Abstract
Purpose:
To evaluate the incidence of failure of trabeculectomy vs. tube shunt glaucoma surgery in eyes of patients with uveitis.
Design:
Multi-center retrospective cohort study.
Participants:
Among 356 eyes of 288 patients with noninfectious inflammatory eye disease undergoing first incisional glaucoma surgery using one of the techniques, 244 eyes had tube shunts (TS) and 112 eyes had trabeculectomy augmented with mitomycin-C (Trab-MMC).
Methods:
Standardized chart review was used to collect clinical data over time retrospectively. Cox regression analyses with adjustment for propensity score and inter-eye correlations were performed to compare the incidence of failure of glaucoma surgery between TS and Trab-MMC.
Main Outcome Measures:
Failure of glaucoma surgery of the first five years postoperatively, defined as: (1) IOP≤ 5 mmHg or >21 mmHg at two consecutive visits at least 90 days apart beginning three months after surgery; or (2) reoperation; or (3) complete blindness (no light perception).
Results:
The median age was 40.3 years (IQR 13.4-57.3) in the TS group and 44.2 years (IQR 29.0-58.9) in the Trab-MMC group. The median pre-glaucoma surgery IOP was 29.0 mm Hg (IQR 21-35.5) in the TS group and 30.0 mm Hg (IQR 20-38) in the Trab-MMC group. Anterior uveitis was most common location of primary inflammation in both the TS group (52.5%) and Trab-MMC group (55.4%). Failure was observed in the TS group in 23.5%, 27.1% and 30.8% cumulatively through 12, 24, and 36 months respectively vs. 16.1%, 25.6% and 30.0% respectively in the Trab-MMC group. In the propensity score adjusted Cox regression analysis, there was no significant difference in failure incidence rate between the TS and trab-MMC groups (adjusted hazard ratio 1.08, 95% CI 0.65-1.78, p=0.77). Success without requirement for IOP-lowering medicines was observed more frequently in the Trab-MMC group.
Conclusions:
TS and Trab-MMC fail frequently with similar incidences when done as the first glaucoma surgery among eyes with uveitis over 5 years of follow-up, but there were more complete successes in the Trab-MMC group compared to the TS group at 12, 24 and 36 months.
Introduction
Glaucoma, a progressive neurodegenerative disease of the optic nerve, is one of the most common causes of irreversible blindness worldwide.1,2 Glaucoma is generally classified into primary and secondary forms, each of which can have open or closed angles.
Uveitic glaucoma is a type of secondary glaucoma that can be very challenging to manage. Patients with uveitis develop glaucoma at a high rate. Fifteen percent of patients with intermediate, posterior and panuevitis develop glaucoma over 10 years when treated with systemic therapies.3 Uveitic glaucoma can have features of open or closed angle glaucomas.4,5 The pathogenesis is heterogeneous, including trabeculitis or blockage of the trabecular meshwork due to inflammatory cells and pigment;6 reduced trabecular outflow facility related to chronic use of local and/or systemic corticosteroids;7 and angle scarring—peripheral anterior synechiae and/or pupillary block due to posterior synechiae.6 In addition, uveitis relies heavily on corticosteroid treatment (local and systemic), which can lead to refractory corticosteroid-response intraocular pressure (IOP) elevation potentially leading to glaucomatous optic neuropathy.3,5,8 In the corticosteroid-response scenario, the angle is open but aqueous fluid is not effectively drained from the eye as a result of exposure to corticosteroids changing gene expression in the trabecular meshwork.9 Because of its multiple potential mechanisms, uveitic IOP elevation and glaucoma can be difficult to control, not infrequently requiring surgery.10
Formal recommendations on which type of glaucoma surgery is preferable in uveitic glaucoma do not exist. Some clinicians recommend tube shunts in eyes with uveitic glaucoma and uncontrolled IOP, on grounds that inflammation is likely to close a simple filtration bleb. Others have recommend conducting trabeculectomy supplemented with intraoperative mitomycin C (MMC) treatment in order to enhance survival.11,12 However, many patients with uveitic glaucoma are young and the long-term outcomes of tube shunts or of MMC treatment in these eyes are not well described. The mechanical stimulation of the intraocular structures from the tube shunt might have undesirable effects over time and therefore a trabeculectomy without an implant could be safer in the long-term for these young patients. For instance, within five years postoperatively nearly 10% needed corneal endothelial transplant surgery with tube shunt placement as compared to <1% with trabeculectomy.13 Data on the effectiveness of the alternative approaches are needed to provide evidence-based decision making about which technique to recommend.
Here, we report the incidence of failure of initial glaucoma surgery in a cohort of uveitic eyes of adult and pediatric patients with non-infectious anterior, intermediate, posterior or panuveitis managed at tertiary uveitis centers in the United States.
Patients and Methods
Study Population
This analysis used data from the Systemic Immunosuppressive Therapy for Eye Diseases (SITE) Cohort Study, a multi-center retrospective cohort study.14–16 The SITE database included patients with non-infectious ocular inflammation from four ocular inflammation tertiary-care centers in the United States. All patients who met eligibility criteria were included into the database from the time each center originated (as early as 1975) until December 2010. Details of the database creation are documented elsewhere.14,15 The study was deemed exempt by the Vanderbilt University Medical Center Institutional Review Board, as the analysis of a previously collected database did not include any identifiers. The study was adherent to the tenents of the Decleration of Helsinki. There were no human or animal subjects used.
For this analysis, inclusion criteria included eyes of patients with non-infectious intraocular inflammation (anterior, intermediate or posterior/panuveitis) which underwent their first incisional glaucoma surgery while under observation in the SITE cohort, surgery occurring in the year 2000 or later, having had preoperative IOP recorded within one month of the glaucoma surgery, and a minimum of one year of post-operative follow-up. Eyes were excluded if they had prior incisional glaucoma surgery, if the inflammation was extraocular (e.g., scleritis, ocular pemphigoid, orbital inflammation), or if a technique other than tube shunt or trabeculectomy with MMC was used. We did not differentiate between eyes with ocular hypertension and glaucoma as we did not have access to visual field testing or the glaucoma specialist documentation. We included them under an umbrella term of “uveitic glaucoma”.
Data Collection:
Data were abstracted by chart review and entered by reviewers into the SITE Cohort Study database.14 Potentially predictive factors collected for this analysis included demographic and clinical characteristics of patients (age, gender, race/ethnicity, smoking status, concurrent use of systemic immunosuppressive agents, associated systemic inflammatory disease) and ocular characteristics of uveitic eyes (primary anatomic location of the ocular inflammation, activity level of inflammation at the visit before surgery, pre-operative IOP, use of IOP lowering medication, and visual acuity).
If an anti-fibrotic [5-Flurouracil (5FU) or MMC] was used for the trabeculectomy, the type of anti-fibrotic was recorded. If the type of anti-fibrotic used was not specified in the chart, the anti-fibrotic was recorded as “unknown”. If the chart specified that no anti-fibrotic was used it was recorded as “no anti-fibrotic .” We excluded all trabeculectomies for the analysis except those done with MMC, as that is method currently employed by glaucoma specialists world wide. The model of tube shunt implanted was incompletely recorded, so the type of tube shunt was not differentiated in the analysis.
Main Outcome Measures:
The primary outcome was failure of incisional glaucoma surgery, over the first five years following trabeculectomy with MMC (trab-MMC) or tube shunt (TS). Failure was defined when the first of the following were observed: 1) IOP >21 mmHg at two consecutive visits at least 90 days apart beginning three months after surgery, 2) IOP ≤ 5 mmHg on two consecutive visits at least 90 days apart beginning three months after surgery, 3) Reoperation for glaucoma at any time after the initial surgery, or 4) complete blindness in the eye (no light perception). To limit surgical decision-making bias, we ran a sub analysis on the first eye for each patient to compare to the whole cohort.
As secondary outcomes, we divided the success into complete and qualified successes at one year, two years and three years. Complete success was achieved when an eye did not reach a failure event and was not on IOP lowering medications at one year, two years and three years. A qualified success was defined as an eye that had not met a failure event but did require the use of IOP- lowering medications by each time point.
Statistical Analysis:
Kaplan-Meier curves were used to demonstrate the time to failure after glaucoma surgery in each surgical group. To evaluate the imbalances in pre-surgery variables between two glaucoma surgery groups, we conducted a propensity score analysis for the TS compared to the Trab-MMC group. Briefly, a propensity score analysis simulates a randomized experiment using observational data. A propensity score is estimated through a logistic regression model for the surgery group, that includes pre-determined variables that may be associated with both the surgery group and outcomes, including age, gender, smoking status, anatomic location of uveitis, activity of inflammation and history of cataract surgery. Cox regression analyses with adjustment for propensity score and inter-eye correlations were then performed to compare the incidence of failure of glaucoma surgery between TS and Trab-MMC. To address decision bias, we performed a propensity score sub analysis of the primary outcomes considering only the first eye surgery. For our secondary outcome, we compared complete and qualified success for each surgery type at one, two and three years with a chi- squared t-test. All the statistical analyses were performed in SAS v9.4 (SAS Institute Inc., Cary, NC).
Results
Characteristics of Study Population
Among the patients included in the SITE database16, there were 837 eyes (662 patients) that underwent a TS or Trab-MMC during the cohort’s observation. Of those, 433 eyes (351 patients) met study criteria (the rest excluded for having extraocular inflammation, not being the first glaucoma surgery, no follow-up, and surgery performed before the year 2000). Of the included eyes, 244 (200 patients) received a TS and 185 eyes (149 patients) underwent trabeculectomy. For our analysis, we chose to include eyes that had the trabeculectomy confirmed as augmented with MMC (112 eyes from 88 patients) as that is the most commonly performed method of trabeculectomy today. We excluded 52 eyes from 46 patients that were recorded as having an “unknown” (or no) anti-fibrotic , patients that had trabeculectomies augmented with 5FU (7 eyes) and those that had no adjunct anti-fibrotic (14 eyes) Table 1 describes the baseline demographic and ocular characteristics of eyes/patients meeting the study inclusion criteria in both surgical groups.
Table 1:
Characteristics of the cohort at the visit preceding glaucoma surgery by treatment group.
| Variable | Tube Shunt | Trabeculectomy with Mitomycin C | p-value | |
|---|---|---|---|---|
| Eyes | Number | 244 | 112 | - |
| Patients | Number | 200 | 88 | - |
| Age at time of glaucoma surgery, years | Mean (SD) | 38.4 (22.87) | 43.4 (19.57) | 0.05 |
| Median (IQR) | 40.3 (13.4-57.3) | 44.2 (29.0-58.9) | 0.25 | |
| Age group, years | 18 and under | 77 (31.6%) | 16 (14.3%) | 0.01 |
| 19-44 | 61 (25.0%) | 41 (36.6%) | ||
| 45-64 | 75 (30.7%) | 38 (33.9%) | ||
| Over 65 | 31 (12.7%) | 17 (15.2%) | ||
| Gender | Female | 159 (65.2%) | 71 (63.4%) | |
| Race/ethnicity | White | 148 (60.7%) | 67 (59.8%) | 0.51 |
| Black | 45 (18.4%) | 26 (23.2%) | ||
| Hispanic | 13 (5.3%) | 7 (6.3%) | ||
| Other | 38 (15.6%) | 12 (10.7%) | ||
| Smoking status | Never Smoker | 178 (73.0%) | 74 (66.1%) | 0.25 |
| Former Smoker | 25 (10.2%) | 20 (17.9%) | ||
| Current Smoker | 33 (13.5%) | 15 (13.4%) | ||
| Unknown | 8 (3.3%) | 3 (2.7%) | ||
| Concurrent daily prednisone or equipotent oral corticosteroid dose | None | 195 (79.9%) | 76 (67.9%) | 0.05 |
| 7.5 mg or lesser | 19 (7.8%) | 14 (12.5%) | ||
| More than 7.5 mg | 30 (12.3%) | 22 (19.6%) | ||
| Concurrent daily topical corticosteroid use | No Drops | 129 (52.9%) | 53 (47.3%) | 0.08 |
| 2 or fewer drops | 46 (18.9%) | 33 (29.5%) | ||
| More than 2 drops | 69 (28.3%) | 26 (23.2%) | ||
| Associated systemic conditions | HLA-B27+ | 16 (6.6%) | 12 (10.7%) | |
| Juvenile Idiopathic Arthritis | 43 (17.6%) | 7 (6.3%) | ||
| Sarcoidosis | 27 (11.1%) | 12 (10.7%) | ||
| Behçet Disease | 4 (1.6%) | 4 (3.6%) | ||
| Anatomic location of inflammation | Anterior uveitis | 128 (52.5%) | 62 (55.4%) | 0.96 |
| Intermediate uveitis | 41 (16.8%) | 17 (15.2%) | ||
| Posterior uveitis | 35 (14.3%) | 15 (13.4%) | ||
| Panuveitis | 40 (16.4%) | 18 (16.1%) | ||
| Pre-surgical visual acuity | Better than 20/40 | 110 (45.1%) | 61 (54.5%) | 0.31 |
| 20/50 - 20/80 | 55 (22.5%) | 17 (15.2%) | ||
| 20/100 - 20/200 | 39 (16.0%) | 17 (15.2%) | ||
| Worse than 20/200 | 40 (16.4%) | 17 (15.2%) | ||
| Presurgical inflammation status | Inactivea/Slightly activeb | 207 (84.8%) | 101 (90.2%) | 0.17 |
| Activec | 37 (15.2%) | 11 (9.8%) | ||
| Pre-surgical intraocular pressure, mmHg | Mean (SD) | 29.0 (10.70) | 30.0 (12.68) | 0.44 |
| Median (IQR) | 30.0 (21.0-35.5) | 30.5 (20.0-38.0) | 0.69 | |
| Pre-surgical IOP-lowering medication use | Yes | 233 (95.5%) | 106 (94.6%) | |
| Prior Cataract Surgery | Yes | 135 (55.3%) | 62 (55.4%) |
SD=standard deviation; IQR=interquartile range; mmHg=millimeters of mercury; IOP=intraocular pressure;
Inactive: The chart indicates there is no inflammation activity on this visit. Examples would include “quiet”, “quiescent”, “stable”, “No cells”. Neither “active” nor “slightly active” as described below are applicable;
Slightly Active: Activity that is minimally present, described also by terms such as mild, few, trace etc;
Active: When there are recorded statements of observing clear signs of inflammation at that visit. Examples would be statements that the inflammation is “active”, that anterior chamber or vitreous cells were observed (1+ or higher), the vitreous haze 1+ or higher was observed, etc.
Of note, there was a significant difference in baseline age between the two groups. The median age was 40.3 years (IQR 13.4-57.3) in the TS group and 44.2 years (IQR 29.0-58.9) in the Trab-MMC group. However, the TS group had a larger percentage of eyes of patients under the age of 18 (31.6%) than the Trab-MMC group (14.3%). The median pre-op IOP was 30 mmHg (IQR 21-36) in the TS group and 30.5 mmHg (IQR 20-38) in the Trab-MMC group. In both groups, approximately 95% of patients were on IOP-lowering medications pre-operatively. Anterior uveitis was more common than intermediate, posterior and panuveitis in both surgical groups (52.5% in the TS group and 55.4% in the Trab-MMC group). The three most common associated systemic inflammatory-related conditions included were HLA-B27+ (6.6% of TS group, 10.7% of Trab-MMC group), Juvenile Idiopathic Arthritis (17.6% of TS group, 6.3% of Trab-MMC group) and Sarcoidosis (11.1% of TS group, 10.7% of Trab-MMC group). The inflammation was graded as active in 15.2% in the TS group and 9.8% in the trab-MMC group. There were significantly more patients using oral corticosteroids in the Trab-MMC group (32.1%) compared to the TS group (20.1%).
Incidence of Failure
Incidence of failure was assessed in each group over the first 5 years from the time of TS or Trab-MMC surgery (Figure 1). By one year, 23.5% of the TS group had met failure criteria and 16.1% of Trab-MMC group met failure criteria. By 2 and 3 years, cumulative failure increased to 27.1% and 30.8% in the TS group, and 25.6% and 30.9 % in the Trab-MMC group, respectively. The TS groups had cumulative failure rate of 36.5% by 5 years, and Trab-MMC group failed at 30.9%. In the propensity score adjusted analysis, we found no significant difference in the risk of failure over 5 years between the TS and trab-MMC groups (adjusted hazard ratio (aHR) 1.08, 95% confidence interval (CI) 0.65-1.78, p=0.77) (Table 2). In a sub analysis of the outcomes considering only the first eye surgery, to reduce decision bias, we found no significant difference in the risk of failure over 5 years (aHR 1.15 95% CI 0.69-1.92, p=0.58).
Figure 1.
Kaplan-Meier curves for time to failure of the tube shunt group or trabeculectomy with mitomycin C (Trab-MMC) group, through five years. The numbers under the x-axis are the number of eyes at risk, and in parentheses are the number of failures at each follow-up time point.
Table 2:
Propensity Score-Adjusted Hazard Ratio (aHR) comparing failure of surgery among eyes undergoing tube shunt or trabeculectomy with mitomycin C (aHR>1.0 indicates a higher failure rate for the tube shunt). The time intervals are up to and including defined year. The variables that contributed to propensity score were age, gender, smoking status, anatomic location of uveitis, activity of inflammation and history of cataract surgery.
| Time period | Adjusted failure rate for glaucoma surgery aHR (95% confidence interval) | p value |
|---|---|---|
| 0- 1 year | 1.38 (0.77 – 2.49) | 0.28 |
| 0-2 year | 1.04 (0.61 – 1.77) | 0.88 |
| 0-3 year | 1.05 (0.64 – 1.74) | 0.84 |
| 0-4 year | 1.08 (0.65 – 1.78) | 0.77 |
| 0-5 year | 1.08 (0.65 – 1.78) | 0.77 |
Of the initial reason for failure in both groups, about half were due to IOP > 21 mmHg on 2 consecutive visits at least 90 days apart by 2 years (23 eyes (41.1%) of the TS group, 10 eyes (43.5%) of the Trab-MMC group) and by 5 years (28 eyes (43.1%) of the TS group, 11 eyes (42.3%) of the Trab-MMC group). Reoperation was the reason for initial failure in 22 eyes (39.3%) by 2 years in the TS group and 4 eyes (17.4%) in the Trab-MMC group; by 5 years it was 24 eyes (36.9%) and 4 eyes (15.4%), respectively. There was no significant difference in the reoperation rates between the TS and Trab-MMC groups by 5 years (HR 2.60, 95% CI 0.86 – 7.88, p= 0.09). Of the 4 reoperations in the trab-MMC group, 3 required a bleb needling and 1 was to repair a bleb leak. Of the 24 reoperated eyes in the TS group, 17 were to revise the TS and 7 were to place a second TS. IOP <5 mmHg on 2 consecutive visits 90 days apart was the cause for failure in 8 eyes (14.5%) by 2 years in the TS group and 9 (39.1%) in the Trab-MMC group; by 5 years it was 10 eyes (15.4%) and 11 eyes (42.3%), respectively. Three eyes met failure by having no light perception (NLP) in the TS group by 2 years, and no other eyes reached NLP by 5 years (Table 3).
Table 3:
Reasons for failure of glaucoma surgery by 2 and 5 years.
| Failure within 2 years following glaucoma surgery | Failure within 5 years following glaucoma surgery | |||||
|---|---|---|---|---|---|---|
| Reason for Failure | Tube Shunt | Trab with MMC | All | Tube Shunt | Trab with MMC | All |
| Number of eyes (%) | 56 (100%) | 23 (100%) | 79 (100%) | 65 (100%) | 26 (100%) | 91 (100%) |
| IOP > 21 mmHg | 23 (41.1%) | 10 (43.5%) | 33 (41.8%) | 28 (43.1%) | 11 (42.3%) | 39 (42.9%) |
| IOP < 5 mmHg | 8 (14.3%) | 9 (39.1%) | 17 (21.5%) | 10 (15.4%) | 11 (42.3%) | 21 (23.1%) |
| Reoperation for Glaucoma | 22 (39.3%) | 4 (17.4%) | 26 (32.9%) | 24 (36.9%) | 4 (15.4%) | 28 (30.8%) |
| Vision to NLP | 3 (5.4%) | 0 (0%) | 3 (3.8%) | 3 (4.6%) | 0 (0%) | 3 (3.3%) |
IOP=intraocular pressure; NLP=no light perception; MMC=mitomycin C
Complete and Qualified Success
At 12 months, 23.0% and 50.3% of eyes in the TS group met criteria for complete and qualified success respectively, compared to 46.0% and 35.6% respectively in the Trab-MMC group. At 2 years, the percentage of complete and qualified success dropped to 21.3% and 42.6% for the TS group and 39.1% and 27.5% in the Trab-MMC group, respectively. By three years, complete and qualified success was 18.4% and 33.6% vs. 37.3% and 18.6% for the TS and Trab-MMC groups, respectively (see Figure 3). There was a significant difference between complete success in the TS and Trab-MMC groups at 12, 24 and 36 months (p <0.0001, p < 0.0005, and p < 0.0005, respectively). Table 4 lists the percentage of patients on 1, 2 or 3+ medications in the qualified success group at 12, 24 and 36 months.
Table 4.
Number of medications in eyes with qualified success at 12, 24 and 36 months.
| Tube Shunt | Trab-MMC | |
|---|---|---|
| 12 Months | n=93 | n=31 |
| 1 medication | 4 (4.3%) | 5 (16.1%) |
| 2 medications | 51 (54.8%) | 19 (61.3%) |
| 3 + medications | 38 (40.9%) | 7 (22.6%) |
| 24 Months | n=65 | n=19 |
| 1 medication | 3 (4.6%) | 1 (5.3%) |
| 2 medications | 36 (55.4%) | 13 (68.4%) |
| 3 + medications | 26 (40.0%) | 5 (26.3%) |
| 36 Months | n=41 | n=11 |
| 1 medication | 0 (0%) | 0 (0%) |
| 2 medications | 25 (61.0%) | 6 (54.6%) |
| 3 + medications | 16 (39.0%) | 5 (45.5%) |
Discussion
This study of the incidence of failure of initial incisional glaucoma surgery for uveitic eyes demonstrated a similar incidence of failure with tube shunts and MMC-enhanced trabeculectomies at all time points through 5 years. While overall failure results were similar, there was a greater level of complete success in trabeculectomies at each time point compared to tube shunts.
El-Saied et al conducted a prospective study including 105 eyes in adults with uveitic glaucoma, randomly assigning patients to trabeculectomy with MMC, Ahmed tube shunts and transscleral cyclophotocoagulation (CPC). The failure rates of all three groups were similar at 2 years, with 60% of Trab-MMC eyes having complete success (no medications) and 26% having failure compared with 69% with complete success in the Ahmed group and 26% having failure.17 That was a higher rate of complete success in the Ahmed group than we found in our TS cohort. The study also found more inflammation in the trabeculectomy and CPC groups than the Ahmed tube group. Though no details were included regarding timing of the inflammation in the postoperative course, the authors postulated the increased inflammation in the trabeculectomy group may have been from the iridectomy, and in the CPC group from the inflammation induced on the ciliary body.
Most prior studies reporting surgical outcomes of uveitic glaucoma have been smaller retrospective case series. The rates of failure in the literature are variable, with our outcomes on average demonstrating similar or worse failure rates to other retrospective reviews.18–23 Eyes of patients included in the SITE cohort all were seen at tertiary uveitis centers, likely with more severe and chronic uveitis than eyes of patients in other studies from glaucoma centers. Therefore, it is understandable that our study had a generally higher rate of failure compared to those in the literature. Outcome definitions also varied amongst the studies. Duration of uveitis prior to entry into the cohort and details on the diagnosis, between ocular hypertension and glaucoma with description of visual function such as visual field and retinal nerve fiber thickness are not available in our database, so we are unable to characterize the severity of disease further.
The landmark Tube Versus Trabeculectomy (TVT) study and the Primary Tube Versus Trabeculectomy (PTVT)24,25 study compared failure rates between the two types of glaucoma surgeries but excluded eyes with uveitis. The PTVT , in which patients had not had prior glaucoma or cataract surgeryes, had a cumulative failture rate of 42% in the tube shunt group and 35% in the trabeculectomy group. In the TVT study, where some patients had a prior trabeculectomy or cataract surgery, tube shunts failed at 29.8% and trabeculectomies failed at 46.9%. All the trabeculectomies in both studies were done with MMC, similar to our study. These study results are not generalizable to uveitic glaucoma, but demonstrate glaucoma surgeries in the best conditions fail at a high rate.
We included a sub analysis dividing complete success from qualified success, based on the need for medications after the glaucoma surgery. We were not surprised to see that eyes which received tube shunts required more IOP-lowering medicines (qualified success), while fewer of the eyes that had trabeculectomies were still on medications (complete success). Trabeculectomies commonly achieve lower IOPs with no or few medications than tube shunts as demonstrated by the PTVT study.25 Though this result was significant in our analysis, we did not perform a separate regression analysis so we were unable to account for the differences between the two groups, which may have favored the trab-MMC group.
For our definition of failure, we chose an upper IOP cutoff of 21 mmHg. We recognize that for uveitic glaucoma (often characterized primarily by very high IOP more than optic nerve damage) some eyes at this pressure may be considered “controlled” to a degree that would be satisfactory to a clinician. We acknowledge no particular level of IOP is the target for all uveitic glaucoma or ocular hypertension patients, but for this analysis we chose a conservative threshold to be sure we did not overestimate the success of glaucoma surgery. In addition, we were unable to capture those patients that may have been a failure when they needed to target IOP lower than 21 mm Hg. We recognize this weakness for the retrospective study but since uveitis is typically a progressive glaucoma with high IOPs we feel this was a reasonable approach when we did not know how many patients had ocular hypertension compared to glaucoma.
Our study did not assess cornea-related complications. We know from recent publications, including a publication from our group,13 that corneal decompensation is not an uncommon occurrence with tube shunt surgery. On average, patients with uveitic glaucoma and ocular hypertension are younger than patients needing surgery for POAG, so long-term side effects are paramount and should be studied further.
There are strengths and limitations to this study. The retrospective nature of the report has inherent limitations, but propensity score analysis assists with virtually balancing known confounders. While unknown confounders might exist, they would have to have very large associations to qualitatively change the results of the analysis, making a qualitative difference in the results based on such unknown confounders unlikely. Due to limitation of our dataset, we were unable to include the presence or absence of prophylactic corticosteroid used in the peri-operative period, in the propensity score. We are unable to determine if this was impactful on the inflammatory control and thus the outcomes of our cohort. Another limitation is our inability to segregate the types of tube shunts implanted, as unfortunately these data were not available. It would be clinically useful to be able to comment on the severity of glaucoma in each patient which could influence failure rates, but unfortunately those details are not available in the database.
In summary, ocular hypertension and glaucoma associated with uveitis are difficult to treat, and the failure of glaucoma surgery is high with either TS or Trab-MMC. Our results suggest that tube shunts and trabeculectomies with MMC fail at similarly high rates over 5 years of follow-up, with a more favorable proportion of Trab-MMC eyes achieving complete success. Patients typically require long term continued use of glaucoma medications and monitoring to maintain IOP and prevent vision loss or reoperation, even after surgery. Considering a much higher incidence of corneal endothelial transplant surgery with tube shunt surgeries13 plus better secondary outcomes in our cohort, Trab-MMC might be considered for use more often than tube shunts.
Figure 2.
Complete success, qualified success and failure of tube shunts and trabeculectomy with mitomycin C at 1 year, 2 years and 3 years. Comparison of complete success: 1 year p<0.0001, 2 years p<0.0005, 3 years p <0.0005.
Acknowledgements:
We want to posthumously acknowledge the contributions of Robert B. Nussenblatt, MD, MPH to this work from the Laboratory of Immunology, National Eye Institute, National Institutes of Health, Bethesda, MD.
Financial support:
National Eye Institute/National Institutes of Health grant 1R21 EY032592-01 (Dr. Gangaputra), 1R01 EY014943 (Dr. Kempen) and unrestricted department support by Research to Prevent Blindness (New York, NY). The funding organizations had no role in the design or conduct of this research.
Potential Conflicts of Interest:
John H. Kempen: Betaliq (Equity Owner); Tarsier Pharma (Equity Owner)
C. Stephen Foster: Aldeyra (Consultant, Grant Support), Allakos (Consultant), Bausch & Lomb (Consultant, Grant Support), Eyegate (Consultant, Grant Support, Stock), Genentech (Consultant), Novartis (Consultant, Grant Support), pSivida (Consultant, Grant Support), Aciont (Grant Support), Alcon (Grant Support, Lecture Fees), Clearside (Grant Support), Dompé (Grant Support), Mallinckrodt (Grant Support, Lecture Fees), Allergan (Lecture Fees), Ocular Therapeutix (Food and Beverage), SUN Pharmaceutical Industries (Food and Beverage), Aerie Pharmaceuticals (Food and Beverage), Horizon Therapeutics (Food and Beverage), Eyepoint Pharmaceuticals (Food and Beverage), Eyevance Pharmaceuticals (Food and Beverage), Alimera Sciences (Food and Beverage).
Eric B. Suhler: Genentech/Roche (Consultant, Research support), Acelyrin (Consultant, Research Support), Gilead (Consultant, Research support), Kriya (Consultant), Alumis (Consultant), Priovant (Research support), Abbvie (Global steering committee).
Jennifer E. Thorne: Canfield (Consultant); Clearside (Consultant); Gilead (Consultant); Guidepoint (Consultant); Roche (Consultant); Sanofi (consultant); Tarsier Pharma (Scientific Advisory Board; Equity Owner); UptoDate (Consultant)
Douglas A. Jabs: Abbvie (Research Funding)
James T. Rosenbaum: Corvus Pharmaceuticals (Employment); Abbvie (Consultant); Gilead (Consultant); Janssen (Consultant); Eyevensys (Consultant); UpToDate (Consultant); Pfizer (Financial Support); Novartis (Consultant); Roche (Consultant); Alcon Research Institute (Financial Support); UCB (consultant); Horizon (Consultant and financial support); Santen (consultant); Celgene (Consultant); Pfizer (Education); UCB (Consultant).
H. Nida Sen: Sanofi Research and Development (Employee)
Grace A. Levy-Clarke: Abbvie (Consultant, Lecture Fees); Allergan (Grant Support); Mallinckrodt (Consultant, Grant Support); Sanofi (Grant Support; Lecture Fees)
R. Oktay Kaçmaz: Santen (Employee)
The other authors indicate no potential conflicts of interest.
Footnotes
Meeting: This manuscript was presented at the 2023 American Glaucoma Society, Austin, TX
The purpose of this study is to look at the outcomes of trabeculectomy and tube shunt surgery in a large cohort of patients with uveitis.
References:
- 1.Quigley HA, Broman AT. The number of people with glaucoma worldwide in 2010 and 2020. Br J Ophthalmol 2006; 90:262–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Tham YC, Li X, Wong TY, Quigley HA, Aung T, Cheng CY. Global prevalence of glaucoma and projections of glaucoma burden through 2040. A Systematic review and meta-analysis. Ophthalmology 2014;121:2081–2090. [DOI] [PubMed] [Google Scholar]
- 3.Kempen JH, Van Natta ML, Friedman DS, Altaweel MM, Ansari H, Dunn JP, Elner SG, Holbrook JT, Lim LL, Sugar EA, Jabs DA, Multicenter Uveitis Steroid Treatment (MUST) Trial and follow-up study research group. Incidence and outcomes of uveitic glaucoma in eyes with intermediate, posterior, or panuveitis followed up to 10 years after randomization to fluocinolone acetonide implant or systemic therapy. Am J Ophthalmol. 2020;219:303–316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Merayo-Lloves J, Power WJ, Rodriguez A et al. Secondary glaucoma in patients with uveitis. Ophthalmologica 1999; 213:300–304. [DOI] [PubMed] [Google Scholar]
- 5.Takahashi T, Ohtani S, Miyata K et al. A clinical evaluation of uveitis-associated secondary glaucoma. Jpn J Ophthalmol 2002; 46:556–562. [DOI] [PubMed] [Google Scholar]
- 6.Moorthy RS, Mermoud A, Baerveldt G, et al. Glaucoma associated with uveitis. Surv Ophthalmol 1997; 41:361 – 394. [DOI] [PubMed] [Google Scholar]
- 7.Sallam A, Sheth HG, Habot-Wilner Z, Lightman S. Outcome of raised intraocular pressure in uveitic eyes with and without a corticosteroid-induced hypertensive response. Am J Ophthalmol. 2009;148(2):207–213.e1. [DOI] [PubMed] [Google Scholar]
- 8.Friedman DS, Holbrook JT, Ansari H, Alexander J, Burke A, Reed SB, Katz J, Thorne JE, Lightman SL, Kempen JH; MUST Research Group. Risk of elevated intraocular pressure and glaucoma in patients with uveitis: results of the multicenter uveitis steroid treatment trial. Ophthalmology. 2013;120:1571–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Liesenborghs I, Eijssen LMT, Kutmon M, Gorgels TGMF, Evelo CT, Beckers HJM, Webers CAB, Schouten JSAG. The Molecular Processes in the Trabecular Meshwork After Exposure to Corticosteroids and in Corticosteroid-Induced Ocular Hypertension. Invest Ophthalmol Vis Sci. 2020. Apr 9;61(4):24. doi: 10.1167/iovs.61.4.24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Writing Committee for the Multicenter Uveitis Steroid Treatment (MUST) Trial and Follow-up Study Research Group, Kempen JH, Altaweel MM, et al. Association Between Long-Lasting Intravitreous Fluocinolone Acetonide Implant vs Systemic Anti-inflammatory Therapy and Visual Acuity at 7 Years Among Patients With Intermediate, Posterior, or Panuveitis. JAMA. 2017;317(19):1993–2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kaburaki T, Koshino T, Kawashima H, Numaga J, Tomidokoro A, Shirato S, Araie M. Initial trabeculectomy with mitomycin C in eyes with uveitic glaucoma with inactive uveitis. Eye (Lond). 2009. Jul;23(7):1509–17. [DOI] [PubMed] [Google Scholar]
- 12.Chow A, Burkemper B, Varma R, Rodger DC, Rao N, Richter GM. Comparison of Surgical Outcomes of Trabeculectomy, Ahmed Shunt, and Baerveldt Shunt in Uveitic Glaucoma. J Ophthalmic Inflamm Infect. 2018; 8:9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Roldan AM, Zebardast N, Pistilli M, Khachatryan N, Payal A, Begum H, Artornsombudh P, Pujari SS, Rosenbaum JT, Sen HN, Suhler EB, Thorne JE, Bhatt NP, Foster CS, Jabs DA, Levy-Clarke GA, Nussenblatt RB, Buchanich JM, Kempen JH; Systemic Immunosuppressive Therapy for Eye Diseases (SITE) Cohort Study Research Group. Corneal Endothelial Transplantation in Uveitis: Incidence and Risk Factors. Am J Ophthalmol. 2022. Apr;236:288–297. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Kempen JH, et al. Methods for identifying long-term adverse effects of treatment in patients with eye diseases: the Systemic immunosuppressive Therapy for Eye Diseases (SITE) Cohort Study. Ophthalmic Epidemiol. 2008; 15:47–55. [DOI] [PubMed] [Google Scholar]
- 15.Kempen J, Daniel E, Dunn JP et al. Overall and cancer related mortality among patients with ocular inflammation treated with immunosuppressive drugs: retrospective cohort study. BMJ. 2009; 339:b2480. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kempen JH, Newcomb CW, Washington TL, Foster CS, Sobrin L, Thorne JE, Jabs DA, Suhler EB, Rosenbaum JT, Sen HN, Levy-Clarke GA, Nussenblatt RB, Bhatt NP, Lowder CY, Goldstein DA, Leiderman YI, Acharya NR, Holland GN, Read DW, Dunn JP, Dreger KA, Artornsombdh P, Begum HA, Fitzgerald TD, Kothari S, Payal AR, Daniel E, Gangaputra SS, Kacmaz RO, Liesegang TL, Pujari SS, Khachatryan N, Maghsoudlou A, Suga HK, Pak CM, Helzlsouer KJ, Buchanich JM, Systemic Immunosuppressive Therapy for Eye Diseases Cohort Study Research Group. Ophthalmology. 2023;130:1258–68. [Google Scholar]
- 17.El-Saied HMA, Abdelhakim MASE. Different surgical modalities for management of uveitic glaucoma: 2 year comparative study. Acta Ophthalmol. 2022. 100:e246–e252. [DOI] [PubMed] [Google Scholar]
- 18.Papadaki TG, Zacharopoulos IP, Pasquale LR, Christen WB, Netland PA, Foster CS. Long-term results of ahmed glaucoma valve implantation for uveitic glaucoma. Am J Ophthalmol 2007; 144:62–69. [DOI] [PubMed] [Google Scholar]
- 19.Audrey C, Burkemper B, Varma R, Rodger DC, Rao N, Richter GM. Comparison of Surgical Outcomes of Trabeculectomy, Ahmed Shunt, and Baerveldt Shunt in Uveitic Glaucoma. J Ophthalmic Inflamm Infect. 2018; 8:9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Bettis DI, Morshedi RG, Chaya C, Goldsmith J, Crandall A, Zabriskie N. Trabeculectomy with Mitomycin C or Ahmed valve implantation in eyes with uveitic glaucoma. J Glaucoma 2015;24:591–599. [DOI] [PubMed] [Google Scholar]
- 21.Shimizu A, Maruyama K, Yokoyama Y, Tsuda S, Ryu M, Nakazawa T. Characteristics of uveitic glaucoma and evaluation of its surgical treatment. Clin Ophthalmol. 2014;8:2383–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Ramdas WD, Pals J, Rothova A, Wolfs RCW. Efficacy of glaucoma drainage devices in uveitic glaucoma and a meta-analysis of the literature. Graefes Arch Clin Exp Ophthalmol, 2019; 257:143–151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Magliyah MS, Badawi AH, Alshamrani AA, Malik R, Al-Dhibi H. The effect of perioperative uveitis control on the success of glaucoma surgery in uveitic glaucoma. Clin Ophthalmol. 2021. 15:1465–1475. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Gedde SJ, Schiffman JC, Feuer WJ, et al. The tube versus trabeculectomy study: design and baseline characteristics of study patients. Am J Ophthalmol. 2005;140(2):275–287. [DOI] [PubMed] [Google Scholar]
- 25.Gedde SJ, Feuer WJ, Lim KS, Barton K, Goyal S, Ahmed II, Brandt JD, for the Primary Tube Versus Trabeculectomy Study Group. Ophthalmology. 2022;129:1344–1356. [DOI] [PMC free article] [PubMed] [Google Scholar]