Fluorouracil Filtering Surgery Study (FFSS)


To determine whether postoperative subconjunctival injections of 5-fluorouracil (5-FU) increase the success rate of filtering surgery in patients at high risk for failure after standard glaucoma filtering surgery.


Filtering surgery adequately lowers intraocular pressure in most glaucoma patients. However, the prognosis is less favorable for aphakic patients with glaucoma or glaucoma in phakic eyes following unsuccessful filtering operations. Failure of filtering surgery is usually attributed to the proliferation of fibroblasts at the filtering site. The use of 5-FU, an antimetabolite, has been shown to inhibit the proliferation of fibroblasts in tissue culture, and in preliminary studies it has increased the success of filtering surgery in a nonhuman primate model.


The Fluorouracil Filtering Surgery Study (FFSS) was a randomized, controlled clinical trial comparing the success rate of standard glaucoma filtering surgery to the success rate of standard surgery with adjunctive 5-FU treatment.

Another element of this study was to evaluate the frequency and severity of possible adverse effects related to 5-FU injections. Detailed preoperative and postoperative examinations of the cornea, lens, and retina were performed. Systemic toxicity was assessed by preoperative and postoperative hematologic studies.

After the investigators performed the filtering surgery and determined that the new outlet channel was working, patients were randomized to receive either 5-FU injections or standard postsurgical care without 5-FU. The patients treated with 5-FU received subconjunctival injections of 5 mg of 5-FU twice daily on postoperative days 1 through 7 and once daily on postoperative days 8 through 14. There were 213 patients recruited into the study, 162 with previous cataract extraction and 51 with previous filtering surgery.

All patients were examined at 1 month, 3 months, 6 months, 1 year, 18 months, and 2 years postoperatively and at yearly intervals thereafter until 5 years postoperatively. Possible concomitant risks of 5-FU treatment, such as toxic effects to the cornea, lens, or retina, were monitored.

Patient Eligibility

Men and women with uncontrolled intraocular pressure greater than 21 mm Hg in one or both eyes despite maximal tolerated therapy and who were aphakic or had undergone previous filtering surgery were eligible to participate.


The data demonstrate improved surgical control of glaucoma using 5-FU in patients at high risk for trabeculectomy failure. At 1 year the cumulative success rates as calculated by survival analysis were 80 percent for the 5-FU group and 60 percent for the standard surgery group; at 3 years the success rates were 56 percent for the 5-FU group and 28 percent for the standard group; at 5 years the success rates were 48 percent in the 5-FU group and 21 percent in the standard. Success was defined as no reoperation for intraocular pressure control and no intraocular pressure over 21 mm Hg at or after the 1-year visit. 

Visual acuity results (logMAR scale) in the 5-FU group were worse than results in the standard therapy group at 1 month; however, the visual acuity change from the qualifying visit was better in the 5-FU group at 1, 2, and 3 years. The difference was not statistically significant at 4 or 5 years. This study showed that, regardless of treatment group, patients with controlled intraocular pressure had less visual acuity loss than patients whose intraocular pressures were not controlled. No difference between treatment groups was found in change in visual field sensitivity. Patients who underwent reoperation showed more visual field loss than those who did not undergo reoperation; the results also suggest that visual field loss is associated with high intraocular pressures. Both treatment groups lost visual acuity and visual field throughout the 5 years of the study.

Risk factors other than treatment that clearly affect success are preoperative intraocular pressure, the number of previous ocular procedures with a conjunctival incision, and the time interval between the last ocular surgery and the study filtering surgery.

The development of a late-onset leak in the filtering bleb was more likely to occur in the 5-FU group (9 percent) than in the standard therapy group (2 percent). No other long-term adverse effects were significantly different between the two groups. Two cases of endophthalmitis developed in the 5-FU group versus one case in the standard group. 

The risk of a suprachoroidal hemorrhage was not related to 5-FU. However, an important finding concerning suprachoroidal hemorrhage after filtering surgery is that the risk of hemorrhage was strongly associated with the preoperative intraocular pressure.

The association between high preoperative intraocular pressure and acute postoperative hypotony was not suspected as a risk factor for the development of suprachoroidal hemorrhage prior to this study. This observation has contributed to a change in ophthalmic surgical practice. Patients with very high preoperative intraocular pressures now undergo trabeculectomies with multiple tightly tied sutures or with releasable sutures in the scleral flap placed to minimize postoperative hypotony. Postoperative argon laser suture lysis or removal of releasable sutures may reduce the likelihood of large postoperative intraocular pressure fall and subsequent hypotony.

The Fluorouracil Filtering Surgery Study Group recommends the use of subconjunctival 5-FU after trabeculectomy in eyes that have undergone previous cataract surgery or unsuccessful filtering surgery. However, because of a higher risk of late-onset wound leaks, which may increase the risk of endophthalmitis, the study group cautions against the routine use of 5-FU in patients with good prognoses. 

Preservation of visual function was associated with intraocular pressure control in both treatment groups, providing additional evidence to support intraocular pressure lowering in patients with glaucoma.


  • The Fluorouracil Filtering Surgery Study Group: Five-year follow-up of the Fluorouracil Filtering Surgery Study. Am J Ophthalmol 121: 349-366, 1996.

  • The Fluorouracil Filtering Surgery Study Group: Three-year follow-up of the Fluorouracil Filtering Surgery Study. Am J Ophthalmol 115: 82-92, 1993.

  • The Fluorouracil Filtering Surgery Study Group: Risk factors for suprachoroidal hemorrhage after filtering surgery. Am J Ophthalmol 113: 501-507, 1992.

  • The Fluorouracil Filtering Surgery Study Group: Fluorouracil Filtering Surgery Study one-year follow-up. Am J Ophthalmol 108: 625-635, 1989.
More important ophthalmic trials