EyeNet Magazine

Rethinking Mitomycin C
By Miriam Karmel

Tips from glaucoma and refractive surgeons on ways to get the most out of this powerful drug while minimizing complications.

Mitomycin C, with its powerful antiproliferative properties, improves the success of a number of ocular surgeries. It also increases the potential for complications, including endophthalmitis and scleral meltdown.

Nevertheless, over the past decade, this antifibrotic, which was first introduced as a systemic chemotherapeutic agent, has become firmly entrenched in the ophthalmologist’s armamentarium. But as its popularity has grown, so has concern. Those familiar with its potential complications argue that MMC should be regarded with a healthy respect—and used sparingly and with care.

“It’s great medicine. It certainly has its place in glaucoma surgery,” said Steven J. Gedde, MD, who acknowledges that the introduction of antifibrotic agents represented a major advance in glaucoma filtering surgery. But doctors should be aware of the potential complications associated with its use, he said, and must consider, on a case-by-case basis, whether an antimetabolite is required. “It shouldn’t be used indiscriminately.” 

In fact, Dr. Gedde, who is using less MMC these days, is one of the leaders of a multicenter, randomized clinical trial to determine which is safer and more effective in patients who have had prior ocular surgery: tube shunts or trabeculectomy with MMC. A major impetus for the trial is the concern over late complications associated with trabeculectomy with MMC. Years of experience, he said, have shown “that not all great things result from MMC.” 

Continuing Popularity

MMC is an antibiotic isolated from Streptomyces caespitosus that prevents DNA synthesis in rapidly dividing cells.

In glaucoma filtering surgery, MMC inhibits fibroblast proliferation and subconjunctival fibrosis, thus preventing the scarring that interferes with bleb formation. But it also causes cell death, induces apoptosis and alters bleb morphology, which leads to complications, including thin walls, denuded epithelium and abnormal stroma that may predispose the patient to bleb infection, hypotony, bleb leaks, endophthalmitis and hypotony maculopathy.

Despite a growing awareness of these complications, the number of people using MMC hadn’t diminished between 1996 and 2002, the years when Richard K. Parrish, MD, surveyed members of the American Glaucoma Society. “The fact of the matter is, mitomycin is tremendously popular,” said Dr. Parrish, another leader of the tube shunt vs. trabeculectomy trial. “Most people are using it.” 

Refractive corneal surgeons have also been attracted to MMC because of its ability to reduce the rate of visually significant corneal subepithelial fibrosis (seen clinically as haze) following RK or PRK.

Need for Caution

However, “horror stories” associated with MMC periodically surface, said corneal specialist Randy J. Epstein, MD. “It has potential to do serious injury to the eye, if you’re not careful. It is radiomimetic. And nobody knows how long it hangs around after it’s administered.”

Although Dr. Epstein has never had a complication that he would ascribe to its use, he said, “I can’t honestly tell you that we know everything about this medication or its long-term side effects. I can’t tell you that 20 years from now, something isn’t going to come up. I would never want anyone to feel that we are cavalier in our use of this medication.”

MMC’s Mixed Blessings


Increases trabeculectomy
success rate

Treats established haze;
can be used to prevent
haze in refractive surgery

Can be used for pterygium
and other corneal surgery


Causes widespread cell death;
induces acellularity and

Alters bleb morphology;
increases risk of complications

Is radiomimetic; long-term
risks are unknown

Dr. Epstein first used MMC to treat a patient who developed idiopathic subepithelial fibrosis following RK. Even after multiple debridements, the patient’s vision was 20/200. He was ready for a corneal transplant when he ran across a paper by Joseph Frucht-Pery, MD, and associates, reporting on the use of topical MMC 0.02 percent four times a day for 10 to 22 days in patients with corneal intraepithelial neoplasia. Within nine weeks, abnormal cells were replaced with biomicroscopically normal epithelium, and the dysplasia did not recur during 12 months of follow-up. Transient toxic side effects resolved with cessation of the drug.1 That paper, said Dr. Epstein, provided a possible precedent for using MMC after refractive surgery, because of the morphological similarity between dysplasia and post-RK haze.

At the same time, he was aware that Roy S. Rubinfeld, MD, who originally reported on patients with scleral necrosis from MMC following bare-sclera resection of pterygia, had subsequently become one of the drug’s strongest advocates for use in pterygium surgery when applied as a single intraoperative application under a conjunctival graft. Dr. Rubinfeld observed that patients who got into trouble with MMC had, in effect, been overdosed. (For years, some surgeons sent patients home with a bottle of MMC after pterygium surgery.) 

“A lot of ophthalmologists don’t appreciate the critical significance of dosing with MMC,” said Dr. Epstein. “They don’t realize that since it’s radiomimetic, the dose that you’re delivering can be quite toxic unless it is carefully controlled.” On the other hand, he argues that a single dose at the time of surgery can be highly effective and is easily controllable.

Dr. Gedde speaks in terms of potency, not toxicity. MMC is “the most potent antimetabolite” in both the positive and negative sense. “We’re not trying to say, ‘Never use it,’” he said. “But we are trying to make people aware that there are potential complications associated with its use. Yes, it improves success. But it also increases complications. How to balance those things out remains a topic that is actively debated within the ophthalmic community.”


Randy J. Epstein, MD

MMC has a good therapeutic index, but because it is such a strong medication there should be specific indications for its use, said Dr. Epstein. “The real area of controversy is, where do you draw the line?” 

He uses MMC in the following instances:

1. Treating patients with established haze following PRK. This “is probably the No. 1 indication,” he said. “Few surgeons would argue against its use in patients with established haze that has not resolved following conventional management [observation or topical corticosteroids].” To treat, manually debride the epithelium and the haze. Next, apply a sterile 6-millimeter circular methylcellulose sponge soaked in MMC (0.02 percent) to the central cornea surface for 12 seconds. Apply with a polishing maneuver for even coverage of the entire surface. Then remove the sponge and irrigate with 30 cubic centimeters of balanced salt solution to abruptly terminate exposure to the MMC. Finally, apply a bandage contact lens and leave it in place until re-epithelialization occurs, usually within three to five days.

2. Treatment of patients with LASIK flap complications, or patients with scarring associated with PTK and PRK. This use of MMC is somewhat controversial, Dr. Epstein said. To treat, avoid manual debridement, so as not to disrupt the abnormally thin flap. To protect the integrity of the flap, use a “laser-scrape” setting (in the United States, this is currently available only on the Visx laser) and perform transepithelial PTK to ablate through the epithelium down to the stroma. Following PTK, a PRK procedure is performed, using 75 percent of the original refraction. MMC is then applied as indicated above.

3. Haze prophylaxis in high myopes and for “late enhancements” through old LASIK flaps. This is very controversial, said Dr. Epstein. His group uses MMC in these situations, but only following a rigorous, informed consent process. “In our experience, it’s much easier to prevent haze than it is to treat it once it forms.” To prevent haze, Dr. Epstein recommends using MMC only on patients who have over 70 micrometers of planned ablation or > 7 D of refractive correction. Treatment parameters are the same as in cases of established haze, except the spherical portion of the PRK treatment amount is cut back 10 percent from the originally planned treatment.   

His group has found that this treatment is effective in preventing haze when performing PRK enhancements over “old” LASIK flaps in patients who come in with myopic regression many years after their original procedure. Lifting these flaps can frequently present a challenge, and there may not be enough residual stromal bed anyway. Some research suggests that performing excimer laser ablation through a LASIK flap without the use of prophylactic MMC results in a high incidence of haze formation.2

Steven E. Wilson, MD

Owing to a nagging uncertainty regarding MMC’s long-term safety, Dr.Wilson opposes the use of MMC to prevent haze. His patients haven’t suffered. In more than 4,000 PRK cases, only four patients developed clinically significant haze, and all resolved spontaneously or with MMC treatment post-PRK.

But Dr. Wilson’s opposition may be an exception. In a January 2005 editorial in Review of Refractive Surgery, he writes, “Many have come to feel that this is a perfectly safe treatment. But is it really?” He notes that hundreds of refractive surgeons have adopted the practice of using topical MMC prophylactically in virtually all surface ablation procedures (PRK and LASEK). This practice persists, despite studies reporting that haze almost never occurs below 6 D of correction. Even in patients with higher correction, studies suggest that haze occurs in only a small percent of cases (0.5 percent to 3 percent).

Lab studies show marked decreases in normal keratocytes in the anterior stroma, with some areas devoid of cells. “What are the long-term effects of this acellularity, say 10 or even 20 years later? That question will remain unanswered for many years, but since these cells maintain collagen and other components in the stroma and the cornea is a living tissue, I have nagging concerns,” he writes.

“The key point,” he said, “is that the vast majority of patients who have PRK or LASEK would never get haze. Therefore, prophylaxis is not needed. Why take the long-term risk when MMC is effective in those few who do get the haze?”

Henry D. Perry, MD

When he uses MMC—in cases of pterygium, ocular pemphigoid and corneal intraepithelial neoplasia—“it’s very effective,” said Dr. Perry. But he regards MMC as a drug of last resort.

Why? Because mitomycin acts as an alkylating agent. “In that sense, it’s like radiation therapy. It creates a permanent injury in the tissues that are being treated. And the injury is essentially forever.” Complications can occur five to 10 years later, he adds. “The concern isn’t effectiveness; it’s the complications years later. The [potential] toxicity is probably greater than with anything else we use in ophthalmology.” 

On the other hand, Dr. Perry is more liberal regarding its use in PRK, though even then he reserves it for higher myopes of –6 D and above. In that case, exposure time is very low—sometimes less than 30 seconds—because the longer MMC is applied, the more toxic it is, he said.

Whenever MMC is used, extra precautions must be taken in terms of its disposal and handling because of its toxicity. “A healthy respect is necessary.”


Richard K. Parrish, MD

While concern persists regarding the complications of endophthalmitis and ischemic blebs associated with MMC, a number of recent changes may reduce those risks, Dr. Parrish said. For starters, surgeons now position filtering blebs superiorly (rather than inferiorly at 6 o’clock) to prevent bleb leaks and the possibility of infection.

One recent advance has been the creation of the fornix-based flap (rather than limbus-based). In this procedure, popularized by Peng Tee Khaw, MD, PhD, the conjunctiva is incised where it attaches to the junction of the cornea in the sclera.3 “This appears to allow the blebs to extend more posteriorly and not to be encapsulated by scar tissue that delimits the posterior extent of the bleb,” Dr. Parrish said.

Caption: MMC alters bleb morphology, which can lead to complications such as the ischemic bleb seen here.

Dr. Khaw also advocates the use of multiple thin sponges and expanding the area treated with MMC to three to four clock hours, between 10 and 2. In the past, treatment covered a smaller area of one to two clock hours, increasing the likelihood that the bleb would not spread out, but rather would stay concentrated in a small zone. “We don’t have great data to prove it, but it’s a widely held impression that by using fornix-based flaps and a broad application of mitomycin on these sponges, you’ll end up with a bleb less likely to get into trouble,” Dr. Parrish said.

Another concern is the role of MMC in pseudophakic eyes. Dr. Parrish is studying which treatment is safer and more effective for reducing pressure in patients who have glaucoma after cataract surgery: implantation of a drainage device or trabeculectomy with antimetabolites? “The answer is yet to be determined.”

George L. Spaeth, MD

Physicians who believe that “every millimeter of pressure lowering is good” are more enthusiastic than ever about MMC, said Dr. Spaeth. “They will use it routinely on all their cases—on every trabeculectomy.”

Other physicians, who are more cautious about the drug’s use, believe that if the first surgery fails without MMC, it’s possible to perform a second. What’s more, the addition of newer medications, like the prostaglandin analogs, has made it possible to get pressures down to 10, 11 or 12 mmHg even with surgery in which the final IOP is in the mid-teens, and sometimes even without any surgery at all. 

“I would say mitomycin has proven itself to have a powerful effect on wound healing in patients having glaucoma surgery. The surgeon has to take into account the goals of the surgery and [consider] whether the very significant risks of mitomycin are justified. And unless it’s clear the patient has the type of glaucoma that’s likely to make the patient go blind, those risks are probably not warranted. Even if the patient has a severe glaucoma, one needs to consider whether it can’t be controlled without mitomycin and with subsequent medical therapy.”

He added, “In my own practice, I believe strongly that safety trumps efficacy. If I were to have a trabeculectomy, there’s no way I would let anybody use mitomycin on me. If the first operation didn’t work, I’d say, ‘Why don’t you use 5-FU?’”  

Dr. Spaeth said, “I virtually never use it [MMC] in routine cases.” He will use it in the following instances:

1. If the risk of failure of filtering surgery is believed to be very high, based on previous experience of the presence of several of the well-known risk factors for failure, “and you are convinced you need a rapid surgical success.”

2. If the previous operation failed, it’s likely the second will fail, too. “So that’s an indication to do something different than the first time around, which may include MMC.”

3. In some dark-skinned patients, because they are more
likely to scar.

4. In some patients with an inflammation, such as uveitis. 

Steven J. Gedde, MD

In recent years, Dr. Gedde has altered his practice regarding MMC. He usually reserves it for eyes that are at higher risk for failure with filtering surgery. This includes eyes that have had previous ocular surgery (such as cataract surgery or failed filtering surgery) and patients who are young, are of African-American descent (because of their more exuberant healing response) or have certain secondary glaucomas, such as uveitic glaucoma. In general, he favors 5-FU in his primary cases. In some primary trabeculectomies, such as in elderly Caucasian patients, he may choose not to use an adjunctive antifibrotic agent.

When applying MMC, he favors the approach advocated by Dr. Khaw, involving a more diffuse application over a large surface area in conjunction with a fornix-based flap. This produces a more diffuse low-lying bleb that may be less prone to develop late-onset bleb leaks and infections.

Finally, Dr. Gedde has been giving greater consideration to alternative surgical approaches. “There are several clinical situations where I believe that a drainage implant is the best surgical approach—for example, eyes with neovascular glaucoma or extensive conjunctival scarring. There are other situations in which it’s unclear to me what’s a better operation. Should it be a trabeculectomy with MMC? Or a drainage implant? Each has its pluses and minuses.” 

1 Am J Ophthalmol 1994;117:164–168.
2 Carones, F. et al. J Cataract Refract Surg 2002;28:2088–2095.
3 Ophthalmology 2003;110:2192–2197.

Meet the Experts

Randy J. Epstein, MD  Professor of ophthalmology at Rush University and CEO of Chicago Cornea Consultants Ltd. Financial interests: None.

Steven J. Gedde, MD  Associate professor of ophthalmology and residency program director at the University of Miami. Financial interests: None.

Richard K. Parrish, MD  Associate dean for graduate medicine and professor of ophthalmology at the University of Miami. Financial interests: None.

Henry D. Perry, MD  Senior founding partner, Ophthalmic Consultants of Long Island. Financial interests: None.

George L. Spaeth, MD  Professor of ophthalmology at Wills Eye Hospital and Jefferson Medical College, Philadelphia. Financial interests: None.

Steven E. Wilson, MD  Director of corneal research and staff refractive surgeon at the Cleveland Clinic’s Cole Eye Institute. Financial interests: None.

About Us Academy Jobs Privacy Policy Contact Us Terms of Service Medical Disclaimer Site Index