Reconstructing Foveola by Foveolar Internal Limiting Membrane Non-Peeling and Tissue Repositioning for Lamellar Hole-Related Epiretinal Proliferation

Differences in the pathogenesis and clinical characteristics between lamellar macular hole (LMH) with and without LMH-associated epiretinal proliferation (LHEP) can have surgical implications. This study investigated the effects of treating LHEP by foveolar internal limiting membrane (ILM) non-peeling and epiretinal proliferative (EP) tissue repositioning on visual acuity and foveolar architecture. Consecutive patients with LHEP treated at our institution were enrolled. The eyes were divided into a conventional total ILM peeling group (group 1, n = 11) and a foveolar ILM non-peeling group (group 2, n = 22). In group 2, a doughnut-shaped ILM was peeled, leaving a 400-μm-diameter ILM without elevated margin over the foveola after EP tissue repositioning. The EP tissue was elevated, trimmed, and inverted into the LMH. Postoperatively, the LMH was sealed in all eyes in group 2, with significantly better best-corrected visual acuity (−0.26 vs −0.10 logMAR; p = 0.002). A smaller retinal defect (p = 0.003), a more restored ellipsoid zone (p = 0.002), and a more smooth foveal depression (p < 0.001) were achieved in group 2. Foveolar ILM non-peeling and EP tissue repositioning sealed the LMH, released the tangential traction, and achieved better visual acuity. The presumed foveolar architecture may be reconstructed surgically. LMH with LHEP could have a combined degenerative and tractional mechanism.

management 7,9,13 . One paper reported a worse visual outcome in patients with LMH and LHEP after surgery 8 , and another reported a positive correlation between the dimensions of the LMH and the area of the LHEP 14 . Postoperative retinal defect, poor recovery of the ellipsoid zone, and even formation of a macular hole have been reported 1,9,15 . It is possible that conventional total ILM peeling does not improve the surgical outcome because of poor reconstruction of the foveolar architecture.
The differences in pathogenesis and clinical characteristics between LMH with LHEP and LMH without LHEP are likely to have surgical implications. In this study, we reviewed the results of vitrectomy for ERM and ILM peeling for LMH with LHEP. We hypothesised that a rational approach of filling the retinal defect with EP tissue and preserving the foveolar ILM but peeling off the surrounding ILM using a foveolar ILM non-peeling technique may improve the surgical outcome.

Results
There was no significant difference in age, sex, or duration of follow-up between the group that underwent conventional total ILM peeling (group 1) and the group that underwent foveolar ILM non-peeling (group 2; Table 1). No patient developed full-thickness macular hole during the postoperative follow-up period.
Surgical outcomes. The preoperative foveal OCT configurations showed no significant difference in the intact ellipsoid zone, retinal defect, or smooth foveal depression between the two groups (Table 1). Postoperative recovery of the ellipsoid zone was significantly better in group 2 than in group 1 (p = 0.002). In group 2, the persistent retinal defect was significantly reduced (p = 0.003) and more postoperative foveal depression was regained (p < 0.001). Figure 1-1 to 1-4 shows the case of a patient with LMH and LHEP in group 2 who underwent foveolar ILM non-peeling surgery with repositioning of LHEP tissue and regained smooth foveal depression without a retinal defect. Figure 1-5 shows the OCT images for a further patient 2 months and 6 months after foveolar ILM non-peeling with LHEP repositioning surgery and demonstrates gradual recovery of foveal depression, a preserved foveolar ILM, and recovery of the ellipsoid zone and the line of the external limiting membrane. The retinal defect disappeared gradually in this case. Figure 2 shows the case of a patient in group 1 who had poor recovery of the ellipsoid zone and a persistent retinal defect.
We observed 1 pseudophakic eye in group 1 and 2 pseudophakic eyes before the surgery. Phacoemulsification with intraocular lens implantation was performed in 2 eyes in group 1 and 4 eyes in group 2 because of the development of cataracts that hindered visual improvement after the vitrectomy surgery.

Discussion
In this report we demonstrate that a combination of LHEP tissue repositioning and foveolar ILM non-peeling is an effective surgical treatment for LMH with LHEP. Our technique may provide a solution for the distinct entity of LMH with LHEP and may explain the variability in visual outcome previously reported after vitrectomy in patients who have LMH without LHEP. Previous reports have shown that conventional total ILM peeling produces worse final visual acuity and more postoperative disruption of the ellipsoid zone in eyes with LMH and LHEP than in eyes that have LMH without LHEP 8,13,16 .
There are two possible explanations for why this combination of surgical techniques is more effective than the LHEP repositioning technique alone for treating LMH with LHEP. The first explanation is that the foveolar ILM non-peeling technique ensures that the LHEP tissue remains securely in position after ILM peeling. If a conventional ILM peeling is performed, the repositioned tissue may be taken away completely or partially along with the ILM that has been peeled off. The second explanation is that this combination of surgical techniques might be more effective because the 360 degrees of tangential traction around the LMH is removed evenly. Given that LHEP is mainly composed of glial cells 6 , which are thought to play a major role in healing of the macula when repositioned into the retinal cleavage 17 , complete release of all the surrounding tangential traction would facilitate  Figure 1-5A-1-5D shows a second case. The first was a 71-year-old man who had a lamellar macular hole (LMH) with lamellar hole-associated epiretinal proliferation (LHEP) in the right eye and a best-corrected visual acuity (BCVA) of 0.2 measured 2 years and 5 months preoperatively. The central foveal thickness (CFT) was 265 μm (1-1A, 1-1B). Epiretinal tissue is shown by the arrow. One month before surgery, a retinal defect was present (1-2A). The arrow shows the proliferative tissue. The BCVA had decreased to 0.1. The CFT had increased to 290 μm (1-2A, 1-2B). One month after foveolar internal limiting membrane (ILM) nonpeeling and LHEP repositioning, the hole was well sealed with foveal depression (1-3A, 1-3B). The CFT was 317 μm. Three months postoperatively, the foveal depression was improved and the CFT had decreased to 258 μm (1-4A, 1-4B). The BCVA was improved to 0.8. The arrowheads mark the margin of the preserved ILM over the foveola (1-4A, 1-4B). The retinal defect was no longer present. A and B are horizontal and vertical optical coherence tomography scans, respectively. The second case (1-5A-1-5D) was a 56-year-old man who had LMHassociated epiretinal tissue with a disrupted ellipsoid zone (EZ) and external limiting membrane (arrowhead, 1-5A). He had a BCVA of 0.2 preoperatively. He underwent foveolar non-peeling internal limiting membrane (ILM) surgery with epiretinal tissue repositioning. Two months postoperatively, the LMH was well sealed with foveal depression, the EZ and external limiting membrane were growing toward the centre of the fovea, and there was a small retinal defect (arrowhead, 1-5B). The epiretinal proliferative tissue had been removed and the foveolar ILM was well preserved (the margins are marked by arrows, 1-5B). One further month later, the EZ and external limiting membrane has recovered almost completely and the retinal defect had disappeared (arrowhead, 1-5C). The BCVA improved to 0.4. Six months later, the BCVA improved to 1.0. The foveolar ILM was still well-preserved with a flat margin (arrows). The foveolar cone was restored gradually when compared with 1-5C with no retinal defect (1-5D). The EZ and external limiting membrane recovered (arrowhead). (2019) 9:16030 | https://doi.org/10.1038/s41598-019-52447-4 www.nature.com/scientificreports www.nature.com/scientificreports/ closure of the hole. The securely repositioned LHEP tissue and gas tamponade may act as a scaffold to promote healing of the glial cells. Our technique will ensure the preservation of the optimal amount of epiretinal proliferation left over during the operation. The traction from 360 degree around can be released completely and evenly so as to achieve a symmetric foveolar architecture. Both the management of epiretinal proliferative tissue and internal limiting membrane are essential in the restoration of foveolar architecture. In our study, smooth foveal depression was found in 86% of eyes treated by this technique.
In regard to the surgical technique, there has been a case report of an ILM inversion technique being used in combination with embedding of the LHEP 18 . When using that technique, an inverted ILM from upper to lower is made to cover the LMH after embedding the LHEP into the LMH. However, the tangential traction of the ILM surrounding the LMH is not released evenly by this technique. This is probably the reason why the foveal depression cannot be restored postoperatively. Furthermore, unlike with the technique devised by Shiraga et al. 17 , our technique involved the usage of an MVR blade to delineate the border of the non-peeling area when using our technique, which ensures preservation of the foveolar ILM. Shiraga's technique described a method by peeling the ILM without delineating the border. In his surgical video he peeled the ILM to the hole margin supposing that the ILM was left behind. The technique was likely to peel off some or all of the internal limiting membrane as well as some of the preserved proliferative tissue.
Recent advances in OCT have revealed that LMH can be classified into two types based on the presence or absence of pathological traction 19,20 . LMHs with pathological retinal traction are classified as tractional LMHs and those without it are classified as degenerative LMHs. LHEP has been thought to be found only in degenerative LMH, with one study reporting that LHEP was present in 80% of eyes with this type of LMH 20 . However, in our study, we found that if the tangential traction of all the ILM surrounding the LMH is released, the foveal depression can be restored evenly with a decrease in central foveal thickness (Fig. 1). Moreover, our present findings suggest that there is a tractional component in the pathogenesis of LMH. If only degenerative component is involved in the pathogenesis the release of the internal limiting membrane traction would not be necessary. The OCT images show incompliance of the inner retinal surface before surgery and released incompliance after surgery (e.g. comparing 1-1 B and 1-4B).
Unlike in some previous reports 1, 13 , we did not find a persistent postoperative retinal defect. There are two possible reasons for this: first, our technique released the tangential ILM traction evenly, allowing for even recovery of the preoperative retinal defect, and second, the repositioned LHEP tissue replaced the missing tissue in the retinal cleavage and blocked repair of glial cells after complete release of the tangential ILM traction. This effect had postoperative benefits following reconstruction of the foveolar architecture in our series.
The fovea itself is a 1500 μm depression in the centre of the macula. The central 350 μm of the fovea (known as the foveola) is located in a retinal capillary-free zone that has a diameter of about 500 μm [21][22][23] . Only photoreceptor (cone) cells and Müller cell processes are present in the foveolar area. Preservation of the foveolar ILM maintains the expanded vitreal processes of the Müller cell. The integrity of the fine processes of the Müller cells www.nature.com/scientificreports www.nature.com/scientificreports/ that envelop the neurons can be preserved to the greatest extent possible using our technique. Figure 3 shows a schematic drawing of the presumed Müller cell cones and foveolar architecture after reconstruction surgery as in the case shown in Fig. 1. The orange outline in Fig. 3A shows the Müller cell cones in the foveola first identified by Gass 22 . Figure 3B shows an OCT image acquired postoperatively for the case shown in Fig. 1  Morescalchi et al. 24 described a technique of peeling of the ILM with foveal sparing for treatment of degenerative lamellar macular hole. The technique adopted by Morescalchi F is different from our foveolar nonpeeling technique. Their technique is similar to Shimada's 25 and involves an ILM peeling starting from further away from fovea and peeled to the centre and trimming the floating flap of ILM to the size of one to two disc size. Their technique leaves the margin elevated and is much larger than ours (our size is smaller than the fovea avascular zone and is approximately 400 μ in diameter). Our technique leaves the foveolar ILM as a foveolar size and a flat margin. Our foveolar ILM nonpeeling is critical in reconstructing the foveolar architecture as a U-shaped foveal depression and layered retinal structures. In Morescalchi's article the cases they presented were shown to have extra and unsmooth proliferative tissue, retinal defect by OCT and micro-scotomas by retinal sensitivity testing post-operatively. The follow-up duration in their study was 6 months. We found that the foveolar structures restored and maintained after an average of follow-up period of 28 months.
There are several limitations to this report, stemming mainly from its retrospective observational design. Other limitations included the relatively smaller number of cases in group 1 and the intact ellipsoid zone, retinal defect, and smooth foveal depression are evaluated subjectively. Furthermore, we have described a technique that probably entails a learning curve. However, we have provided serial surgical photographs (Fig. 4) and a surgical recording (Supplementary Information, Video S1) for reference.
In conclusion, our results suggest that combining the LHEP repositioning technique with foveolar ILM non-peeling might be an effective treatment for LMH with LHEP. The findings of this research also suggest that ILM traction exists in this disease entity, despite it having been thought to be degenerative in nature.

patients and Methods
In this observational case series, we retrospectively reviewed the medical records of 31 consecutive patients who underwent vitrectomy for LMH with LHEP at our institution from January 2013 to December 2016. The study was approved by the National Taiwan University Hospital Research Committee and adhered to the tenets of the Declaration of Helsinki. All patients who participated in the study gave written informed consent after receiving an explanation of the nature of the study and its possible consequences. Patients who had LMH with LHEP underwent surgery if best-corrected visual acuity (BCVA) was worse than 20/40. The eyes were divided into 2 groups: Group 1 (11 eyes) received vitrectomy, EP tissue repositioning, total ILM peeling, and air-fluid exchange, and Group 2 (22 eyes) received vitrectomy, a doughnut-shaped ILM was peeled, leaving a 400-μm-diameter ILM over the foveola after EP tissue repositioning, and air-fluid exchange.
Basic demographic data were collected retrospectively by chart review for each patient. The data collected included age, sex, surgical history, operative methods used, and duration of follow-up. BCVA measurements were made using a Snellen acuity chart and converted to the logMAR scale for all data analyses.

Surgical technique.
A standard three-port pars plana vitrectomy was performed in all cases by the same surgeon (TCH) using a 23-gauge or 25-gauge vitrectomy system. After core vitrectomy, the posterior hyaloid was completely removed with or without the assistance of triamcinolone acetonide. Diluted indocyanine green dye (initially 25 mg in 20 mL of 5% glucose and diluted 10 further times with 5% glucose) was then used to stain the ILM for 30 seconds. We then applied a surgical technique whereby the ERM is removed by picking up the margin between the non-stained ERM area and the stained ILM area using a bent MVR blade. After peeling to the margin of the LMH, the yellowish LHEP material is left unpeeled. The LHEP material is then trimmed with a 23-gauge or 25-gauge vitrectomy cutter to an appropriate size to fill the retinal defect. A foveolar ILM non-peeling technique is then performed with the aid of an MVR blade, microscissors and forceps to leave a 400-μm-diameter ILM over the foveola (see Fig. 4 and Video S1 in the Supplementary Information) with a preserved ILM margin that is not elevated. The authors have used a similar surgical technique to treat early stage 2 macular hole and myopic traction maculopathy 26,27 . At the end of surgery, air-fluid exchange and 15% perfluoropropane (C3F8) were used as gas tamponade. All subjects were advised to maintain a prone position for 7 days. In the total peeling group the LHEP material is elevated and saved as much as possible. Conventional total ILM peeling is then performed without preserving the foveolar ILM.
Statistical analysis. The Mann-Whitney U test was used to compare the preoperative BCVA, postoperative BCVA, and perioperative BCVA gain. Wilcoxon's signed-rank test was used to compare the preoperative and postoperative BCVA in each study group. The statistical analyses were performed using SPSS version 22.0 software (IBM Corp., Armonk, NY). Statistical significance was defined as p < 0.05.