The results of the present study clearly demonstrated a significant increase in the CMT until 3 months after cataract surgery in both diabetic and non-diabetic eyes. Moreover, there was a significant increase in the aqueous flare value after cataract surgery in diabetic eyes; this confirmed the findings of previous reports [3,7,8]. A recent large real-world study indicated that the incidence of CME peaked at 3-6 months after surgery in patients with diabetes; this finding is nearly consistent with our finding, although previous studies have reported that CME developed within 4-6 weeks after cataract surgery in most cases [16,17]. From these findings, we reconfirmed persistent postoperative inflammatory activity even in minimally invasive surgery with a small incision performed by experienced surgeons.
The aqueous flare values were significantly higher in the patients with diabetes than in those without diabetes during the follow-up period, and the CMT values tended to be higher in patients with diabetes than in those without diabetes; however, the difference was not significant. As previously reported , antiphlogistic eye drops are considered useful and necessary for reducing postoperative inflammation in the anterior chamber, particularly in patients with diabetes. Some researchers stated that, in patients with diabetes, the incidence of CME after cataract surgery, even in the absence of diabetic retinopathy, is more frequent because the BRB is already damaged before surgery . Several reports suggested that, in diabetic eyes, the elevated levels of various inflammatory mediators in the aqueous humor may play a role in the breakdown of the BRB [38,39]. However, in the present study, the effect of increased aqueous flare on the retina was considerably small. Minimally invasive, uncomplicated surgery conducted by an experienced surgeon induced minimal changes in the BRB, and diabetes had almost no effect on the retina. We also found that the baseline CMT value was not significantly different between the diabetic and non-diabetic groups. This indicates that the effect of diabetes on the retina and BRB was not apparent from the beginning. In the retina that is minimally affected by diabetes, the influence of surgery may be similar to that in the retina without diabetes. Furthermore, the mean HbA1c level of patients was 6.95%±0.76%, and diabetes was relatively well controlled with mild or no retinopathy in all patients. These factors may have resulted in the lack of significant differences in the CMT between the two groups.
The main finding of our study was the lack of a significant difference in CT after cataract surgery in both groups and between diabetic eyes and non-diabetic eyes during the follow-up period. Our findings confirmed the correlation between CT and cataract surgery reported by Brito et al.  and Falcão et al.  In these studies, cataract surgery did not induce choroidal changes, and the CT was free of the inflammation of the anterior chamber and inner retina caused by BRB disturbance. The absence of change in CT might be due to the fact that the retinal capillaries have a BRB and are autoregulated, whereas the choroidal capillaries are not autoregulated and thus behave differently from the retinal vessels.
Although there are many reports on CT in diabetic patients, the effect of diabetes on the choroid remains to be established [26-29,31]. The general consensus is that diabetes causes a reduction in CT. Some authors have suggested that reduced CT in diabetes is due to the loss or drop out of the choriocapillaris, relative vasoconstriction, and reduced blood flow into the choriocapillaris . Choroidal vessels provide nutrients to the RPE and outer retinal layers, and the reduced CT at the fovea, accompanied by retinal hypoxia, may cause the onset of diabetic retinopathy . Retinal hypoxia increases vascular endothelial growth factor expression in RPE, pericytes, and microvascular endothelial cells, and may result in breakdown of the BRB, which induces diabetic retinopathy or maculopathy [27,41,42].
We revealed that while the SCT tended to be thinner in patients with diabetes than in those without diabetes throughout the follow-up period, the difference was not statistically significant. A possible reason for this result is that the eyes included in this study had no retinopathy or were in the early stage of retinopathy. In the early stage of diabetes, the diabetic angiopathic changes tend to be less severe in the choroidal capillaries and therefore cannot be detected by SD-OCT . Lee et al suggested that the functional integrity of the choroid in diabetic retinopathy is preserved after initial damage because the choroid is relatively resistant to the effects of diabetic retinopathy owing to a surplus of choroidal vessels and sufficient blood flow .
Odrobina et al reported that the SCT in eyes with postoperative CME was significantly thinner than that in fellow eyes without CME . Moreover, other researchers have reported that the choroid is also thinner in eyes with DME than in eyes without DME [36,44]. Odrobina et al speculated that the mechanisms may be similar to those in diabetic retinopathy. Since there is no retinal vasculature in the foveal region, impairment of the choriocapillaris may cause severe functional damage to the retinal tissue in the fovea.
In this study, none of the patients developed CME during the follow-up period, and the CMT increase was relatively small. The CMT values tended to be higher in patients with diabetes than in those without diabetes; however, the difference was not significant. Furthermore, 26 of 33 eyes of patients with diabetes did not exhibit retinopathy. In addition, the sample size of this study was relatively small. These factors may explain why no significant changes were observed in the SCT before and after surgery. We considered that in patients with diabetes whose retinopathies were in the relatively early stage, cataract surgery did not affect the SCT and the choroidal influence on the retina was undetectable.
This study had limitations such as a short follow-up period and a relatively small sample size. There were also measurement errors in the analysis of the SCT, as the SCT measurements were performed manually. In addition, we analyzed only the mean SCT and mean CMT, leading to a lack of segmentation of these parameters. There is great variability even in the SCT of healthy eyes. Studies of healthy eyes have revealed variations in CT with age, axial length, and even time of day, with diurnal variation in the SCT . We excluded the eyes with the shortest and longest axial lengths, and all examinations were performed at almost the same time of day; however, we could not exclude all of the parameters that may affect SCT measurement.