Twenty-Four-Month Outcome of Gonioscopy-Assisted Transluminal Trabeculotomy (GATT) in Eyes With Prior Corneal Transplant Surgery

Purpose: Glaucoma is a well-known sequelae of corneal transplant surgery and is a leading cause of visual loss in this patient group. We evaluated the performance and safety of gonioscopy-assisted transluminal trabeculotomy (GATT) in this population. Design: Noncomparative retrospective case series. Participants: Consecutive eyes of patients receiving the GATT procedure for uncontrolled intraocular pressure (IOP) following corneal transplant surgery from 2016 to 2019. Methods: Retrospective analysis of eyes with a history of prior corneal transplant undergoing GATT at Glaucoma Associates of Texas between 2016 and 2019 was performed. Data included IOP, patient demographics, preoperative and postoperative medications, preoperative and postoperative corneal procedures, complications, and need for reoperation for IOP control. Main Outcome Measures: IOP reduction and medication use following the procedure. Results: Thirty-nine eyes of 32 patients with prior corneal transplant surgery underwent a GATT procedure. Prior corneal surgery included penetrating keratoplasty (59.0%), Descemet’s stripping endothelial keratoplasty (35.9%), Descemet’s membrane endothelial keratoplasty (2.6%), and deep anterior lamellar keratoplasty (2.6%). Patient age ranged from 24 to 94 years (mean 68.0 y) with 44% female and 81% Caucasian. The majority of patients had secondary open angle glaucoma (64.1%). There was a significant reduction in IOP and number of medications at all postoperative time points after censoring patients requiring reoperation to control IOP (P<0.001). At 24 months the pressure decreased from baseline of 30.9±11.5 to 13.9±4.7 mm Hg. Medications decreased from 4.2±1.0 medications at baseline to 0.6±1.0 at 24 months. Visual acuities decreased significantly over the first postsurgical month (all P<0.05), but these recovered at subsequent follow-up visits with 2-Snellen line improvements exceeding losses from month 3 to 36. Seven eyes required reoperation for uncontrolled glaucoma at a median of 8.5 months (range: 1.6 to 16.2 mo) after GATT. The cumulative proportion of eyes undergoing repeat cornea surgery was 2.6%, 2.6%, and 14.3% at 12, 24, and 36 months post-GATT, respectively. Conclusions: This case series describes a group of glaucoma patients, with a history of prior corneal surgery, that were safely and successfully treated with GATT. While classically traditional glaucoma surgeries are considered the standard of care for eyes following corneal transplant surgery, GATT should be considered as a reasonable, safe and effective alternative for surgically lowering IOP.

Results: Thirty-nine eyes of 32 patients with prior corneal transplant surgery underwent a GATT procedure. Prior corneal surgery included penetrating keratoplasty (59.0%), Descemet's stripping endothelial keratoplasty (35.9%), Descemet's membrane endothelial keratoplasty (2.6%), and deep anterior lamellar keratoplasty (2.6%). Patient age ranged from 24 to 94 years (mean 68.0 y) with 44% female and 81% Caucasian. The majority of patients had secondary open angle glaucoma (64.1%). There was a significant reduction in IOP and number of medications at all postoperative time points after censoring patients requiring reoperation to control IOP (P < 0.001). At 24 months the pressure decreased from baseline of 30.9 ± 11.5 to 13.9 ± 4.7 mm Hg. Medications decreased from 4.2 ± 1.0 medications at baseline to 0.6 ± 1.0 at 24 months. Visual acuities decreased significantly over the first postsurgical month (all P < 0.05), but these recovered at subsequent follow-up visits with 2-Snellen line improvements exceeding losses from month 3 to 36. Seven eyes required reoperation for uncontrolled glaucoma at a median of 8.5 months (range: 1.6 to 16.2 mo) after GATT. The cumulative proportion of eyes undergoing repeat cornea surgery was 2.6%, 2.6%, and 14.3% at 12, 24, and 36 months post-GATT, respectively.
Conclusions: This case series describes a group of glaucoma patients, with a history of prior corneal surgery, that were safely and successfully treated with GATT. While classically traditional glaucoma surgeries are considered the standard of care for eyes following corneal transplant surgery, GATT should be considered as a reasonable, safe and effective alternative for surgically lowering IOP.
Key Words: secondary glaucoma, gonioscopy-assisted transluminal trabeculotomy, trabecular meshwork stripping, corneal transplantation: glaucoma surgery, steroid-induced glaucoma (J Glaucoma 2022;31:54-59) G laucoma is a well-known sequelae of corneal transplant surgery and is one of the leading causes of visual loss in this group. 1 The incidence of postoperative intraocular pressure (IOP) elevation in eyes after penetrating keratoplasty (PKP), Descemet's stripping endothelial keratoplasty (DSEK/DSEAK), deep anterior lamellar keratoplasty (DALK), or Descemet's membrane endothelial keratoplasty (DMEK) ranges from 15% to 35%. [2][3][4][5][6][7][8][9][10][11][12] Further, posttransplant glaucoma is the second leading cause of graft failure after rejection. 13 In the acute postoperative period after DSEK or DMEK, an elevation in IOP is often the result of angle closure because of air bubble dislocation behind the iris and pupillary block. This can be successfully managed with pupillary dilation, positioning and medical therapy. Delayed postoperative elevations of IOP are the focus of this study and are caused by other mechanisms and must be managed differently. Proposed mechanisms of delayed elevated IOP after corneal transplant include outflow dysfunction because of corticosteroid use, collapse of the iridocorneal angle, postoperative inflammation, and synechial angle closure. [14][15][16] IOP elevation after corneal transplant can be managed successfully with medications and laser trabeculoplasty for many patients. However, there are some whose IOP remains inadequately controlled and are at risk for irreversible optic nerve damage without surgical intervention. Glaucoma surgery in these patients traditionally includes trabeculectomy, glaucoma drainage device or cyclodestructive procedures. There are case series of these procedures in post-transplant eyes that show varying rates of success at lowering IOP but also higher risk of complications including shorter time to graft rejection, greater likelihood of multiple rejection episodes and greater chance of graft failure. 13,17 A microinvasive surgical option for post-transplant eyes with an open angle is a gonioscopy-assisted transluminal trabeculotomy (GATT) procedure. This procedure allows for a circumferential trabeculotomy through an ab DOI: 10.1097/IJG.0000000000001949 interno approach to canulate and circumnavigate Schlemm's canal with either a blunt-tip suture or flexible iTrack microcatheter (Nova Eye Medical, Kent Town, SA). 18 With this approach the conjunctiva is spared, and the complications associated with blebs and drainage devices are avoided. In a study by Grover et al, 18 79 eyes with secondary open angle glaucoma underwent a GATT procedure, and at 24 months there was an average decrease in IOP of 14.1 mm Hg (49.8%) on an average of 2 fewer medications.
There is 1 case report of successfully using GATT in a post-PKP steroid-induced glaucoma patient. 19 We report 24-month outcomes in a series of eyes with prior corneal transplants treated with GATT.

METHODS
A retrospective chart review was performed for all patients who underwent a GATT procedure who had a preexisting corneal transplant (PKP, DSEK, DMEK, or DALK). The GATT procedure was performed by the physicians at Glaucoma Associates of Texas (OUS, MRB, MEE, DGG, and HLK) between January 2016 through December 2019. This study followed the tenants of the Declaration of Helsinki and was approved by the affiliated hospital's institutional review board. Written consent was obtained from all patients before surgery.
All patients had a preoperative exam that included evaluation of their corneal graft status (clear, edematous, or scarred) as well as gonioscopy, documenting an open angle and identifiable angle landmarks. The GATT procedure has been previously reported but a brief summary will be described. 20 A temporal paracentesis is created with a 15-degree blade. Cohesive viscoelastic (sodium hyaluronate) is injected into the anterior chamber. A 23-gauge needle paracentesis track is created in either the superonasal or superotemporal quadrant. A thermally blunted 5-0 Prolene suture (Ethicon) or illuminated iTrack microcatheter (Nova Eye Medical) is inserted into the anterior chamber through the nasal entry track. The patient's head and microscope are rotated and a Swan-Jacob gonio lens is placed on the eye to view the nasal angle structures. A goniotomy is created by incising the nasal trabecular meshwork with a 25-gauge microvitreoretinal blade. Microsurgical forceps are then used to feed the distal tip of the suture or microcatheter into Schlemm's canal through the goniotomy site. The microsurgical forceps are used to advance the suture or catheter around the canal 360 degrees until the distal tip emerges at the other end of the goniotomy. The end is then retrieved and externalized while pulling gentle traction on the trailing end to cleave the meshwork, thus creating a circumferential 360-degree ab interno trabeculotomy. The viscoelastic is removed with bimanual irrigation aspiration. The anterior chamber is occasionally left with a 15% to 30% fill of a cohesive viscoelastic to help tamponade bleeding and protect from early postoperative hypotony that would result in additional blood reflux. Postoperative steroid and antibiotic drops are given at the surgeon's discretion. If the suture does not circumnavigate the canal for 360 degrees, a partial circumferential trabeculotomy can be created by grasping the proximal end of the suture next to the initial goniotomy site and pulling the suture along the arc of the canal as it cleaves it open. This usually opens the canal for several clock hours until the surgeon approaches the distal tip. This type of partial circumferential trabeculotomy is possible when the suture is in the canal for several clock hours but will not advance to the initial goniotomy site. When the suture or iTrack only traverses the canal for 1 or 2 clock hours, the device should be reinserted through a separate paracentesis to see if the canal can be cannulated from the other direction, which is often the case. The goal of each procedure is to open as much canal as possible.
In this study, the GATT procedure was done predominantly as a standalone procedure, but there was one performed in combination with cataract extraction/intraocular lens insertion and one that included goniosynechiolysis. During goniosynechiolysis, a 4-mirror gonioprism was placed on the eye to visualize the angle. Peripheral anterior synechiae were manually dissected off with viscoelastic and micro forceps or a spatula if needed.
Postoperatively, all patients were instructed use topical broad-spectrum antibiotics and topical steroids. The antibiotics were stopped at postoperative week 1. The topical steroids were tapered to the patient's baseline level at the surgeon's discretion. Topical pilocarpine was applied once at nighttime for a varying time frame in some patients with the aim of potentially keeping the cleaved open trabecular shelf functional to allow the exit of aqueous through collector channels. The patient's IOP was managed per the surgeon discretion as well. Data was collected for postoperative day 1, week 1, month 1, month 3, month 6, month 12, and month 24. Visual acuity, IOP, medications, and complications were collected at each visit. The corneal graft status and need for additional surgery (glaucoma or cornea) was obtained.

Statistical Methods
Continuous variable measurements were summarized with means, SDs, and ranges. Binary and polychotomous data were summarized with percentages. Time to surgery for both IOP control and repeat cornea surgery was assessed with Kaplan-Meier methods and cumulative proportions were reported with 95% confidence intervals (95% CI). The statistical significance of post-GATT reductions in IOP and medications required for IOP control was evaluated with the paired t test. Postoperative changes in visual acuity were assessed with the paired Wilcoxon test.
Preoperatively, mean IOP was 30.9 ± 11.5 mm Hg on 4.2 ± 1.0 medications. Thirty-three eyes (84.6%) were on topical steroids preoperatively. The median number of months from corneal surgery to GATT was 24.8 (interquartile range: 11.0 to 63.6) months with a range of 4.4 months to 14 years. Ten eyes (25.6%) had surgery < 12 months from corneal surgery, 46.2% between 12 and 60 months, 20.5% 61 months to 10 years, and 7.7% over 10 years from the corneal procedure.
GATT as a standalone procedure was performed in all eyes except 2: one had GATT combined with phacoemulsification and the other with goniosynechiolysis. 82.1% of patients underwent GATT with an iTrack catheter and 17.9% with a suture. The total amount of canal opened at the time of surgery was 360 degrees in 34 eyes and 180 to 359 degrees in 5 eyes.
There were no intraoperative complications noted. Gonioscopy was performed postoperatively and confirmed at least 180 degrees of a visible trabecular shelf after GATT in most patients.
There was significant reduction in IOP and number of medications at all postoperative time points after censoring patients requiring reoperation to control IOP (P < 0.001). At 24 months the pressure decreased from baseline of 30.9 ± 11.5 to 13.9 ± 4.7 mm Hg. Medications decreased from 4.2 ± 1.0 medications at baseline to 0.6 ± 1.0 at 24 months. Of the patients reaching 24 months, 88% remained on chronic topical steroid for graft maintenance ( Table 2).
Among the entire study cohort, 4 eyes underwent repeat corneal surgery. One eye that required a tube shunt 4 months after GATT developed corneal edema and needed corneal surgery 32 months post-GATT. The corneal edema was attributed to the tube shunt in this case. Another eye with a DSEK had pre-existing corneal edema at the time of GATT surgery which did not resolve after IOP improvement and required a repeat DSEK 6 months after GATT and a subsequent DMEK 2 years later. The 2 other eyes were stable for over 3 years before developing corneal edema and undergoing repeat corneal surgery. The length of time from last transplant before GATT to rejection in these four eyes was 22 months, 38 months, 5 years, 8 months, and 15 years. The cumulative proportion of eyes undergoing repeat cornea surgery was 2.6%, 2.6%, and 14.3% at 12, 24, and 36 months post-GATT, respectively (Fig. 2).
Visual acuities decreased significantly over the first postsurgical month (all P < 0.05), but these recovered at subsequent follow-up visits with 2-Snellen line improvements exceeding losses from month 3 to 36. Table 3 presents visual acuity results over follow up for this cohort. Of the 4 eyes with visual decrease from baseline at 12 months, 2 eyes had deteriorated from wet age-related macular degeneration with clear corneas, and 1 eye had a tube shunt with corneal decompensation and inflammation.

DISCUSSION
Treating IOP elevation and glaucoma in eyes with prior corneal transplantation surgery is extremely challenging. The goal is to prevent damage to the optic nerve through lowering IOP while at the same time, minimizing any potential insult to the corneal graft tissue.
This retrospective case series demonstrates that GATT provides a clinically and statistically significant reduction in both IOP and number of medications in eyes with glaucoma following corneal surgery. At 24 months, the mean IOP of  the cohort was 13.9 mm Hg achieving a 56.2% reduction from baseline on 3.1 fewer medications. Previously published studies on trabeculectomy and tube shunts after a corneal transplant surgery report successful IOP control at 2 years around 50% after trabeculectomy and 62% to 96% after glaucoma drainage device. 17 The success after GATT in this study is comparable with 79.7% controlled IOP without additional glaucoma surgery at 24 months.
Compared with trabeculectomy and drainage implants, GATT is less invasive, spares the conjunctival and has a more favorable safety profile. The tube versus trabeculectomy study found that trabeculectomy carries greater risk of bleb-related complications such as hypotony, choroidal effusion, bleb leak, cystoid macular edema, flat chamber, suprachoroidal hemorrhage, endophthalmitis, and even corneal edema. 21 Gedde et al 21 also found late tube shunt complications to include diplopia, choroidal effusion, cystoid macular edema, tube erosion, endophthalmitis, tube obstruction, chronic iritis, corneal edema, and flat chamber. The most common adverse event noted in this study was hyphema which is expected with the GATT procedure. Most hyphemas resolved within the first month. Importantly, there were no cases of Descemet's detachment or graft dehiscence in any of these cases.
Persistent corneal edema is seen in 16% of the eyes with tube shunts and 9% of the trabeculectomies. 21 In this study, 10% of eyes had early postoperative corneal edema after GATT with all but 1 resolving by 9 months after surgery.   Long-term graft survival was observed in patients following GATT with 2.6% needing repeat corneal surgery at 24 months and 14.3% at 36 months. There was no difference in the surgery, amount of hyphema, or early postoperative management in the 4 patients that ultimately required repeat corneal surgery. In comparison, corneal graft failure at 2 years has been reported to be 40% after trabeculectomy and 35% to 74% after glaucoma drainage device. 17 Studies looking at endothelial cell loss after glaucoma drainage implant surgery found 36.8% to 50.1% central and peripheral endothelial cell loss following Baerveldt glaucoma implants at 5 years. 22,23 GATT does not leave any foreign body in the anterior chamber and thus there is no risk of ongoing endothelial damage because of a device left in the anterior segment of the eye. Similar excisional goniotomies with the Kahook dual blade (New World Medical, Rancho Cucamonga, CA) and the Trabectome (Microsurgical Technology, Redmond, WA) were associated with endothelial cell loss of 3.4% at 18 months and no further loss at 36 months. 22,24,25 Moreover, the authors are aware of the challenges of trying to fit post-PKP patients with a contact lens, especially after a trabeculectomy or a tube shunt. The conjunctiva sparing nature of GATT allows the patient to return safely to contact lens wear without the fear of an increased risk of endophthalmitis or tube erosion.
In addition, most corneal transplant patients still require topical corticosteroid use to prevent rejection, and historically, trabeculotomy has an excellent track record for controlling IOP with steroid-induced glaucoma. 26 The majority of cases were performed with an illuminated iTrack microcatheter (Nova Eye Medical) to assist in identifying the location of the leading end of the threading device at all times. This was preferred because of the occasional difficulty viewing the angle because of obscuration at the graft host interface and the need to complete the procedure quickly without excessive instrumentation in the anterior chamber.
This analysis has all the limitations of a retrospective study. There was potential selection bias since the decision to perform a GATT was at the surgeon's discretion. Those eyes with extensive peripheral synechiae or poor visualization through the cornea graft were poor candidates for GATT and other surgeries were recommended. The retrospective nature is dependent on accurate data acquisition and coding. The surgical technique was not standardized (suture vs. iTrack, variation in amount of canal opened, and amount if any of cohesive viscoelastic left at the conclusion of the case). Further, postoperative drop management was left to the discretion of the surgeon. The goal of surgery for most eyes was IOP reduction, but for 6 eyes (15.4%) who had controlled IOP on 4 or more medications, the goal was to reduce the medication burden. This made it difficult to define success and failure based solely on pressure reduction, so instead the Kaplan-Meier curve reflects the number of eyes that required additional glaucoma surgery (Fig. 1) and repeat corneal surgery (Fig. 2). The study is also limited by size and follow up. While 39 eyes of 32 patients is still a small study, this is the largest to date looking at GATT in eyes with prior corneal transplants. Twenty eyes (51.3%) were accounted for at the 24-month follow up (Table 2).
Patients with prior corneal transplant surgery are a challenging group to care for, especially when the IOP begins to increase. While traditional glaucoma surgeries were considered the standard of care for glaucomatous eyes with prior corneal surgery, the authors feel GATT surgery can and should be considered as a reasonable, safe and effective alternative for surgically lowering IOP.
This case series describes a group of glaucoma patients, the largest to our knowledge, with a history of prior corneal surgery, that were safely and successfully treated with GATT, a circumferential ab interno conjunctival sparing trabeculotomy.