Article Date: 11/1/2012

Surgical Precision

Management of ProliferativeVitreoretinopathy

Epiretinal membranes pose specific challenges requiring special techniques.


Vitreoretinal traction in proliferative vitreoretinopathy cases can be divided into several subtypes, based on the location and orientation of collagen fibers foreshortened by the hypocellular contraction process. There is no three-dimensional “core” vitreous in PVR cases. The most common configuration is a multilayer, planar confluence of the anterior and posterior vitreous cortex, contiguous with the vitreous base and parallel to the iris plane.

It is best practice to remove this confluence first, using the highest possible cutting rate (5,000-7,500 cuts/minute) and relatively low vacuum level and avoiding withdrawing the cutter while the vitreous is engaged.

The second step is removal of what I have termed anterior loop traction (Figure 1). This configuration results from hypocellular contraction of radial vitreous collagen fibers extending anteriorly from the posterior aspect of the vitreous base. This process pulls the equatorial retina anteriorly, causing marked foreshortening of the retina.

Steve Charles, MD, FACS, FICS, is clinical professor of ophthalmology at the University of Tennessee College of Medicine in Memphis. Dr. Charles reports significant financial interest in Alcon. He can be reached via e-mail at


Figure 1. While tight anterior loop traction can be resected with scissors, using 25-g instruments can obviate that need.

Referring to this stage of the surgery as “peripheral vitreous shaving” does not adequately describe the purpose of this step. Twenty-five-gauge cutters are ideal for this purpose, as they fit into the narrow gap between the posterior edge of the vitreous base and the ora serrata.

Often there is additional anterior-posterior vitreoretinal traction due to vitreous adherence to fibrotic lens capsule and/or vitreous extending into the anterior chamber or limbal incisions.


Epiretinal membranes in PVR cases are usually managed by membrane peeling rather than shearing with vitreous cutters or scissors because of relatively low adherence to the retina. I developed forceps membrane peeling to avoid the retinal breaks that can be caused by peeling ERMs with bent needles, picks, and membrane scrapers.

Peeling with picks or scrapers requires peeling in an outside-in direction, as well as finding the edge during which the tip of the peeling instrument may engage and tear the retina. In contrast, forceps membrane peeling is performed in an inside-out direction by grasping the ERM at its apparent epicenter.

The epicenter of the star folds is the virtual intersection of the radial folds (Figure 2). Radial folds can also be caused by linear ERM on the crest or by incarceration in a subretinal fluid drainage site or trauma site.

Forceps membrane peeling requires forceps that close first at the very tip of the forceps, not starting at the fulcrum, with the tips finally coming into contact as the blades completely close. I use the Alcon 25-g DSP ILM forceps for all ERM membrane peeling because they incorporate this design.

An additional advantage of forceps membrane peeling over picks and scrapers is that it is a single-step technique; the peeling device is also used to remove the ERM from the eye. This technique works well when peeling under silicone oil for recurrent PVR or ERM surgery, as I described in my previous column on interface vitrectomy (January/February 2012 Retinal Physician).

Perfluorocarbon Liquids

I do not use perfluorocarbon liquids for retinal stabilization during ERM removal because they add cost and all too frequently result in subfoveal PFC or residual droplets, or both. Retina rigidified by ERMs, coupled with the presence of retinal breaks or a retinectomy, is a setup for subretinal PFC.


As I described in a previous column (May 2012 Retinal Physician), retinectomy (Figure 3) is preferred over relaxing retinotomy when residual retinal foreshortening prevents reattachment after removal of all vitreoretinal traction and apparent ERMs.

The need for retinectomy is best determined by what I term the reattachment experiment: Slow, simultaneous internal drainage of subretinal fluid, combined with fluid-air exchange. The process should be terminated immediately if subretinal air appears, indicating the need for further vitreoretinal traction or ERM removal, subretinal surgery, or, finally, retinectomy.

Bimanual Surgery

I do not perform bimanual surgery, defined as two active tools and chandelier illumination. I do, however, often use the endoilluminator as a second tool, to hold back retina during peeling, shaving, or subretinal band removal, as well as using it for blunt dissection to open folds.


Figure 2. The epicenter of the star folds is the virtual intersection of the radial folds.

Visualization and Illumination

The flat fundus contact lens is better than noncontact (BIOM, Resight, etc.) for peeling ERM in PVR cases because it eliminates all corneal asphericity and improves both lateral and axial resolution. If the equatorial retina is highly elevated, the flat lens can be used for the entire peeling process.


Figure 3. Retinectomy is preferred over relaxing retinotomy when residual retinal foreshortening prevents reattachment after removal of all vitreoretinal traction and apparent ERMs.

In most cases, contact-based wide-angle visualization is required for viewing the vitreous base region and retinal breaks. Contact-based wide-angle visualization not only eliminates corneal asphericity, but it also increases the field of view by 10°, compared to noncontact viewing, usually eliminating the need for scleral depression. Chandelier illumination produces diffuse illumination, which reduces visualization of the vitreous and of transparent ERMs.

Combined Scleral Buckles

I have not used scleral buckles combined with vitrectomy for more than 15 years, as this approach offers no advantages over retinectomy and results in longer operating times, more pain, more use of general anesthesia, 2.75 D of induced myopia on average, conjunctival scarring, and strabismus. RP

Retinal Physician, Volume: 9 , Issue: November 2012, page(s): 64 65