Fig. 11.1
SS-OCT images showing the sequence of macular hole formation, (a) beginning with vitreomacular traction. (b) Evolution from vitreomacular traction to idiopathic macular hole by formation of intraretinal channel. Image obtained two months after image (a) and shows elevation of photoreceptors (white arrow). Patient was operated on two weeks later for full-thickness macular hole. (c) Three-dimensional image of macular hole formation. (d) Cystic spaces (white star) seen prior to the formation of an outer lamellar macular hole. (e) Appearance of outer lamellar macular hole (white hash), just preceding full-thickness macular hole formation. (f) Full-thickness macular hole in the same eye
We also have very little information about the role of the choroid. Our group found that choroidal thickness did not change in any quadrant in eyes with full-thickness macular hole (FTMH) as compared with fellow eyes and healthy control (Fig. 11.2). However, in eyes with full-thickness macular holes, the lamina suprachoroidea was more often irregular and suprachoroidal space was more frequently visible when compared to healthy controls as previously described by Michalewska et al. [6]. Postoperatively the lamina suprachoroidea has a tendency to become more regular.
Fig. 11.2
(a) Swept source OCT of a macular hole before surgery (upper image) and after surgery (lower image). The white lines represent the regularity of the choroidoscleral boundary (CSB). Ch choriocapillaries; HL Haller layer; SL Sattler layer. (b) Swept source OCT of a macular hole before surgery (upper image) and after surgery (lower image). The white arrows indicate an irregular outer choroidoscleral boundary. the irregularities remained after surgery. (c) The arrows indicate the suprachoroidal space in an eye with full-thickness macular hole before (upper image) and after (lower image) surgery. (d) Fellow eye of an eye with a full-thickness macular hole. Posterior hyaloid is partially detached but adheres to the optic nerve. Irregular fovea contour and Drusen are visible. The enlarged area presents suprachoroidal layer and suprachoroidal space
The fact that changes in the outer choroidoscleral boundary are more frequently observed in fellow eyes of FTMH than in healthy controls may indicate that the choroid has a role in FTMH etiopathogenesis. The normalization of the outer choroidoscleral boundary after vitrectomy may additionally indicate a role of the choroid in the healing process of FTMH [6].
11.1.1 Spontaneous Closure of Macular Holes
In very rare cases, the macular hole may close spontaneously without any surgery (Fig. 11.3). In these eyes, closure of macular hole begins at the inner layers of the retina. Our observations of these cases allow us to suggest that if the edges of macular hole are not completely smooth, spontaneous healing may occur [7].
Fig. 11.3
(a) An SS-OCT scan from June 2015 and (b) SS-OCT scan from December 2015. No surgery was performed. Posterior hyaloid detachment is visible on both sides (arrows)
11.1.2 Surgical Closure of Macular Holes
After surgery, the macular hole closes during the first day in about 55% of cases and in 65–75% within 48 hours [8, 9]. This has been presented with a specially prepared OCT device, which allows OCT scans to be performed whilst the patient remains in a prone position [8]. Kikushima demonstrated that SS-OCT allows retina scans as early as 20 min after surgery in a gas-filled eye [9]. Because we perform the inverted ILM flap technique with an air tamponade in idiopathic macular hole cases, we can obtain good quality scans with the SS-OCT device after a few days. This is possible because air disappears quickly in the first days after surgery. We can observe U-type and V-type or irregular closure at the one-week follow-up visit. The best functional results are achieved with U-type closure [10]. Macular hole closure always begins in the inner retina layers (Fig. 11.4).
Interestingly, we may see at the one-week visit that macular hole is closed with ILM only (Fig. 11.4d). This may happen in 15–30% of cases (Nawrocki, unpublished data). We would like to suggest that those holes would not close or would stay flat open if a traditional approach was performed. Flat open macular hole appearance is defined as edges of the hole being flat on the pigment epithelium while bare pigment epithelium is visible in the center of the fovea (Fig. 11.4e). In the past this was considered as macular hole closure, but in our opinion such cases should rather be considered as surgical failures because visual acuity shows no improvement in follow-up.
In 2008, using spectral domain OCT, our group described different retina structure pathologies after macular hole surgery [10]. The following retinal abnormalities were observed: photoreceptor defect defined as linear lack of photoreceptors in the subfoveal area with normal retinal reflectivity; cysts in outer retinal layers; nerve fiber layer defect; elevation of all retinal layers in fovea; and retinal pigment epithelial defects. The study was performed after vitrectomy with ILM peeling and gas for idiopathic FTMH. Even at that time, we suggested some regeneration process occurred in the foveal architecture.
Due to the high resolution of SS-OCT, we can sometimes visualize the ILM flap covering the macular hole at the first post-operative follow-up (Fig. 11.5). After employing the inverted ILM flap technique, it is interesting that in many cases, such as U-type closure (Fig. 11.4a), we do not see any sign of the inverted ILM flap itself on the surface of the retina, whereas in V-type closure or irregular closure cases, we see some presence of ILM (Fig. 11.4b, c).
Fig. 11.4
(a), U-shape closure. Photoreceptor defect is visible in the fovea (arrow). (b) V-shape closure. ILM is visible on the surface of the retina (arrow). No photoreceptor layer is visible. (c) Irregular closure. ILM visible on the surface of fovea, photoreceptor layer almost intact. (d) ILM flap closure. Macular hole closed with the ILM only (arrow), no retina tissue visible in the center of fovea. (e) Flat open macular hole
Fig. 11.5
3-D SS-OCT image showing an ILM flap on the surface of a macular hole
In some cases, during the first control after surgery the macular hole is closed with ILM flap only. We refer to these cases as “flap closure.” During the next weeks or months of follow-up, we may observe almost complete regeneration of retinal tissue under the flap of ILM, and we suggest that this also confirms that ILM, as a base membrane, serves as a scaffold for retinal tissue to proliferate or migrate in the foveal area (Fig. 11.6).
Fig. 11.6
(a) Macular hole preoperative view. (b) One-week postoperative view, macular hole closed with ILM flap only (arrow), visual acuity 20/100. (c) Six-week post-operative control, remnants of the ILM are visible (arrow), foveal layers of the retina are visible, defect in photoreceptor layer, visual acuity 20/50. (d) Twelve months after surgery defect in photoreceptor layer decreased in size (arrow). Visual acuity was 20/30. Please note the increase of nerve fiber layer defects during follow up at the temporal side of the fovea (arrowhead)
During follow-up the ILM flap may become less visible but in some cases, after complete closure of macular hole, the ILM flap may change its position or even float above the surface of the retina. Interestingly, in macular holes initially closed with ILM only, we observe reappearance of retinal layers during the follow-up period and even the reformation of photoreceptor layers, even if the flap itself has become somewhat detached from the retinal surface (Fig. 11.7).
Fig. 11.7
(a), One week after surgery we see the macular hole closed with ILM flap (arrow). (b) One-month follow-up: we see reappearance of inner foveal layers and detachment of ILM flap from the surface of retina (arrow). (c) Two-year follow-up: almost complete recovery of foveal architecture with free-floating ILM (arrow)
After surgery, or in rare cases with spontaneous resolution, macular hole closure always begins in the inner layers of the retina. During follow-up we observe healing of the inner and outer layers of the retina and photoreceptor layers. The process of regeneration lasts many months. We should usually reserve our judgment until 1 year after surgery, and in many cases complete reformation of the retina layers is not observed until even later. It is interesting to observe regeneration of retina tissue during follow-up. It was once thought that nervous tissue does not regenerate. We now know better, and looking at multiple examples after macular hole surgery we clearly see that as retina layers reappear the retina architecture normalizes and this correlates with improvement in visual acuity. These changes are clearly visualized with SS-OCT (Fig. 11.8).
Fig. 11.8
(a) SS-OCT Preoperative appearance of a macular hole. (b) One-week postoperative B-scan shows closure of macular hole with ILM flap, some tissue appears on the outer surface of ILM, cystic space in the outer fovea is present. No defects in the nerve fiber layer. Visual acuity 20/100. (c) Two-month follow-up shows recovery of inner retinal layers defect in the photoreceptor layer is visible, visual acuity 20/40. (d) Two-year follow-up, almost complete recovery of retina tissue can be seen. The ILM flap is barely visible. A minimal defect in photoreceptor layer in fovea still persists. Visual acuity 20/30. Please note increased defects of nerve fiber layer during follow-up at the temporal side of the fovea (arrows)
With the new device we were also able to present that, contrary to the regeneration of foveal architecture, we see new defects in the nerve fiber layers as shown in Fig. 11.8c, d. These defects are known as dissociated nerve fiber layer (DONFL) [11]. DONFL is defined as indentation of inner retinal layer or formation of small dimples visible in the retinal nerve fiber layer, and can appear even 6–12 months after surgery. The most recent modification of the inverted ILM flap technique, called “the temporal inverted ILM flap technique,” allows us to reduce the number and area of nerve fiber layer defects [2]. With this technique we observe DONFL only on the temporal side of the fovea where ILM had been peeled. We do not see them between the fovea and the optic nerve where ILM was not peeled.