You are on page 1of 6

Anti-inammatory Therapy after Selective

Laser Trabeculoplasty
A Randomized, Double-Masked, Placebo-Controlled
Clinical Trial
Delan Jinapriya, MD,1 Mark DSouza, MD,1 Hussein Hollands, MD, MSc,1 Sherif R. El-Defrawy, MD, PhD,1
Isabella Irrcher, PhD,1 Donald Smallman, MD,1 James P. Farmer, MD,1 John Cheung, MD,1 Todd Urton, MD,1,2
Andrew Day, MSc,2 Xiaoquin Sun, MSc,2 Robert J. Campbell, MD, MSc1,3

Purpose: To investigate the effect of anti-inammatory therapy on selective laser trabeculoplasty (SLT)
outcomes.
Design: Randomized, double-masked, placebo-controlled trial.
Participants: Patients with primary open-angle or pseudo-exfoliation glaucoma.
Methods: Patients undergoing SLT were randomized to receive placebo (articial tears), prednisolone
acetate 1%, or ketorolac tromethamine 0.5% eye drops 4 times per day for 5 days commencing immediately
after SLT.
Main Outcome Measures: Change in intraocular pressure (IOP) from baseline to the 1-month post-SLT visit.
Results: Mean change in IOP at the 1-month primary outcome time point, as well as all other time points, was
not signicantly different among groups (P 0.99). Likewise, a repeated-measures, mixed-effects model did not
nd signicant differences in IOP outcome at the 1-month time point (P 0.95). The IOP was reduced in all
groups at the 1-month post-SLT time point and all other time points, and no signicant differences were found
between groups using separate unadjusted cross-sectional analyses of variance (P > 0.15 for analyses at all time
points). Treatment failure rates were not different among groups (P 0.75), and at 1 year after SLT, the per-
centage of patients maintaining a 20% IOP reduction ranged from 18% to 22% in the 3 study groups.
Conclusions: Anti-inammatory therapy after SLT does not seem to substantially inuence the IOP-lowering
effect of SLT. In this study of patients with low baseline IOP, SLT showed limited efcacy in achieving a sustained
reduction in IOP. Ophthalmology 2014;121:2356-2361 2014 by the American Academy of Ophthalmology.

Selective laser trabeculoplasty (SLT) effectively lowers complications, the degree of inammation seen after SLT is
intraocular pressure (IOP) with minimal risk of complica- minimal. In light of the likelihood that inammatory mediators
tions and is a widely used and effective treatment for are important in the positive clinical effect of SLT, high-level
glaucoma.1e5 Despite its widespread use, the mechanisms evidence is needed to inform the use of anti-inammatory
by which SLT induces an IOP-lowering effect remain only medication after SLT.1,6,8 However, to date, only limited data
partially understood.1,3e8 have been reported from randomized trials and observational
The IOP-lowering effect of SLT is premised on selective studies have been inconclusive.3,6,7,11,12 Indeed, a recent
photothermolysis, in which laser interacts with pigmented American Academy of Ophthalmology Ophthalmic Tech-
components of trabecular meshwork cells to facilitate aqueous nology Assessment identied the need for research into
outow.9 In particular, SLT induces the release of chemotactic pharmacologic methods that could enhance the response to
and vasoactive agents, including the inammatory mediators trabeculoplasty.13 Thus, we conducted a randomized, double-
interleukin-1a, interleukin-1b, and tumor necrosis factor-a.1 masked, placebo-controlled trial to determine whether post-
These mediators have numerous effects, including macro- laser anti-inammatory therapy modies the effect of SLT.
phage recruitment, matrix metalloproteinaseeinduced extra-
cellular matrix remodeling, and improved permeability of
Schlemms canal endothelial cells.1,8,10 Methods
Many studies of SLT have described the routine use of anti-
inammatory ophthalmic medications for a short period after Study Design and Overview
laser treatment.1,4e7 However, the utility of this practice has This randomized, double-masked, placebo-controlled trial was
not been validated. Although high levels of intraocular conducted at the Hotel Dieu Hospital in Kingston, Ontario,
inammation over extended periods of time can cause Canada, and adhered to the tenets of the Declaration of Helsinki.

2356  2014 by the American Academy of Ophthalmology http://dx.doi.org/10.1016/j.ophtha.2014.07.017


Published by Elsevier Inc. ISSN 0161-6420/14
Jinapriya et al 
Anti-inammatory Therapy after SLT

Enrollment Randomized (n = 125)

Tears Prednisolone acetate Ketorolac tromethamine


(placebo) (steroid) (NSAID)
Allocation Received allocated Received allocated Received allocated
intervention (n = 38) intervention (n = 46) intervention (n = 41)

Not available for 1 month Not available for 1 month Not available for 1 month
Follow-Up at 1
follow-up (n = 1) follow-up (n = 3) follow-up (n = 4)
Month

Total included in primary Total included in primary Total included in primary


Analysis
outcome end point outcome end point outcome end point
(n = 37) (n = 43) (n = 37)

Figure 1. Consolidated Standards of Reporting Trials diagram. NSAID nonsteroidal anti-inammatory drug.

All patients provided written, informed consent before any study- or data acquisition, were used to assign patients to treatment arms
related procedures. The Queens University and Afliated Teach- and to assign 1 eye to be enrolled if both eyes were undergoing
ing Hospitals Health Sciences Research Ethics Board approved the laser treatment. Commercially available medication bottles were
study before enrolling study participants. The study is listed on covered with opaque labels over the entire bottle and labeled only
ClinicalTrials.gov, under the identier NCT00485108. with instructions on how often to administer drops.

Study Population and Inclusion/Exclusion Criteria


Selective Laser Trabeculoplasty Treatment
Patients were recruited from study ophthalmologists practices Procedure, Baseline Examinations, and Follow-Up
within the Department of Ophthalmology at Queens University,
Kingston, Ontario, Canada. Eligible patients were aged 18 years or In all cases, the SLT procedure consisted of 50 pulses to the
older with a diagnosis of primary open-angle glaucoma (POAG) or inferior 180 of the trabecular meshwork. Power was initially set at
pseudo-exfoliation glaucoma (PEXG), without previous laser tra- 0.8 mJ and titrated to produce occasional small bubbles. All
beculoplasty. Patients receiving SLT as initial glaucoma therapy or patients were treated with a single brimonidine drop before laser.
as an adjunctive therapy were included. In addition, to be eligible, Patients were assessed twice on separate days before SLT and
patients had to be capable of providing consent and returning for follow-up examinations scheduled at 1 hour, 1 day, 1 week, 1
follow-up visits. Only 1 eye per patient was enrolled in the study. month, 3 months, 6 months, and 1 year after SLT.
Patients were excluded if they had a history of uveitis or previous
incisional glaucoma surgery.
Examination Procedures
Random Treatment Assignment The IOP measurement protocol followed the methods described
Patients undergoing SLT were assigned to 1 of 3 post-laser treat- in the Ocular Hypertension Treatment Study, and calibrated
ment arms in blocks of 6 to maintain balanced groups: (1) car- tonometers were used throughout the study.10 Measurements at
boxymethyl cellulose sodium 1.5% articial tears (placebo group) all study time points were taken by an ophthalmologist or a
1 drop 4 times daily for 5 days after SLT; (2) prednisolone acetate senior resident involved with the study. All follow-up IOP
1% (steroid group) 1 drop 4 times daily for 5 days after SLT; or (3) measures were the average of 2 measurements taken
ketorolac tromethamine 0.5% (nonsteroidal anti-inammatory drug sequentially at each visit. Baseline IOP was dened as the
[NSAID] group) 1 drop 4 times daily for 5 days after SLT (Fig 1). average of the 4 measurements taken on 2 separate days before
Randomization was stratied by (1) surgeon; (2) glaucoma subtype laser treatment.
(POAG or PEXG); and (3) whether SLT was the initial glaucoma Anterior chamber inammation was graded by a study
therapy or an adjunctive treatment. ophthalmologist or study senior resident. Assessments were made
Sequentially numbered sealed envelopes, prepared by an on the basis of the Standardization of Uveitis Nomenclature
unmasked study coordinator who was not involved with treatment working group grading scheme for anterior chamber cells.14

2357
Ophthalmology Volume 121, Number 12, December 2014

Table 1. Patient Demographics and Baseline Characteristics

Articial Tears (Placebo) Prednisolone Acetate (Steroid) Ketorolac Tromethamine (NSAID)


n [ 38 n [ 46 n [ 41
Age (yrs) 71.911.5 69.310.1 70.611.6
Female sex (%) 50 50 63
Baseline IOP (mmHg) 19.14.6 19.64.6 18.53.5
Study eye diagnosis (%)
POAG 87 70 83
PEXG 13 30 17
No. of medications at baseline 1.41.4 1.51.3 1.61.3
Timing of SLT (%)
Initial therapy 34 43 26
Adjunctive therapy 66 57 73

IOP intraocular pressure; NSAID nonsteroidal anti-inammatory drug; PEXG pseudoexfoliation glaucoma; POAG primary open-angle glaucoma;
SLT selective laser trabeculoplasty.
Data are reported as mean  standard deviation unless otherwise specied.

Outcome Measures was not at least 20% below baseline or (2) requiring additional
IOP-lowering therapy. Cause-specic failure rates were estimated
The primary study outcome was the absolute change in IOP from by the life table method. The log-rank test was then used to test for
baseline to the 1-month time point after SLT. Additional therapies differences in all-cause treatment failure rates. All statistical ana-
were not permitted before this follow-up time point. In a secondary lyses were carried out using SAS version 9.2 (SAS Inc., Cary, NC).
analysis, IOP changes from baseline at other time points also were
compared between treatment arms.
Time to treatment failure, dened as not maintaining a 20% IOP Results
reduction from baseline or requiring additional glaucoma therapy
(at the treating ophthalmologists discretion), was assessed as a In total, 125 patients were randomized to placebo (n 38), steroid
secondary outcome. Finally, anterior chamber inammation was (n 46), or NSAID (n 41). The Consolidated Standards of
graded according to the Standardization of Uveitis Nomenclature Reporting Trials ow diagram illustrating patient allocation to
grading scheme.14 groups is presented in Figure 1. Baseline patient characteristics
were well balanced (Table 1). Mean age across all 3 groups was
Sample Determination and Power 70.5 years (standard deviation, 11 years). Mean baseline IOP
across all groups was 19.1 mmHg (standard deviation, 4.3 mmHg).
By assuming that our primary outcome (absolute change in IOP
from baseline to the 1-month post-laser assessment) had a within- Intraocular Pressure
group standard deviation of 4 mmHg, using the F-test from anal-
ysis of variance (ANOVA), 40 patients per arm provide at least Mean change in IOP at the 1-month visit, the primary outcome,
85% power at alpha 0.05 (2-sided) to reject the null hypothesis was not signicantly different among groups (P 0.99). At all
that all 3 arms had the same mean change if the actual change in other time points, mean change in IOP was not signicantly
IOP differed by at least 3 mmHg between the 2 most extreme arms. different among groups (P > 0.1). The adjusted between-arm
differences in IOP lowering from baseline are shown in Figure 2.
Statistical Analysis After adjusting for baseline IOP and the study stratication factors
(physician, diagnosis, and SLT as initial vs. adjunctive treatment),
The difference between arms in IOP change from baseline to the repeated measures mixed-effects model did not show any sig-
follow-up time points was estimated by a linear mixed-effects nicant differences between groups at any time point. An analysis
model for longitudinal data.15 The model included the baseline IOP based on percentage of IOP reduction from baseline showed
measurement and the trials 3 stratication factors (physician, analogous results (not shown). Boxplots of raw IOP measurements
diagnosis, and initial vs. adjunctive treatment) as xed covariates. from each group over time are shown in Figure 3. These data show
The estimated adjusted between-group differences were then that IOP was reduced in all groups at the 1-month post-SLT time
plotted with 95% condence intervals (CIs). The TukeyeKramer point and all other time points, with no signicant differences
adjustment was applied to the CIs to account for the 3 pairwise between groups found with separate unadjusted cross-sectional
between-group comparisons at each time point.16 The model used ANOVA analyses (P > 0.15 for analyses at all time points).
an unstructured covariance to account for within-subject correla-
tion between the 5 post-baseline assessments.17 The conclusions Time to Treatment Failure
were conrmed by a simple unadjusted cross-sectional analysis
comparing the mean and variance between arms separately at each The cumulative survival plot for treatment success is shown in
time by a 1-way ANOVA with Levenes test for homogeneity Figure 4. The treatment failure rates were not signicantly different
variance. The ManteleHaenszel mean test score was used to between treatment arms (P 0.75). At the primary end point of 1
compare our secondary outcome, anterior chamber inammation, month, the treatment failure rates in the placebo, steroid, and
between arms at 1 hour, 2 days, and 1 month after SLT. NSAID arms were 59% (95% CI, 44e75), 44% (95% CI, 31e60),
In the secondary analysis examining time to treatment failure, and 51% (95% CI, 37e67), respectively. One patient who failed
failure was dened a priori as (1) an IOP at month 1, 3, 6, or 12 that required trabeculectomy surgery during follow-up (NSAID group).

2358
Jinapriya et al 
Anti-inammatory Therapy after SLT

Figure 2. Adjusted between-arm differences in intraocular pressure (IOP) reduction from baseline. The expected mean difference is represented by the
diamond. The boxes represent nominal 95% condence interval (CI). The whiskers extend to the TukeyeKramer adjusted simultaneous 95% CI accounting
for the 3 between-arm pairwise comparisons. P values (p) represent the between-group comparisons at each time point. D2 day 2 after selective laser
trabeculoplasty (SLT); H1 hour 1 after SLT; M1 month 1 after SLT; M3 month 3 after SLT; M6 month 6 after SLT; NSAID nonsteroidal anti-
inammatory drug; Y1 year 1 after SLT.

Anterior Chamber Inammation In our study, SLT had a smaller IOP-lowering effect than
that observed in some previous reports.3,7,18e20 Moreover,
Mean anterior chamber inammation scores at 1 hour, 2 days, and initial reductions in IOP were often not sustained. Conse-
1 month after SLT treatment were not signicantly different be-
tween any of the treatment arms (P > 0.2 for all comparisons). By quently, by the 1-year post-SLT time point, approximately
1 month after SLT, anterior chamber inammation was observed in 80% of patients in all groups had failed to maintain a 20%
only 1 patient. reduction in IOP. This low success rate may be a result of
the low baseline IOP in our study, which strongly inuences
SLT success.4,12 Indeed, the efcacy of SLT in our study
Discussion was similar to that observed by Song et al,21 who also
studied a population with relatively low baseline IOPs.
Despite the widespread use of anti-inammatory medication Nevertheless, baseline IOP effects were nondifferential and
after SLT, high-level data supporting this practice have been would not be expected to inuence our comparisons be-
lacking. In this randomized clinical trial, we found that the tween groups.
use of anti-inammatory therapy after SLT did not inuence
the IOP-lowering effect of SLT when compared with pla- Study Limitations
cebo. This nding was consistent among patients with both
POAG and PEXG and among patients receiving SLT as To our knowledge, our study is the largest double-masked,
initial or adjunctive glaucoma therapy. placebo-controlled trial to evaluate this clinically important
Although previous observational studies that included question. The use of a clinically relevant primary outcome
groups receiving different post-laser anti-inammatory ther- chosen a priori is an important strength of our study. Stratied
apy were not designed to address the question of efcacy, they randomization and standardized examination and assessment
do provide some information.1,3,6,12 In particular, McIlraith procedures were additional strengths. Our study has limitations
et al6 found no difference in IOP reduction between patients that warrant mention. First, a larger study could provide more
receiving ketorolac and prednisone after SLT. Realini et al11 precise measures of effect. Second, although subgroup results
randomized eyes of patients undergoing bilateral SLT to suggest that our ndings apply to both POAG and PEXG, and
prednisolone acetate versus no therapy. Although this study to SLT as an initial and adjunctive therapy, the subgroup data
did not detect differences in outcome between treated and should be interpreted with caution because of the limited power
untreated eyes, the study included only 25 patients.6,7,10 of our study to detect clinically important differences in

2359
Ophthalmology Volume 121, Number 12, December 2014

Figure 3. The intraocular pressure (IOP) (mmHg) over time by treatment arm. Boxes represent the interquartile range (i.e., middle half) of the data.
Lines and diamonds inside the boxes represent the median and mean, respectively. Whiskers extend to the most extreme values up to 1.5 times the
interquartile range beyond the interquartile range. Circles are used to plot values beyond whiskers. P values (p) are based on separate unadjusted 1-way
analyses of variance comparing means at each time point. B1 baseline 1; B2 baseline 2; D2 day 2 after selective laser trabeculoplasty (SLT); H1
hour 1 after SLT; M1 month 1 after SLT; M3 month 3 after SLT; M6 month 6 after SLT; NSAID nonsteroidal anti-inammatory drug; Y1
year 1 after SLT.

subgroup analyses. Third, it is possible that patients could assessment carried out by an investigator who had been present
determine the medication received from the cap color or at the initial SLT. Thus, it is possible that the investigator could
medication consistency. However, few patients would be ex- have recalled the treatment assignment. Finally, in our study
pected to be aware of the specic differences. Of note, study we evaluated 180 SLT treatments only, and the generaliz-
evaluators did not see the medication bottles at follow-up visits. ability to 360 treatments is not known.
However, some patients may have had their follow-up visit In conclusion, our ndings suggest that the therapeutic
mechanism of SLT may be less dependent on inammatory
1
mediators than some studies have postulated.1,8 However,
although we studied commonly used doses of widely used
anti-inammatory medication, our results may not be
0.8 generalizable to other drugs and alternative doses. This
should be the focus of future research. Further studies will
be needed to assess the impact of anti-inammatory treat-
ment on SLT responses in subtypes of glaucoma other than
Proportion Surviving

0.6

Placebo
POAG and PEXG, such as pigment dispersion glaucoma.
Steroid The results of this randomized, double-masked, placebo-
0.4 NSAID controlled study suggest that the IOP-lowering effect of SLT
is not substantially inuenced by the use of post-laser anti-
inammatory medications.
0.2

References
0
0 3 6 9 12
Time (months)
1. Latina MA, de Leon JM. Selective laser trabeculoplasty.
Figure 4. The selective laser trabeculoplasty (SLT) success survival curves. Ophthalmol Clin North Am 2005;18:40919. vi.
P 0.75 by log-rank test comparing time to treatment failure between 2. Campbell RJ, Bell CM, Gill SS, et al. Subspecialization in
groups. NSAID nonsteroidal anti-inammatory drug. glaucoma surgery. Ophthalmology 2012;119:22703.

2360
Jinapriya et al 
Anti-inammatory Therapy after SLT

3. Nagar M, Ogunyomade A, OBrart DP, et al. A randomised, 12. Mao AJ, Pan XJ, McIlraith I, et al. Development of a pre-
prospective study comparing selective laser trabeculoplasty diction rule to estimate the probability of acceptable intraoc-
with latanoprost for the control of intraocular pressure in ular pressure reduction after selective laser trabeculoplasty in
ocular hypertension and open angle glaucoma. Br J Oph- open-angle glaucoma and ocular hypertension. J Glaucoma
thalmol 2005;89:14137. 2008;17:44954.
4. Hodge WG, Damji KF, Rock W, et al. Baseline IOP predicts 13. Samples JR, Singh K, Lin SC, et al. Laser trabeculoplasty
selective laser trabeculoplasty success at 1 year post-treatment: for open-angle glaucoma: a report by the American
results from a randomised clinical trial. Br J Ophthalmol Academy of Ophthalmology. Ophthalmology 2011;118:
2005;89:115760. 2296302.
5. Melamed S, Ben Simon GJ, Levkovitch-Verbin H. Selective 14. Standardization of Uveitis Nomenclature (SUN) Working
laser trabeculoplasty as primary treatment for open-angle Group. Standardization of uveitis nomenclature for reporting
glaucoma: a prospective, nonrandomized pilot study. Arch clinical data. results of the First International Workshop. Am J
Ophthalmol 2003;121:95760. Ophthalmol 2005;140:50916.
6. McIlraith I, Strasfeld M, Colev G, Hutnik CM. Selective laser 15. Verbeke G, Molenberghs G. A Model for Longitudinal
trabeculoplasty as initial and adjunctive treatment for open- Data. In: Linear Mixed Models for Longitudinal Data.
angle glaucoma. J Glaucoma 2006;15:12430. New York: Springer-Verlag; 2000:239.
7. Latina MA, Sibayan SA, Shin DH, et al. Q-switched 532-nm 16. Kramer CY. Extension of multiple range tests to group means
Nd:YAG laser trabeculoplasty (selective laser trabeculo- with unequal numbers of replications. Biometrics 1956;12:
plasty): a multicenter, pilot, clinical study. Ophthalmology 30910.
1998;105:208290. 17. Littell RC, Pendergast J, Natarajan R. Modelling covariance
8. Alvarado JA, Alvarado RG, Yeh RF, et al. A new insight into structure in the analysis of repeated measures data. Stat Med
the cellular regulation of aqueous outow: how trabecular 2000;19:1793819.
meshwork endothelial cells drive a mechanism that regulates 18. Lanzetta P, Menchini U, Virgili G. Immediate intraocular
the permeability of Schlemms canal endothelial cells. Br J pressure response to selective laser trabeculoplasty. Br J
Ophthalmol 2005;89:15005. Ophthalmol 1999;83:2932.
9. Ayala M, Chen E. The inuence of topical prostaglandin an- 19. Gracner T. Intraocular pressure response to selective laser
alogues in inammation after selective laser trabeculoplasty trabeculoplasty in the treatment of primary open-angle glau-
treatment. J Ocul Pharmacol Ther 2012;28:11822. coma. Ophthalmologica 2001;215:26770.
10. Stein JD, Challa P. Mechanisms of action and efcacy of 20. Damji KF, Bovell AM, Hodge WG, et al. Selective laser tra-
argon laser trabeculoplasty and selective laser trabeculoplasty. beculoplasty versus argon laser trabeculoplasty: results from a
Curr Opin Ophthalmol 2007;18:1405. 1-year randomised clinical trial. Br J Ophthalmol 2006;90:
11. Realini T, Charlton J, Hettlinger M. The impact of anti-in- 14904.
ammatory therapy on intraocular pressure reduction 21. Song J, Lee PP, Epstein DL, et al. High failure rate associated
following selective laser trabeculoplasty. Ophthalmic Surg with 180 degrees selective laser trabeculoplasty. J Glaucoma
Lasers Imaging 2010;41:1003. 2005;14:4008.

Footnotes and Financial Disclosures


Originally received: November 30, 2013. This study was funded by the Glaucoma Research Society of Canada. The
Final revision: February 3, 2014. sponsors of this study had no role in the design and conduct of the study;
Accepted: July 9, 2014. collection, management, analysis, and interpretation of the data; prepara-
Available online: September 15, 2014. Manuscript no. 2013-1972. tion, review, or approval of the manuscript; and the decision to submit for
1
Department of Ophthalmology, Queens University and Hotel Dieu publication. The opinions, results, and conclusions reported in this article
are those of the authors and are independent from the funding sources.
Hospital, Kingston, Ontario, Canada.
2 Abbreviations and Acronyms:
Clinical Evaluation Research Unit, Kingston General Hospital, Kingston,
Ontario, Canada. ANOVA analysis of variance; CI condence interval;
IOP intraocular pressure; NSAID nonsteroidal anti-inammatory
3
Institute for Clinical Evaluative Sciences, Ontario, Canada. drug; PEXG pseudo-exfoliation glaucoma; POAG primary open-
Financial Disclosure(s): angle glaucoma; SLT selective laser trabeculoplasty.
The author(s) have made the following disclosure(s): D.J.: received Correspondence:
payment for CME talks for Allergan, Alcon, and Pzer. H.H.: received Robert J. Campbell, MD, MSc, Department of Ophthalmology, Hotel Dieu
payment for the development of educational presentations from Novartis.
Hospital, 166 Brock Street, Kingston, Ontario, Canada, K7L 5G2. E-mail:
D.S.: consultant to Tear Science. rob.campbell@queensu.ca.

2361