Trabeculectomy Versus EX-PRESS Shunt Versus Ahmed Valve Implant: Short-term Effects on Corneal Endothelial Cells


To evaluate short-term changes in corneal endothelial cells after trabeculectomy, EX-PRESS device implantation, and Ahmed valve implantation for the treatment of primary open-angle glaucoma.


Prospective, interventional, comparative case series with contralateral eye control study.


We prospectively evaluated the changes in number, density, and shape of the corneal endothelium cells in 128 eyes of 64 patients divided into 3 groups depending on the treatment received. Corneal specular microscopy was performed with a noncontact specular microscope preoperatively and at 1 and 3 months after surgery. The changes at each time point were compared with those of the control group, which consisted of 32 contralateral glaucomatous eyes receiving antiglaucoma medications without any previous glaucoma surgery.


In the subjects who underwent trabeculectomy, corneal endothelial cell density (ECD) significantly decreased by 3.5% ( P = .012, paired t test) at 1 month and 4.2% ( P = .007) at 3 months after surgery, compared to the baseline values. In the Ahmed valve group ECD did not change at 1 month after surgery and had a significant 3.5% decrease at 3 months ( P = .04). In the patients who underwent EX-PRESS implantation and in the control group ECD did not change either at 1 month or at 3 months after surgery ( P > .05).


EX-PRESS shunt, compared to trabeculectomy and Ahmed valve, seems to be a safer procedure regarding the risk of endothelial cell loss. For this reason, it may be the treatment of choice in patients with significant low corneal ECD before surgery or other risk factors for corneal damage.

Glaucoma is a slowly progressive disease affecting the optic nerve, characterized by degeneration of retinal ganglion cells, loss of the retinal nerve fiber layer, and thinning of the neuroretinal rim, with consequent damage to the visual field and vision decline. It is a leading cause of irreversible blindness worldwide. In Europe, glaucoma is the cause of blindness in 10.6%–13.5% of cases and is present at the onset of approximately 4% of cases of moderate and severe vision impairment. The level of intraocular pressure (IOP), old age, family history of glaucoma, African-Caribbean ethnicity, use of topical or systemic corticosteroids, and central corneal thickness are considered major risk factors, although optic nerve damage is seen in subjects with normal IOP (normal-tension glaucoma).

Glaucoma treatment aims to slow the progression of the disease and preserve visual function without affecting quality of life. Currently, lowering IOP is the mainstay of glaucoma treatment. Medical therapy with antiglaucomatous drugs usually represents the first approach. The progression of perimetric damage and uncontrolled IOP, in spite of maximum tolerated medical therapy and laser trabeculoplasty, are some of the indications for surgical treatment. Nowadays, filtering surgery is the most frequently used technique for surgical glaucoma treatment. It comprises both posterior and anterior approaches, which are divided into penetrating (ie, trabeculectomy, EX-PRESS device implant) and nonpenetrating techniques (ie, deep sclerectomy, viscocanalostomy, angular implants).

Trabeculectomy is the most commonly performed glaucoma operation worldwide. Antifibrotic agents, such as mitomycin C (MMC) and fluorouracil, may be applied intraoperatively or via perioperative subconjunctival injection to reduce the episcleral proliferative response, which blocks the aqueous outflow channel. Glaucoma drainage implants are artificial devices inserted surgically into the anterior chamber (AC) to drain the aqueous toward the orbital surface of the eye. The Ahmed glaucoma valve is one of the most commonly used drainage implants. It consists of a silicone tube connected to a silicone sheet valve, which allows unidirectional liquid flow. The EX-PRESS shunt is a nonvalved stainless steel tube that is inserted under a partial-thickness scleral flap to connect the AC to the subconjunctival space.

Although glaucoma surgery has made significant progress in terms of safety, complications are still possible. We studied the possibility of faster postoperative loss of corneal endothelial cells, following a filtering procedure. A decrease in the number of corneal endothelial cells is a physiological phenomenon due to aging. Cell density ranges from almost 3000 cells/mm 2 in a young adult to just over 1000 cells/mm 2 in patients over 80 years old. Several risk factors are known to accelerate this loss, including surgery, antiscarring agents, argon laser iridotomy, and glaucoma itself. Corneal decompensation was reported as a late postoperative complication after trabeculectomy and tube shunt surgery. In long-term follow-up studies of Ahmed valve implantation, corneal decompensation has been reported to occur in up to 30% of patients. The aim of this study was to evaluate short-term changes in corneal endothelial cells after trabeculectomy, EX-PRESS shunt, and Ahmed glaucoma valve implantation for the treatment of primary open-angle glaucoma (POAG).


In this prospective, interventional, comparative case series with contralateral eye control study, we enrolled 64 subjects affected by POAG and requiring surgical treatment. The study was retrospectively approved by the Ethical Review Board of the University of Pisa (Comitato Etico, Pisa University, Pisa, Italy). A power analysis was executed using the effect size from the results of a previous study and indicated that 22 patients were required to detect a 2.6% endothelial cell density (ECD) decrease with a power of 80% and a significance level of 0.05.

The patients were divided into 3 groups depending on the type of glaucoma surgery received. Group 1 included 22 eyes of 22 patients who underwent trabeculectomy, Group 2 included 24 eyes of 24 patients who underwent EX-PRESS Glaucoma Filtration Device (Alcon, Fort Worth, Texas, USA) implantation, and Group 3 included 18 eyes of 18 patients who underwent Ahmed glaucoma valve (New World Medical, Inc, Rancho Cucamonga, California, USA) implantation. We also included a unique control group (32 eyes) to measure the contralateral eyes of subjects with glaucoma who were receiving antiglaucoma medication without any previous surgery, except for cataract surgery at least 180 days before enrollment. The choice of the type of treatment for the particular patient was made using agreed guidelines. For those patients that had received long-term topical therapy for inflamed conjunctiva and were at higher risk of bleb fibrosis, the Ahmed valve was preferred. Patients whose eyes had a wider angle at gonioscopy were assigned to the EX-PRESS shunt group; otherwise, trabeculectomy was performed.

Inclusion criteria were perimetric POAG, preoperative uncontrolled IOP despite maximum tolerated medical therapy, best-corrected visual acuity (BCVA) equal or above 24 letters using Early Treatment of Diabetic Retinopathy Study (ETDRS) methodology, and the ability to return in the 4 months following surgery for scheduled visits. Exclusion criteria were allergy to medication used during and after surgery, previous ocular surgery except for cataract surgery at least 180 days before enrollment, eyes receiving a topical anhydrase inhibitor, any corneal disease, presence of an intraocular lens in the anterior chamber, postoperative complications such as flat anterior chamber or need to refill the anterior chamber, and significant comorbid disease that could interfere with the follow-up.

Baseline data, including age at surgery, sex, type of glaucoma, glaucoma medications, number of previous intraocular surgeries, and lens status, were recorded. All operations were performed by the same experienced surgeon with a standard technique after obtaining informed consent.

Before surgery, and 1 month and 3 months after surgery, a complete ocular examination was performed, including BCVA (ETDRS optotype), IOP measurement (Goldmann applanation tonometry), slit-lamp examination, and corneal specular microscopy on the central area with a noncontact specular microscope (Tomey EM-3000, Nagoya, Japan). Noncontact specular microscopy is a standard technique used to assess corneal ECD and morphology; it has the advantage of being a simple, repeatable, and noninvasive examination. Specular microscopy was performed by the same expert physician and endothelial cell data were based on the average of 3 measurements. The following parameters were analyzed: ECD, coefficient of variation (CV) of cell area (polymegathism), hexagonality (pleomorphism), and central corneal thickness (CCT).

Statistical analysis was performed using SPSS software (SPSS Inc, Chicago, Illinois, USA). Homogeneity of variance was assessed with Levene’s test. Paired t tests were used to compare ECD , polymegathism, pleomorphism, and CCT before and after surgery. The comparison of changes in ECD among groups did not fulfill the assumptions of homogeneity of variance, so the Mann-Whitney U test was used. Differences were considered significant when P < .05.


Subject characteristics are summarized in Table 1 . Patient age, sex, lens status, and the number of previous intraocular surgeries did not differ significantly across the trabeculectomy, EX-PRESS shunt, Ahmed valve, and control groups.

Table 1

Trabeculectomy Versus EX-PRESS Shunt Versus Ahmed Valve Implant and Corneal Endothelial Cells: Demographic Characteristics of Patients Enrolled in the Study

Trabeculectomy EX-PRESS Ahmed Control P
Number of eyes 22 24 18 32
Mean age ± SD (y) 64.6 ± 9.3 63.6 ± 7.5 69.2 ± 8.4 66.4 ± 7.7 .150 a
Sex (male/female) 12/10 15/9 7/11 18/14 .490 b
Lens status, no. eyes (%) .873 b
Phakia 14 (63.6) 14 (58.3) 11 (61.1) 22 (68.7)
Aphakia or pseudophakia 8 (36.4) 10 (41.7) 7 (38.9) 10 (31.3)
No. of past intraocular surgeries, mean ± SD 0.8 ± 0.9 0.7 ± 0.7 0.9 ± 0.7 0.6 ± 0.5 .368 a

a One-way analysis of variance.

b χ 2 test.

After surgery with all 3 procedures there was a significant decrease in IOP and number of antiglaucoma medications. Conversely, changes in IOP and number of antiglaucoma medications were not significant in the control group ( Table 2 ). Performing a Pearson correlation, we did not find any correlation between preoperative IOP and ECD decrease after surgery (Group 1 P = .091; Group 2 P = .972; Group 3 P = .435), nor did we find a correlation between preoperative number of medications and ECD decrease (Group 1 P = .846; Group 2 P = .566; Group 3 P = .959).

Table 2

Intraocular Pressure (mm Hg) and Number of Antiglaucoma Medications Before and After Trabeculectomy, EX-PRESS Shunt, and Ahmed Valve Implantation

Time of Examination Trabeculectomy (n = 22) EX-PRESS (n = 24) Ahmed (n = 18) Control (n = 32)
IOP (mm Hg) Eye Drops IOP (mm Hg) Eye Drops IOP (mm Hg) Eye Drops IOP (mm Hg) Eye Drops
Baseline 26.6 ± 3.4 3.0 ± 0.7 26.5 ± 4.5 3.1 ± 0.8 27.7 ± 3.2 3.2 ± 0.7 19.9 ± 2.6 2.7 ± 0.8
1 month 19.9 ± 3.5 a 1.5 ± 1.1 a 19.4 ± 4.2 a 1.3 ± 1.0 a 18.9 ± 3.9 a 1.2 ± 0.9 a 19.4 ± 3.2 2.6 ± 0.9
3 months 20.3 ± 4.0 a 1.3 ± 0.9 a 20.0 ± 3.0 a 1.3 ± 0.9 a 19.5 ± 3.5 a 1.0 ± 0.9 a 19.5 ± 2.8 2.6 ± 0.8

Data are presented as mean ± standard deviation.

a P < .05 compared with baseline (paired t test).

Among subjects who underwent trabeculectomy, corneal ECD was significantly decreased by 3.5% 1 month after surgery ( P = .012, paired t test) compared to baseline values; 3 months after surgery ECD was significantly decreased by 4.2% ( P = .007). Among subjects who underwent Ahmed valve implantation, ECD decreased by 1.4% ( P = .020) 1 month after surgery and by 3.5% 3 months after surgery ( P = .043). Among patients who underwent EX-PRESS implantation and in the control group, ECD did not change either 1 month or 3 months after surgery ( P > .05) ( Table 3 ).

Jan 6, 2017 | Posted by in OPHTHALMOLOGY | Comments Off on Trabeculectomy Versus EX-PRESS Shunt Versus Ahmed Valve Implant: Short-term Effects on Corneal Endothelial Cells

Full access? Get Clinical Tree

Get Clinical Tree app for offline access