Long-Term Continuous Assessment of Internal Limiting Membrane Filling Induced Super-Large Macular Hole Healing


To assess the long-term continuous anatomical and functional healing process of a super-large full-thickness macular hole (FTMH) after internal limiting membrane (ILM) filling and air tamponade.


Prospective, interventional case series.


Pars plana vitrectomy with ILM filling and air tamponade was performed in eyes with idiopathic super-large FTMH (> 900 µm). Patients were divided into 3 groups according to the macular hole (MH) diameter: group A, < 1000 µm; group B, 1000 to 1100 µm; and group C, > 1100 µm. The MH closure pattern was analyzed using optical coherence tomography. The preoperative and postoperative macular hole sensitivity, parafoveal sensitivity, and fixation status were assessed using a Microperimeter-3. The preoperative and postoperative best corrected visual acuity were measured. A monthly follow-up was conducted for 15 months postoperatively.


Ninety-seven eyes from 96 consecutive patients with super-large FTMH were included in the study. Primary closure was seen in 93 of 97 eyes (95.88%), in which 88 of 97 eyes (90.72%) had a closed MH 1 week after surgery . The MH closure pattern continuously improved and significantly differed in the 3 groups ( P < .05). Significant improvements in best corrected visual acuity ( P < .01), parafoveal sensitivity ( P < .05), and fixation stability ( P < .01) were observed in all groups.


Super-large FTMH (> 900 µm) closed promptly within 1 week postoperatively after ILM filling and air tamponade. The morphological and functional improvement lasted over an extended period postoperatively. No central scotoma enlargement was found regardless of the closure patterns, and there was significant retinal sensitivity and fixation status improvement.


A full-thickness macular hole (FTMH) is a vertical defect arising from the internal limiting membrane (ILM) and extending down to the retinal pigment epithelium (RPE) in the foveal retina. In 1991, Kelly and Wendel first described vitrectomy to treat macular hole (MH) defects. The first report of ILM peeling specifically for a MH was described by Eckardt and associates in 1997. Nowadays, the general anatomical closure rate of a FTMH is about 90% with pars plana vitrectomy (PPV) and ILM peeling; however, the closure rate of a large FTMH is still unsatisfactory. With an increase in hole diameter, the closure rate significantly decreases. The literature has reported relatively low closure rates of 50% to 88% in FTMHs > 400 µm. Ch’ng and associates reported a closure rate of 76% in FTMHs > 650 µm. In addition, a larger FTMH also indicates a higher reopening rate and worse visual outcome, and the success rate of reoperation has been found to be 46.7% in failed or reopening cases.

The inverted ILM flap technique was introduced by Michalewska and associates in 2010 for the treatment of large FTMHs. As initially described, the technique involved circumferential ILM peeling with a frill of ILM left around the hole; this frill was trimmed and inverted into the hole using forceps. Since this initial description, several variations of the technique have been proposed, including creating a flap on one side of the hole only and folding it over the hole as a flat sheet. Although the single-sheet “cover” variant may achieve better anatomical restoration and vision, the original ILM flap “filling” technique offers advantages in the closure of larger FTMHs. Rossi and associates showed that the original ILM filling technique was more efficacious in closing FTMHs > 630 µm. Previous observations from the current institution found the ILM filling technique more preferable in FTMHs > 900 µm, which were defined as super-large FTMH. Moreover, for reopened cases with previous ILM peeling, the ILM filling technique may be a better choice for reoperation.

Despite the preferable results of the ILM filling technique in closing super-large FTMHs, a question arises as to whether the visual prognosis is affected by the filling of the non-neural retinal sheet in the hole. Whether ILM filling will lead to the enlargement of postoperative central scotoma is a topic of constant concern. The long-term recovery of the foveal structure and function after hole closure is still unclear, especially in super-large FTMHs. There have been few studies to date on the progressive evolution of the MH closure pattern and the related visual function after the ILM filling technique. The purpose of this study was to evaluate the long-term change in the MH closure pattern and visual function after the ILM filling technique was conducted in the context of super-large FTMHs.



A prospective interventional study of patients with super-large idiopathic FTMHs was conducted. A super-large FTMH was defined as a FTMH > 900 µm in minimum linear diameter. The cases with previously failed operations were included. Exclusion criteria were defined as follows: high myopia, including axial length > 26.00 mm or refractive error > 26.00 diopters (D); traumatic MH; chronic MH (duration of symptoms > 6 months); history of retinal detachment surgery; history of uveitis; history of any other fundus diseases or intravitreal injection; and any visually significant ocular pathology accompanying the defect. This study adhered to the tenets of the Declaration of Helsinki, and ethics approval was obtained from the ethics committee of the Zhongshan Ophthalmic Center (2022KYPJ027).

All patients underwent complete eye examination preoperatively with best corrected visual acuity (BCVA) using Snellen charts; intraocular pressure using noncontact tonometry (Topcon); axial length of the eyeball using IOL Master Biometry (Carl Zeiss); anterior segment examination using slit-lamp biomicroscopy; dilated posterior segment examination using ophthalmoscopy; fundus color photography (fundus camera, TRC-50; Topcon); spectral domain optical coherence tomography (SD-OCT; Carl Zeiss); and retinal sensitivity and fixation stability/shift using a Microperimeter-3 (MP-3; Nidek Co.). The follow-up time was weekly for 1 month and monthly thereafter up to 15 months.

The diagnosis of a super-large idiopathic FTMH was confirmed by a detailed clinical history, thorough ocular examination, and OCT images. The diameter of each FTMH was measured by minimum linear diameter, which was measured at the narrowest hole point in the mid-retina using the OCT caliper function as a line drawn roughly parallel to the RPE. Patients were divided into 3 groups according to the diameter of the FTMH as follows: group A, < 1000 µm; group B, 1000 to 1100 µm; and group C, > 1100 µm. The FTMH was considered closed if there was evidence of any tissue bridging the preoperative FTMH, regardless of retinal layer integrity (but not only ILM).

Microperimetry assessment was conducted in a dark room with the fellow eye patched. The examination was performed by 2 experienced operators. The MP-3 took high-definition color images of the fundus to observe structures of each MH. A standardized grid pattern was used for testing, which consisted of 45 stimuli points arranged concentrically in the central point and 4 circles that occupied an area of 12° angle of view (in diameter) within the central macula. The fixation target was a 1° diameter red cross, the white background illumination was set at 1.27 cd/m 2 and the duration of the stimulus was 200 ms. In the measurement of retinal sensitivity, the follow-up pattern was used to ensure the selected dots were located in the same position in every examination ( Figure 1 ). Macular hole sensitivity (MHS) was defined as the sensitivity values of stimuli points within and at the margin of the MH. Parafoveal sensitivity (PFS) was defined as sensitivity values of stimuli points within 2° beyond the hole margin (without the hole margin). The changes in MHS and PFS reflected the change of the central scotoma. Fixation stability was graded by the built-in software as follows: stable fixation: ≥ 75% of the fixation points fell inside a 2° circle around the point of fixation; relative stability: ≥ 75% of the fixation points fell inside a 4° circle (but < 75% were inside the 2° circle); and instability: < 75% of the points fell inside the 4° circle. The assessment of change in preferred retinal locus (PRL) location, which represented the fixation shift, was performed with the differential map analysis feature of the MP-3. When proper calibration is used, this feature shows the Euclidean distance in degrees between the centroids of the fixation clusters of the 2 examinations.


MP-3 images of a patient with super-large full-thickness macular hole preoperatively and 1 week postoperatively. A standardized grid pattern consisted of 45 stimuli points concentrically arranged around a central point and 4 circles that occupied an area of a 12° angle of view (in diameter) in the central macula. A . Microperimeter-3 images preoperatively. B . Microperimeter-3 images 1 week postoperatively. The follow-up pattern was used to ensure that the selected dots were located in the same position. The white circle shows the margin of the macular hole before surgery. The white dotted cross shows the position of the preoperative ruler.


One surgeon (J.M.) operated on all eyes of all patients under local anesthesia. All participants underwent a standard 3-port 25-gauge PPV (Alcon). The epiretinal membrane was peeled if present. Indocyanine green staining was performed, and the ILM was peeled in a circular fashion with a diameter of about 3 optic discs. During circumferential peeling, the ILM was not completely removed from the retina but was left attached to the edges of the MH. The peripheral piece of the ILM was trimmed with a vitreous cutter to about one-third of the MH diameter, and the central part of the ILM was left in place. The hanging remnants of the ILM flap were then folded in multiple layers and deliberately “stuffed” or “packed” into the MH defect using a forceps. Fluid-air exchange was applied before this “filling” manipulation, so the pressure of air could stabilize the free ILM remnants, which could readily be filled into the hole. Sodium hyaluronate gel was used to cover the ILM flap to stabilize it. In cases of reoperation, where the ILM had previously been peeled, the MH was filled with free ILM flap peeled from the remnant ILM. At the end of surgery, the vitreous cavity was filled with air. The peripheral retina was carefully inspected, and degeneration areas or tears were photocoagulated with argon laser if present. All the patients were ordered to strictly keep face-down posturing for 5 days. Patients with cataracts underwent combined phacoemulsification with in-the-bag IOL implants during PPV surgery.


Statistical analysis was performed using SPSS software (SPSS for Windows, version 22.0). A paired-samples 2-tailed t test was performed to determine differences in preoperative and postoperative BCVA in all groups. An independent-samples t test was undertaken to compare the BCVA values among the groups. Fisher exact test was used to test the significance of MH closure patterns among the groups. The χ 2 test was used to test the significance of free ILM remnants among the groups. The Wilcoxon rank sum test was used to test the significance of the change in retinal sensitivity (MHS and PFS) in all the groups. Fisher exact test was used to test the significance of the change in fixation stability in all the groups. The Dunnet t test was used to test the significance of a different PRL in all the groups. A hierarchical multiple regression was performed to investigate the relationship between the MH diameter, MH closure pattern, retinal sensitivity (MHS and PFS), fixation stability, and postoperative BCVA. A P -value < .05 was considered statistically significant.


A total of 97 eyes from 96 consecutive patients with super-large FTMHs were recruited in the Retina Department of Zhongshan Ophthalmic Center of Sun Yat-sen University from September 2014 to August 2020. Primary closure was seen in 93 of 97 eyes (95.88%), in which 88 of 97 eyes (90.72%) had a closed MH 1 week after surgery. Four of 97 eyes (4.12%; 1 eye from group B and 3 eyes from group C) had no ILM observed at 1 week postoperatively, probably because of dislocation of the ILM flap; they were excluded from the analysis. These 4 eyes underwent reoperation, and anatomical success was ultimately achieved. Six eyes (1 from group A, 3 from group B, and 2 from group C) were lost to follow-up. Ultimately, 87 of 97 eyes (89.69%) from 86 patients had complete follow-up of 15 months postoperatively, including 27 eyes in group A (MH < 1000 µm), 31 eyes in group B (MH 1000-1100 µm) and 29 eyes in group C (MH > 1100 µm). The demographics and baseline information of the patients are listed in Table 1 . The 3 groups did not significantly differ in terms of baseline information.


Baseline Characteristics of the Patients ( N = 86) and Affected Eyes ( N = 87)

Total Group A Group B Group C P Value
Patients, n 86 26 31 29 < .05
Gender, male/female 46/40 13/13 17/14 16/13 < .05
Age, years (mean, range) 59.20,
< .05
Eyes, n 87 27 31 29 < .05
Lens, eyes, n
Phakic 53 17 18 18 < .05
Aphakic 0 0 0 0 < .05
Pseudophakic 34 10 13 11 < .05
History of MH surgery, eyes, n
No 76 25 26 25 < .05
One failed surgery 6 1 2 2 < .05
Two failed surgeries 5 1 3 2 < .05

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Sep 11, 2022 | Posted by in OPHTHALMOLOGY | Comments Off on Long-Term Continuous Assessment of Internal Limiting Membrane Filling Induced Super-Large Macular Hole Healing

Full access? Get Clinical Tree

Get Clinical Tree app for offline access