AUTHOR AND EDITOR INFORMATION

Section 1 of 11  

• Authors and Editors
• Introduction
• Indications
• Relevant Anatomy
• Contraindications
• Workup
• Treatment
• Complications
• Outcome and Prognosis
• Future and Controversies
• References

INTRODUCTION

Section 2 of 11  

• Authors and Editors
• Introduction
• Indications
• Relevant Anatomy
• Contraindications
• Workup
• Treatment
• Complications
• Outcome and Prognosis
• Future and Controversies
• References

Glaucoma remains the leading cause of blindness following penetrating keratoplasty (PKP) because it causes irreversible optic nerve damage and graft failure. The management of penetrating keratoplasty and glaucoma (PKPG) remains controversial mainly because of the high risk of graft failure associated with the treatment.

Recent developments in the management of glaucoma, including newer classes of drugs, such surgical procedures as trabeculectomy with mitomycin-C, glaucoma drainage devices (GDDs), and cyclodestructive procedures with Nd:YAG and diode lasers have increased the options available to the clinician in the management of PKPG. This article addresses the etiology, diagnosis, and treatment of PKPG.

History of the Procedure

In 1969, Irvine and Kaufman reported the high incidence of increased intraocular pressure (IOP) following PKP. They reported a mean maximum pressure of 40 mm Hg in aphakic transplants and 50 mm Hg in combined transplants and cataract extraction in the immediate postoperative period. Since then, numerous authors have reported on the incidence and management of PKPG.

Frequency

The incidence of PKPG varies from 9-31% in the early postoperative period (Foulks, 1987; Wilson, 1990; Karesh, 1983) and from 18-35% in the late postoperative period (Irvine, 1969; Foulks, 1987; Karesh, 1983; Chien, 1993; Goldberg, 1981).

Etiology

Important risk factors for glaucoma in patients undergoing PKP include aphakic bullous keratopathy, combined PKP and intracapsular cataract extraction, preexisting glaucoma, perforation, and previous keratoplasty (Irvine, 1969; Foulks, 1987; Wilson, 1990; Karesh, 1983; Chien, 1993; Goldberg, 1981; Kirkness, 1988). Goldberg et al (1981) reported a 29% incidence of increased IOP in the early postoperative period and 30% in the late postoperative period in patients with aphakic bullous keratopathy. The same authors also reported the incidence of increased IOP to be higher in patients with repeat grafts (45% in the early postoperative phase and 52% in the late postoperative phase) and those with preexisting glaucoma (71% of all patients with preexisting glaucoma developed elevated IOP in the early postoperative course).

Kirkness (1988) reported a higher incidence of glaucoma in patients undergoing keratoplasty following corneal perforation, especially that following suppurative keratitis, probably from peripheral anterior synechiae and angle closure. Other studies have found traumatized eyes and older patients to be at increased risk for developing PKPG (Irvine, 1969; Foulks, 1987; Wilson, 1990; Karesh, 1983; Chien, 1993; Goldberg, 1981; Kirkness, 1988). The possible causes of PKPG are included in the lists below.

Causes for elevated IOP in the early postoperative period are as follows:

• Postoperative inflammation
• Viscoelastic substances
• Use of alpha-chymotrypsin
• Wound leak with angle closure
• Hyphema
• Operative technique
  • Tight suturing and long bites with compression of the angle
  • Larger recipient bed with same size donor button
  • Increased peripheral corneal thickness
• Pupillary block glaucoma
• Preexisting glaucoma
• PKP in aphakic eyes secondary to mechanical angle collapse

Causes for elevated IOP in the late postoperative period are as follows:

• PKP in aphakic eyes
• PKP combined with cataract extraction
• Chronic angle-closure glaucoma
• Preexisting glaucoma
• Steroid-induced glaucoma
• Graft rejection with glaucoma
• Ghost cell glaucoma
• Misdirected aqueous or ciliary block (malignant) glaucoma

Pathophysiology

The etiology of PKPG is probably multifactorial and may be related to distortion of the angle with collapse of the trabecular meshwork, suturing technique, postoperative inflammation, and peripheral anterior synechiae. The usual factors that contribute to postoperative glaucoma, such as preexisting glaucoma, postoperative inflammation, use of viscoelastic substances, and steroid-induced glaucoma (Foulks, 1987), should be considered. Other factors that are peculiar to patients who have undergone keratoplasty exist. Olson and Kaufman (1977), using a mathematical model, proposed that the elevated IOP following keratoplasty in an aphakic patient might be the result of angle distortion secondary to a roll of excess compressed tissue in the angle.

Because of edema and inflammation, trabecular meshwork function is compromised. According to them, factors that contribute to angle distortion include tight suturing, long bites (more compressed tissue), larger trephine sizes, smaller recipient corneal diameter, and increased peripheral corneal thickness. Conversely, less tight wounds, smaller trephine sizes, donor corneas larger than the recipient, thinner recipient corneas, and larger overall corneal diameter tend to alleviate the angle distortion.

On the other hand, Zimmerman et al (1978) have proposed mechanical collapse of the trabecular meshwork in aphakic grafts to be the key problem leading to glaucoma. They postulated that the trabeculum needs posterior fixation afforded by the ciliary body–lens support system and an anterior support afforded by the Descemet membrane. In aphakia, the posterior support is relaxed but not critically. With keratoplasty, this loose roll of tissue relaxes the anterior support and leads to partial trabecular collapse. This theory might explain the high incidence of glaucoma in aphakic grafts when compared to phakic transplants. In perfusion studies of aphakic eyes, these authors have shown that full-thickness sutures that approximate the Descemet membrane were not associated with alterations of facility of outflow. On the other hand, midstromal bites were associated with a 37% reduction in the outflow facility.

Supporting this theory, Zimmerman et al (1978) have shown that oversized donor buttons (0.5 mm larger than the host bed) in aphakic patients reduced the incidence of glaucoma. The effect was more obvious when an 8-mm donor button was used in 7.5 mm host bed. Similar results were found by Bourne (1982) with oversized corneal buttons. However, these results were not reproduced in other studies. Perl et al (1981) showed no protection against postkeratoplasty glaucoma by using oversized grafts (0.5 mm) in any study group (ie, aphakics, pseudophakics, phakics). None of the authors addressed such variables as trephination technique, suturing technique, and corneal rigidity (which may lead to inconsistent button size). Lass (1979) proposed that postkeratoplasty glaucoma could be related to the development of fine peripheral anterior synechiae. A floppy atrophic iris may lead to a higher incidence of peripheral anterior synechiae formation, which can be prevented by iris suturing or iridoplasty (Cohen, 1982).

Clinical

The diagnosis of PKPG is made based on IOP measurements in the early postoperative period, and, in the late postoperative period, it is based on IOP, optic disc changes, and progressive visual field changes. Patients with extremely high IOPs might present with graft edema and/or failure.

INDICATIONS

Section 3 of 11  

• Authors and Editors
• Introduction
• Indications
• Relevant Anatomy
• Contraindications
• Workup
• Treatment
• Complications
• Outcome and Prognosis
• Future and Controversies
• References

Indications are covered in other sections.

RELEVANT ANATOMY

Section 4 of 11  

• Authors and Editors
• Introduction
• Indications
• Relevant Anatomy
• Contraindications
• Workup
• Treatment
• Complications
• Outcome and Prognosis
• Future and Controversies
• References

Relevant anatomy is covered in other sections.

CONTRAINDICATIONS

Section 5 of 11  

• Authors and Editors
• Introduction
• Indications
• Relevant Anatomy
• Contraindications
• Workup
• Treatment
• Complications
• Outcome and Prognosis
• Future and Controversies
• References

Contraindications are covered in other sections.

WORKUP

Section 6 of 11  

• Authors and Editors
• Introduction
• Indications
• Relevant Anatomy
• Contraindications
• Workup
• Treatment
• Complications
• Outcome and Prognosis
• Future and Controversies
• References

Diagnostic Procedures

• Accurate measurement of IOP in patients with keratoplasty can be difficult. IOP in the early postoperative period, when the corneal surface is irregular, can be measured with the Mackay-Marg electronic applanation tonometer, the Pneumatic applanation tonometer, or the Tono-Pen.
• • If the graft surface is smooth, the epithelium is intact, and the mires are regular, then Goldmann applanation can be used to measure the IOP. Marked corneal astigmatism causes an elliptical fluorescein pattern. To obtain an accurate reading with the Goldmann applanation tonometer, the clinician should rotate the prism so that the red mark on the prism holder is set at the least curved meridian of the cornea (along the negative axis). Alternately, 2 pressure readings, taken 900 apart, can be averaged.
  • The accuracy of applanation tonometry is reduced in certain situations, such as corneal edema, scars, bloodstaining, or any condition that thickens or alters the cornea. Corneal epithelial edema and stromal edema predispose to inaccurately low readings, whereas pressure measurements taken over a corneal scar will be falsely high. Thin corneas result in underestimation, and thick corneas result in overestimation. Tonometry performed over a soft contact lens and following scleral buckling procedures gives falsely low values.
  • The Pneumatic tonometer has a pressure-sensing device that consists of a gas-filled chamber covered by a Silastic diaphragm. The gas in the chamber escapes through an exhaust vent. As the diaphragm touches the cornea, the gas vent is reduced in size and the pressure in the chamber rises. Because this instrument applanates only a small area of the cornea, it has the advantage of measuring the IOP in the presence of corneal scars, corneal edema, or when only a small portion of the cornea is visible (large tarsorrhaphy). In addition, in patients with neurotrophic corneas, it is possible to measure the IOP with minimal disturbance of the epithelium.
  • In cases with complete tarsorrhaphy, an attempt must be made to measure the IOP by digital palpation. Measuring the IOP in the normal eye using one of the standard techniques (eg, Goldmann applanation tonometer) is helpful; then, perform digital palpation on both eyes. IOP should be measured on every visit following PKP.
  • Optic disc changes should be monitored in all cases of elevated IOP. This can be completed either by serial disc photography (where possible) or by the same observer serial optic disc diagrams. Visual fields may be difficult to perform in patients with corneal graft, especially in the early postoperative period. In patients with reasonable vision, Humphrey visual fields or Goldmann visual fields should be performed.

Histologic Findings

Chronic elevations of IOP potentially compromise the graft endothelial function, leading to graft failure. The degree and duration of elevated IOP has been shown to result in significant endothelial cell loss. Following an acute attack of angle-closure glaucoma, 10-33% endothelial cell loss has been reported (Bigar, 1982), and 77% cell loss has been reported in eyes with acute angle-closure glaucoma lasting more than 12 days. Morphologic changes in the endothelial cells, such as vacuolization, loosening of cell junctions, blebbing, disruption of the plasma membrane, exkaryocytosis, and loss of whole cells, have been observed in experimentally induced acute glaucoma (Svedbergh, 1975). Corneal sensation also is noted to be decreased in cases of angle-closure glaucoma.

TREATMENT

Section 7 of 11  

• Authors and Editors
• Introduction
• Indications
• Relevant Anatomy
• Contraindications
• Workup
• Treatment
• Complications
• Outcome and Prognosis
• Future and Controversies
• References

Medical therapy

Medical management (topical drops or systemic pills) continues to be the first line of treatment in cases of PKPG. Currently available antiglaucoma medications include beta-adrenergic blocking agents (eg, timolol, betaxolol), adrenergic agents (eg, epinephrine, dipivefrin), alpha2-adrenergic agonists (eg, brimonidine, apraclonidine hydrochloride), miotics (eg, pilocarpine, echothiophate iodide, carbachol), prostaglandin analogues (eg, latanoprost), topical carbonic anhydrase inhibitors (eg, dorzolamide, brinzolamide), and systemic carbonic anhydrase inhibitors (eg, acetazolamide, methazolamide, dichlorphenamide).

Beta-adrenergic blocking agents have been the cornerstone of glaucoma treatment for the last 2 decades. They act by decreasing the aqueous production, and they have no effect on the outflow pathways. Lass and Pavan-Langston (1979) demonstrated the efficacy of timolol in the treatment of PKPG, even in the presence of chronic angle closure. Adverse effects of beta-blockers include, but are not limited to, superficial punctate keratopathy, corneal anesthesia, and damage to the ocular surface by decreasing the aqueous layer production rate and impairing the quantity and quality of the mucus layer of the tear film, resulting in a dry eye state. All these may have an adverse effect on the graft epithelium that might compromise graft function.

Adrenergic agents can help in lowering the IOP but should be used with caution in aphakic and pseudophakic patients because they can produce cystoid macular edema. Brimonidine tartrate 0.2%, a relatively selective alpha2-adrenergic agonist, is better tolerated than apraclonidine hydrochloride and has been shown to be a safe drug in the long-term control of the IOP. Apraclonidine 0.5% is a potent anterior segment vasoconstrictor and is useful both to prevent anterior chamber bleeding during the operation and to control the pressure spike resulting from such a bleed. In those patients with a high risk of bleeding, 1 drop of apraclonidine 0.5% is suggested 1 hour before surgery and 12 hours postoperatively.

Miotics should be used with caution in this patient population. They can be useful in patients with open angles but may have very little effect in the presence of significant angle closure caused by peripheral anterior synechiae. Miotics can induce uveitis by breakdown of the blood-aqueous barrier, and they can initiate graft rejection. In one report (Beebe, 1990), 3 out of 4 grafts had graft rejection within 3 days to weeks after the start of the miotic agents. Three of these patients were on phospholine iodide, and 1 patient was on pilocarpine. Discontinuation of the miotics with intensive corticosteroid treatment reversed the rejection in 1 patient, while irreversible graft failure developed in the other 2 grafts. In aphakic patients, miotics can increase the risk of a retinal detachment.

Topical carbonic anhydrase inhibitors (eg, dorzolamide, brinzolamide) have similar ocular hypotensive efficacy as betaxolol 0.5% and are not associated with clinically significant electrolyte disturbances or systemic adverse effects seen with systemic carbonic anhydrase inhibitors. However, they should be used with caution in patients with PKPG, especially in those with a past history of graft rejection and/or with limited endothelial cell counts (Konowal, 1999); they can contribute to an irreversible corneal decompensation, especially in patients with compromised endothelial function. Systemic carbonic anhydrase inhibitors are useful in the treatment of pressure spikes in the immediate postoperative period. Their long-term use is limited because 30-50% of patients experience adverse effects, such as paresthesias, tinnitus, nausea, gastrointestinal disturbances, fatigue, depression, anorexia, and weight loss.

Prostaglandin analogues, such as latanoprost, appear to decrease IOP by increasing the uveoscleral outflow and can be used with beta-blockers and carbonic anhydrase inhibitors. Latanoprost should be used with caution in patients with a history of herpes simplex keratitis because it recently has been reported to induce recurrent herpetic infection both in humans (Wand, 1999) and in the rabbit model (Kaufman, 1999). In patients with aphakia and pseudophakia, latanoprost has been reported to cause cystoid macular edema, and in patients with a past history of uveitis, it should be used with caution (Ayyala, 1998).

The benefits of pressure reduction with topical glaucoma medications should be weighed against potential adverse effects. Apart from the specific adverse effects listed above, benzalkonium chloride (BAC), the preservative used in most topical glaucoma medications (0.01% concentration BAC), can cause severe surface toxicity. One drop of 0.02% BAC or several drops of 0.01% BAC can have toxic effects on the corneal epithelium. These effects include cell wall damage and destruction of the corneal epithelial microvilli, leading to increased permeability of the corneal epithelium (Fraunfelder, 1989). The acidic pH of some of the topical drops (eg, Cosopt, 5.8; dorzolamide, 5.6), apart from causing burning sensation, also may be toxic to the corneal epithelium.

In patients who are allergic to the preservatives, preservative-free drugs, such as an Ocudose form of timolol maleate, should be used. Also, pilocarpine powder can be reconstituted with balanced salt solution locally by the pharmacy without any preservative.

In case of steroid-responsive glaucoma, tapering the dose of steroid drops (prednisolone acetate 1%) to the minimum required or replacing stronger steroid drops, such as prednisolone acetate, with steroids with less tendency to increase the IOP, such as topical fluorometholone, loteprednol etabonate 0.5% or 0.2%, and rimexolone 1% or with cyclosporin A 0.5% topical drops (in combination with topical glaucoma medications), may help in controlling the pressure. The following table documents the disadvantages of using some of the topical glaucoma medications in patients with PKPG.

Possible Disadvantages of the Various Glaucoma Medications in Patients With PKPG

Surgical therapy

Surgical treatment may be in the form of argon laser trabeculoplasty; glaucoma-filtering procedures, such as trabeculectomy and GDDs; or various cyclodestructive procedures.

Argon laser trabeculoplasty

Argon laser trabeculoplasty (ALT) can result in 10-40% reduction in the IOP in primary open-angle glaucoma in the short term. The efficacy of ALT depends on the clinical characteristics of the patients and the type of glaucoma treated. The IOP-lowering effect tends to diminish between 1.5-4 years postoperatively with only a 40-50% success rate at 5 years (Shingleton, 1987). The effects of ALT following keratoplasty are variable and may be tried as a short-term measure in patients with open angles and clear grafts with moderately elevated IOP (20-25 mm Hg) on glaucoma medications.

Recommended setting: Blue green or all green argon laser with a 50 µm spot size, 0.1 seconds, 50 spots in 180°, and power 400-1000 W. The end point is to visualize blanching or bubble formation. The laser is applied at the junction of the pigmented and the nonpigmented trabecular meshwork.

Complications: Pressure spikes and iritis may occur, both of which can trigger graft rejection.

Trabeculectomy

Conventional trabeculectomy without antimetabolites (5-fluorouracil [5-FU]) and alkylating agents (mitomycin-C) in patients with PKPG has a high failure rate (Gilvarry, 1989) secondary to limbal conjunctival scarring from previous surgery, extensive peripheral synechiae, aphakia, and extremely shallow anterior chamber.

The introduction of 5-FU and mitomycin-C has increased the success rate of trabeculectomies, especially in patients with complicated glaucoma. These agents appear to increase the success rate by inhibiting the fibroblast proliferation and enhancing the formation of filtering blebs. Five milligrams of 5-FU in 0.1 cc is given as a subconjunctival injection in the immediate postoperative period for 10-14 days. Apart from the inconvenience of daily injections, 5-FU injections are associated with a high rate of corneal epithelial toxicity, corneal ulceration, corneal perforation, and stem cell failure, which could prove to be disastrous to the graft (Knapp, 1987). Because of corneal toxicity, 5-FU should be used with caution in patients with PKPG.

Intraoperative local application of mitomycin-C (0.2-0.4 mg applied for 1-4 min) has significantly improved the success rate of filtering surgery for glaucoma (Skuta, 1992). Apart from the convenience of single application at the time of surgery, mitomycin-C trabeculectomy has no demonstrable toxicity on the corneal epithelium. However, mitomycin-C trabeculectomy may result in thin cystic bleb formation and an increased risk of bleb-related infection (Ayyala, 1997). The reported success rate in IOP control with mitomycin trabeculectomy in patients with PKPG is 67-91% and that of graft failure was 12-18% (Ayyala, 1998; Figueiredo, 1996). The bleb failure rate is higher when trabeculectomy is combined with additional surgical procedures, such as cataract surgery and vitrectomy (WuDunn, 1999).

Recommendations: Trabeculectomy with mitomycin-C can be attempted in patients with no or limited superior limbal conjunctival scarring, absence of extensive peripheral anterior synechiae, aphakia, and extremely shallow anterior chambers. Avoid this procedure in patients with contact lens usage because it can predispose them to bleb infection. Avoid shallow or flat anterior chamber in the postoperative period because this could compromise the graft endothelium. Avoid combining trabeculectomy with other intraocular procedures because it compromises the success rate of the trabeculectomy. Avoid 5-FU in patients with sick epithelium and persistent epithelial defects. Watch for Dellen formation that can trigger thinning of the adjacent graft cornea, leaking blebs, and bleb-related infections.

Glaucoma drainage devices

GDDs create an alternate aqueous pathway by channeling aqueous from the anterior chamber through a long tube to an equatorial plate that promotes bleb formation. In 1987, Kirkness first reported the use of GDDs in PKPG. Even though GDDs appear to control glaucoma in a high percentage of patients in all published series (71-96%, with an average of 84.8%), it appears to be associated with a high incidence of graft failure in the range of 10-51% (average of 36.2%) (Rapuano, 1995; Beebe, 1990; Ayyala, 1998; Kirkness, 1987; Topouzis, 1999). The etiology of graft failure probably is multifactorial. The presence of underlying chronic inflammation, extensive peripheral synechiae, and multiple previous surgeries may compromise the graft. The introduction of a GDD into the anterior chamber also may be associated with increased inflammation and may further compromise the graft.

The timing of GDD surgery is another factor that can contribute to graft failure. In the series published by Beebe et al (1990) and Rapuano et al (1995), a trend exists toward a higher incidence of graft failure when GDD surgery was performed after PKP. GDD surgery–related complications, including inflammation, shallow anterior chamber with iris graft endothelial touch, and tube-endothelial touch, could possibly contribute to graft failure. Meticulous surgery should avoid the complications of flat anterior chamber and tube-endothelial touch. The use of high-dose steroids for 3-6 months in the postoperative phase may help in controlling and suppressing inflammation.

The choice of the GDD in the treatment of patients with PKPG depends upon the case and the surgeon. Currently, 4 main GDDs are available; the Ahmed (Ayyala, 1998) and the Krupin implants offer resistance to the outflow in the form of a sheet valve and a slit valve, respectively, and Molteno and Baerveldt (Siegner, 1995) have no resistance to the outflow and, thus, may lead to hypotony. However, this problem can be overcome with the use of the ripcord technique.

The advantages of the valved implants, especially that of the Ahmed glaucoma valve, appears to be that of easy insertion following 1-quadrant dissection and low incidence of hypotony in the immediate postoperative phase. However, the Ahmed valve is associated with a high incidence of the hypertensive phase (as much as 80%), 1-3 months after the operation, which may need needling with 5-FU injections (Ayyala, 1998). On the other hand, GDDs with a larger surface area, such as the double-plate Molteno and Baerveldt (Siegner, 1995), appear to exhibit a lesser incidence of the hypertensive phase and may achieve slightly lower IOP. The overall success rate and other complications, including corneal decompensation, appear to be similar to all GDDs (Rapuano, 1995; Sherwood, 1993; Topouzis, 1999; Ayyala, 1998).

The authors prefer the Ahmed glaucoma valve in patients with mild-to-moderate glaucomatous optic nerve damage and larger surface area implants, such as the double-plate Molteno and the Baerveldt implant, with more significant pressure reduction in patients with more advanced optic nerve damage. The surgeon should take intraoperative precautions to decrease the incidence of postoperative hypotony and shallow anterior chambers. Topical steroids should be used liberally and for prolonged periods of time following the operation.

Recommendations: Precede or combine the GDD operation with the corneal transplantation in those patients with preexisting glaucoma. Use topical steroids for at least 3 months in the postoperative phase to avoid graft rejection.

Ahmed glaucoma valve: This valve is easy to insert, is a 1-quadrant dissection, has less operative time compared to other GDD operations, and has a low incidence of hypotony in the postoperative period. However, it does have a higher incidence of hypertensive phase in the postoperative phase that might need additional glaucoma medications or needling of the bleb.

Double-plate Molteno and Baerveldt implants: These implants require more extensive dissection and more operative time. A stent must be used to avoid the postoperative hypotony and shallow anterior chamber. The larger surface area of the endplate results in larger blebs and lower IOPs. These implants are reserved for patients with advanced glaucomatous optic nerve damage.

Complications of GDD surgery include graft rejection and failure, conjunctival erosion, prolonged hypotony, tube-endothelial touch, tube obstruction, tube failure, retinal detachment, tube plate extrusion, epithelial down growth, and infection.

Cyclodestructive procedures

Cyclodestructive procedures aim to control the IOP by decreasing aqueous humor production by destroying part of the ciliary body. Cyclocryotherapy (Binder, 1975), transscleral cyclophotocoagulation with Nd:YAG (Threlkeld, 1995), and semiconductor diode laser (Youn, 1998) are the various cyclodestructive procedures that can be performed on patients with intractable PKPG. The reported success rates and the complications following any of these procedures appear to be similar (Binder, 1975; Threlkeld, 1995; Youn, 1998). The individual surgeon must decide which 1 of the 3 to use, depending on the availability of the instruments and the lasers. The overall success rate in controlling the IOP is 60-80%.

The major complications from any of these procedures are the risk of graft rejection and loss of vision. The authors prefer the semiconductor diode laser to the procedures for the following reasons. The diode laser, with a wavelength of 810 nm, has lower scleral transmission than the Nd:YAG laser (1064 nm) but greater absorption by melanin. Also, because semiconductor diode lasers have solid-state construction, they have the advantages of portability, durability, and smaller size compared with Nd:YAG lasers.

Recommendations include the following:

Cyclocryotherapy: The glaucoma cryoprobe is placed for 1 minute, 3 mm behind the corneoscleral limbus. Six burns are made with equidistant spots involving the inferior 180° of the globe at a temperature of -60°C to -80°C. The superior half of the globe almost never is treated. The treatment is repeated in exactly the same fashion when indicated. Avoid doing 3600 cyclodestruction to decrease the risk of hypotony.

YAG cyclophotocoagulation: The use of Nd:YAG (Microruptor 11, LASAG, Thun, Switzerland) in a maximally defocused position is recommended. Approximately 15 evenly spaced burns are placed 1-1.5 mm from the limbus for 1800°. The recommended mean energy level is 4.1-9.3 joules. Postoperative pain medication and topical steroids are indicated. Low energy settings are preferred in patients previously treated with cyclocryotherapy or filtering procedure to avoid hypotony. Repeated applications may be necessary before adequate control is achieved.

Diode laser cyclophotocoagulation: The recommended power settings with the diode laser are 1750-3000 mW, with a 2-second exposure time. An initial power setting of 1750 mW is increased or decreased by 250-mW increments until it is 250 mW below that producing an audible popping sound.

Complications include decrease in the Snellen visual acuity (22-56%), graft failure (11-65%), persistent hypotony (5-10%), anterior uveitis, epithelial defects, loss of vision, severe pain, phthisis bulbi, hyphema, hypopyon, intractable pain, sympathetic ophthalmia, scleral thinning, and vitreous hemorrhage.

Summary

Uncontrolled IOP after PKP can result in graft failure and visual loss. IOP should be monitored on a regular basis after corneal transplantation. Uncontrolled IOPs should be treated aggressively. Any patient with preexisting glaucoma must be evaluated carefully prior to the corneal transplants.

Patients with uncontrolled IOPs or patients with borderline IOP control on 2 or more medications may be treated with either mitomycin-C trabeculectomy or GDD surgery prior to or combined with the planned corneal transplant. This is based on the fact that multiple studies have documented preoperative glaucoma to be a high-risk factor for the development of PKPG (Irvine, 1969; Foulks, 1987; Wilson, 1990; Karesh, 1983; Chien, 1993; Goldberg, 1981; Kirkness, 1988; Reinhard, 1997) and higher incidence of graft failure following glaucoma operations following PKP (Rapuano, 1995). PKPG not responding to medications should be treated surgically. Mitomycin-C trabeculectomy is the safest operation both in terms of IOP control and graft survival. Literature favors a combined mitomycin-C trabeculectomy with corneal graft operation in patients with preexisting glaucoma who need a corneal transplant (Rapuano, 1995; Beebe, 1990; Ayyala, 1998; Figueiredo, 1996; WuDunn, 1999; McDonnell, 1988; Topouzis, 1999).

Additional surgical procedures should be avoided if possible at the time of the trabeculectomy because this is associated with a higher incidence of trabeculectomy failure (WuDunn, 1999).

GDD surgery is the preferred operation over other surgical options in patients with PKPG who have extensive limbal conjunctival scarring, shallow anterior chamber, extensive peripheral anterior synechiae, and failed trabeculectomy. GDD surgery appears to be superior to cyclodestructive procedures in patients who have failed mitomycin-C trabeculectomy or where mitomycin-C trabeculectomy is contraindicated (ie, patients who wear contact lenses). The graft failure rate following GDD surgery and cyclodestructive procedures may be the same, but there appears to be a higher incidence of permanent visual loss and hypotony following cyclodestructive procedures (Ayyala, 2000). Thus, cyclodestructive measures should be reserved for patients who have failed all other interventions. Randomized, prospective studies are needed to determine which of the currently available treatment options should be the treatment of choice in the postkeratoplasty glaucoma population.

Preoperative details

Preexisting glaucoma is frequently more difficult to treat following keratoplasty in both aphakic and pseudophakic eyes (Irvine, 1969). Preexisting glaucoma also is noted to be a risk factor for graft failure (Reinhard, 1997). Reinhard et al (1997) estimated the 3-year graft survival rate in patients with a preoperative history of glaucoma to be 71% in contrast to 89% without such history. Some studies suggest a higher incidence of graft failure following glaucoma operation performed after PKP (Rapuano, 1995). Hence, in this patient population, it is recommended that the glaucoma operation either precede or be combined with PKP.

Intraoperative details

During PKP, such measures as using an oversized donor button (0.5 mm), deep bites, goniosynechialysis in the presence of peripheral anterior synechiae, iridoplasty (iris-tightening procedure) in cases of a floppy iris, removal of viscoelastic material at the end of the operation, and careful wound closure to prevent postoperative wound leaks are useful in reducing the incidence of postoperative glaucoma.

Postoperative details

In the postoperative phase, judicious use of steroids controls the inflammation and prevents peripheral anterior synechiae. Cycloplegics (when indicated) keeps the pupil mobile and prevents pupillary block glaucoma.

Follow-up

For excellent patient education resources, visit eMedicine's Glaucoma Center. Also, see eMedicine's patient education articles Glaucoma Overview, Glaucoma FAQs, and Understanding Glaucoma Medications.

COMPLICATIONS

Section 8 of 11  

• Authors and Editors
• Introduction
• Indications
• Relevant Anatomy
• Contraindications
• Workup
• Treatment
• Complications
• Outcome and Prognosis
• Future and Controversies
• References

See Surgical therapy.

OUTCOME AND PROGNOSIS

Section 9 of 11  

• Authors and Editors
• Introduction
• Indications
• Relevant Anatomy
• Contraindications
• Workup
• Treatment
• Complications
• Outcome and Prognosis
• Future and Controversies
• References

The surgical success rate of the 3 procedures (ie, trabeculectomy with mitomycin-C, GDD surgery, cyclodestructive procedure) in controlling the IOP to less than 21 mm Hg is similar (70-75%). The prognosis for graft survival is less clear. The lowest incidence of graft failure follows trabeculectomy (10-20%), as compared to GDD surgery (10-50%) and cyclodestructive procedure (20-50%). Therefore, the long-term prognosis for graft survival appears to be 40-60% in patients with PKPG.

FUTURE AND CONTROVERSIES

Section 10 of 11  

• Authors and Editors
• Introduction
• Indications
• Relevant Anatomy
• Contraindications
• Workup
• Treatment
• Complications
• Outcome and Prognosis
• Future and Controversies
• References

Eye care professionals should be educated to monitor the IOP in all patients following PKP. Patients with PKPG who are not responding to medications should be treated surgically.

Controversy still exists as to which of the 3 surgical procedures is the best initial treatment option in terms of graft survival. In addition, the timing of the surgery (ie, whether to perform the surgery prior to, combined with, or after the corneal transplant operation) is still not clear. Some authors recently recommended placement of the GDD into the posterior chamber combined with vitrectomy. These authors believe that placing the tube behind the lens-iris diaphragm decreases the risk of graft failure. Similarly, diode laser cycloablation is believed to result in less inflammation and more precise ciliary process destruction. However, definitive evidence is still lacking in both situations. Randomized, prospective studies are needed to determine which of the currently available treatment options should be the treatment of choice in the postkeratoplasty glaucoma population.

REFERENCES

Section 11 of 11

• Authors and Editors
• Introduction
• Indications
• Relevant Anatomy
• Contraindications
• Workup
• Treatment
• Complications
• Outcome and Prognosis
• Future and Controversies
• References

• Ayyala RS, Cruz DA, Margo CE, et al. Cystoid macular edema associated with latanoprost in aphakic and pseudophakic eyes. Am J Ophthalmol. Oct 1998;126(4):602-4. [Medline].
• Ayyala RS, Bellows AR, Thomas JV, Hutchinson BT. Bleb infections: clinically different courses of "blebitis" and endophthalmitis. Ophthalmic Surg Lasers. Jun 1997;28(6):452-60. [Medline].
• Ayyala RS, Pieroth L, Vinals AF, et al. Comparison of mitomycin C trabeculectomy, glaucoma drainage device implantation, and laser neodymium:YAG cyclophotocoagulation in the management of intractable glaucoma after penetrating keratoplasty. Ophthalmology. Aug 1998;105(8):1550-6. [Medline].
• Ayyala RS, Zurakowski D, Smith JA, et al. A clinical study of the Ahmed glaucoma valve implant in advanced glaucoma. Ophthalmology. Oct 1998;105(10):1968-76. [Medline].
• Ayyala RS. Penetrating keratoplasty and glaucoma. Surv Ophthalmol. Sep-Oct 2000;45(2):91-105. [Medline].
• Beebe WE, Starita RJ, Fellman RL, et al. The use of Molteno implant and anterior chamber tube shunt to encircling band for the treatment of glaucoma in keratoplasty patients. Ophthalmology. Nov 1990;97(11):1414-22. [Medline].
• Bigar F, Witmer R. Corneal endothelial changes in primary acute angle-closure glaucoma. Ophthalmology. Jun 1982;89(6):596-9. [Medline].
• Binder PS, Abel R Jr, Kaufman HE. Cyclocryotherapy for glaucoma after penetrating keratoplasty. Am J Ophthalmol. Mar 1975;79(3):489-92. [Medline].
• Bourne WM, Davison JA, O''Fallon WM. The effects of oversize donor buttons on postoperative intraocular pressure and corneal curvature in aphakic penetrating keratoplasty. Ophthalmology. Mar 1982;89(3):242-6. [Medline].
• Chien AM, Schmidt CM, Cohen EJ, et al. Glaucoma in the immediate postoperative period after penetrating keratoplasty. Am J Ophthalmol. Jun 15 1993;115(6):711-4. [Medline].
• Cohen EJ, Kenyon KR, Dohlman CH. Iridoplasty for prevention of post-keratoplasty angle closure and glaucoma. Ophthalmic Surg. Dec 1982;13(12):994-6. [Medline].
• Figueiredo RS, Araujo SV, Cohen EJ, et al. Management of coexisting corneal disease and glaucoma by combined penetrating keratoplasty and trabeculectomy with mitomycin-C. Ophthalmic Surg Lasers. Nov 1996;27(11):903-9. [Medline].
• Foulks GN. Glaucoma associated with penetrating keratoplasty. Ophthalmology. Jul 1987;94(7):871-4. [Medline].
• Fraunfelder FT. Drugs used primarily in ophthalmology. In: Meyer SM, ed. Drug-induced Ocular Side Effects and Drug Interactions. Philadelphia: Lea & Febiger;1989:476-77.
• Gilvarry AM, Kirkness CM, Steele AD, et al. The management of post-keratoplasty glaucoma by trabeculectomy. Eye. 1989;3 ( Pt 6):713-8. [Medline].
• Goldberg DB, Schanzlin DJ, Brown SI. Incidence of increased intraocular pressure after keratoplasty. Am J Ophthalmol. Sep 1981;92(3):372-7. [Medline].
• Irvine AR, Kaufman HE. Intraocular pressure following penetrating keratoplasty. Am J Ophthalmol. Nov 1969;68(5):835-44. [Medline].
• Karesh JW, Nirankari VS. Factors associated with glaucoma after penetrating keratoplasty. Am J Ophthalmol. Aug 1983;96(2):160-4. [Medline].
• Kaufman HE, Varnell ED, Thompson HW. Latanoprost increases the severity and recurrence of herpetic keratitis in the rabbit. Am J Ophthalmol. May 1999;127(5):531-6. [Medline].
• Kirkness CM, Moshegov C. Post-keratoplasty glaucoma. Eye. 1988;2 Suppl:S19-26. [Medline].
• Kirkness CM. Penetrating keratoplasty, glaucoma and silicone drainage tubing. Dev Ophthalmol. 1987;14:161-5. [Medline].
• Knapp A, Heuer DK, Stern GA, Driebe WT Jr. Serious corneal complications of glaucoma filtering surgery with postoperative 5-fluorouracil. Am J Ophthalmol. Feb 15 1987;103(2):183-7. [Medline].
• Konowal A, Morrison JC, Brown SV, et al. Irreversible corneal decompensation in patients treated with topical dorzolamide. Am J Ophthalmol. Apr 1999;127(4):403-6. [Medline].
• Lass JH, Pavan-Langston D. Timolol therapy in secondary angle-closure glaucoma post penetrating keratoplasty. Ophthalmology. Jan 1979;86(1):51-9. [Medline].
• McDonnell PJ, Robin JB, Schanzlin DJ, et al. Molteno implant for control of glaucoma in eyes after penetrating keratoplasty. Ophthalmology. Mar 1988;95(3):364-9. [Medline].
• Olson RJ, Kaufman HE. A mathematical description of causative factors and prevention of elevated intraocular pressure after keratoplasty. Invest Ophthalmol Vis Sci. Dec 1977;16(12):1085-92. [Medline].
• Perl T, Charlton KH, Binder PS. Disparate diameter grafting. Astigmatism, intraocular pressure, and visual acuity. Ophthalmology. Aug 1981;88(8):774-81. [Medline].
• Rapuano CJ, Schmidt CM, Cohen EJ, et al. Results of alloplastic tube shunt procedures before, during, or after penetrating keratoplasty. Cornea. Jan 1995;14(1):26-32. [Medline].
• Reinhard T, Kallmann C, Cepin A, et al. The influence of glaucoma history on graft survival after penetrating keratoplasty. Graefes Arch Clin Exp Ophthalmol. Sep 1997;235(9):553-7. [Medline].
• Sherwood MB, Smith MF, Driebe WT Jr, et al. Drainage tube implants in the treatment of glaucoma following penetrating keratoplasty. Ophthalmic Surg. Mar 1993;24(3):185-9. [Medline].
• Shingleton BJ, Richter CU, Bellows AR, et al. Long-term efficacy of argon laser trabeculoplasty. Ophthalmology. Dec 1987;94(12):1513-8. [Medline].
• Siegner SW, Netland PA, Urban RC Jr, et al. Clinical experience with the Baerveldt glaucoma drainage implant. Ophthalmology. Sep 1995;102(9):1298-307. [Medline].
• Skuta GL, Beeson CC, Higginbotham EJ, et al. Intraoperative mitomycin versus postoperative 5-fluorouracil in high- risk glaucoma filtering surgery. Ophthalmology. Mar 1992;99(3):438-44. [Medline].
• Svedbergh B. Effects of artificial intraocular pressure elevation on the corneal endothelium in the vervet monkey (Cercopithecus ethiops). Acta Ophthalmol (Copenh). Dec 1975;53(6):839-55. [Medline].
• Threlkeld AB, Shields MB. Noncontact transscleral Nd:YAG cyclophotocoagulation for glaucoma after penetrating keratoplasty. Am J Ophthalmol. Nov 1995;120(5):569-76. [Medline].
• Topouzis F, Coleman AL, Choplin N, et al. Follow-up of the original cohort with the Ahmed glaucoma valve implant. Am J Ophthalmol. Aug 1999;128(2):198-204. [Medline].
• Wand M, Gilbert CM, Liesegang TJ. Latanoprost and herpes simplex keratitis. Am J Ophthalmol. May 1999;127(5):602-4. [Medline].
• Wilson SE, Kaufman HE. Graft failure after penetrating keratoplasty. Surv Ophthalmol. Mar-Apr 1990;34(5):325-56. [Medline].
• WuDunn D, Alfonso E, Palmberg PF. Combined penetrating keratoplasty and trabeculectomy with mitomycin C. Ophthalmology. Feb 1999;106(2):396-400. [Medline].
• Youn J, Cox TA, Herndon LW, et al. A clinical comparison of transscleral cyclophotocoagulation with neodymium: YAG and semiconductor diode lasers. Am J Ophthalmol. Nov 1998;126(5):640-7. [Medline].
• Zimmerman T, Olson R, Waltman S, Kaufman H. Transplant size and elevated intraocular pressure. Postkeratoplasty. Arch Ophthalmol. Dec 1978;96(12):2231-3. [Medline].
• Zimmerman TJ, Krupin T, Grodzki W, Waltman SR. The effect of suture depth on outflow facility in penetrating keratoplasty. Arch Ophthalmol. Mar 1978;96(3):505-6. [Medline].

Article Last Updated: Jun 6, 2006