In this study, we have reported similar effectiveness after administration of the 0.7-mg DEX implant in non-vitrectomized and vitrectomized eyes. The mean foveal thickness and mean SFCT both improved from baseline by 1 month after treatment with a DEX implant, and the improvement remained statistically significant throughout the 3-month study, in both groups. There was no difference between foveal thickness and BCVA logMAR between non-vitrectomized and vitrectomized eyes from baseline to month 3; however, there were statistically significant differences in SFCT between non-vitrectomized and vitrectomized eyes at month 3 (the difference of least square means: 31.07 (standard error: 7.41; 95% confidence interval [16.56, 45.59]), P < 0.0001) and the trend from baseline to month 3 (the difference of least square means: 10.57 (standard error: 2.58; 95% confidence interval [5.52, 15.62], P < 0.0001). The peak effectiveness for the vitrectomized group was during the first and the second month by the value of SFCT. The slope of the SFCT would ascend from month 2 to month 3 in the vitrectomized group, while the value at month 3 was still significantly different compared with the baseline.
The following discussion focuses on these three components: the pharmacokinetics after vitrectomy, Ozurdex®, and SFCT.
1. The pharmacokinetics after vitrectomy
When the eye became vitrectomized, the vitreous humor would be removed, and less-viscous liquid would fill the space, increasing intravitreal circulation. The pharmacokinetics of the delivered drug would be significantly changed, with the clearance rate being faster in the vitrectomized group. These findings suggest that sustained drug delivery with an implant such as Ozurdex® could be particularly useful in vitrectomized eyes for treating DME.
2. Ozurdex®
Corticosteroids have raised interest in the treatment of DME due to their anti-inflammatory effects and because they inhibit the synthesis of VEGF and reduce vascular permeability. The Ozurdex® implant does not need to be removed as the copolymer slowly biodegrades into carbon dioxide and water and is absorbed over time. Chang-Lin et al previously published their results on the pharmacokinetics and pharmacodynamics of Ozurdex® [10]. DEX was detected in the retina and vitreous humor for 6 months, with peak concentrations during the first 2 months. After 6 months, DEX was below the limit of quantitation; however, the pharmacokinetics and pharmacodynamics of Ozurdex® in vitrectomized eyes should be discussed, which concerns the duration of every injection to control the DME well. Iglicki et al. investigated the efficacy of intravitreal dexamethasone (IVD) implant in DME and reported that the improvement in visual acuity was more pronounced in naïve eyes when compared with the eyes refractory to anti-VEGF drugs [11]. In another study, Iglicki et al. investigated the long-term effect of the IVD implant on the severity and progression of non-proliferative DR and reported that both the progression and the severity of DR and proliferative DR decreased in eyes which received DEX injection [12]. Kim et al. investigated the effect of the IVD implant on eyes with DME resistant to anti-VEGF drugs and reported that the implant causes an increase in BCVA and a decrease in the central fovea and SFCT thickness [13].
3. Sub-foveal choroidal thickness (SFCT) change
Studies investigating biomarkers and predictive factors in DME have been conducted as reported in the literature [14], revealing that the absence of sub-retinal fluid, hyper-reflective foci and integrity of the inner segment-outer segment layer could be OCT biomarkers for better functional success in eyes that received IVD implant injection for DME treatment [15]. As for the choroid, it is the nutrient source of the outer retina and its structure consists of blood vessels. It provides thermoregulation, affects the retina due to changes in its thickness while regulating the secretion of various mediators [16]. These mediators could be the VEGF and proinflammatory cytokines that might also play important roles in the pathogenesis of DME. In a prospective study, Mathis et al. reported that the increase in choroidal thickness in eyes with DME is an indicator for exudative recurrence that regressed with anti-VEGF or IVD implant injection [2]. In the pathogenesis of the active phase of DME, not only VEGF but proinflammatory cytokines might also have a role, which could cause vasodilatation and increase choroidal thickness [17].
In our study, a statistically significant thinning developed in the mean foveal thickness and SFCT in the follow-up period after IVD implant injection in vitrectomized and non-vitrectomized diabetic eyes with DME. These results might have occurred due to the decrease in vasodilator effect of proinflammatory cytokines and mediators.
However, the SFCT showed a trend of ascent in the vitrectomized group from month 2 to month 3 in our study, which was not found in foveal thickness and BCVA logMAR. We opined that even under DEX treatment, the faster clearance rate in the vitrectomized group provided a hint that the effect of DEX will disappear earlier, which allowed the choroidal vessels to thicken due to the VEGF-mediated vasodilation of the choroidal vessels. Choroid conserves its capacity for expansion, acting as a vascular reservoir, called on at the time of recurrence. Because of the farther distance of the choroid to the vitreous cavity than fovea, DEX will lose its “protective effect” to the choroid first; therefore in our study, SFCT showed early trend of increase than did foveal thickness from month 2 to month 3.
Small sample and short follow-up period are the limitations of this study. Moreover, previous studies have shown that SFCT might decrease after the PRP [18]. To avoid any choroidal modifications associated with laser photocoagulation, patients who underwent PRP and focal laser therapy within the last 3 months were excluded before and during the study. SFCT was measured manually, but we had no access to an automatic analysis software program, although a high level of reproducibility was noted between three blinded observers, and re-evaluation took place if any value was too extreme. More comprehensive long-term randomized clinical studies are needed to investigate the effect of dexamethasone implant on choroidal thickness.