to the outer retina layers through the DR progression. (22) The contribution of activated microglia to the progression of BRVO was also shown in animal model.(23)
The zonulae adherents between the photoreceptors and Muller cells, creating the ELM, can block the transfer of macromolecules, and that the healthy ELM restricts the migration through the outer retinal layers. A breakdown of ELM permits these inflammatory molecules through the outer layers of retina and leads to both photoreceptor disorganization which is in line with the results of a previous report, documenting the outer retinal discontinuity in eyes with retinal vein occlusion. (24) We hypothesized that the inflammatory microenvironment in the outer retinal layer might be responsible for the damage of photoreceptor status. We observed that the eyes with received only IVR had significantly more HRDs in the outer retinal layers and more ELM disruption and inner segment/outer segment disruption at the final visit. The pathologic association of increased HRDs in outer retinal layers with disruption of the ELM and EZ and poor visual acuity, may be a clinical prognostic marker of outer blood-retinal breakdown and consequent photoreceptor dysfunction. The abovementioned conclusions may also describe the fact why visual acuity does not always increase after intravitreal treatment, even in the decrease of CMT.
The increased VEGF expression contributes to the pathology of ME due to BRVO, and various intravitreal anti-VEGF injections have been widely used for treatment.(1, 25) In addition to increased VEGF expression, inflammatory cytokines and chemokines also play a crucial role in the pathogenesis of ME. The inflammatory cytokines like soluble intercellular adhesion molecule 1, interleukin 6 (IL-6), monocyte chemotactic protein 1 (MCP-1) are elevated in the aqueous humor of patients with BRVO, which enhance vessel permeability and play an essential role in the pathogenesis of ME. (1) After intravitreal injection, CMT was more decreased in the triamcinolone acetonide (IVTA) group compared with the bevacizumab group. IL-6, interferon-γ-inducible protein 10, MCP-1, Human Platelet Derived Growth Factor-AA, and VEGF were significantly decreased in the IVTA group, but only VEGF in the intravitreal bevacizumab group.(26) Corticosteroids repress the production of prostaglandins and leukotrienes, decreasing edema within a variety of mechanisms, essentially suppressing macrophage activity, vasoconstrictive effect, reduction of lymphokine, and VEGF. In opposite, anti-VEGF agents have been observed to reduce hyperpermeability through a decrease in the production of VEGF.(27) Therefore, IVD implants, which inhibit inflammatory cytokines, may be more effective as a therapeutic option for ME in patients with BRVO.
Our study confirmed the outcomes made in previous studies that HRDs correlated negatively with visual acuity.(28–31) Do et al reported that the final BCVA was associated with the baseline number of outer retinal HRDs in the IVB group. However, in the IVD group, the authors did not observe the correlation with the baseline number of outer retinal HRDs. They explained the disparity because the IVD injection reduced outer retinal HRDs more effectively than the IVB and suggested that IVD could resolve the inflammatory components more effectively than IVB with better visual outcomes.(9)
Hwang et al. also evaluated the correlation between the number of HRDs and the therapeutic responsiveness of bevacizumab or dexamethasone implant. They found that the number of HRDs on OCT can be a predictive prognostic factor of the treatment response to bevcizumab injection or IVD implant. A higher number of HRDs and higher rate of OPL disruptions was observed on SD-OCT in bevacizumab nonresponders than in responders in their studies. They supported that dexamethasone implant may be more effective in treating diabetic ME or retinal vein occlusion eyes with many HRDs and OPL disruptions on OCT.(30) In current study, the final BCVA and the presence intact ELM and EZ were higher in eyes received IVD than in eyes with only IVR treatment.
Chatziralli et al. described that the amount and location of HRDs are independent factors of worse final BCVA in patients with ME due to retinal vascular diseases, such as DR and BRVO. The decline of the numbers of HRDs was not influenced by the decision of treatment option (IVR or IVD) used to decrease the ME. In addition, they found that the number of HRDs was associated with the status of EZ and ELM. (31) In contrast, in our study we observed that the final visual acuity and the number of HRDs in outer retinal layer were higher in patients who received IVD. We speculate that these different results may be due to the difference in the patient groups included.
The limitation of our study is its retrospective and nonrandomized nature. In addition, the manual measurement and classification of the position of the HRDs may have introduced a subjective element. However, the strengths of our study are the relatively homogenous involvements of patients with ME due BRVO, treated by two agents and combined treatment.
In conclusion, this study demonstrated that the outer HRDs are independent factors associated with the final BCVA in patients with ME due to DR and BRVO, suggesting HRDs as a potential biomarker of poor final visual outcome. Furthermore, the reduction of HRDs are more prominent in patients received dexamethasone implant treatment. The reduction in HRDs at 12 months and better BCVA after IVD intimates that the HRDs should be considered as inflammatory markers in the follow-up of CME in BRVO. Thus, IVD injection could be more appropriate for patients with higher HRDs after BRVO.