Article Date: 1/1/2006

Debating the Pros and Cons of 23-g vs. 25-g Vitrectomy

The pros of 25-g vitrectomy

When 25-g, trans-conjunctival, sutureless vitrectomy was introduced, the stated advantages were less inflammation, decreased operating times, less damage to the conjunctiva, and generalities such as "less invasive." The potential negatives discussed at that time included limited flow rates because of the small lumen and wound leaks because of the sutureless closure. Limited flow rates implied that diabetic traction retinal detachments, dense vitreous hemorrhages, silicone oil procedures, and dislocated lens material could not be surgically approached with this technology.

Many surgeons noted faster visual recovery, particularly in epimacular membrane cases, probably because of less astigmatism and less ocular surface inflammation. Because 25-g vitrectomy reduces conjunctival damage it was found to be ideal for patients with filtering procedures (present or future) and patients with ocular surface disorders. Less conjunctival damage is produced by 25-g systems than 23-g.


Soon after introduction of sutureless vitrectomy, the potential wound leak issue was shown to be somewhat of a problem with significant hypotony and volume enhancements reported in the initial series. I introduced the concept of using a 1/3 fluid-air exchange so the interfacial tension of the air bubble would "seal" the wounds until fibrin deposition and finally wound healing took place. This technique was later modified to utilize a larger air bubble so the inferotemporal sclerotomy would be covered by the bubble when the patient was seated or standing. Using fluid-air exchange has resulted in a very low incidence of hypotony, no reoperations for wound suturing, and no volume enhancements. Some surgeons not using fluid-air exchange are recently reporting a very low incidence of hypotony as well.

Initially, 25-g vitrectomy was primarily used for epimacular membranes, macular holes, vitreomacular traction, as well as some unproven and possibly ineffective procedures such as, branch vein decompression (sheathotomy) and radial optic neurotomy. As 25-g vitrectomy techniques were applied to a broader variety of disease states requiring more peripheral vitrectomy, tool flexure was found to be a significant problem. Optimal sclerotomy positioning with respect to the supraorbital rim and bridge of the nose, head rotation during surgery, contact-based wide-angle visualization which requires minimal eye rotation, and scleral depression eliminated most problems due to tool flexure. The second generation Alcon (Fort Worth, Texas) endoilluminators and laser probes are 57% stiffer than the first generation, and the Alcon cutter is 20% stiffer than the electric cutter (Bausch & Lomb, Rochester, NY), which reduces tool flexion. The second generation Alcon DSP forceps are 3 mm shorter and significantly stiffer than the first generation. The second generation Alcon endoilluminator has a 78% divergence angle vs. 44% for the first generation, which facilitates peripheral viewing with less probe angulation and also has 2X the light transmission. Articulating laser probes address the issue of treating peripheral retina breaks without obliquity, but eliminate probe flexion as well. Alcon will introduce 25-g articulating laser probes in the near future.


Many surgeons as well as vitrectomy system manufacturers seem to have the misconception that higher flow rates are better. Lower flow rates are always safer and are an issue only if they significantly increase operating times, which has not been the case with 25-g systems. High-speed cutters were initially promoted using vague terms like cutting efficiency or better cutting when in fact the actual advantage is port-based flow limiting; a term I defined. Port-based flow limiting is inherently produced by smaller lumens. I have defined pulse flow as the volume of fluid that goes through the port with each open-close cycle; port-based flow limiting reduces pulse flow. Reduced pulse flow increases fluidic stability, thereby reducing cutter-induced motion of detached retina enabling safer removal of peripheral vitreoretinal traction. In addition, port-based flow limiting fluid surge after sudden elastic deformation of dense epiretinal membranes through the port, facilitating safer conformal cutter delamination for diabetic traction retinal detachments. The crucial point is that 25-g cutter lumens are smaller than 23-g and therefore produce better fluidics performance because of increased port-based flow limiting.

Improved techniques, technology and a broader array of tools have enabled expanded indications for 25-g vitrectomy. I enlarge 1 sclerotomy to introduce, infuse (silicone reoperations), or remove silicone oil. I never remove silicone oil to reoperate for redetachment due to proliferative vitreoretinopathy, epimacular membranes, or recurrent diabetic traction detachment. Forceps membrane peeling, scissors segmentation and delamination, retinectomy, drainage of subretinal fluid, and laser endophotocoagulation can all be easily performed "under" silicone oil. This technique of enlarging 1 wound, which I call 20/25 vitrectomy, is also ideal to remove dislocated lens material, perform fragmenter lensectomy in adults, or remove intraocular foreign bodies.


Many problems with 23-g vitrectomy relate to the larger wound required; unlike 25-g, conjunctival displacement is not sufficient to prevent wound leaks so a scleral tunnel is required. Many surgeons have observed that 23-g wounds leak if the scleral tunnel is too short. Some surgeons have noted intraoperative scleral folds extending into the posterior pole if the scleral tunnel is too long. Although tools and technique changes have attempted to address this problem, many surgeons still have difficulty locating the sclerotomy made with the stiletto when subsequently inserting the cannula. Endoscopic studies of Frank Koch, MD, have shown more vitreous reflux and incarceration in the cannula than with 25-g systems. Significant subconjunctival hemorrhage is also more frequent with 23-g than with 25-g surgery.

In summary, the 2 stated advantages of 23-g over 25-g are said to be higher flow rates, which is actually a disadvantage, and less tool flexion, which is not an issue with second generation tools and techniques. Use of 25-g systems eliminates the many wound related problems associated with 23-g vitrectomy.

Steve Charles, MD is clinical professor of ophthalmology at the University of Tennessee, College of Medicine, Memphis. Dr. Charles is a consultant to Alcon Surgical. He can be reached by e-mail at

Editor's Note: Bausch & Lomb recently introduced its next-generation 25-g platform, the TSV25 System. In comparison with earlier generations: The entry site alignment system requires less insertion force, and the improved trocar-cannula system reduces the potential for buckling. The light pipe offers 50% greater light output and a 30% larger diameter field. A new stiffening sleeve provides greater control. According to B&L, the new high-speed vitrectomy electric cutter generates 47% more aspiration flow than a 25-g pneumatic cutter with less vibration. "Higher cut rates mean, essentially, you are creating a safer working environment inside the eye," explains David Chow, MD.

The pros of 23-g vitrectomy

Figure 1: 23-g vitrectomy technique.
A: The conjunctiva is displaced with the toothed pressure plate (Dutch Ophthalmic USA, Kingston, NH).
B: Caliper measurements are not needed as the central opening of the circular pressure plate is 3.5mm from the edge.
C: The 45° 23-g stiletto blade (BD Medical-Ophthalmic Systems, Franklin Lakes, NJ) is inserted at a 20° to 30° angle through the conjunctiva and sclera at the pars plana, creating a tunneled sclerotomy.
D: Next, without releasing the pressure plate, the metal trocar is inserted through the wound using a blunt inserter.
E: The inserter can be removed by stabilizing the trocar at its flanged lip with the notched edge of the pressure plate.
F: Configuration once the trocars have been inserted. Removal of the trocars is performed by grasping the flanged lip with the trocar forcep, followed by brief pressure over the sclerotomy site with a cotton swab

Several problems are associated with 25-g surgery, some of which, but not all, have been addressed by alteration in instrument design. These problems include overly flexible instruments, limited functionality of forceps and scissors, decreased flow rates, and the initial attribute of 25-g surgery: the small hole. While the small hole produced in the eye wall means that suture closure is not usually needed, the small hole is a liability. Produced by perpendicular access, a 25-g sclerotomy provides a direct pathway out of the eye for intraocular fluids if the scleral opening does not seal.

Solutions to some of these problems would seemingly involve using slightly larger gauge instruments — 20-g instruments are good and 25-g are too small to be
effective — so a compromise between these 2 sizes would seem best. The 23-g instruments have been used in pediatric cases for years, and there is a long-term manufacturing experience in this bore size. The remaining problem is the sclerotomy. In almost all forms of ocular surgery an incision into the eye is made at an angle to the wall of the eye. This produces an angled tunnel through the wall of the eye that provides greater transmural wall strength in the area of the incision and has the potential to self-seal (Figure 1). The last new facet is the use of precision-made metal cannulas instead of the plastic tubing cannulas employed in 25-g surgery. So 23-g vitrectomy is the sum of improvements made to 25-g surgery.



Larger instrument diameters are like a currency for an instrument designer. One can spend the currency on increased wall thickness for improved stiffness, increased central diameter for better flow, or both. In addition, wider diameter tubes have greater structural strength and stiffness for any given wall thickness. This means that a 23-g cutter will be stiffer than a 25-g cutter using the same wall thickness. The second consequence is the increased inner diameter. Flow rates through a central opening or a tube are related to the radius to the fourth power. Any small increase in the diameter of the "pipe" for the fluid to flow through causes a big increase in the flow rates. Larger central diameters also mean backflush instruments can passively aspirate fluid and light pipes provide plenty of light (Figure 2).


There are inherent limits to surgical instrument capabilities given size constraints, and higher instrument functionality can trade off with smaller size. For instance, the radius of curvature of intraocular scissors is limited by bore size. The 25-g curved intraocular scissors cannot have a pronounced curve or they will not fit through the plastic cannula. The blunted curve renders them much less effective than their 23- or 20-gauge counterparts. Surgeons who cannot use the right tool for the job are forced to compensate with other methods.

Another Achilles heel of 25-g vitrectomy systems is the flexibility of the instruments. Flexibility of the vitrector is the principal concern (Figure 3). As mentioned, given the same wall thickness, wider diameter tubes are stiffer. Instruments inserted through sclerotomies should behave like oars in oarlocks, with movements at the surgeon's end mirrored at the instrument tip. In addition, the surgeon uses the instruments to affect rotation of the eye to expose the periphery for visualization and instrument access. The flexibility of 25-g instruments limits these movements. There is a disconnect in moving the instruments within the eye, particularly when attempting to trim peripheral vitreous or in performing a capsulectomy. Moving the eye is also made more complex because of bending of the instruments. Surgeons may attempt to partially compensate by stabilizing the instrument with an extra finger close to the sclerotomy to reduce bending.

Figure 2: 23-g total view light pipe

This deficiency does not come without a price. There is a general agreement that peripheral vitrectomy with 25-g devices is not as complete as with 20- or 23-g systems owing to increased instrument flexibility. Ibarra noted that a "more prominent residual vitreous skirt could cause significant anterior vitreoretinal traction and subsequent retinal tears or detachments beyond the immediate postoperative period with an increased incidence of 25-g cases."1 His series reported a subsequent detachment rate of 2.2%. Gupta reported an incidence of intraoperative retinal tears with 25-g systems of 2.9%, but no postoperative detachments.2 Fujii also described a 2% postoperative detachment rate.3 Another factor possibly contributing to the apparent increased rate of retinal detachments following 25-g surgery is decreased illumination from smaller bore light pipes causing missed breaks intraoperatively.2

Overall, stiff 23-g rather than flexible 25-g instruments behave like traditional 20-gauge instruments in the eye offering improved surgical capability. A full complement of 23-g surgical instruments is available (Figure 4). Surgeons are able to approach more complex cases with 23- than with 25-g instrumentation.


Figure 3: Vitrector flexibility.
The extreme flexibility of 25-g cutters compared to 20- and 23-g cutters is demonstrated. A syringe filled with 20 mL of water was suspended from the cutter port via a 6-0 silk suture on each of a 20-gauge (A), 23-g (B), and 25-g (C) cutter held horizontally.

Another principal benefit to the surgeon for choosing a 23- over a 25-g system is a significant improvement in fluidics. The term fluidics is used to describe the interaction between the vitrectomy machine, the cutter, flow-rates, and the surgeon. Early in the development of vitrectomy, cutters had slow cut rates and there was a significant lag between reducing the depression on the foot pedal and having an actual reduction in suction at the opening of the cutter. This made shaving vitreous or removing proliferation from the surface of the retina very difficult. The advent of high-speed cutters with improvements in machine technology changed vitrectomy surgery. It is now much easier to shave the vitreous base and even possible to repair diabetic traction detachments using only the vitrector (Figure 5).

The term fluidics carries with it a concept of dynamic range. For diabetic traction detachments, we need to have low flow rates with a precise ability to cut. Using the vitrector to reliably create a posterior-vitreous detachment requires a high-net flow rate. Both 23- and 25-g vitrectomy approaches are quite good at low-flow fluidics. Each can be used to shave the vitreous base and to remove proliferations associated with diabetes, for instance. Both 23 and 25-g vitrectomy are available on the Accurus (Alcon Labs, Fort Worth, Texas) and Millennium (Bausch & Lomb, Rochester, N.Y.) machines. However, the Millennium 23-g system currently utilizes the pneumatic drive, which only has a cut rate of 750, slower than the 25-g system. On the Accurus there is little to differentiate the systems in low-flow fluidics from the machine standpoint. The separation between 23- and 25-g vitrectomy approaches becomes most evident in the high flow domain on both machines. Peak flow rates thorough a 25-g vitrectomy system are limited by the small internal diameter of the cutter and the infusion cannula. In 23-g surgery both the cutter and the infusion cannula have improved designs affording higher flow rates. In this sense we use flow as a tool. Our ability to grasp and remove a posterior-vitreous face depends on the amount of suction, the diameter of the cutter opening, and, because thin vitreous sheets do not readily occlude openings, the flow rate. This goes beyond glibly saying we can wait the minute or so longer with 25-g to perform a core vitrectomy.4 Since we want to be able to use flow as a tool, the reduced high-flow fluidics of 25-g vitrectomy is a serious handicap.


Closure of sclerotomies is essential to prevent postoperative wound leakage. The traditional 20-g sclerotomy has a 1.15 mm width, requiring sutures. In 25-g surgery, the trocar, which is shaped like a hypodermic needle, leaves roughly 0.5 mm hole in the sclera. When initially inserted into tissue, the trocar's shape forms acute angles in the sclera. As the trocar is further advanced into tissue, the shape becomes rounder and the angles less acute. The trocar largely produces the opening by stretching and ripping of the tissue rather than cutting.

Figure 4: 23-g internal limiting membrane peel.
Sequential intraoperative photographs (A-B) showing peeling of the internal limiting membrane using 23-g end-gripping forceps. The internal limiting membrane was elevated without the use of stains and peeled through the macula.

To improve on this approach, 23-g surgery uses an angled blade that produces a 0.72 mm slit-like opening.5 The cannula is then inserted into this opening. The incision used for 23-g surgery is made at an angle to the wall of the eye, like the angled incisions used in cataract surgery. This angled incision does not present a direct perpendicular opening from the vitreous to the outside of the eye. Instead a tunnel incision is created that has the potential to self-seal because the IOP helps close the tunnel.

There is evidence in the literature that non-tunneled 25-g sclerotomies frequently do not self-seal. Gupta indicated that 22% (9 of 41) of fluid-filled eyes that underwent 25-g vitrectomy required sclerotomy suture placement or suffered postoperative hypotony.6 In a review of 140 consecutive 25-g cases, Lakhanpal and colleagues reported that 7.1% of eyes required a suture for adequate closure and 3.8% demonstrated postoperative choroidal detachment.7 The authors did note that many of these cases were early in the 25-g experience and that half of the cases that required a suture had undergone wound enlargement to allow for larger instrumentation.7 Another series reported a 14% incidence of hypotony requiring additional gas or saline tamponade within 2–6 hours postoperatively.8 Hypotony, even if transient, is not a benign condition and may increase the postoperative risks for serious complications including retinal or vitreal incarceration, suprachoroidal hemorrhage, and endophthalmitis.9,10


Figure 5: 23-g diabetic membrane dissection.
Sequential intraoperative photographs (A-D) showing dissection of diabetic membranes without the use of intraocular scissors. The membranes were elevated with the use of suction and then delaminated with the vitrector for en bloc resection.

Systems for 25-g vitrectomy suffer from several fundamental flaws. Leaky sclerotomies, inferior fluidics, and flimsy instruments are inherent to the system. Adding bells and whistles by upgrading instruments will not change basic design problems. Twenty three-gauge vitrectomy fixes these problems. Tunneled sclerotomies are better sealing, which avoids suturing and hypotony. Greater instrument stiffness allows for more complete anterior vitrectomy and manipulation of the globe. Robust fluidics allow for greater aspiration rates to generate vitreous detachment and more efficient core vitrectomy. These factors improve both the surgical performance and safety profile of the vitrectomy system. In sum, 23-g vitrectomy systems enjoy the best of both 20- and 25-g worlds. RP


1. Ibarra MS, Hermel M, Prenner JL, Hassan TS. Longer-term outcomes of transconjunctival sutureless 25-gauge vitrectomy. Am J Ophthalmol. 2005;139:831-836

2. Fujii GY, De Juan E Jr, Humayun MS, et al. Initial experience using the transconjunctival sutureless vitrectomy system for vitreoretinal surgery. Ophthalmology. 2002;109:1814-1820

3. Gupta OP, Weichel ED, Fineman MS, et al. Postoperative complications associated with 25-gauge pars plana vitrectomy. *Abstract citation

4. Fujii GY, De Juan E Jr, Humayun MS, et al. A new 25-gauge instrument system for transconjunctival sutureless vitrectomy surgery. Ophthalmology. 2002;109:1807-1812

5. Eckardt C. Transconjunctival sutureless 23-gauge vitrectomy. Retina. 2005;25:208-211.

6. Gupta OP, Weichel ED, Fineman MS, et al. Postoperative complications associated with 25-gauge pars plana vitrectomy. *Abstract citation

7. Lakhanpal RR, Humayun MS, de Juan E Jr, et al. Outcomes of 140 consecutive cases of 25-gauge transconjunctival surgery for posterior segment disease. Ophthalmology. 2005;112:817-824.

8. Gupta A, Gonzales CR, Lee SY. Transient post-operative hypotony following transconjunctival 25 gauge.Invest Ophthalmol Vis Sci. 2003;44:Abstract 2026.

9. Meyer CH, Rodrigues EB, Schmidt JC, Horle S, Kroll P. Sutureless vitrectomy surgery. Ophthalmology. 2003 Dec;110(12):2427-8

10. Lam DS, Yuen CY, Tam BS, Cheung BT, Chan WM. Sutureless vitrectomy surgery. Ophthalmology. 2003 Dec;110(12):2428-9

Richard Spaide, MD and Howard Fine, MD are retina specialists at the Vitreous, Retina, Macula Consultants of New York. Neither author has financial interest in the information presented in this article. Dr. Spaide can be reached by e-mail at

Retinal Physician, Issue: January 2006