We investigated the effects of anti-VEGF therapy on visual improvements in DME patients with extended IRCs and SRD, which have a negative impact on retinal function.13–14 Our data demonstrate that the logMAR BCVA is successfully improved in these patients, which is in accordance with the results of previous large pivotal clinical trials.19–21 Numerous studies have shown that most retinal micromorphologic changes documented with OCT imaging are associated with retinal function, so we focused on the prognostic value of OCT data from patients with DME with extended IRCs and SRD.22–24
CFT is commonly used as a secondary outcome measure for DME,19–20 but CFT can change out of proportion to the visual response in the treatment of DME25; thus, it is not a reliable indicator of visual acuity. Deak et al.26 showed only a weak correlation between central retinal thickness and visual outcomes, especially at long-term follow-up. Otani et al.27 revealed that CFT was negatively correlated with the logMAR BCVA in patients with DME (r = − 0.23), and that EZ integrity and ELM integrity were positively correlated with visual function in these patients. Our study showed no significant correlation between CFT and the logMAR BCVA at 12 months in patients with DME with extended IRCs and SRD. However, the extent of EZ disruption and ELM disruption were strongly correlated with the logMAR BCVA at 12 months.
Increased VEGF has been founded to correlate with the ELM and EZ disruption which affects visual acuity.28 In our cohort, the baseline extent of EZ disruption was more serious than the baseline extent of ELM disruption, which revealed that EZ disruption might occur prior to ELM disruption. Another study showed that administration of intravitreal anti-VEGF agents could lead to sequential restoration of ELM and EZ disruption.29 For the first time, we investigated the pertinence of baseline OCT imaging characteristics and final photoreceptor status in DME with coexisting extended IRCs and SRD after intravitreal ranibizumab treatment.
The EZ corresponds to the ellipsoid of the photoreceptor cell, which is rich in mitochondria that consume large amounts of energy. SRDs decrease photoreceptor cell metabolism and result in a high degree of EZ disruption.30 In the present study, the extent of EZ disruption at 12 months was positively correlated with the area ratio of SRD spaces. Large cysts in the ONL have a negative impact on photoreceptor cells and disrupt EZ integrity, which irreversibly damages visual function.31 Meanwhile, we found that extended IRCs adjacent to SRD spaces caused focal absence of EZ reflectivity on SD-OCT. Restoration of the EZ defect was significantly evident in patients with CME and was dependent on the pattern of DME on SD-OCT,32 which might explain the irrelevancy between EZ disruption at 12 months and the baseline area ratio of IRCs in the present study.
The ELM, which serves as a barrier against macromolecules and as a prerequisite for an intact EZ in DME,33–34 is formed by tight junctions between glial Müller cells and photoreceptor cells. Occludin is organized between glial Müller cells and photoreceptors and is a key component of tight junctions.34 The Müller glia intracellular edema36 and dysfunction37 in DME are indicated by the hyporeflective cystoid edema spaces on OCT images. In DME, the swollen glial Müller cells, as well as their lost occludin content, can lead to ELM disruption.37 The improved drainage function of glial Müller cells in response to anti-VEGF therapy, which is indicated by the decreased height of cystoid edema,38 might help to resolve ELM disruption. The present study showed that the baseline area ratio of IRCs is predictive of ELM disruption at 12 months in DME patients with extended IRCs and SRD who undergo anti-VEGF therapy.
Considering that the association of SRD and IRCs with EZ/ELM disruption determine visual outcomes, we assessed the value of baseline OCT characteristics in predicting the visual outcomes of DME patients with extended IRCs and SRD after anti-VEGF therapy. The predictive value of SRD for visual outcomes in patients with DME is controversial.39–40 However, our study showed that the baseline area ratio of SRD had positive predictive value for determining final visual outcomes. Large cysts did not offer any prognostic significance, except for cysts with ellipsoid zone loss or large cysts without intervening bridges,41–43 which might explain the absence of any correlation between the baseline IRC ratio and the logMAR BCVA in the present study. The relative importance of ELM and EZ integrity in predicting visual acuity is unclear. In the present study, baseline ELM status, but not baseline EZ, was positively correlated with the logMAR BCVA after anti-VEGF therapy.44 This contradicts a previous study demonstrating that the status of EZ integrity is the sturdiest predictor of final BCVA.45
This study has some limitations that should be noted. First, this study was a retrospective study with a small sample size, so the findings will need to be validated in a larger cohort in the future. During the 12-month period, the influence of cataract progression on the logMAR BCVA was difficult to exclude. The effect of other baseline OCT characteristics, including disorganization of the inner retinal layers and hyperreflective foci, on EZ/ELM disruption and visual outcomes remain to be investigated. Further prospective, large-sample studies are needed to evaluate the predictive value of these OCT biomarkers for determining visual outcomes in DME patients with extended IRCs and SRD.