Diabetes mellitus (DM) is an important a health problem that threatening societies today. It is estimated that 463 million people have DM in 2019 and could reach 578 million by 2030 and 700 million by 2045 (20). DME is a common loss of vision in diabetic patients and develop in 7% of patients (21). Risk factors for the development of DME are generally similar to those of diabetic retinopathy (22).
The most widely accepted priority treatment for DME today is injection of intravitreal anti-vascular endothelial growth factor (anti-VEGF) drugs (23). As the availability of more user-friendly microsurgical instruments and high incision speed devices increases, there is a tendency for surgical intervention to diabetic retinopathy and DME in earlier stage. PPV surgery is thought to have positive effects on DME therapy by reducing VEGF concentration in the eye and increasing oxygenation or nutrient diffusion of the retina (24, 25).
In Academy, the benefit of ILM peeling for the patients undergoing PPV for DME is still a dilemma (26, 27). In terms of visual acuity, there are publications that argue ILM peeling is beneficial (28), useless (8, 11) and harmful (15) in DME. Khurieva-Sattler et al. compared ILM peeling or intravitreal triamcinolone acetonide (IVTA) injection in addition to PPV in patients with diffuse DME in a non-randomized prospective study and reported more permenant results in terms of visual and anatomical improvement in the group whose ILM was peeled after the 4th month (28). Figueroa et al. investigated the surgical and anatomic results of PPV in patients with diffuse nontractional DME. They compared ILM peeling and IVTA injection in patients undergoing PPV and reported that the anatomical and functional success achieved in the early period in both groups did not continue in the long term and there was no statistically significant difference between the groups (8). In a controlled clinical study, Kumar et al. compared ILM peeling or grid laser application in addition to PPV in patients with diffuse DME. Although there was no difference between the groups in the first 6 months, they reported that the decrease in foveal thickness and macular volume was statistically significantly higher in the group with ILM peeling in the following period (15). Aboutable reported in a study on patients with diffuse DME without epimacular membranes, ILM peeling reduced foveal thickness and did not improve visual acuity improvement compared to non-peeling (11). Bahadir et al. reported in their randomized controlled study that ILM peeling did not cause a significant improvement in visual acuity compared to non-peeling in patients who underwent PPV for DME treatment (16). The results of our study indicate that ILM peeling during PPV is useful for macular function and may accelerate improvement in visual acuity. One-year results of our study showed that the improvement in BCVA in the first and third months was higher and the need of IVD implants postoperatively was lower in eyes with ILM peeling.
Clinical studies in recent years have shown that frequent anti-VEGF injection therapy in the treatment of DME patients rebust efficacy and safety (29–31). There is limited evidence today about the benefit and risk of continuous injections of anti-VEGF in eyes that have not responded adequately to anti-VEGF therapy before (32). However, it is still unclear in academy how many times anti-VEGF injection had to be made in order to talk about resistance in DME management. In our study was defined resistant DME as the minimal improvement in central retinal thickness (< 15%) to at least 4 treatments (three of which were anti-VEGF injection) in last 6 months (18). In patients with proliferative diabetic retinopathy, the risk of intraoperative and postoperative hemorrhage has been reported to be reduced with intravitreal anti-VEGF treatment 3 days before the operation (33–35). For these reasons, we also applied intravitreal anti-VEGF injection 3 days before PPV to all cases in our study.
In our study, we preferred IVD implant injection in the treatment of resistant DME. IVD implant nowadays could be used for the treatment of pathologies that may have inflammation in the pathogenesis such as central retinal vein occlusion, posterior non-infectious uveitis and macular edema due to diabetic retinopathy (36, 37). Inflammation could play a critical role in the pathogenesis of DME. Many studies have shown that in patients with DME, proinflammatory mediators (cytokines, chemokines, growth factors) can be found in higher concentrations in aqueous humor and vitreous gel (38, 39). These findings may explain the effectiveness of IVD implant with anti-inflammatory properties in the treatment of DME. There are publications in the literature reporting that IVD implant injection in eyes that had previously undergone PPV was not associated with an increased risk (39–42). There are studies report that combined application of vitrectomy and IVD implant is safe and effective in terms of anatomical and functional improvement in DME treatment (43, 44). We have never faced any uncontrolled increase of intraocular pressure in our patients. Castro-Navaro et al. reported that IVD implant is effective even in resistant cases in DME treatment (45). The results of our study also support this situation. After PPV operation IVD implant injection alone was successful in almost all cases with resistant DME.
Other than DME, ILM peeling is often applied to reduce the force of tangential traction on the retinal surface for the treatment of macular hole or pucker. In recent years, many clinical studies have been published in the literature reporting that ILM peeling might be harmful for macula. Halfter et al. showed that the absence of ILM triggered irreversibly retraction of the endfeet of the neuroepithelial cells from the inner surface of the retina and the formation of an irregularly thickened ganglion cell layer (46). In a controlled randomized study conducted by Ripandelli et al. investigated the anatomical and functional results of ILM peeling in patients with idiopathic macular pucker. At the end of the one-year follow-up period, they reported that, the mean retinal sensitivity in the central region was higher and showed faster recovery in the non-peeling group, but the absolute number of microscomas was higher in the peeling group (14). In our study, the improvement in BCVA in the first and third months in the ILM peeled group was statistically significantly higher than ILM non-peeled group. In addition, the mean IVD requirement was statistically significantly higher in the ILM non-peeled group. Our findings oppose the hypothesis that ILM peeling may damage the anatomy and function of the macula with resistant DME.
This study has some limitations. Firstly, small sample size and short follow-up time can be meaningful. Second, considering ‘resistant DME’ as non-responding to 4 consecutive treatments alone is a controversial subject. Long-term randomized clinical trials with comprehensive outcomes are needed to evaluate the safety and effectiveness for the treatment of resistant DME.