When Should You Peel the ILM in Macular Hole Surgery?

When Should You Peel the ILM in Macular Hole Surgery?

Brian C. Joondeph, MD, FACS

Macular hole surgery is approaching its 20-year anniversary, as the first report of successful closure by Kelly and Wendel appeared in 1991.1 Initial treatment consisted of a vitrectomy with creation of a posterior vitreous detachment placement of a gas bubble with at least one week of face-down positioning.

The next steps in the evolution of macular hole surgery were peeling of the internal limiting membrane and shorter durations of face-down positioning. A large series of 500 cases of macular hole surgery, half with ILM peeling and half without, showed a comparable rate of hole closure, reopening and visual acuity between the two groups, suggesting that ILM peeling may not be necessary.2

A literature review of macular hole case series showed a closure rate of approximately 80% without ILM peeling and 96% with ILM peeling,3 suggesting that ILM peeling does improve the rate of macular hole closure. Nonetheless, ILM peeling is an added surgical maneuver with potential risks as well as increased surgical time. There are also increased costs with ILM peeling, including the possible use of adjunctive membrane stains and disposable ILM forceps or membrane scrapers. The question remains as whether or not the ILM needs to be peeled in order to successfully close a macular hole, or in which cases, specifically, it needs to be peeled.


To approach this problem, let us look at the surgical steps involved in macular hole surgery. These include: (1) vitrectomy; (2) creation of a PVD, if it is not already present; (3) ILM peeling; (4) fluid/air/gas exchange. Steps 1 and 4 are the same regardless of what is peeled or not peeled. Instead, the focus will be on the two middle steps.

Determining the presence or absence of a PVD pre-operatively is generally possible with optical coherence tomography testing, particularly high-definition OCT. A schisis cavity may be present in some cases, giving the appearance of a complete PVD over the macula. In actuality, there may be posterior hyaloid remnants on the surface of the retina. If these remnants are not seen and removed at the time of surgery, the hole is unlikely to close. Visualization agents, which will be discussed shortly, may assist in the identification of remaining posterior hyaloid. Absence of staining, however, does not guarantee absence of posterior hyaloid on the retinal surface. Staining can be variable — intense in some eyes and faint in others.

Peeling of the ILM is the most technically challenging portion of the surgical procedure. This includes the creation of an edge and complete peeling of the ILM around the macular hole. Again, visualization agents can greatly assist in the visualization and identification of the ILM. Added instruments are needed for ILM peeling. These could include a pick or bent MVR blade used to create an edge in the ILM, a diamond-dusted membrane scraper, also used to create an edge in the ILM, and finally intraocular forceps used to complete the peeling. A membrane stain will often be used as well, such as triamcinolone, ICG or trypan blue.

In the current era of increased attention to and scrutiny of costs, these added agents and possible use of disposable instruments will add to the cost of the case. The benefit of membrane stains is that they ensure complete ILM removal with any remnants stained and clearly visible. A discussion of these agents is beyond the scope of this article, but there are several good reviews.4,5 The benefit of ILM peeling is that it ensures complete removal of the posterior hyaloid or any overlying epiretinal membranes, since any hyaloid or membrane remnants on the surface of the ILM would be removed along with the ILM.

There are several arguments against ILM peeling. One is that the added manipulation increases the risk of iatrogenic macular damage. This could include retinal breaks caused by the process of creating and elevating an edge of the ILM. The counter-argument is that membrane staining agents improve visualization of the ILM, making it easier to identify and grasp. In the hands of an experienced vitreoretinal surgeon, the ILM should be no more difficult to peel than an epiretinal membrane. One might also argue that the added manipulation increases the risk of a peripheral retinal break. This complication, however, is usually the result of the induced PVD and resulting traction along the vitreous base and not the ILM peeling.

Another argument against ILM peeling is the risk of toxicity of the staining agents. Much has been published about potential toxicity issues, both in vivo and in vitro, and these will not be reviewed in this article. My opinion is that toxicity is rare and unpredictable, occurring in only a very small percentage of patients. This is despite using a standard surgical approach, with no difference between cases in light exposure, dye concentration, peeling technique, etc. This suggests that toxicity is an idiosyncratic reaction without an easily identifiable cause. The ILM can also be peeled without a visualization agent to avoid concerns of toxicity, especially if the ocular media and surgeon's view are particularly clear. This would avoid dye toxicity concerns but runs the risk of incomplete peeling, which could lead to failure of macular hole closure. Light toxicity is another concern, whether from the fiber optic probe or from the surgical microscope. The added time and light exposure from ILM peeling may have toxic retinal effects, although this is observed clinically on a rare and sporadic basis, much like toxicity from staining agents.

Specific types of macular holes have lower closure rates. Hole size and chronicity may influence closure rates. Larger (>400 microns) and chronic (>6 months) macular holes should have ILM peeling if possible to increase the chance of success. Highly myopic eyes with posterior pole staphyloma should also have ILM peeling to increase the surgical success rate. Traumatic macular holes are another category with a lower success rate that benefit from ILM peeling.

ILM peeling may reduce the duration of face-down positioning required for macular hole closure. With ILM peeling, five or fewer days of face-down positioning may be adequate to effect hole closure. Tornambe has demonstrated a high success rate with no face-down positioning with ILM peeling.6 If the patient may not be able to comply with several days of face-down positioning, it is important to peel the ILM to achieve hole closure without positioning.

Figure 1. In a 54-year-old woman, macular hole reopened (A) one month after surgery due to residual traction (B).

That leaves the more routine macular holes — those of average size and shorter duration. I prefer to peel the ILM on these cases routinely, regardless of whether the macular hole is stage 2, 3 or 4. Peeling the ILM eliminates all tangential traction around the edges of the hole, the process believed to contribute to macular hole formation. It also ensures removal of any hyaloid remnants or epiretinal membranes that could otherwise be missed. This provides the patient with the highest chance of surgical success.


Case 1: A 54-year-old myopic woman underwent macular hole repair several years ago. At the time, ICG dye was not available and triamcinolone was used to "stain" or identify the ILM. In spite of initial hole closure, the hole reopened (Figure 1A) one month after surgery. There was incomplete posterior hyaloid removal, resulting in residual traction that reopened the macular hole (Figure 1B). This underscores the benefit of complete ILM removal as a means of ensuring complete removal of all posterior hyaloid remnants. Staining with triamcinolone, while useful in identifying ILM, is not a true stain and ILM remnants may be left behind if the triamcinolone particles are irrigated from the retinal surface, giving the appearance of a complete peel.

Case 2: A 65-year-old woman underwent macular hole repair using trypan blue staining of the ILM. Her macular hole was of average size, with elevated edges and cystoid macular edema. A posterior vitreous detachment was visible above the macula. She had trypan blue-assisted ILM peeling, which resulted in complete closure of her macular hole and restoration of normal macular architecture. Brisk staining of the ILM with trypan blue and ease of ILM identification was noted, which allowed for complete peeling without leaving any remnants of posterior hyaloid on the macular surface. RP


1. Kelly NE, Wendel RT. Vitreous surgery for idiopathic macular holes. Arch Ophthalmol. 1991;109:654-659.
2. Noffke AS, Sarrafizadeh R, Williams GA, et al. Macular hole surgery with or without internal limiting membrane (ILM) stripping: a comparison of visual results, hole closure, and hole re-opening. Paper presented at Annual Meeting of the Association for Research in Vision and Ophthalmology; May 5-10, 2002; Fort Lauderdale, FL.
3. Mester V, Kuhn F. Internal limiting membrane removal in management of full-thickness macular holes. Retina Specialists of Alabama Web site. Accessed October 1, 2010.
4. Mathis AE. Visualization agents for vitreoretinal surgery. Retinal Physician. 2010; 7(6):45-48.
5. Joondeph BC. Use of membrane blue in ILM and ERM peeling. Retinal Physician. 2009;6(7):54-56.
6. Heath G, Rahman R. Combined 23-gauge, sutureless transconjunctival vitrectoy with phacoemulsification without face down posturing for the repair of idiopathic macular holes. Medscape Web site. Accessed October 1, 2010.

Brian C. Joondeph, MD, FACS, is a retinal surgeon in practice with Colorado Retina Associates, PC, in Denver. He reports no financial interest in any products mentioned in this article. Dr. Joondeph may be reached via e-mail at