Retained cellulose sponge after trabeculectomy with adjunctive subconjunctival mitomycin C

observed for two months. During that time, visual acuity had decreased to 20/200. There was override of the automated lamellar keratoplasty donor tissue onto the host cornea superi­ orly, and the graft appeared to be thin and slightly hazy (Fig. 2). Therapeutic options of­ fered to the patient included repeat, homoplastic automated lamellar keratoplasty and pene­ trating keratoplasty. The patient elected to undergo penetrating keratoplasty. One month after the procedure, with sutures in place, bestcorrected visual acuity was 20/60. These two cases illustrate a serious complica­ tion of in situ keratomileusis. Although some surgeons use suture fixation of the corneal len­ ticule, others use a sutureless technique. Re­ cently, some surgeons have used a hinge tech­ nique, in which the lenticule is not entirely removed from the host bed, but left attached at one end. Which of these techniques affords the greatest safety, while not adversely altering refractive results, requires further investiga­ tion. Surgeons performing automated lamellar keratoplasty should be aware of this potentially devastating complication.