In myopic eyes, MHRD is one of the most severe disorders that threaten vision [1,2,3,4, 18]. Michalewska et al. [15, 16] introduced the inverted ILM flap technique to facilitate MH closure and improve visual outcomes in large MHs. Kuriyama et al. [4] extended the technique to the treatment of MHRD. Using the inverted ILM flap technique, the MH closure rate increased to 75%–100% [4, 14, 28]. Blood components have recently been explored as an adjuvant to achieve better MH outcomes in eyes with large or refractory MHs [17,18,19,20,21]. Lai et al. [18] developed a novel approach using ILM flap technique combined with ABC and reported MH closure rates of 96% when treating MHRD. These reports hypothesized that an ABC could not only act mechanically as a glue to enhance adhesion of the ILM and reduce the risk of postoperative flap dislocation, but also act physiologically as a potential autologous adjuvant to augment healing processes [17,18,19,20, 22]. However, no studies have compared the anatomical or functional outcomes of the ILM flap technique without or with an ABC in patients with MHRD. Therefore, our study was performed to address this gap.

The inverted ILM flap technique improved MH closure in patients with MHRD [4, 14,15,16]. ILM mechanically compensates for the shortening of the retina in the posterior pole [29], serves as a scaffold for astrocytes and Müller cell gliosis [15], provides activated Müller cells and neurotrophic factors to encourage MH healing [30], and facilitates glial cell proliferation to result in a favorable foveal configuration [31]. MH closure was achieved in 69 eyes (86%) in our study. Postoperative OCT revealed that the rate of MH closure with HBT was greater in the ILM flap with ABC group than in the ILM flap group, with borderline statistical significance. The slightly elevated hyperreflective tissue representing the macular plug after surgery with blood application was observed in several studies [18, 21]. A flat flap may promote MH closure with a relatively normal foveal contour [4, 14,15,16, 21]. However, blood application might have a negative effect [18, 20]. The underlying mechanism may involve the following: autologous blood contains various growth factors (GFs), such as vascular endothelial GF, platelet-derived GF, and epidermal GF, which facilitate tissue regeneration [17, 20, 22]. These GFs activate multiple transduction signal pathways when in contact with disintegrated neuroretinal tissue [22] to promote MH healing. They increase proliferation, differentiation, and migration of retinal glial cells and RPE cells [20, 22], which are crucial for retinal repair, angiogenesis, and regeneration [17, 20, 22]. The trophic factors and GFs also coordinate various communication pathways, including the fibrin formation network, to restore retinal integrity, and thus provide a provisional extracellular matrix for harboring retinal cells [32]. ILM peeling shaves the basal membrane of Müller cells, thereby acting as a stimulus for proliferation [19]. The ILM flap provides sufficient fibrosis and induces prolonged foveal glial proliferation [4, 14,15,16, 21]. After blood application, the ILM flap and ABC mixture form a macular plug that completely seals the entire MH (Fig. 3). The GFs in autologous blood further facilitate glial and fibroblast proliferation with the help of ILM acting as a bridge [18, 20, 21]. Excessive gliosis, induced by these synergistic effects, could harm retinal neurons and promote fibrotic tissue formation [20, 21], which may impair reconstitution of the foveal microstructures and hinder functional recovery [18, 21]. Furthermore, the excessive gliosis that fills the MH may prevent reconnections among the retinal layers and affect the recovery of microvascular blood flow [21]. In turn, inadequate microvascular blood flow may further hinder restoration of the foveal microstructure [21]. Moreover, the migration of blood into the subretinal space may release free radicals and proinflammatory substances that affect the regeneration of retinal neurons [33]. Given that the use of the inverted ILM flap alone already provides good stability, ABC could be more beneficial for recurrent MHRD in which an intrinsic instability of a free ILM flap or an autologous retinal transplantation flap may require the adjunction of ABC.

Representative ultra-widefield fundus photography and optical coherence tomography (OCT) images demonstrating the postoperative follow-up of a patient who underwent the inverted internal limiting membrane (ILM) flap technique with an autologous blood clot. a1, a2 Ultra-widefield fundus photography and OCT revealed type II macular hole (MH)-associated retinal detachment before surgery. The visual acuity was counting fingers at 30 cm. b1 One day after the surgery, the blood clot is layered over the inverted ILM flap to form a macular plug closing the MH. b2 The retina was well attached. c1 The blood clot persisted for 1 month after surgery. c2 The ILM flap and blood clot complex sealed the MH. d1 The blood clot was completely absorbed 2 months after surgery. d2 The MH was closed, and hyperreflective bridging tissue (HBT) appeared at the fovea. e1, e2 Four months after surgery, the size of HBT was stable and a defect in the outer retinal layer was visible. f1, f2 Eight months after surgery, the HBT remained, and the outer retinal integrity did not recover. The patient’s visual acuity at the final follow-up examination remained counting fingers at 30 cm

MH closure is pivotal to the final visual outcome in MHRD [4, 14,15,16]. In this study, the high anatomical success rate also resulted in favorable postoperative functional recovery, and vision was improved or remained stable in 70 eyes (86%) at the final follow-up examination. The ILM flap may act as a bridge for naked cones [28], facilitate the regeneration of Müller cells [28], and create a microenvironment conducive to the repositioning of photoreceptors in direct proximity with the fovea [15], all of which contribute to visual improvement [34]. Here, we compared the visual outcomes of the ILM flap technique with or without ABC for MHRD. The postoperative improvement in BCVA was significantly better in the ILM flap group than in the ILM flap with ABC group. Multiple regression analysis revealed that the preoperative BCVA was positively associated but the inverted ILM flap technique with ABC was negatively associated with the postoperative improvement in BCVA. There are some possible explanations for this. First, combining the inverted ILM flap technique with blood application likely stimulated gliosis through a synergistic process more strongly than the ILM flap technique alone [18, 20, 21]. The increased risk of excessive gliosis could have a detrimental effect by impairing reconstitution of the foveal microstructures and create unfavorable conditions for photoreceptor repositioning [21, 35], hindering further visual recovery [18]. The contraction of excessive glial tissue may cause centripetal movement of photoreceptors [19]. The excessive glial tissue may also increase the risk of cytotoxic damage to retinal ganglion cells [36]. Second, the leakage of blood into the subretinal space may have a toxic effect on photoreceptors due to the release of free radicals and proinflammatory substances that hamper the regeneration of retinal neurons [33]. Third, the persistence of the macular plug [18], which mainly consists of excessive glial tissue, may cause central visual disturbances, scotomas, or distorted vision [14]. In addition, the patients in the ILM flap with ABC group complained of central visual occlusion that lasted for more than 1 month, until the blood clot was completely resolved. Therefore, we think that the ABC not only leads to a smaller improvement in BCVA in the long term but also has a short-term impact on postoperative visual quality.

Regarding the surgical procedure, other than applying blood as an adjuvant, the major modification in this study involved using an inverted ILM flap to cover the surface of the MH instead of inserting the ILM to fill the MH. This modification could prevent the instruments from touching the RPE and choroid, reducing the risk of severely damaging these vulnerable tissues in highly myopic eyes [18].

Several limitations of our study should be addressed. First, the follow-up period was probably too short to observe full recovery of the foveal microstructures. The functional recovery may occur gradually, over several years after MH surgery [20]. Second, this was a retrospective observational study, and the use of ABC was not randomly selected. Thirdly, it was difficult to evaluate the extent of myopic degeneration before the formation of MHRD, which may confound the visual outcomes of the two techniques. Finally, fresh blood is more visible and accessible to manipulate. Other potential blood components, such as platelets, could be explored in future studies.