▪ Any glaucoma patient potentially requiring PKP warrants careful surgical and medical management. This 48-year-old black man has uveitic glaucoma and IOP that was controlled for 2 years after trabeculectomy in the left eye, at which time the bleb failed. The superonasal bleb is described as diffusely ischemic, nonvascular, and nonfunctional. The first question that arises is whether the bleb could have been made functional by needling with adjunctive 5-fluorouracil (5-FU) injections. Our preference is to try to resurrect a failed bleb before performing a second procedure. The choice of a second procedure includes trabeculectomy or a glaucoma implant. In this case, a superotemporal trabeculectomy with MMC and 2 releasable sutures, later removed, was performed and remained successful.
One year postoperatively, the visual acuity was 20/25 with 1.25 D of astigmatism. A year later, vision decreased subjectively and corneal topography showed a marked increase in astigmatism, by approximately 13.0 D. This astigmatism has increased further.
Corneal topography is consistent with a diagnosis of keratoconus in both eyes. The cone in the right eye is sufficiently inferior to maintain 20/20 acuity; however, acuity in the left eye is 20/60−.
Trabeculectomy can produce corneal astigmatic changes. Excessive cautery, the size of the scleral flap, tight wound closure, IOP, and the location of the ostium all affect post trabeculectomy astigmatism. Regular and irregular astigmatism can be seen; however, they usually resolve by 6 to 12 months postoperatively. Often, astigmatic changes are underestimated by manual keratometry and are more apparent with computerized topography. In a literature search, we found no publications on the effect of trabeculectomy in the keratoconus patient, although we postulate that the astigmatic effects of glaucoma surgery may be exacerbated in a patient with existing corneal ectasia.
We prefer to manage these patients jointly with our cornea colleagues. If this patient is unable to tolerate a contact lens and wishes to proceed with keratoplasty, a decision must be made on how to manage the glaucoma. In this case, the patient appears to have well-controlled pressure on 1 medicine and has a well-functioning bleb. He is black and has uveitis, so there will likely be significant inflammation associated with surgery. We would treat him with a short course of perioperative oral steroids and every 1 hour and prednisolone acetate 1% postoperatively. We would also give subconjunctival 5-FU injections totaling about 25 to 35 mg in the first 2 postoperative weeks, irrespective the appearance of the bleb, in an attempt to keep the bleb functioning. These injections may delay healing of the corneal wound, but repeat glaucoma surgery bears a worse prognosis for the graft.
In a different scenario, when the patient has poorly controlled glaucoma and a poorly functioning bleb, we would consider placing a glaucoma tube shunt at the time of the PKP. We prefer to use Baerveldt implants placed through the pars plana if the patient is aphakic or pseudophakic. We have found that a tube placed in the anterior chamber can damage the cornea. We have a retina specialist perform a pars plana vitrectomy at the time of the PKP. If a cataract is present, we would also perform cataract extraction. If the lens is clear, we would be forced to place the tube in the anterior chamber and would position the tip well away from the cornea. There must be enough conjunctiva to cover an implant, but we would make every effort to avoid doing transscleral cyclophotocoagulation in an eye with good visual potential, while endocyclophotocoagulation in a patient with uveitis might produce more inflammation after keratoplasy than one would wish.