Macular Hole Surgery

Abstract

Macular hole surgery is one of the most rapidly changing fields in vitreoretinal surgery, the authors discuss the recent acknowledgments and surgical options. Macular holes are classified, and surgical techniques are described in order to have the most successful procedure. Diagnostic tools and surgical instruments improvement allow surgeons to face difficult cases with a variety of surgical options unknown until a few years ago and is mandatory nowadays to approach the different patients with a broad mind.

1. Introduction

A macular hole (MH) is a full-thickness defect of the neurosensory retina involving the fovea Figures 1 and 2. The prevalence of macular holes is estimated at 0.1% in individuals aged 40 years or older and 0.8% in those aged over 74 years. Idiopathic macular holes account for up to 85% of all macular holes. Other causes include blunt trauma, high myopia, macular schisis, macular telangiectasia type 2, wet age-related macular and surgical trauma [1, 2]. Most patients are females over 65 years of age and may also be seen with myopic eyes [3].

Figure 1.

Full-thickness macular hole OCT cross-sectional image.

Figure 2.

Intraoperative fundus pictures of a full-thickness macular hole.

2. Pathogenesis of idiopathic MH

The posterior vitreous cortex exhibits anteroposterior and tangential traction forces on the fovea [4]. The role of vitreomacular tractions in the formation of the MH is supported by the reduced incidence of bilateral MH in patients with a posterior vitreous detachment (PVD) in the fellow eye as a PVD significantly reduces the risk of MH formation [5].

3. Clinical presentation

Patients with macular hole usually present with decreased vision, central scotoma and metamorphopsia. Slit-lamp bio-microscopy with 78, 90 D or a fundus contact lens is the best way to visualise the hole clinically. The size of macular hole, status of the vitreous, presence of epiretinal membrane (ERM), degenerative changes of the Retinal Pigment Epithelium (RPE), overlying operculum and presence of surrounding cuff of fluid should be reported. Traumatic macular holes can be identified by their ragged and irregular margins and can be easily differentiated from idiopathic macular holes clinically as well as by the history of previous trauma. Full-thickness macular holes can be differentiated from pseudo-holes or lamellar holes by the Watzke-Allen test or laser aiming beam test but these clinical tests have been superseded by the common use of ocular coherence tomography.

4. Investigations

4.1 Ocular coherence tomography

Swept-Source Ocular Coherence Tomography (SS-OCT) gives a high-resolution image of the vitreoretinal interface, neurosensory retina, retinal pigment epithelium and choroid. This investigation identifies surrounding epiretinal membrane, the cuff of subretinal fluid and intraretinal cystic change, vitreo-foveal adhesion, operculum or pseudo-operculum and status of RPE. It also allows accurate calculation of various macular hole indices that have a role in giving a prognosis.

4.2 Macular hole indices

1. Macular Hole height (MHH): Vertical length between RPE and the highest point of the hole

2. Base diameter: Measured at the level of RPE

3. Minimum linear diameter (MLD): Minimum dimension

4. Macular hole inner opening: Distance between innermost layer

5. Tractional hole index (THI): MHH/MLD

6. Macular hole index (MHI): MHH/Base diameter

7. Hole forming factor (HFF): (left arm length of macular hole + right arm length of macular hole)/base diameter

Greater MHI and THI are linked to better post-operative functional results. MHI relates to external limiting membrane (ELM) restoration, while MLD and BD are to anatomical closure (Figure 3).

Figure 3.

Preoperative macular hole indices.

5. Fundus autofluorescence

This test also helps with prognosis. The absence of foveal pigments in macular hole leads to hyper autofluorescence of healthy RPE. Poor autofluorescence from RPE due to pigmentary or other degenerative changes in the RPE is associated with poor visual prognosis.

5.1 Classification

Idiopathic full-thickness macular holes are produced by traction exerted by the posterior vitreous cortex on the neurosensory retina in the central macular area. Disinsertion of the overlying glial plug may result in the formation of a foveal “cyst.” Further traction with removal of the cyst roof produces a free-floating operculum on the posterior vitreous cortex. The neurosensory retina then displaces centrifugally as a full-thickness retinal hole.

Gass described MHs according to clinical evolution:

1. Stage 1a: (Impending Hole): 100–200-micron, foveolar detachment- Yellow Spot

2. Stage 1b: (Occult Hole): 200–300-micron, foveal detachment- Yellow Ring

3. Stage 2: Small full-thickness macular hole less than 400 microns

4. Stage 3: Full-thickness macular hole of more than 400 microns with or without an operculum. No Posterior Vitreous Detachment (PVD)

5. Stage 4: Full-thickness macular hole with complete PVD.

A new classification introduced by the International Vitreomacular Traction Study Group was facilitated by high-resolution OCT imaging and was defined as: vitreomacular adhesion (VMA), vitreomacular traction (VMT) and macular hole (Table 1) [2].

VMA	Size: focal (<1500 mm) or broad (>1500 mm)
VMT	Size: focal (<1500 mm) or broad (>1500 mm)
Full-thickness Macular hole	Size: small (250 mm),; medium (>250 e < 400 mm),; large (>400 mm
Vitreous	Status of vitreous: with or without VMT primary or secondary

Table 1.

International Vitreomacular traction study group classification.

Evidence from a large surgical series of 1483 primary MH repairs reported by Steel et al. [6] treated with vitrectomy, ILM peel, and gas or air tamponade noted that a linear diameter of 500 μm or less was the threshold for successful hole closure. Other studies demonstrated a similar correlation between MH size and successful hole closure, and it was suggested that the definition of large MH’s should be changed to ≥500 μm [7, 8, 9]. The evolution of holes is variable, with a tendency to progress with time in 74% of cases within 2 years [10]. Small holes may close spontaneously in about 5% of cases and it may be reasonable to closely observe small holes with good vision in case of spontaneous resolution [11].

High myopia is a well-recognised risk factor for unsuccessful MH repair, with an axial length (AL) > 26 mm or refraction higher than −6 Diopters making successful hole closure less likely [12]. Anatomical success is reduced as the degree of myopia increases and ranges from 91.7% (AL: 26–29.9 mm) to 0% in eyes with AL > 30 mm [13].

6. Pharmacologic treatment

The only pharmacologic treatment available to date as an alternative to surgical treatment is Ocriplasmin (OCP). This treatment is approved in cases of symptomatic VMA including VMA with MH less than 400 microns. While Ocriplasmin releases VMT in the majority of eyes, the success rate for macular hole closure remains limited and is in the range of 40–50% in eyes with a small-diameter macular hole but decreases to 15–20% in the medium-sized macular hole [14, 15]. Floaters can still be troublesome after OCP and there is a side effect profile that has resulted in most vitreoretinal surgeons still favouring a formal surgical procedure over pharmacological vitreolysis with OCP.

7. Surgical treatment

7.1 Timing

Jaycock et al. [16] reported a closure rate of 94% among patients having surgery within 1 year of the onset of symptoms. The anatomical success was noted to be reduced to 47% when surgery was postponed more than 12 months. Macular holes operated after 1 year yield poor functional improvement even if anatomic closure is obtained. Limited visual improvement may be obtained after successful closure of chronic MH and is linked with the MH duration [17]. One possible reason for a failed surgical outcome in chronic MH surgery may be due to a strong adhesion of the photoreceptor layer of the retina to the underlying RPE [18, 19, 20].

7.2 Evaluation of outcome of surgery

It is difficult to accurately compare the results of different studies due to the lack of a uniform method for determining the integrity of the outer retinal layers after MH vitrectomy. The rapid development of the OCT has facilitated images with a very high resolution. This has allowed accurate interpretation of the outer retinal layers and the ability to compare them to histological sections of the retina.

The outer retinal layers can be divided into 4 lines or bands [21] and the presence of well-defined bands is associated with good vision. The first band arises from the External Limiting membrane (ELM). The second band is referred to as the inner segment – outer segment (IS-OS) intersection of the photoreceptors. The third band has been described as the COST (cone outer segment tips), intermediate line and more recently the interdigitation zone (IZ). The fourth band corresponds to the RPE. Iwasaki et al. [22] also described a method in which the ELM recovery rate and the EZ recovery rate were evaluated in the treated groups.

Caprani et al. [23] describe the restoration of outer retinal layers from the external limiting membrane (ELM), inner segment/outer segment junction (IS-OS), and cone outer segment tips (COST) to retinal pigment epithelium (RPE) after IMH surgery and its relation to visual acuity. They classified the layers as either present or absent after MH surgery with ILM peeling. They also observed the ELM was the first layer to be restored in the healing process, while the integrity of the ellipsoid zone was present in 53.5% of patients at 3 months and in 73.91% at 6 months.

7.3 Surgical technique

Surgical treatment was pioneered in 1990 by Kelly and Wendel [24] by performing 20 gauge pars plana vitrectomy, PVD induction, intraocular gas tamponade and postoperative face-down positioning. They reported an anatomical closure rate of 58% and 73% and visual improvement in 42% and 55% of cases in two consecutive reports.

Eckardt [25] in 1997, introduced the concept of internal limiting membrane (ILM) peeling in the management of macular holes. Peeling of the internal limiting membrane relieves the tangential traction caused by glial cells and improves the anatomical and visual success rates of macular hole surgery.

A Cochrane review in 2013 [26] showed better visual results, anatomical closure rates and lower re-operation rates in patients who underwent ILM peeling when compared to those who underwent pars plana vitrectomy alone.

7.4 Microincision vitreoretinal surgery (MIVS)

The advent of smaller gauge and better instrumentation, tissue staining during ILM removal, combined phaco-vitrectomy surgery and reduced or no postoperative face-down positioning are all factors that have facilitated surgery and made it technically less challenging as well as being less obtrusive and difficult for the patient.

Shift to small gauge instrumentation in the last 20 years may have resulted in a slightly increased duration of the vitrectomy operation due to the increased cut rates and slight reduction of fluid passage in the eye. MIVS has however caused a significant reduction in complications like entry site retinal breaks and vitreous traction-related retinal damage (Table 2).

	PRO	CONS
20G	Instrument stiffness, illumination; Easy PVD induction	Scleral wound needs suturing; Retinal and entry site breaks
23G	Safer vitreous removal, efficient aspiration for PVD induction	Wounds often require suturing
25G	Sutureless wounds, very controlled vitreous removal	Wounds rarely require suturing; Slightly greater flexibility of instruments
27G	Sutureless wounds, safe vitreous base shaving; Fast healing	Difficult PVD induction, flexible instruments and less illumination. Longer vitrectomy time, sometimes difficult to grasp the membrane.

Table 2.

Comparison of 20-G, 23-G, and 25-G vitreous cutters and instrumentation.

7.5 Gas tamponade

Filling the vitreous cavity with gas encourages hole closure by preventing fluid from accessing the hole and may facilitate the formation of a glial plug that can contract and help with hole closure. The buoyancy effect of the gas is also thought to play a role by direct pressure on the hole and hence the reason that face-down posturing has been recommended in the postoperative period. Gas mixtures used in different percentages are C3F8, C2F6, SF6 and air. Recent studies have shown that longer-acting gases like C3F8 and C2F6 may not be needed as SF6 and air show similar anatomic results in terms of hole closure and better patient compliance [27, 28]. Silicone oil is rarely used as a tamponade agent and needs a further surgical procedure to remove the oil.

7.6 Postoperative prone posturing

Facedown posturing (FDP) has been recommended because the gas bubble with its surface tension forces may support the apposition of the MH edges and also provide a scaffold for the migration of glial cells and blocking fluid entry into the hole.

The force of the gas bubble is greatest at the apex of the arc of contact to the retinal surface and diminishes from this point [29].

Postoperative face-down positioning night and day for a week was recommended as a critical step for many years to increase the buoyancy force exerted on the posterior pole but studies have subsequently shown that prolonged face-down posturing may not be needed especially for smaller holes. Recent publications report closure rates for small- to medium-size MHs at around 95% in both postured and non-postured groups indicating that prolonged posturing is not necessary for MH closure after surgery. The authors concluded that face-down posturing is not necessary for medium-sized MHs [30, 31, 32]. Large holes >400 microns may benefit from prolonged postoperative posturing, but studies still show inconclusive results [33].

Ye et al. [34] in a meta-analysis of five randomised controlled trials compared MH surgeries with ILM peeling with postoperative FDP versus those with non-supine posturing (NSP). The MH closure rate was higher in the FDP group, with a significant difference in the closure rate for MH with size >400 μm, but not for those <400 μm.

Eckardt et al. [35] considered stopping the FDP as soon as the OCT confirmed the closure but in case of non-closure on day 3, it required a second procedure on day 5 or 6 to facilitate hole closure. It has been noted that increased duration between the first and second surgery when there is failed primary repair may result in a worse visual and anatomical outcome [36, 37, 38].

Nearly all patients having vitrectomy surgery with a gas bubble will develop cataract. A pilot study by P R Simcock and S Scalia [39] showed that combining lens removal at the time of vitrectomy resulted in greater space for the gas bubble and better tamponade and patients had successful hole closure without having to adopt a prone posture as well as the advantage of not having to return for further cataract surgery.

Chakrabarti et al. [40] described the use of autologous gluconated blood (AGBL) in the macular area for treating MH without the use of intraocular gas or prone positioning. They report no significant side effects and 26 patients with large MH’s, obtained 100% closure using an inverted ILM flap and AGBL to encourage macular healing with good functional outcome. The technique involves a preparation of AGBC before the surgery (1 mL of 5% glucose added to 2 mL of autologous blood) then an ILM flap was created and AGBC on top to create a macular plug. This technique allowed the patients to adopt a comfortable position without gas tamponade.

8. Staining

Vital dyes were introduced more than 20 years ago to allow better visualisation and removal of the vitreous and staining and removal of epiretinal membranes and the internal limiting membrane [41, 42].

8.1 Triamcinolone acetonide

Triamcinolone crystalline microparticles are trapped in the vitreous gel facilitating gel removal by improving visualisation and determining if there has been a vitreous separation from the retina. The microparticles also tend to settle on membrane surfaces creating a demarcation between the peeled and remaining membrane.

8.2 Indocyanine green (ICG) and Infracyanine green (IfCG)

ICG green stains the ILM because of its affinity to laminin and collagen type IV within the ILM. ILM peeling was popularised as a result of this stain as it was a technically challenging procedure to perform without the stain. It has however been criticised due to complications associated with cytotoxic and phototoxic effects on the exposed RPE. Infracyanine Green at a concentration of 0.5 mg/ml also allows good visualisation of the ILM but with a better safety profile.

8.3 Trypan blue

Trypan blue 0.2% is also used to stain the ILM or ERM during vitreoretinal surgery and is often injected after a fluid air exchange to increase the staining of the membrane (Figure 4).

Figure 4.

Trypan blue is injected under the air bubble.

8.4 Brilliant blue G

The brilliant blue stain has been used for selective ILM staining. It was approved in the European Union in 2007 as Brilliant Peel. Brilliant Blue G formulation uses 4% polyethylene glycol (PEG) as a carrier for the dye and prevents the blue stain from diluting in the vitreous cavity. Since the dye is heavier than water it settles at the posterior pole with a standard injection technique.

9. Dye injection techniques

There are different ways to protect the RPE cells during dye injection in MH surgery:

1. gentle injection of the dye (this is the most common way that is used now as Brilliant Blue is very safe to use and not harmful to the RPE)

2. placing substances such as sodium hyaluronate over the MH.

3. perflourocarbons liquids (PFCL),

4. autologous blood

The most common techniques in dye injection are:

1. The “dry method” consists of removing the balanced salt solution (BSS) in the vitreous cavity by a fluid-gas exchange before dye injection. The technique has the advantage of concentrating the dye in the posterior pole, exposing the retinal surface to a higher concentration of dye in the vitreoretinal interface

2. The “wet method” while the surgeon injects the dye in a fluid-filled vitreous cavity. The amount of dye in contact with the retinal surface is lower because it is diluted by the fluid in the vitreous cavity. The wet method is safer and faster but may be less effective in ILM staining, particularly in highly myopic eyes. If a poor stain is encountered, it is often worth re-staining as there may be greater stain uptake on the second attempt.

10. Macular hole closure

A macular hole is said to be closed when there is flattening and reattachment of the hole rim along the whole circumference of the macular hole (Figures 5 and 6).

Figure 5.

Full-thickness macular hole preop cross-sectional image and measurements.

Figure 6.

Macular hole closure after successful vitrectomy, ILM peeling and gas tamponade.

Two types of closure have been defined [43].

• Type 1 closure: No interruption in the foveal retinal tissue above the RPE layer (Figures 7 and 8).

• Type 2 Closure: Interruption of inner retinal tissue and central RPE exposure.

Figure 7.

Full-thickness macular hole with operculum.

Figure 8.

Macular hole closure after surgical treatment.

Imai et al. [44] gave another OCT-based classification of macular hole closure and divided it into three types:

• Type “U”- Normal foveal contour on OCT

• Type “V”- Steep foveal contour

• Type “W”- Foveal defect of neurosensory retina

The various lines seen in the outer retinal on OCT scanning have been mentioned earlier in this chapter, namely the External limiting membrane (ELM), Inner segment/Outer segment junction (IS/OS junction) or Ellipsoid zone (EZ), Cone outer segment tips (COST) or Interdigitation zone (IZ) and Retinal pigment epithelium (RPE).

Structural changes in the macular region after surgery are correlated to photoreceptor alterations [45, 46, 47, 48]. They can be related to the outer retinal bands seen on OCT scanning. The ELM is the first retinal layer to recover, followed by the IS/OS, and lastly the COST (Figure 8).

Min Woo Lee et al. [49] analysed the process of recovery of the ELM, Outer nuclear layer (ONL), and EZ after surgical treatment. They tried to identify the factors affecting changes in visual acuity and those associated with EZ recovery. They divide full-thickness MH healing into the following stages: the first stage occurred when the traction from ILM to the neuroretina was released after surgery resulting in the resolution of intraretinal cysts, this phase occurred when the inner retinal layers grew to the centre of the macular hole and two edges were connected by forming a tissue bridge. At this stage, SRF could still be found on B-scan. Finally, the SRF resolves and photoreceptors begin remodelling, which could lead to restoration of the ellipsoid area (Figure 9).

Figure 9.

Macular hole closure: Preoperative optical coherence tomography (OCT) and 10 months postoperative OCT.

11. Role of the internal limiting membrane

The ILM is the most superficial layer of the retina and is formed by the Muller cell footplates together with a fibrous component. This membrane is 400 nm thick at the retinal periphery, rising to about 1400 nm in the macular region [50]. This structure provides greater mechanical strength than retinal cell layers, being responsible for 50% of retinal stiffness. The rationale for its removal is to relieve all tangential traction around the macular hole and improve hole closure rates [51, 52, 53].

During macular hole surgery, central vitrectomy is usually followed by PVD induction by placing the cutter on aspiration mode in the proximity of the optic disk until a Weiss ring is noticed and a wave of circumferential vitreous separation is seen. Vitreous removal with a high-speed cutter to the retinal periphery is then performed. At this stage, any epiretinal membranes (ERM) surrounding the macular hole should be stained and removed up to the borders of the hole. If no ERM is identified, then most surgeons would routinely remove the ILM. The advent of tissue stains and dedicated surgical instrumentations allows the identification and removal of the inner limiting membrane and this procedure can be described as total peeling (TP) when the ILM around the MH is removed or foveal sparing peeling (FSP).

12. ILM peeling technique

12.1 Initiation

The creation of an ILM flap is a crucial step in ILM peeling. The ideal starting point has been suggested as about 1000 microns above or below the fovea. A small ILM tear is created on the retinal surface by scraping the retinal surface with a diamond-dusted or using picks, bent MVR blades or with dedicated ILM forceps using a pinch and peel technique. Many surgeons utilise custom-designed micro forceps to pinch and lift the ILM and create a small flap to then grasp and customise the direction and shape of the desired flap (Figure 10).

Figure 10.

The finesse SHARKSKIN ILM forceps.

12.2 Flap creation

Vitreoretinal forceps and the diamond-dusted membrane scraper are valuable tools for flap creation [54]. The majority of surgeons tend to peel an ILM area of about one disk diameter around the fovea, releasing enough retinal tissue in order to allow macular hole closure (Figure 11). A broader area of ILM up to 3 disk diameters may be removed to improve retinal compliance and subsequent retinal closure but it is still unclear regarding the exact dimension of ILM removal needed.

Figure 11.

ILM flap folded over the macular hole.

12.3 Side effects

ILM peeling is one of the most demanding procedures in ophthalmic surgery requiring a high level of manual dexterity. Retinal damage can occur at the initial ILM flap creation point resulting in retinal haemorrhages and nerve fibre layer damage as well as iatrogenic eccentric retinal holes have been reported [55]. A peculiar retinal alteration in the area where the ILM is removed has been described as Disassociated Optic Nerve Fibre Layer (DONFL) with a characteristic change in inner retinal morphology and appears related to Muller cell end plate damage [56, 57, 58].

12.4 ILM peeling variants

12.4.1 Foveal sparing peeling

This technique entails ILM removal but sparing a circular area of 400 microns around the MH rim. In a study by HO [59] foveal sparing resulted in better visual acuity and postoperative anatomy, based on the concept that Muller cells are important for maintaining foveal architecture. Muller cells may improve light transmission to the photoreceptors [60]. In a study by Morescalchi [61], a total of 46 eyes had macular hole surgery and were randomly allocated to complete or foveal-sparing peeling. The latter group demonstrated a greater increase in foveal sensitivity following surgery. Muller cells preservation may therefore provide a better functional outcome after macular hole surgery [62, 63].

12.4.2 ILM abrasion

Mahajan [64] reported a different procedure to reduce trauma during ILM removal by using a diamond-dusted membrane scraper (DDMS) in circumferential and centripetal motions around the MH. This procedure was thought to encourage glial cell activation, losing ILM attachment to the underlying retinal tissue and ultimately encouraging MH closure.

ILM peeling was initially described in 1997 by Eckardt [25]. The rationale for ILM peeling was the removal of residual adherent vitreous cortex remnants, thereby increasing retinal compliance. ILM serves as a scaffold for cellular proliferation, and its removal should encourage MH closure. Peeling the ILM ensures the thorough removal of any tangential tractional components implicated in the development of macular holes. The removal of a potential scaffold for the re-proliferation of myofibroblasts may reduce the possibility of late reopening of surgically closed holes. Furthermore, peeling off the ILM is also believed to stimulate wound healing at the macula, possibly by inducing local expression of growth factors that promote glial repair (Video 1, http://bit.ly/433B3XW).

The Cochrane database of systemic reviews concluded in 2013 that there was enough evidence to support the positive effects of ILM peeling for stages 2–4 idiopathic MH’s to improve the primary anatomical hole closure rate, although no clear benefit was found for small holes [24].

ILM peeling increases the likelihood of successful macular hole closure, but it has been suggested that the extra manipulation of this membrane at the time of surgery could be harmful to the retina.

Swelling of the arcuate nerve fibre layer after internal limiting membrane peeling was described by Clark [65]. Electrophysiologic studies using focal macular electroretinogram showed a delayed recovery of the b-wave 6 months after macular hole surgery in eyes that underwent ILM peeling compared to those without peeling [66].

Although anatomical closure achieved after complete ILM peeling was associated with improved visual outcomes, the rate of anatomic closure was inversely correlated with the extent of ILM peeling actually achieved. It was suggested that excessive unsuccessful attempts at ILM peeling might enhance anatomic success (possibly through enhanced promotion of glial healing) at the expense of poorer visual outcomes, presumably resulting from damage to inner retinal elements [67].

Inner limiting membrane peeling appears to improve the rate of anatomical closure, but its effect on visual outcome is less predictable and unsuccessful attempts to peel the ILM are associated with poor visual outcome. While ILM peeling may be performed for full-thickness macular holes of any stage, it is more commonly reserved for stage 3 or 4 holes, long-standing holes, those that have failed to close, or those that have re-opened following conventional surgery.

ILM peeling is thought to be beneficial for macular hole closure and in particular for large holes, however, it may also cause side effects on retinal anatomy and functionality. Retinal changes described after ILM peeling included inner retinal dimpling [68], dissociated retinal nerve fibre layer (RNFL), [69] and reduced parafoveal retinal thickness [70].

13. Internal limiting membrane peel extension

ILM peeling has improved surgical success in MH surgery. The ILM can be removed using forceps, diamond dusted scraper or other surgical tools and it has been suggested that at least 2 disc diameter (DD) of ILM should be removed around the fovea. Different surgeons perform this delicate procedure using peel radii from 0.5 to 3 DD. Bae et al. [71] compared peel sizes of 0.75 DD versus 1.5 DD and showed that a larger ILM peel lessens postoperative metamorphopsia. Modi et al. [72] have shown similar results in MH closure rates with 3 mm versus 5 mm peel sizes. No consensus on optimal peel width exists, but case reports of wide peels having success in large MHs have led many surgeons to peel ILM up to the arcades [73].

13.1 Macular holes >400 microns

In macular holes <400 microns conventional ILM peeling provided better functional outcomes compared with the inverted flap technique and should be advocated [74].

FTMH above 400 microns have a reduced rate of success with standard surgery, and other techniques such as ILM flaps and retinal expansion may be preferred for these macular holes [7]. Vitrectomy with the inverted ILM flap technique seems to be effective surgery for large idiopathic and myopic MHs, improving both functional and anatomical outcomes in a study by Rizzo [75] and surgical closure in all patients is reported in a study by Yamashita [76].

The practice of using the ILM on top of a macular hole has many effects, such as a bandage, isolating the retinal hole from the vitreous fluids, as a scaffold stimulating glial tissue proliferation, and will all encourage hole closure. In a study by Michalewska et al. [19] the use of ILM flap increased the rate of MH closure up to 98% for large MHs. This technique has also resulted in a 100% macular hole closure rate in myopic MH’s reported in some studies [77, 78].

ILM flaps can be divided into two groups, inverted and non-inverted flaps:

Inverted flaps

• the folded inverted flap

• temporal inverted flap,

• nasal inverted flap (Texas Taco technique)

• superior inverted flap

• the cabbage leaf flap

• SWIFT (superior wide base flap transposition)

These flaps require the presence of ILM still hinged around the hole or near the hole in order to prevent displacement of ILM in the vitreous cavity. Michalewska [79] describes a procedure where the ILM is engaged with ILM forceps and removed almost in its entirely around the macular hole but a tiny residual attachment is left in place. Folded ILM is then packed inside the MH rather than a flap covering it. ILM is massaged into the MH from all sides until it becomes inverted and may be described as an ‘ILM plug’. At a microscopic level, the presence of this plug may hamper outer retinal layer healing and prevent visual acuity recovery by interfering with photoreceptor reconstitution [80]. ILM tissue however is also known to work as a scaffold for tissue proliferation, promoting photoreceptors restoration and providing guidance for the correct positioning of the cells [81]. Rossi [82] stated that by using the fill technique, the ILM acts as a filler, glue, and scaffold all at the same time.

Shin [83] first introduced a true flap technique (single-layered flap of the ILM) for covering MH’s (larger than 400 μm) with the assistance of perfluoro-n-octane (PFO) in 2014. Perflurocarbon liquid allows ILM flap stabilisation during fluid-air exchange by reducing dislocation or flap loss.

The temporal inverted internal limiting membrane flap was described by Michalewska [79]. The nasal ILM was left in place to protect the tissue from surgical trauma and lessen the occurrence of dissociated optic nerve fibre layer (DONFL).

The opposite approach has also been described and named the “Texas taco” [84]. This procedure involves peeling the nasal ILM and then it is folded temporally to cover the MH.

Ghassemi et al. [85] investigated the results of different ILM flap directions. The techniques used were a hemi circular ILM peel with a temporally hinged inverted flap, a circular ILM peel with a temporally only hinged inverted flap and a circular ILM peel with a superior inverted flap. Similar results were obtained with all flap directions.

Aurora [86] described the Cabbage Leaf Inverted Internal Limiting Membrane Flap technique. Three inverted ILM flaps sealed the hole looking like cabbage leaves. These flaps were connected to the edge of the MH, trimmed and flipped over the MH, one above the other like seen with cabbage leaves.

In cases when the central ILM had already been removed during previous surgery, Tabandeh [87] described the SWIFT flap procedure. An ILM flap is fashioned from superior residual ILM then a narrow strip of residual ILM forms the base of the flap, which is positioned horizontally. The ILM flap is inverted over the macular hole and covers the MH and the retina inferior to the MH Figures 12 and 13.

Figure 12.

SWIFT (superior wide-based FLAP transposition) FLAP.

Figure 13.

Macular holes covered with inverted ILM.

Leisser [88] described a technique where a temporal ILM flap was prepared while the residual ILM around the MH was peeled to the rim of the MH, after which the ILM flap was positioned in an inverted fashion over the MH (Figures 14 and 15).

Figure 14.

Temporal residual ILM peduncolated flap.

Figure 15.

Macular hole covered.

The pedunculated flap technique creates an ILM flap that covers the macular hole from the border of the previous ILM peel. The flap should be large enough to cover the region of the pre-existing ILM peel as well as the macular hole.

Non-inverted flaps

• Pedicle ILM transposition

• Retracting door

• Free flaps

14. Pedicle ILM transposition

Hu [89] suggested a non-inverted flap provides a more physiological scaffold for healing. The pedicle is created by circular ILM peeling in the macular area but leaving a hinge attached superiorly to pivot the mobilised ILM to cover the hole. This method reduces flap loss during fluid-air exchange observed using free flaps [90].

15. Internal limiting membrane retracting door

Finn [91] performed surgery on myopic MH’s using a hinged ILM flap retracted to cover the MH. This procedure addressed the issues of relaxing the stiff ILM in myopic eyes together with providing the scaffold needed for retinal tissue healing (Figures 16 and 17).

Figure 16.

ILM retracting door flap.

Figure 17.

Macular hole covered with non-inverted ILM.

16. Autologous ILM free flap

Eyes with an extensive ILM peel may benefit from a free ILM flap. Keeping the ILM in place during air-fluid exchange is never easy, but a small amount of heavy liquid or viscoelastic like Viscoat, can stabilise the flap [92, 93]. Patients who have a persistent MH hole following earlier surgery with ILM peeling may benefit from the use of an ILM-free flap [94]. Peripheral peeled ILM is placed on the MH during redo surgery. Successful results may be obtained in the closure of large MHs with this technique, but this procedure may cause damage to the retinal pigment epithelium in the fovea and alterations of the photoreceptor layers [95]. Prolonged exposure of the RPE to the staining fluids and stained ILM may also cause chemical damage to the RPE, as detected by De Novelli [96].

17. Lens capsule flap transplantation (LCFT)

When there is no ILM available, the lens capsule may be used as an alternative tool to encourage MH closure in refractory cases [97]. The procedure entails staining the anterior (AC) or posterior capsule (PC). In phakic patients, combined cataract surgery was performed, and the AC was used, and in pseudophakic patients, the PC. The flap was trimmed and positioned within the hole. Anatomic and functional results are promising but the challenge of flap dislocation remains an issue.

18. Autologous neurosensory retinal transplant (ART)

In 2016, Grewal and Mahmoud introduced the use of an autologous full-thickness retinal free flap for closure of refractory myopic MHs. They applied endolaser and diathermy in a circular pattern around a 2-disc diameter area of the retina and using a bimanual approach with vertical scissors and forceps obtained a retinal free flap. Instillation of perfluoro-noctane heavy liquid over the flap was followed by a direct PFC-silicone oil exchange [98, 99, 100, 101, 102].

19. Human amniotic membrane (hAM) transplantation

Rizzo [103] describes the use of a human amniotic membrane (hAM) for MH treatment. A 2 mm disk of hAM grasped by forceps under fluid or perfluorocarbon was transplanted into the subretinal space. It is speculated that the hAM stimulates retinal pigment epithelium (RPE) cells division.

Kuriyan [104] described the use of commercially available human amniograft from Bio-Tissue (Tissue Tech, Miami) using sub- and pre-retinal placement. A dermal punch may be used to trim the tissue for hole placement. The chorion side is supposed to face the retinal pigment epithelium. Once in place inside or on top of the MH, the fluid-air exchange was performed.

20. Autologous platelet concentrate

Gaudric [105] explored the effects of autologous platelet concentrate (APC) in macular hole closure by injecting some APC in the hole at the end of a vitrectomy and obtained an improvement in anatomical closure rate. In a study comparing ILM peeling versus ILM peeling plus platelet-rich plasma (PRP) a significant improvement in anatomic and functional results was reported in eyes that had an application of PRP [106]. Several growth factors and cytokines are released by platelets and platelet-rich plasma has been used in various medical conditions [107]. Autologous PRP has been used in more complex holes such as myopic and refractory MH’s with encouraging anatomical and functional results [108]. PRP derived from a patient’s peripheral blood requires special tools, which may not be always available. One study found low rates of closure in refractory MHs with autologous blood [109] but a further study found high rates of closure when combining the inverted ILM flap technique with autologous blood for large MHs [110].

21. Subretinal blebs

The concept underlying this technique is that it is thought to increase retinal compliance by releasing the adhesions of photoreceptors to the retinal pigment epithelium (RPE).

The ILM is peeled in the usual manner at the sites of the injection. A small-gauge subretinal cannula (usually 38-41gauge) is used to inject a small volume of fluid under the retina, and usually few blebs are created by injecting BSS into the subretinal space. A confluent perifoveal serous detachment is induced, using a Tano diamond-dusted scraper or a Flex Loop to help massage the sub-retinal fluid until the retina surrounding the macular hole is detached. A fluid-air exchange is performed, and gas or silicone oil is used. This procedure has been shown to be a viable option for the closure of recurrent or persistent holes [20, 111, 112].

Multiple entry sites into the retina may be avoided as suggested by Felfeli [113] using a silicone extrusion cannula to inject fluid through the macular hole. Once the central retina has been detached by refluxing fluid into the macular hole the margins of the hole are once more re-opposed by carefully massaging them and this is followed by a fluid-air exchange and then gas tamponade. In a series of 39 complex cases, this technique resulted in a 95% closure rate. This procedure is especially useful where chorioretinal scarring is present at the macula resulting in extra adhesion of the retinal layers.

22. Retinal relaxing incisions

Charles [114] reported using retinal relaxing incisions in six eyes to release tangential traction and increase retinal elasticity. This procedure does result in damage to the neuroretina and potentially the underlying RPE and visual improvement were limited to 3 cases.

Reiss [115] described 7 patients treated with five radial full-thickness incisions and gas tamponade. Using this procedure, the investigators reported 100% anatomic success in patients with refractory MH’s.

The need to perform a deep incision in the retina with no damage to the underlying retinal pigment epithelium and choroid makes this procedure more technically challenging than others and with potentially greater risk.

23. Dragging and peeling

Peng in 2020 described a procedure called ILM dragging and peeling [116]. Two horizontal ILM strips were removed in the upper and lower quadrant of the macula. An upper ILM flap was created and pulled down towards the edge of the MH to try and reduce the size of the hole using the adhesion between the ILM and underlying retina. A similar procedure was performed by creating a lower ILM flap which was pulled upwards, also try and reduce the size of the hole by a similar mechanism. Once these tangential forces were applied to minimise the hole size a standard circular ILM peel was created. This surgical procedure derives from the belief that MH mobility is critical for MH closure and it may be that in some cases peeling alone may not give enough mobility to the retina [117]. The authors did not report significant complications and MH closure rate was 96.2%.

24. Adjuvants with ILM techniques

24.1 Autologous blood

Autologous gluconated blood has been used with inverted ILM flaps as a macular bandage, and one study showed initial surgical success in all patients with MH > 500 without gas tamponade or postoperative positioning [118].

24.2 Heavy fluids

Per-fluro-octane (PFO) has been applied to the MH to stabilise the free or inverted ILM flap in the correct position until the end of the fluid-air exchange [119, 120, 121]. PFO could also help flatten the retina whilst the ILM flap is being created by providing a degree of counteraction. Due to its vapour pressure, a small bubble of residual PFO can be removed by evaporation instead of using irrigation, thus reducing the risk of flap displacement [92, 122].

24.3 Viscoelastic

Viscoelastic can facilitate macular hole surgery in a variety of ways. When used before ILM peeling it reduces the toxicity of the dye to the retina. When applied over the ILM flap it can act to reduce the displacement of the flap and keep it in the correct position. It can also act as a binder to stabilise the flap. The functions of adhesive viscoelastic (Viscoat, Alcon) were investigated in a study by Song [123]. Viscoelastic was injected into the MH and then ILM was stained with ICG. The ILM below the hole was removed but the ILM above the hole was used to create an inverted flap. Supplemental viscoelastic was then injected into the surface of the inverted ILM flap prior to the fluid air exchange. This technique resulted in anatomical and functional recovery in highly myopic patients with large MHs.

25. Summary

There are many techniques available for the treatment of macular holes. The authors preferred technique for small to medium-sized macular holes is combined phaco-vitrectomy with 360-degree ILM peel without face down posture but with the patient avoiding lying on his or her back for 1 week. Face-down posturing is reserved for larger holes. With small holes and good vision, a period of observation is an option as these holes can spontaneously improve and so avoid surgery.

With the many techniques available it is possible to postulate a flowchart for patient treatment.

Recent, small Stage 2 MH’s may be managed by PPV and gas tamponade although a short period of observation is an option as some can spontaneously improve and avoid surgery.

MH <400 μm with ERM, PPV with ILM peel is suggested.

MH >400 μm or chronic, inverted ILM flap is preferable.

MH >700 μm, primary ART (Autologous neurosensory retinal transplant).

Refractory holes where ILM has been peeled or difficult ILM flap, consider a free ILM flap or hinged ILM flap. If ILM is not available ART, human amniotic membrane graft and lens capsule are all options depending on the availability of tissue and surgical experience (Figure 18).

Figure 18.

Macular hole surgery suggested approach.