Longitudinal changes in retinal microvasculature after panretinal photocoagulation in diabetic retinopathy using swept-source OCT angiography

Longitudinal changes in retinal microvasculature after panretinal photocoagulation in diabetic retinopathy using swept-source OCT angiography

The pathophysiology of early DR, prior to the development of significant vasculopathy, has not been clearly defined. Several studies indicate that DR progression is accompanied by significant retinal microvascular changes and neurodegeneration14,15. However, the correlation between these pathologies and their contribution to clinically visible retinopathy is still unclear. In the early stages of DR, hypoperfusion may contribute to pro-inflammatory changes such as leucocytic adherence to the retinal capillaries and vaso-occlusion; resulting in capillary dropout and the development of a progressive, irreversible ischemic hypoxia16. Although FA remains the gold standard for the visualization of the retinal neovascularization in PDR, the mechanism of large vessel hemodynamic alteration affecting macular microvascular distribution, has not been established yet. Recent advances, like higher-axial-resolution OCTA, have the potential to generate superior axial resolution and the capability to segment the microvascular network from different retinal layers. The OCTA system has demonstrated its superiority in quantification of microvascular density, capillary dropout, and FAZ area in diabetic patients17. Moreover, several studies have reported the OCTA parameters of macular perfusion state in the different stages of DR, with or without diabetic macular edema (DME)18,19.In argon laser photocoagulation, the laser energy is absorbed by the retinal pigment epithelium (RPE) and generates thermal energy to the outer retina. Heat damage in the inner retinal layers causes edema due to increased vascular permeability and presents as increased retinal thickness20. As expected, the authors observed a mild increase in macular thickness during the 12 months post PRP. Previous studies have demonstrated large vessel constriction and overall blood flow decrease, following the treatment of PDR with PRP, by means of a host of techniques4,9,10. Color Doppler imaging, laser interferometry, and laser Doppler flowmetry have been used to obtain quantitative data on retinal and choroidal circulation in eyes with DR, before and after PRP21. In addition, laser speckle flowgraphy has been used to measure the reduced retinal blood flow, during and 6 months after the PRP22. However, previous reports on choroidal blood flow at the macula area, following PRP treatments, have been contradictory21,23.Recent advances like OCTA permit the quantification of macular capillary modulations across the superficial and deep layers, in response to PRP. This finding has not been reported by previous studies, which were limited to the study of large vessel hemodynamics. Therefore, we performed serial follow-up OCTA imaging to compare with baseline measurement by 12 months post PRP. Fawzi et al.24 reported an overall increase in the OCTA flow metrics of all capillary layers in the macula, 6 months after PRP. Mirshahi et al.25 reported that foveal vascular density, measured using OCTA, increased 3 months after PRP. In the current study, decrease of PD and VLD, 1 month after PRP implies the effect of acute inflammation of the retinal tissue, subsequent to laser treatment. Consequently, we observed an overall increase in PD and VLD at both capillary plexuses, 12 months after PRP and there was a significant difference, when compared to the measurements taken 1 months after PRP. Possible underlying mechanism for the improved flow in the remaining macular capillaries could be re-establishment of macular microvasculature, from regression of peripheral neovascularization or intraretinal microvascular abnormalities (IRMA). A representative case of PDR treated with PRP was shown in Fig. 3A. As mid-peripheral neovascularization and IRMA disappeared, capillary perfusion improved and NPA decreased after 6 months (Fig. 3B). Conversely, our analyses did not reveal any significant alteration in the FAZ area following PRP, which was consistent with previous literature25,26. Lorusso et al.26 in a similar study, investigated the alteration of OCTA parameters following PRP. Contradictory to our results, they did not observe any changes in the PD and FAZ area. This discrepancy can be due to the different methods employed to measure OCTA parameters and the diverse follow-up periods. Primarily, we focused on the macular microvascular status, without considering the well-known vasoconstrictive effect on larger vessels, following PRP treatment. By excluding the major branches of retinal vessels on OCTA, we calculated the capillary density selectively as the percentage of vascular voxels on en face projection angiograms. Furthermore, the duration of follow-up in the current study extended up to 12 months after PRP, compared to previous studies with relatively shorter duration of follow-ups, such as one to 3 months after PRP. In the present study, the measured values of PD and VLD increased continuously, across the 12-month follow-up period, and the difference was found to be statistically significant.Figure 3(A) Representative SS-OCTA images for PDR cases that had mid-peripheral neovascularization and IRMAs (red circles). (B) Three months after PRP treatment, neovascularization and IRMAs disappeared and re-established with retinal capillaries. Overall nonperfusion area (blue color coded) reduced and capillary perfusion density and vessel length density improved after PRP.The exact mechanism of the early decrease in PD and VLD, following PRP, is still unknown. One possible explanation for this reduction is that the capillary flow following PRP might be associated with early PRP-induced inflammation. After sufficient time, the inflammation probably subsides and subsequently, the resumption of the autoregulatory functions of retinal vasculature occurs. PRP is known to cause the upregulation of pro-inflammatory cytokines, such as intercellular adhesion molecule-1 (ICAM-1) and monocyte chemotactic protein-1 (MCP-1), vascular endothelial growth factor (VEGF) which in turn stimulates vascular permeability and capillary dropout27. Hence, it can be naturally hypothesized that these changes negatively add up to the pre-existing condition of DR associated maculopathy, further reducing the perfusion of the posterior pole, increasing the macular ischemia and enlarging the FAZ area. Qualitative and quantitative abnormalities such as reduced perfusion within the posterior pole and increase of the macular ischemia have been described in subjects with uveitis, in recent OCTA studies, which demonstrated a significantly lower parafoveal capillary density in the SCP layer of eyes with uveitis, when compared to healthy eyes28. Abnormalities in retinal vasculitis included capillary dropout or loss of the SCP layer29. Flow indices in perifoveal area were decreased with OCTA, suggestive of ischemia; this correlated with a greater severity of Birdshot chorioretinopathy30.A recent study reported the longitudinal outcomes regarding the changes in NPA, following anti-VEGF treatment in DR. Reperfusion of vessels or capillary network was not detected in NPA using two imaging techniques (FA and SS-OCTA), following three anti-VEGF injections, in eyes with DR31. Similarly, in this study, PRP did not significantly change NPA in 12 × 12 mm field of SS-OCTA, across the 12 months after PRP.The optimal laser protocol for PRP and the recommended end point for laser treatment have not been determined yet, because the reaction to PRP varies from patient to patient. Some earlier studies have tried to determine the prognostic factors, which influence the clinical outcomes of PRP. Grunwald et al. suggested using the restored response to the hyperoxic challenge as a tool to determine the success of PRP24. Hammer et al.32 demonstrated a trend towards lower retinal venular oxygen saturation in PDR patients who underwent PRP, compared to treatment-naïve PDR. They also suggested that retinal oxygen saturation could act as an early indicator of PRP treatment response and can be a valuable tool for the individualized treatment33. Furthermore, it was found that increasing retinal vascular arteriolar oxygen saturation, following PRP, trended towards a lower risk of active PDR34. Earlier studies have found a decrease in fractal dimension in patients with PDR, who underwent PRP35. Torp TL et al.36 found an increase in the arteriolar vessel caliber, which could represent a positive response to the PRP treatment, with a lower hypoxic load on the retina and a subsequent autoregulatory arteriolar dilation. The present study demonstrates that an increase in the parafoveal PD 6 months after PRP, can independently predict the progression of PDR, and might be a potential non-invasive OCTA marker of DR activity.A major limitation of this study is the relatively small sample size, which may not be sufficient to find a definite temporal relationship between PRP and microvascular changes. Moreover, the current study used 3 × 3 mm scans for the calculation of PD and VLD, which were not wide enough to include the vascular arcades or the entire posterior pole. Rabiolo A et al.37 reported that interrater reliability for the quantified vessel density was relatively poorer in 6 × 6 mm, 12 × 12 mm than 3 × 3 mm scan in DR. The decrease of image resolution to larger angiocubes could have been strongly affected by the visualization of small capillaries or blocking artifacts, especially when a threshold is applied. Future research, which employs SS-OCTA devices with higher scan speed, density, with wider range of scans might provide more reliable information on the subject. Finally, the effect of PRP can vary considerably, depending on undetermined factors including spot size, number and the extent of PRP. Therefore, all the procedures were performed by a single operator, in order to minimize the effect of confounding factors in the study.In summary, the current study observed that PRP treated eyes showed significant changes of macular capillaries, with increased PD and VLD at both capillary layers, suggesting an effective perfusion of the posterior pole. We assume that redistribution of blood flow from the periphery to the macular region results in reorganization of capillary networks. In view of the fact that the postoperative changes of macular PD closely reflect the disease activity, an individualized treatment approach might be available. This might be helpful in determining the necessity of additional PRP or anti-VEGF injections, in the early phase after PRP, to prevent the progression of DR.

Via Source link