TEMPORAL INVERTED INTERNAL LIMITING MEMBRANE FLAP TECHNIQUE VERSUS CLASSIC INVERTED INTERNAL LIMITING MEMBRANE FLAP TECHNIQUE A Comparative Study ZOFIA MICHALEWSKA, MD, PHD,* JANUSZ MICHALEWSKI, MD, PHD,* KAROLINA DULCZEWSKA-CICHECKA, MD,* RON A. ADELMAN, MD, MPH,† JERZY NAWROCKI, MD, PHD* Purpose: To determine if reducing the area of internal limiting membrane (ILM) peeling in the inverted ILM flap technique results in satisfactory outcomes for the repair of large Stage IV idiopathic macular holes. Methods: Prospective comparative interventional study of 87 consecutive eyes. Participants were randomized into two treatment groups. In Group A, the classic inverted ILM flap technique was performed. In Group B, a modification of this procedure, the temporal inverted ILM flap technique, was performed. In the modified inverted ILM flap technique, peeling of the ILM was restricted to the temporal side of the fovea only—the macular hole was then covered with the temporal ILM flap. Results: There was no significant difference in initial and final visual acuities between Groups A and B. In both groups, defects in photoreceptors and the external limiting membrane decreased with time. Successive postoperative examinations revealed an increasing number of patients with the dissociated optic nerve fiber layer appearance, although this was less frequent in Group B (modified ILM flap) than in Group A. Conclusion: The study results indicate that the temporal inverted ILM flap technique is as effective as the classic inverted ILM flap technique for the repair of large Stage IV macular holes. RETINA 35:1844–1850, 2015
T
he authors first introduced the inverted internal limiting membrane (ILM) flap technique for the repair of large Stage IV idiopathic full-thickness mac-
ular holes (FTMH) in 2009.1 In this method, ILM is not completely removed, but a small remnant is left on the margin of FTMH to cover it. Initial reports suggested that this technique might produce superior anatomical and functional results than the conventional vitrectomy with ILM peeling for the repair of macular holes. Previous studies of vitrectomy with ILM peeling reported that the best postoperative functional results were associated with the U-shape closure, which is most similar to the healthy foveal contour and is present in 45% of cases on postoperative optical coherence tomography (OCT). Flat macular hole margins with bare retinal pigment epithelium (i.e., flat/ open macular holes) are observed in 19% to 39% of Stage 3 and 4 FTMH, postoperatively.2–4 Although considered an anatomical success, flat/open macular
From the *Ophthalmic Clinic “Jasne Blonia”, Lodz, Poland; and †Ophthalmic Department, Yale University School of Medicine, New Haven, Connecticut. Presented at Vail Vitrectomy Meeting 2013, European VitreoRetinal Society (EVRS) Annual Meeting 2014, American Society of Retina Specialists (ASRS) 2014, World Ophthalmology Congress 2014, and American Academy of Ophthalmology (AAO) 2014. None of the authors have any financial/conflicting interests to disclose. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.retinajournal.com). Reprint requests: Zofia Michalewska, MD, PhD, Klinika Okulistyczna “Jasne Blonia,” Rojna 90, Lodz 91-162, Poland; email: [email protected]
1844
TEMPORAL INVERTED ILM FLAP TECHNIQUE MICHALEWSKA ET AL
holes are generally associated with worse functional results than other closure types.2–4 After the introduction of the inverted ILM flap technique for macular hole closure, the number of cases with flat/open macular holes was reduced to 0%.5 In the same prospective randomized study, the authors achieved a closure rate of 98% when the inverted ILM flap technique was performed compared with an 88% closure rate with conventional vitrectomy and ILM peeling.5 Multiple groups reported similar satisfactory results with this technique, expanding its use to myopic macular holes and repeat macular hole surgery.6–9 Internal limiting membrane peeling remains a method of choice for the repair of FTMH; however, this technique has been associated with dissociated optic nerve fiber layer (DONFL).10–13 In blue filter photography, DONFL is visible as hatched retina with striae along the optic nerve fibers, which corresponds to dimples in inner retinal layers visible in OCT. DONFL was reported to appear 1 to 3 months postoperatively.10–13 We endeavored to maintain good surgical results while minimizing iatrogenic trauma associated with ILM peeling by decreasing the area of peeled ILM. The aim of this study is to compare the original inverted ILM flap technique with a modified form where ILM is peeled from the temporal side of the fovea only. Accordingly, we call this modified procedure the temporal inverted ILM flap technique.
received local anesthesia, and one surgeon (J.N.) performed all operations. Immediately before surgery, the surgeon drew a ball from a covered box; if the ball was white, the surgeon performed the original inverted ILM flap technique (Figures 1 and 2); if black, the surgeon removed ILM from only the temporal side of the fovea and covered the fovea by inverting this temporal ILM flap (Figures 3 and 4). This section provides a detailed description of the temporal inverted ILM flap technique. To start, we performed 20-gauge core vitrectomy and trypan blue staining (0.06% solution of Membrane Blue or Membrane Blue–Dual, Dorc, Zuiland, the Netherlands) for 1 minute. The surgery was performed with either the Associate system (Dorc) or the OS-3 device (Oertli, Berneck, Switzerland) and a peristaltic pump. A
Methods This was a prospective comparative interventional study with institutional review board approval. Patients signed an informed consent. As is the clinical standard, all patients underwent a complete ophthalmic evaluation, including best-corrected visual acuity, slit-lamp examination, and ophthalmoscopy. Per study protocol, a masked examiner measured best-corrected visual acuity on Early Treatment Diabetic Retinopathy Study charts. Spectral domain optical coherence tomography (SOCT Copernicus HR; Optopol, Poland) or swept-source optical coherence tomography (DRIOCT; Topcon, Japan) was performed to confirm the diagnosis of macular hole. The minimum and maximum diameters of the hole were measured, and only those patients who had macular holes with a minimum diameter exceeding 400 mm were included in this study. Patients with severe retinal pathology such as proliferative diabetic retinopathy, high myopia, exudative age related macular degeneration, and retinal detachment were excluded. Patients were randomized into two treatment groups. To minimize confounding factors, all patients
1845
Fig. 1. Scheme of the “inverted ILM flap technique.”
1846 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
2015 VOLUME 35 NUMBER 9
Fig. 2. Surgical photograph of the inverted ILM flap on the surface of the macular hole.
Fig. 4. Surgical photograph of the temporal ILM flap on the surface of the macular hole.
contact lens (Alcon, Geneva, Switzerland) was placed for macular surgery. If present, the surgeon first peeled an epiretinal membrane. Next, ILM forceps were used to grasp and peel the ILM off at the temporal side of the macular hole in an area of about two diameters of the optic nerve. During this peeling, the ILM was not removed completely from the retina but instead was left attached to the temporal edge of the macular hole, then
inverted and gently coaxed over the macular hole until adequate coverage was achieved. At the end of surgery, the vitreous cavity was filled with air. Patients were instructed to spend 3 to 4 days in prone position (see Video 1, Supplemental Digital Content 1, http://links. lww.com/IAE/A336). Patients were seen and examined at 1 week, 1, 3, 6, and 12 months postoperatively. At each visit, patients underwent visual acuity measurement, slit-lamp examination, indirect ophthalmoscopy and were imaged with spectral domain optical coherence tomography or swept-source optical coherence tomography. During each postoperative visit, the foveal contour and the presence of photoreceptor (i.e., ellipsoid zone [EZ]) defects, external limiting membrane (ELM) defects, and DONFL were estimated. Postoperative foveal contour was characterized according to its crosssectional morphology on spectral domain optical coherence tomography into the various categories, U-shape, V-shape, W-shape (irregular), and earlier not described “flap closure” (Figure 5C). Flap closure was recognized when a thin hyperreflective band of tissue extended between the inner surface of the retina on both sides of the previous FTMH. SigmaStat 3.5 for Windows was used for statistical analysis. Results
Fig. 3. Scheme of the “temporal inverted ILM flap technique.”
Among 87 eyes, 43 were randomized to Group A (inverted ILM flap technique) and 44 were randomized to Group B (temporal inverted ILM flap technique). At baseline, there were no significant differences between Group A and Group B in regard to preoperative visual acuity (P = 0.28), minimum macular hole diameter (P = 0.81), and base macular hole diameter (P =
TEMPORAL INVERTED ILM FLAP TECHNIQUE MICHALEWSKA ET AL
1847
0.4). In both groups, preoperative visual acuity correlated negatively with minimum diameter (P = 0.003) and base diameter (P = 0.008). Demographic data are presented in Table 1. Functional Results After surgery, patients in both groups experienced a statistically significant improvement in visual acuity (P , 0.001). In Group A, the average visual acuity was 1.01 logMAR 1 week after surgery; the average improved to 0.66 logMAR at the 3-month and 6-month postoperative visits, and then to 0.4 logMAR at 12 months. The average visual acuity for patients in Group B was 1.03 logMAR at the first postoperative visit (1 week). Visual acuity at 3, 6, and 12 months after surgery was measured as 0.63, 0.68, and 0.45 logMAR, respectively. Statistical analysis confirmed that there was no significant difference in visual acuity between the two groups at postoperative visits (t-test). Statistical analysis revealed that final visual acuity was associated with initial visual acuity (P , 0.001), minimum diameter (P = 0.05), and base diameter of FTMH (P = 0.001). Anatomic Results
Fig. 5. A. Full-thickness macular hole. B. One week after surgery, macular hole almost completely covered with the inverted ILM flap (flap closure). Red arrow: Inverted ILM flap covering the macular hole. C. One month after surgery, macular hole completely covered only with the inverted ILM flap (flap closure). Red arrow: Inverted ILM flap covering the macular hole. D. Three months after surgery, macular hole is closed.
Closure of FTMH was achieved in all cases; however, three eyes in each group required repeat macular hole surgery with reapproximation of the inverted ILM flap on the surface of the FTMH. In 93% of eyes in Group A (40/43) and 93% of eyes in Group B (41/44), macular hole closure was observed after the initial surgery. Initial failure of FTMH closure was associated with retroversion of the flap during fluid/air exchange (see Video 2, Supplemental Digital Content 1, http://links.lww.com/IAE/A337). Surgical failure did neither correlate with size of the macular hole nor with any other factor analyzed in this study. All repeat surgeries were performed between 1 and 7 months after the initial surgery. During repeated surgeries, we elevated the ILM flap again from the retinal surface and we covered the macular hole with it. The vitreous cavity was filled with silicone oil, which was removed 2 months later. No reopenings of FTMH were noted during the follow-up period. Foveal contour. One week after surgery, 14% of cases in Group A and 16% of cases in Group B were noted to have macular hole closure by only the inverted ILM flap, deemed “flap closure” (Figure 5A). In group B, at the 3-month postoperative mark, flap closure remained in 14% of Group A cases and only in 6% of Group B cases. By 12 months after surgery, flap closure was observed in 3% of cases in
1848 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
2015 VOLUME 35 NUMBER 9
Table 1. Demographic Data
Age, years Male:female Initial visual acuity, logMAR Minimum FTMH diameter, mm Base FTMH diameter, mm
Group A
Group B
P
67 5:34 0.95 533 (range, 400–763) 826 (range, 420–1,286)
66 11:29 1.02 544 (range, 400–841) 894 (range, 440–2,203)
0.3 1.0 0.28 0.81 0.4
Group A and in none of the eyes from Group B (Figure 5, B–D). At the 12-month postoperative visit, we also noted a lower percentage of U-shape closures in Group A (62%) than in Group B (71%). These eyes had presented with the best preoperative and postoperative visual acuities. Twelve months after surgery, flat/open closure was not noted in any eye. The varying types of foveal contour changes that were seen at follow-up are summarized in Table 2. Foveal morphology. Ellipsoid zone defects were noted in nearly 100% of cases at the 1-week postoperative visit; however, by 12 months after surgery, EZ defects had decreased to 57%. Lasting EZ defects were seen more often in eyes that had a larger initial base diameter of the macular hole (P = 0.02). Average visual acuity was significantly worse for patients who had EZ defects at 12 months compared with those who did not: 0.48 logMAR versus 0.3 logMAR (P = 0.03). Similarly, ELM defects were noted in all cases 1 week after surgery; however, this decreased to approximately a quarter of cases at 12 months postoperatively. Patients with lasting ELM defects had significantly worse final visual acuity than those without: 0.52 versus 0.3 logMAR at 12-month follow-up (P = 0.04). These results were reflected in retinal morphology changes observed on consecutive postoperative spectral domain optical coherence tomography images; EZ and ELM defects were noted to decrease with time. Restoration of the ELM preceded EZ restoration in 46% of eyes in Group A and 30% of eyes in Group B. In the remaining eyes, simultaneous restoration of the two layers was observed.
Group A tended to have a higher rate of DONFL appearance than Group B at the 12-month postoperative visit (84% vs. 64%, P = 0.06). Additionally, among Group B patients, DONFL was limited to the temporal side of the fovea where ILM peeling occurred. For patients in Group A, DONFL was noted diffusely around the fovea, also in the area of peeled ILM. The number of eyes with DONFL increased over time (Table 3); however, the appearance of DONFL was not associated with any significant changes in visual acuity (P = 0.3).
Discussion The results from this study demonstrated that the original inverted ILM flap technique and the temporal inverted ILM flap technique produce similar closure rates and visual acuity improvements for large Stage IV FTMH. Additionally, the temporal inverted ILM flap technique in which the ILM is removed only from the temporal side of the macular hole tended to be associated with fewer cases of DONFL. Our group first presented the inverted ILM flap technique in 2009.1 It is a surgical maneuver primarily designed for the treatment of challenging macular hole cases such as large Stage IV FTMH. The concept behind this technique is to not completely remove the peeled ILM but to leave a small remnant of it at the surface of the macular hole to act as a hinge. This remnant is then inverted on its point of attachment in such a fashion that the part primarily oriented to the
Table 2. Changes of the Fovea Contour During 12-Month Follow-up in the Groups Operated With the “Inverted ILM Flap Technique” (Group A) and “Temporal Inverted ILM Flap Technique” (Group B) 1 Week Closure U shape V shape W shape Flat/open Flap Open
Group A, Group B, % % 44 22 5 5 14 10
38 17 17 3 16 9
3 Months P 0.7 0.73 0.1 0.9 0.97 0.9
Group A, Group B, % % 47 19 17 — 14 3
49 15 24 — 6 3
6 Months P 0.9 0.8 0.7 — 0.37 —
Group A, Group B, % % 60 18 8 — 11 3
64 14 16 — 3 3
12 Months P 0.87 0.83 0.4 — 0.29 —
Group A, Group B, % % 62 22 10 — 3 —
71 17 12 — — —
P 0.5 0.73 0.9 — 0.78 —
TEMPORAL INVERTED ILM FLAP TECHNIQUE MICHALEWSKA ET AL
1849
Table 3. Defects of Particular Retinal Layers During Consecutive Follow-up Visits 1 Week Group A, Group B, % % EZ defects ELM defects RNFL defects
3 Months P
Group A, Group B, % %
6 Months P
Group A, Group B, % %
12 Months P
Group A, Group B, % %
P
98 100
100 100
0.88 —
91 57
87 72
0.8 0.2
84 48
72 48
0.2 0.8
57 24
57 25
0.82 0.88
21
18
0.93
71
56
0.2
74
56
0.2
84
64
0.06
vitreous cavity is made to face toward the retinal pigment epithelium, whereas the part initially touching retinal layers is oriented to the vitreous cavity (Figure 1, A and B). The exact mechanism of improved surgical results using this technique is not precisely understood. One theory is that the ILM may act as a scaffold for glial cells to proliferate, thus enhancing closure of the macular hole. Another explanation might be that ILM serves as a barrier. It disables the entrance of fluid from the vitreous cavity to the macular hole. Additionally, once vacuum is created inside the macular hole sealed by the flap, retinal pigment epithelium outflow is altered. As it was first described for the repair of large Stage IV macular holes, the inverted ILM flap technique has found additional indications. In recent studies, the authors and Kuriyama et al proved that this technique is also successful for the repair of FTMH associated with high myopia.4,5 These cases, however, are extremely challenging. Another group has proposed using the inverted ILM flap in optic disk pitassociated maculopathy with promising results.14 Finally, even if there is complete removal of the ILM flap during the initial surgery, it has been shown that ILM may be transplanted from a different location in the retina, allowing closure in cases of primary failure.6 Internal limiting membrane peeling significantly improves the closure rate of FTMH.15 However, despite its advantages, several authors have reported the appearance of dark striae on autofluorescence imaging in areas of previous ILM peeling; these striae correspond to swelling of the inner retinal layers (swelling of the arcuate nerve fiber layer) on OCT.16.Swelling of the arcuate nerve fiber layer is visible shortly after surgery but disappears in approximately 1 month. Swelling of the arcuate nerve fiber layer is followed by the formation of small dimples in the retinal nerve fiber layer (RNFL), which are visible on OCT and referred to as dissociated optic nerve fiber layer (DONFL) appearance.11,16 The risk of RNFL dissociation has not been shown to decrease with time, and in fact, authors have observed new RNFL dehis-
cence up to 6 months postoperatively.11,17 We confirmed this finding in the extended follow-up; additionally, new RNFL dehiscence was observed in both groups beyond 6 months (up to 12 months) postoperatively. These changes had no significant effect on visual acuity, however. The frequency of DONFL was higher in eyes from Group A than in eyes from Group B. Moreover, in Group B, DONFL was observed on the temporal side of the fovea only, which suggests that the changes are indeed induced by ILM peeling. Temporal ILM flap technique minimizes unnecessary trauma to the NFL, which may be advantageous. Skill level varies significantly from surgeon to surgeon. A minimalistic approach in maximizing hole closure is important to avoid RNFL defects and eccentric holes postoperatively. Overall, patients in Group B tended to have a lower incidence of DONFL. Importantly, functional results and closure rates did not differ between Group A and Group B, indicating that it is not necessary for the ILM flap to be as large as in the original inverted ILM flap technique. The ultimate goal of each surgery is to achieve FTMH closure while minimizing iatrogenic complications. Importantly, DONFL has not been definitively associated with a decrease of visual acuity or microperimetry changes11; however, there may be unknown effects on function that are not yet known. It is possible that RNFL dehiscence weakens the retinal structure and thus DONFL shows the healing process. The original inverted ILM flap technique was established by the authors with the primary aim of improving surgical results in challenging cases of FTMH repair. This study focused on maintaining a high percentage of FTMH closures while minimizing surgical trauma induced by ILM peeling. This study reports results from a modification of the original inverted ILM flap technique; in the temporal inverted ILM flap technique, the ILM is removed from the temporal side of the fovea, up to the margins of the FMTH, and inverted to cover the macular hole while leaving the nasal ILM in place. Thus, it decreases the risk of surgical trauma in the area of papillomacular bundle. Temporal inverted ILM flap technique
1850 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
achieves satisfactory anatomical and functional results, as well as reduction of DONFL appearance. Key words: macular hole, ILM peeling, inverted ILM flap, vitrectomy, temporal inverted ILM flap.
9.
10.
References 1. Michalewska Z, Michalewski J, Nawrocki J. Macular hole closure after vitrectomy: the inverted flap technique. Retina Today 2009;3:73–74. 2. Michalewska Z, Michalewski J, Cisiecki S, Nawrocki J. High speed, high resolution spectral optical coherence tomography after macular hole surgery. Graefes Arch Clin Exp Ophthalmol 2008;246:823–830. 3. Imai M, Iijima H, Gotoh T, Tsukahara S. Optical coherence tomography of successfully repaired idiopathic macular holes. Am J Ophthalmol 1999;128:621–627. 4. Kang SW, Ahn K, Ham DI. Types of macular hole closure and their clinical implications. Br J Ophthalmol 2003;87:1015–1019. 5. Michalewska Z, Michalewski J, Adelman RA, Nawrocki J. Inverted internal limiting membrane flap technique for large macular holes. Ophthalmology 2010;117:2018–2025. 6. Mahalingam P, Sambhav K. Surgical outcomes of inverted internal limiting membrane flap technique for large macular hole. Indian J Ophthalmol 2013;61:601–603. 7. Michalewska Z, Michalewski J, Dulczewska- Cichecka K, Nawrocki J. Inverted internal limiting membrane flap technique in macular hole associated with pathological myopia. Retina 2014;34:664–669. 8. Kuriyama S, Hayashi H, Jingami Y, et al. Efficacy of inverted internal limiting membrane flap technique for the treatment of
11.
12.
13.
14.
15.
16.
17.
2015 VOLUME 35 NUMBER 9
macular hole in high myopia. Am J Ophthalmol 2013;156: 125–131. Morizane Y, Shiraga F, Kimura S, et al. Autologous transplantation of the internal limiting membrane for refractory macular holes. Am J Ophthalmol 2014;157:861–869. Tadayoni R, Paques M, Massin P, et al. Dissociated optic nerve fiber layer appearance of the fundus after idiopathic epiretinal membrane removal. Ophthalmology 2001;108:2279–2283. Ito Y, Terasaki H, Takahashi A, et al. Dissociated optic nerve fiber layer appearance after internal limiting membrane peeling for idiopathic macular holes. Ophthalmology 2005;112: 1415–1420. Michalewska Z, Michalewski J, Nawrocki J. Continuous changes in macular morphology after macular hole closure visualized with spectral optical coherence tomography. Graefes Arch Clin Exp Ophthalmol 2010;248:1249–1255. Mitamura Y, Ohtsuka K. Relationship of dissociated optic nerve fiber layer appearance to internal limiting membrane peeling. Ophthalmology 2005;112:1766–1770. Mohammed OA, Pai A. Inverted autologous internal limiting membrane for management of optic disc pit with macular detachment. Middle East Afr J Ophthalmol 2013;20: 357–359. Ï Christensen UC, Krryer K, Sander B, et al. Value of internal limiting membrane peeling in surgery for idiopathic macular hole stage 2 and 3: a randomised clinical trial. Br J Ophthalmol 2009;93:1005–1015. Clark A, Balducci N, Pichi F, et al. Swelling of the arcuate nerve fiber layer after internal limiting membrane peeling. Retina 2012;32:1608–1613. Spaide RF. Dissociated optic nerve fiber layer appearance after internal limiting membrane removal is inner retinal dimpling. Retina 2012;32:1719–1726.
T
he authors first introduced the inverted internal limiting membrane (ILM) flap technique for the repair of large Stage IV idiopathic full-thickness mac-
ular holes (FTMH) in 2009.1 In this method, ILM is not completely removed, but a small remnant is left on the margin of FTMH to cover it. Initial reports suggested that this technique might produce superior anatomical and functional results than the conventional vitrectomy with ILM peeling for the repair of macular holes. Previous studies of vitrectomy with ILM peeling reported that the best postoperative functional results were associated with the U-shape closure, which is most similar to the healthy foveal contour and is present in 45% of cases on postoperative optical coherence tomography (OCT). Flat macular hole margins with bare retinal pigment epithelium (i.e., flat/ open macular holes) are observed in 19% to 39% of Stage 3 and 4 FTMH, postoperatively.2–4 Although considered an anatomical success, flat/open macular
From the *Ophthalmic Clinic “Jasne Blonia”, Lodz, Poland; and †Ophthalmic Department, Yale University School of Medicine, New Haven, Connecticut. Presented at Vail Vitrectomy Meeting 2013, European VitreoRetinal Society (EVRS) Annual Meeting 2014, American Society of Retina Specialists (ASRS) 2014, World Ophthalmology Congress 2014, and American Academy of Ophthalmology (AAO) 2014. None of the authors have any financial/conflicting interests to disclose. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.retinajournal.com). Reprint requests: Zofia Michalewska, MD, PhD, Klinika Okulistyczna “Jasne Blonia,” Rojna 90, Lodz 91-162, Poland; email: [email protected]
1844
TEMPORAL INVERTED ILM FLAP TECHNIQUE MICHALEWSKA ET AL
holes are generally associated with worse functional results than other closure types.2–4 After the introduction of the inverted ILM flap technique for macular hole closure, the number of cases with flat/open macular holes was reduced to 0%.5 In the same prospective randomized study, the authors achieved a closure rate of 98% when the inverted ILM flap technique was performed compared with an 88% closure rate with conventional vitrectomy and ILM peeling.5 Multiple groups reported similar satisfactory results with this technique, expanding its use to myopic macular holes and repeat macular hole surgery.6–9 Internal limiting membrane peeling remains a method of choice for the repair of FTMH; however, this technique has been associated with dissociated optic nerve fiber layer (DONFL).10–13 In blue filter photography, DONFL is visible as hatched retina with striae along the optic nerve fibers, which corresponds to dimples in inner retinal layers visible in OCT. DONFL was reported to appear 1 to 3 months postoperatively.10–13 We endeavored to maintain good surgical results while minimizing iatrogenic trauma associated with ILM peeling by decreasing the area of peeled ILM. The aim of this study is to compare the original inverted ILM flap technique with a modified form where ILM is peeled from the temporal side of the fovea only. Accordingly, we call this modified procedure the temporal inverted ILM flap technique.
received local anesthesia, and one surgeon (J.N.) performed all operations. Immediately before surgery, the surgeon drew a ball from a covered box; if the ball was white, the surgeon performed the original inverted ILM flap technique (Figures 1 and 2); if black, the surgeon removed ILM from only the temporal side of the fovea and covered the fovea by inverting this temporal ILM flap (Figures 3 and 4). This section provides a detailed description of the temporal inverted ILM flap technique. To start, we performed 20-gauge core vitrectomy and trypan blue staining (0.06% solution of Membrane Blue or Membrane Blue–Dual, Dorc, Zuiland, the Netherlands) for 1 minute. The surgery was performed with either the Associate system (Dorc) or the OS-3 device (Oertli, Berneck, Switzerland) and a peristaltic pump. A
Methods This was a prospective comparative interventional study with institutional review board approval. Patients signed an informed consent. As is the clinical standard, all patients underwent a complete ophthalmic evaluation, including best-corrected visual acuity, slit-lamp examination, and ophthalmoscopy. Per study protocol, a masked examiner measured best-corrected visual acuity on Early Treatment Diabetic Retinopathy Study charts. Spectral domain optical coherence tomography (SOCT Copernicus HR; Optopol, Poland) or swept-source optical coherence tomography (DRIOCT; Topcon, Japan) was performed to confirm the diagnosis of macular hole. The minimum and maximum diameters of the hole were measured, and only those patients who had macular holes with a minimum diameter exceeding 400 mm were included in this study. Patients with severe retinal pathology such as proliferative diabetic retinopathy, high myopia, exudative age related macular degeneration, and retinal detachment were excluded. Patients were randomized into two treatment groups. To minimize confounding factors, all patients
1845
Fig. 1. Scheme of the “inverted ILM flap technique.”
1846 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
2015 VOLUME 35 NUMBER 9
Fig. 2. Surgical photograph of the inverted ILM flap on the surface of the macular hole.
Fig. 4. Surgical photograph of the temporal ILM flap on the surface of the macular hole.
contact lens (Alcon, Geneva, Switzerland) was placed for macular surgery. If present, the surgeon first peeled an epiretinal membrane. Next, ILM forceps were used to grasp and peel the ILM off at the temporal side of the macular hole in an area of about two diameters of the optic nerve. During this peeling, the ILM was not removed completely from the retina but instead was left attached to the temporal edge of the macular hole, then
inverted and gently coaxed over the macular hole until adequate coverage was achieved. At the end of surgery, the vitreous cavity was filled with air. Patients were instructed to spend 3 to 4 days in prone position (see Video 1, Supplemental Digital Content 1, http://links. lww.com/IAE/A336). Patients were seen and examined at 1 week, 1, 3, 6, and 12 months postoperatively. At each visit, patients underwent visual acuity measurement, slit-lamp examination, indirect ophthalmoscopy and were imaged with spectral domain optical coherence tomography or swept-source optical coherence tomography. During each postoperative visit, the foveal contour and the presence of photoreceptor (i.e., ellipsoid zone [EZ]) defects, external limiting membrane (ELM) defects, and DONFL were estimated. Postoperative foveal contour was characterized according to its crosssectional morphology on spectral domain optical coherence tomography into the various categories, U-shape, V-shape, W-shape (irregular), and earlier not described “flap closure” (Figure 5C). Flap closure was recognized when a thin hyperreflective band of tissue extended between the inner surface of the retina on both sides of the previous FTMH. SigmaStat 3.5 for Windows was used for statistical analysis. Results
Fig. 3. Scheme of the “temporal inverted ILM flap technique.”
Among 87 eyes, 43 were randomized to Group A (inverted ILM flap technique) and 44 were randomized to Group B (temporal inverted ILM flap technique). At baseline, there were no significant differences between Group A and Group B in regard to preoperative visual acuity (P = 0.28), minimum macular hole diameter (P = 0.81), and base macular hole diameter (P =
TEMPORAL INVERTED ILM FLAP TECHNIQUE MICHALEWSKA ET AL
1847
0.4). In both groups, preoperative visual acuity correlated negatively with minimum diameter (P = 0.003) and base diameter (P = 0.008). Demographic data are presented in Table 1. Functional Results After surgery, patients in both groups experienced a statistically significant improvement in visual acuity (P , 0.001). In Group A, the average visual acuity was 1.01 logMAR 1 week after surgery; the average improved to 0.66 logMAR at the 3-month and 6-month postoperative visits, and then to 0.4 logMAR at 12 months. The average visual acuity for patients in Group B was 1.03 logMAR at the first postoperative visit (1 week). Visual acuity at 3, 6, and 12 months after surgery was measured as 0.63, 0.68, and 0.45 logMAR, respectively. Statistical analysis confirmed that there was no significant difference in visual acuity between the two groups at postoperative visits (t-test). Statistical analysis revealed that final visual acuity was associated with initial visual acuity (P , 0.001), minimum diameter (P = 0.05), and base diameter of FTMH (P = 0.001). Anatomic Results
Fig. 5. A. Full-thickness macular hole. B. One week after surgery, macular hole almost completely covered with the inverted ILM flap (flap closure). Red arrow: Inverted ILM flap covering the macular hole. C. One month after surgery, macular hole completely covered only with the inverted ILM flap (flap closure). Red arrow: Inverted ILM flap covering the macular hole. D. Three months after surgery, macular hole is closed.
Closure of FTMH was achieved in all cases; however, three eyes in each group required repeat macular hole surgery with reapproximation of the inverted ILM flap on the surface of the FTMH. In 93% of eyes in Group A (40/43) and 93% of eyes in Group B (41/44), macular hole closure was observed after the initial surgery. Initial failure of FTMH closure was associated with retroversion of the flap during fluid/air exchange (see Video 2, Supplemental Digital Content 1, http://links.lww.com/IAE/A337). Surgical failure did neither correlate with size of the macular hole nor with any other factor analyzed in this study. All repeat surgeries were performed between 1 and 7 months after the initial surgery. During repeated surgeries, we elevated the ILM flap again from the retinal surface and we covered the macular hole with it. The vitreous cavity was filled with silicone oil, which was removed 2 months later. No reopenings of FTMH were noted during the follow-up period. Foveal contour. One week after surgery, 14% of cases in Group A and 16% of cases in Group B were noted to have macular hole closure by only the inverted ILM flap, deemed “flap closure” (Figure 5A). In group B, at the 3-month postoperative mark, flap closure remained in 14% of Group A cases and only in 6% of Group B cases. By 12 months after surgery, flap closure was observed in 3% of cases in
1848 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
2015 VOLUME 35 NUMBER 9
Table 1. Demographic Data
Age, years Male:female Initial visual acuity, logMAR Minimum FTMH diameter, mm Base FTMH diameter, mm
Group A
Group B
P
67 5:34 0.95 533 (range, 400–763) 826 (range, 420–1,286)
66 11:29 1.02 544 (range, 400–841) 894 (range, 440–2,203)
0.3 1.0 0.28 0.81 0.4
Group A and in none of the eyes from Group B (Figure 5, B–D). At the 12-month postoperative visit, we also noted a lower percentage of U-shape closures in Group A (62%) than in Group B (71%). These eyes had presented with the best preoperative and postoperative visual acuities. Twelve months after surgery, flat/open closure was not noted in any eye. The varying types of foveal contour changes that were seen at follow-up are summarized in Table 2. Foveal morphology. Ellipsoid zone defects were noted in nearly 100% of cases at the 1-week postoperative visit; however, by 12 months after surgery, EZ defects had decreased to 57%. Lasting EZ defects were seen more often in eyes that had a larger initial base diameter of the macular hole (P = 0.02). Average visual acuity was significantly worse for patients who had EZ defects at 12 months compared with those who did not: 0.48 logMAR versus 0.3 logMAR (P = 0.03). Similarly, ELM defects were noted in all cases 1 week after surgery; however, this decreased to approximately a quarter of cases at 12 months postoperatively. Patients with lasting ELM defects had significantly worse final visual acuity than those without: 0.52 versus 0.3 logMAR at 12-month follow-up (P = 0.04). These results were reflected in retinal morphology changes observed on consecutive postoperative spectral domain optical coherence tomography images; EZ and ELM defects were noted to decrease with time. Restoration of the ELM preceded EZ restoration in 46% of eyes in Group A and 30% of eyes in Group B. In the remaining eyes, simultaneous restoration of the two layers was observed.
Group A tended to have a higher rate of DONFL appearance than Group B at the 12-month postoperative visit (84% vs. 64%, P = 0.06). Additionally, among Group B patients, DONFL was limited to the temporal side of the fovea where ILM peeling occurred. For patients in Group A, DONFL was noted diffusely around the fovea, also in the area of peeled ILM. The number of eyes with DONFL increased over time (Table 3); however, the appearance of DONFL was not associated with any significant changes in visual acuity (P = 0.3).
Discussion The results from this study demonstrated that the original inverted ILM flap technique and the temporal inverted ILM flap technique produce similar closure rates and visual acuity improvements for large Stage IV FTMH. Additionally, the temporal inverted ILM flap technique in which the ILM is removed only from the temporal side of the macular hole tended to be associated with fewer cases of DONFL. Our group first presented the inverted ILM flap technique in 2009.1 It is a surgical maneuver primarily designed for the treatment of challenging macular hole cases such as large Stage IV FTMH. The concept behind this technique is to not completely remove the peeled ILM but to leave a small remnant of it at the surface of the macular hole to act as a hinge. This remnant is then inverted on its point of attachment in such a fashion that the part primarily oriented to the
Table 2. Changes of the Fovea Contour During 12-Month Follow-up in the Groups Operated With the “Inverted ILM Flap Technique” (Group A) and “Temporal Inverted ILM Flap Technique” (Group B) 1 Week Closure U shape V shape W shape Flat/open Flap Open
Group A, Group B, % % 44 22 5 5 14 10
38 17 17 3 16 9
3 Months P 0.7 0.73 0.1 0.9 0.97 0.9
Group A, Group B, % % 47 19 17 — 14 3
49 15 24 — 6 3
6 Months P 0.9 0.8 0.7 — 0.37 —
Group A, Group B, % % 60 18 8 — 11 3
64 14 16 — 3 3
12 Months P 0.87 0.83 0.4 — 0.29 —
Group A, Group B, % % 62 22 10 — 3 —
71 17 12 — — —
P 0.5 0.73 0.9 — 0.78 —
TEMPORAL INVERTED ILM FLAP TECHNIQUE MICHALEWSKA ET AL
1849
Table 3. Defects of Particular Retinal Layers During Consecutive Follow-up Visits 1 Week Group A, Group B, % % EZ defects ELM defects RNFL defects
3 Months P
Group A, Group B, % %
6 Months P
Group A, Group B, % %
12 Months P
Group A, Group B, % %
P
98 100
100 100
0.88 —
91 57
87 72
0.8 0.2
84 48
72 48
0.2 0.8
57 24
57 25
0.82 0.88
21
18
0.93
71
56
0.2
74
56
0.2
84
64
0.06
vitreous cavity is made to face toward the retinal pigment epithelium, whereas the part initially touching retinal layers is oriented to the vitreous cavity (Figure 1, A and B). The exact mechanism of improved surgical results using this technique is not precisely understood. One theory is that the ILM may act as a scaffold for glial cells to proliferate, thus enhancing closure of the macular hole. Another explanation might be that ILM serves as a barrier. It disables the entrance of fluid from the vitreous cavity to the macular hole. Additionally, once vacuum is created inside the macular hole sealed by the flap, retinal pigment epithelium outflow is altered. As it was first described for the repair of large Stage IV macular holes, the inverted ILM flap technique has found additional indications. In recent studies, the authors and Kuriyama et al proved that this technique is also successful for the repair of FTMH associated with high myopia.4,5 These cases, however, are extremely challenging. Another group has proposed using the inverted ILM flap in optic disk pitassociated maculopathy with promising results.14 Finally, even if there is complete removal of the ILM flap during the initial surgery, it has been shown that ILM may be transplanted from a different location in the retina, allowing closure in cases of primary failure.6 Internal limiting membrane peeling significantly improves the closure rate of FTMH.15 However, despite its advantages, several authors have reported the appearance of dark striae on autofluorescence imaging in areas of previous ILM peeling; these striae correspond to swelling of the inner retinal layers (swelling of the arcuate nerve fiber layer) on OCT.16.Swelling of the arcuate nerve fiber layer is visible shortly after surgery but disappears in approximately 1 month. Swelling of the arcuate nerve fiber layer is followed by the formation of small dimples in the retinal nerve fiber layer (RNFL), which are visible on OCT and referred to as dissociated optic nerve fiber layer (DONFL) appearance.11,16 The risk of RNFL dissociation has not been shown to decrease with time, and in fact, authors have observed new RNFL dehis-
cence up to 6 months postoperatively.11,17 We confirmed this finding in the extended follow-up; additionally, new RNFL dehiscence was observed in both groups beyond 6 months (up to 12 months) postoperatively. These changes had no significant effect on visual acuity, however. The frequency of DONFL was higher in eyes from Group A than in eyes from Group B. Moreover, in Group B, DONFL was observed on the temporal side of the fovea only, which suggests that the changes are indeed induced by ILM peeling. Temporal ILM flap technique minimizes unnecessary trauma to the NFL, which may be advantageous. Skill level varies significantly from surgeon to surgeon. A minimalistic approach in maximizing hole closure is important to avoid RNFL defects and eccentric holes postoperatively. Overall, patients in Group B tended to have a lower incidence of DONFL. Importantly, functional results and closure rates did not differ between Group A and Group B, indicating that it is not necessary for the ILM flap to be as large as in the original inverted ILM flap technique. The ultimate goal of each surgery is to achieve FTMH closure while minimizing iatrogenic complications. Importantly, DONFL has not been definitively associated with a decrease of visual acuity or microperimetry changes11; however, there may be unknown effects on function that are not yet known. It is possible that RNFL dehiscence weakens the retinal structure and thus DONFL shows the healing process. The original inverted ILM flap technique was established by the authors with the primary aim of improving surgical results in challenging cases of FTMH repair. This study focused on maintaining a high percentage of FTMH closures while minimizing surgical trauma induced by ILM peeling. This study reports results from a modification of the original inverted ILM flap technique; in the temporal inverted ILM flap technique, the ILM is removed from the temporal side of the fovea, up to the margins of the FMTH, and inverted to cover the macular hole while leaving the nasal ILM in place. Thus, it decreases the risk of surgical trauma in the area of papillomacular bundle. Temporal inverted ILM flap technique
1850 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
achieves satisfactory anatomical and functional results, as well as reduction of DONFL appearance. Key words: macular hole, ILM peeling, inverted ILM flap, vitrectomy, temporal inverted ILM flap.
9.
10.
References 1. Michalewska Z, Michalewski J, Nawrocki J. Macular hole closure after vitrectomy: the inverted flap technique. Retina Today 2009;3:73–74. 2. Michalewska Z, Michalewski J, Cisiecki S, Nawrocki J. High speed, high resolution spectral optical coherence tomography after macular hole surgery. Graefes Arch Clin Exp Ophthalmol 2008;246:823–830. 3. Imai M, Iijima H, Gotoh T, Tsukahara S. Optical coherence tomography of successfully repaired idiopathic macular holes. Am J Ophthalmol 1999;128:621–627. 4. Kang SW, Ahn K, Ham DI. Types of macular hole closure and their clinical implications. Br J Ophthalmol 2003;87:1015–1019. 5. Michalewska Z, Michalewski J, Adelman RA, Nawrocki J. Inverted internal limiting membrane flap technique for large macular holes. Ophthalmology 2010;117:2018–2025. 6. Mahalingam P, Sambhav K. Surgical outcomes of inverted internal limiting membrane flap technique for large macular hole. Indian J Ophthalmol 2013;61:601–603. 7. Michalewska Z, Michalewski J, Dulczewska- Cichecka K, Nawrocki J. Inverted internal limiting membrane flap technique in macular hole associated with pathological myopia. Retina 2014;34:664–669. 8. Kuriyama S, Hayashi H, Jingami Y, et al. Efficacy of inverted internal limiting membrane flap technique for the treatment of
11.
12.
13.
14.
15.
16.
17.
2015 VOLUME 35 NUMBER 9
macular hole in high myopia. Am J Ophthalmol 2013;156: 125–131. Morizane Y, Shiraga F, Kimura S, et al. Autologous transplantation of the internal limiting membrane for refractory macular holes. Am J Ophthalmol 2014;157:861–869. Tadayoni R, Paques M, Massin P, et al. Dissociated optic nerve fiber layer appearance of the fundus after idiopathic epiretinal membrane removal. Ophthalmology 2001;108:2279–2283. Ito Y, Terasaki H, Takahashi A, et al. Dissociated optic nerve fiber layer appearance after internal limiting membrane peeling for idiopathic macular holes. Ophthalmology 2005;112: 1415–1420. Michalewska Z, Michalewski J, Nawrocki J. Continuous changes in macular morphology after macular hole closure visualized with spectral optical coherence tomography. Graefes Arch Clin Exp Ophthalmol 2010;248:1249–1255. Mitamura Y, Ohtsuka K. Relationship of dissociated optic nerve fiber layer appearance to internal limiting membrane peeling. Ophthalmology 2005;112:1766–1770. Mohammed OA, Pai A. Inverted autologous internal limiting membrane for management of optic disc pit with macular detachment. Middle East Afr J Ophthalmol 2013;20: 357–359. Ï Christensen UC, Krryer K, Sander B, et al. Value of internal limiting membrane peeling in surgery for idiopathic macular hole stage 2 and 3: a randomised clinical trial. Br J Ophthalmol 2009;93:1005–1015. Clark A, Balducci N, Pichi F, et al. Swelling of the arcuate nerve fiber layer after internal limiting membrane peeling. Retina 2012;32:1608–1613. Spaide RF. Dissociated optic nerve fiber layer appearance after internal limiting membrane removal is inner retinal dimpling. Retina 2012;32:1719–1726.
