Selective Laser Trabeculoplasty versus

Argon Laser Trabeculoplasty in Glaucoma

Patients Treated Previously with 360

Selective Laser Trabeculoplasty
A Randomized, Single-Blind, Equivalence Clinical Trial
Cindy Hutnik,1 Andrew Crichton,2 Bryce Ford,2 Marcelo Nicolela,3 Lesya Shuba,3 Catherine Birt,4 Enitan Sogbesan,5
Karim Damji,6 Michael Dorey,6 Hady Saheb,7 Neil Klar,1 Hui Guo,1 William Hodge, MD, PhD1

Purpose: The effectiveness of selective laser trabeculoplasty (SLT) was compared with argon laser trabe-
culoplasty (ALT) in a randomized clinical trial for patients with medically uncontrolled open-angle glaucoma who
have previously received 360
SLT.
Design: An active equivalence parallel armed randomized control trial.
Participants: Patients with open-angle glaucoma including pigmentary dispersion syndrome and pseu-
doexfoliation syndrome were enrolled into the study from 7 different sites across Canada.
Methods: One setting of 180
of either SLT or ALT was assigned randomly and applied to each participant.
Main Outcome Measures: The change in intraocular pressure (IOP) from baseline to 12 months was
compared between the 2 groups.
Results: A total of 132 patients were recruited, 2 of which dropped out early in the study, leaving 130 patients
who completed the study as per protocol. For those, the study’s primary outcome was calculated. The IOP
change at 1 year in comparison to baseline for SLT vs. ALT was found to be different by 0.33 mmHg between the
2 groups (3.16 for SLT and 2.83 for ALT) and was not statistically significant (P ¼ 0.71) Further analysis, though,
showed that SLT had a significantly lower IOP reduction at early time points: 1 week and 1 month, but this effect
was lost by 3 months. Corresponding to this finding was the strong trend for ALT to fail more quickly than SLT.
Although repeatable, the first repeat SLT reduced IOP to only about half compared with initial SLT treatment.
Conclusions: The comparison at 12 months following the laser therapy showed that both modalities lowered
the IOP with approximately 3 mmHg, yet essentially all of the time-to-failure analyses favored SLT over ALT. The
repeat SLT effect was found to be half of the initial treatment. Ophthalmology 2018;-:1e10 Crown
Copyright ª 2018 Published by Elsevier Inc. on behalf of the American Academy of Ophthalmology

Argon laser trabeculoplasty (ALT) was the standard-of-care
method of laser therapy to the trabecular meshwork (TM)
for open-angle glaucoma until 2005.1,2 Argon laser trabe-
culoplasty is effective, but its most significant problem is
that its effectiveness decreases with retreatment because the
tissue it targets (the TM) is changed by the laser, rendering
repeat treatments less effective.3 Subsequent to its
introduction in 2005, selective laser trabeculoplasty (SLT)
has emerged as the more popular trabeculoplasty laser.
There are many potential advantages to SLT, but to date,
these advantages are not supported with the best evidence.
The most important potential clinical advantage of SLT is
based on the premise that it causes less damage to the
tissue it targets.1 The nanosecond pulse duration of the
SLT is a key factor in preventing collateral thermal
damage because it is less than the thermal relaxation time
of the melanin chromophore. The absence of histologic or
architectural change in SLT-treated eyes has the theoret-
ical advantage of allowing effective repeated laser treat-
ments, which was not possible with ALT. Indeed, if correct,
this would be a very significant clinical advantage of SLT
and is the focus of this randomized clinical trial (RCT).
The first phase 1/2 results from treatment of patients with
SLT were reported in 1998 and established safety and ef-
ficacy of the procedure.4 The first small RCT that studied
SLT vs. ALT was published in 1999 by some of our
authors.5 This was a small trial that found and showed
equivalence in efficacy of the 2 lasers. The definitive RCT
confirming efficacy equivalence was published in 2006 by
the same authors.6 Baseline variables that predict SLT
success also have been published.7 This study found that
the most important predictor of SLT success was baseline

Crown Copyright ª 2018 Published by Elsevier Inc. on behalf of the American https://doi.org/10.1016/j.ophtha.2018.09.037 1
Academy of Ophthalmology ISSN 0161-6420/18
 Ophthalmology Volume -, Number -, Month 2018
intraocular pressure (IOP). The results of SLT treatment
after ALT therapy were published in 2007 by our group.8
Prediction rules that estimate the probability of SLT
success were published in 2008 and 2011.9,10 Finally, we
recently published the results of an RCT examining the
efficacy and side effects of SLT vs. ALT in patients with
pseudoexfoliation syndrome (PXF), again demonstrating
equivalence.11
Regarding SLT retreatment, in 2009, Hong et al12 first
reported the efficacy of repeat SLT in a case series of 44
eyes with open-angle glaucoma. In those patients, IOP
was controlled successfully for at least 6 months after the
first 360
SLT treatment, and they underwent a second 360

treatment. The IOP reduction was significantly greater for
the first SLT than the second one during the 1- to 3-month
follow-up, with an average value of 5 mmHg for the first
SLT and 2.9 mmHg for the second SLT. Avery et al13
conducted a similar case series using 42 eyes with
primary open-angle glaucoma (POAG). No significant dif-
ference was found between the first and the second treat-
ments with an IOP drop of approximately 4 mm Hg. Khouri
et al14 retrospectively reviewed the electronic medical
records from 45 eyes that underwent 2 360
SLT
treatments with up to 24 months of follow-up. The initial
SLT yielded a significantly greater IOP reduction than the
repeated one at 12 months, but the effect was the same at 1,
18, and 24 months. In 2014, Ayala15 reported the first RCT
for evaluating the IOP-lowering effect of repeat SLT. Pa-
tients (n = 80) who had undergone previous 180
SLT were
allocated randomly to receive another 180
SLT either at the
same TM area or the untreated TM area. Intraocular pressure
reduction at 2 hours, 1 month, 3 months, or 6 months after
laser treatment was not significantly different between the
primary and repeat SLT groups. Recently, Polat et al16 re-
treated patients with 360
of SLT after an initial 360

treatment. Thirty-eight eyes in the study were diagnosed
with POAG, PXF, or pigment dispersion syndrome. The
mean IOP reduction at each time point throughout the 24-
month follow-up ranged from 2.9 to 5.7 mmHg for the
initial SLT and 2.3 to 4.4 mmHg for the repeat SLT without
significant difference between the 2 SLT procedures. In
2016, Durr and Harasymowycz17 reported a case series
evaluating repeat 360
SLT on IOP control. Thirty-eight
Table 1. Potential Clinical Scenarios and Interpretations of this Randomized Control Trial
Selective Laser Trabeculoplasty Argon Laser Trabeculoplasty
Decrease (mmHg) Decrease (mmHg)
6e7 6e7
2e3 2e3
e1 to 1 e1 to 1
6e7 0e3
0e3 6e7
ALT ¼ argon laser trabeculoplasty; IOP ¼ intraocular pressure; SLT ¼ selective laser trabeculoplasty.
2
eyes with POAG, normal-tension glaucoma, or PXF were
included in the study. Intraocular pressure reduced
1.82.2 mmHg and 2.23.7 mmHg at 15 months, which
was not a significant difference (P ¼ 0.53). One of the
difficulties in interpreting these case series is that the
concomitant use of medications for glaucoma cannot be
controlled for in either a randomization methodology or in a
generalized linear model analysis. Hence, confounding by
indication permeates these studies and makes interpretation
difficult.
In summary, according to most of the available infor-
mation, repeat SLT is comparable with primary SLT
regarding IOP reduction, although not all studies show the
same IOP-lowering effect from the first and repeat SLT.
Both IOP spike and inflammation after laser treatment were
rare. However, only 1 study was an RCT and had a very
short follow-up. The other studies were all case series car-
ried out by retrospective chart review. The sample size of
most of the previous studies was small to moderate.
Therefore, more studies with better study design for evalu-
ating the efficacy of repeat SLT are needed.
In this trial, all patients had been treated previously with
360
of SLT. Then, they were randomized to either 180
of
SLT or 180
of ALT treatment. The 1-year IOP change from
baseline is the primary outcome. This question and outcome
were chosen to help understand several issues and to
examine several clinical scenarios as follows, which are
summarized in Table 1.
Our studies on previous SLT and ALT work indicate that
equivalence in lowering IOP is probable.5,6 Thus, we chose
to design an equivalence study. Nevertheless, even with
IOP-lowering equivalence between the 2 lasers, SLT has
theoretical advantages over ALT, but clinical advantages
remain to be ascertained.
Methods
Trial Design
We chose an active equivalence parallel-arm RCT based on the
results of our earlier clinical trial work with SLT and ALT. We
registered the study at clinicaltrials.gov (identifier, NCT01687465).
The study protocol was approved by the Western University
Interpretation
Consistent with first SLT IOP-lowering effect. Both lasers effective after initial
SLT. No change to angle from initial SLT is likely.
Both lasers effective after initial SLT, but only at approximately half the effect of
initial SLT. Initial SLT must cause some irreversible angle change, even if not
apparent histologically.
Neither laser is effective after initial SLT. The initial SLT laser must change the
angle in as permanent a way as ALT does, even if not histologically evident.
After initial SLT, SLT is effective, but ALT not or only partially effective. The
SLT angle changes of first laser do not affect future SLT treatments only.
There is a laser-specific change to the angle. Like ALT, repeatability reduces
effectiveness, but only with the same laser.
 Hutnik et al 
SLT vs. ALT in Patients with Previous SLT
Research Ethics Board (file no., 103238). This study was con-
ducted in accordance with the ethical standards of the Western
University Research Ethics Board and the 1964 Helsinki declara-
tion. The patient was masked to treatment, but the clinician was not
masked. In keeping with an effectiveness-type clinical trial, we
chose to include a generalizable study population, permissive
eligibility criteria, an easily administered treatment protocol, and
outcomes that are relevant to patient care. This effectiveness-type
RCT maximizes generalizability; that and ethical priorities neces-
sitated that any glaucoma medications used would not be washed
out in this trial. Substantive glaucoma surgical trial design and
reporting issues, including IOP measurements, were followed
based on the World Glaucoma Society consensus
recommendations.18
Inclusion Criteria
Inclusion criteria were as follows. All patients were enrolled from
one of the practices participating in this study and were 18 years
of age or older. Patients were diagnosed with open-angle glau-
coma, including pigmentary dispersion syndrome and PXF to
increase the generalizability of the study. Previous 360
SLT had
failed in all patients based on determining a target pressure for
each individual patient. This target IOP can be defined as the
upper limit of a stable range of measured IOPs deemed likely to
retard further optic nerve damage.19,20 In all patients, an IOP
measurement of more than the target pressure on at least 2 suc-
cessive occasions separated by 1 month constituted failure in light
of previous SLT. All patients had 2 sighted eyes, defined as best-
corrected visual acuity of 20/200 or better in the absence of an
advanced visual field defect.
Exclusion Criteria
Exclusion criteria were as follows. Patients with any evidence of
secondary open-angle glaucoma (other than pigmentary dispersion
syndrome and PXF) or narrow-angle glaucoma (where the anterior
TM is not visible over 360
); advanced visual field defect in the
eye being considered for treatment (defined as a scotoma within
10
of fixation or split fixation on Humphrey visual field 24-2, full-
threshold program); previous nonlaser glaucoma surgery in the eye
being considered for treatment; intraocular surgery anticipated in
the 12 months after treatment; any corneal disease obscuring
adequate visualization of the anterior chamber TM or reliable
applanation tonometry; and present treatment with topical or sys-
temic steroids or anticipated treatment with systemic steroids in the
6 months after treatment were excluded.
Randomization
Patients were recruited by each center’s participating glaucoma
physician(s). Patients were randomized by a randomized block
design. The randomization was computer generated using the
uniform distribution and was carried out by a centralized clinical
trial randomization unit (The London Health Research Kidney
Research Unit; Dr. Amit Garg, head) that was in no other way
involved in the execution of the study. The intervention type was
determined by this randomization method. Patients were random-
ized only after informed consent was obtained and just before the
laser application to reduce the chance of randomized patients not
participating in the study. Patients were masked to the allocation,
but the physician was not.
Trial Interventions and Protocol
The intervention was to apply 1 setting of SLT or 1 setting of ALT.
After instillation of 1% apraclonidine, patients then were treated
with either SLT or ALT according to a blocked randomization
schedule, with blocks of 4, 6, or 8 chosen randomly by a computer-
generated algorithm, with each block type having an equal chance
of being chosen. Generally, the inferior 180
of the angle was
treated. One hundred eighty degrees was chosen conservatively
because the effect of repeat laser had not been studied in this
context and in this type of RCT before. Argon laser trabeculoplasty
was performed using 50 applications of 50-mM spot size,
0.1-second duration, and an average power ranging from 400 to
600 mW directed through an antireflective coated Goldman lens to
produce blanching or occasional bubble formation in the anterior
TM. Selective laser trabeculoplasty was performed with the Selecta
7000 using 50 nonoverlapping applications, with a spot size of
400 mM (centered on the TM) and pulse duration of 3 nanoseconds.
The initial energy used was 0.8 mJ. The energy was adjusted until
bubble formation appeared and then was decreased by 0.1 mJ for
the remainder of the treatment. Average power during treatment
ranged from 0.8 to 1.4 mJ. A drop of 1% apraclonidine was
instilled in all treated eyes after laser treatment.
Frequency and Duration of Follow-up
After treatment, patients were sent home with instructions that
prednisolone acetate 1% be instilled in the treated eye 4 times daily
for 5 days. Patients were evaluated at 1 hour, 1 week, and at 1, 3, 6,
and 12 months after surgery. At all follow-up examinations, best-
corrected vision, anterior chamber reaction, IOP, and cup-to-disc
ratio were recorded. Gonioscopy to assess angle pigmentation
and peripheral anterior synechiae was performed at baseline and at
12 months after laser treatment. At every postoperative visit, every
attempt was made to keep the topical medication as constant as
possible.
Primary Outcome and Its Measure
The primary outcome was the change in IOP from the baseline visit
to the 12-month visit. The Goldman applanation tonometer was
used, and it was calibrated weekly. Applanation was performed at
approximately the same time of day (1 hour) as the original
baseline IOP.
Sample Size
The sample size calculation was based on the IOP change from
baseline to the 12-month follow-up. Equivalence was defined if the
95% confidence interval (CI) of the mean difference between 2
treatment groups was within e3 mmHg and þ3 mmHg. This was
chosen as the equivalence window because the standard deviation
of normal IOP variation is 2.6 mmHg.21 The margin was decided in
advance by the study group based on the clinically meaningful
difference along with previous clinical and statistical outcomes
and feasibility. Assuming a 90% chance (power) with type I
error rate of 0.05 that a 95% CI can exclude a difference of
more than 3 mmHg, the trial would need to recruit 117 eyes in
total. The process of recruitment, including the assessment,
enrollment, and randomization of participants into this RCT, was
documented in a Consolidated Standards of Reporting Trials
(CONSORT) flow diagram (Fig 1).
Primary Analysis
The primary outcome was analyzed based on the intention-to-treat
principle and was the comparison of IOP change at 12 months
between the 2 groups. The mean difference between the 2 laser
treatments and the 95% CI was derived by an independent-sample t
test. Before an independent-sample t test was conducted, normality
of the samples of the 2 groups was determined by visualizing the
3
 Ophthalmology Volume -, Number -, Month 2018
Enrollment Assessed for eligibility
(n = 166)
Randomized (n = 139)
Allocation
Allocated to SLT (n = 69)
Follow-Up
Lost to follow-up (no contact) (n = 2)
Discontinued intervention (as per physician
decision) (n = 3)
Wrong randomization (n = 1)
Analysis
Analyzed (n = 64)
Figure 1. Consolidated Standards of Reporting Trials (CONSORT) diagram showing progress of patients through the study. ALT ¼ argon laser trabe-
culoplasty; SLT ¼ selective laser trabeculoplasty.
distribution histogram. Homogeneity of variance between the 2
treatment groups was analyzed using the F test.
For the secondary outcomes, an independent-sample t test was
used for the continuous outcomes. For the binary outcomes, results
were presented as risk difference and relative risk with 95% CI. As
a secondary analysis, we performed a generalized linear model
multivariate analysis. However, the results were identical to those
of the univariate analysis. Hence, the results shown are from the
univariate analysis unless otherwise indicated.
Time to Failure
We also performed 3 time-to-failure analyses (KaplaneMeier
curves and Cox proportional hazards testing) with the definitions of
failure being (1) IOP reduction by less than 20%, (2) IOP reduction
by 3 mmHg or less, and (3) any medication addition or repeat angle
laser or interventional surgery needed. Reintervention was per-
formed when the patient’s individual target pressure was no longer
met.
Subgroup Analysis and Secondary Outcomes
Secondary outcomes included repeating the primary analysis with
POAG patients only; change in IOP from baseline to every
4
Excluded (n = 27)
♦ Not meeting inclusion criteria (n = 11)
♦ Declined to participate (n = 4)
♦ Unable to travel (n = 1)
♦ Medical disease (n = 1)
♦ Other reasons (n = 10)
Allocated to ALT (n = 70)
Lost to follow-up (no contact) (n = 1)
Discontinued intervention (as per physician
decision) (n = 2)
Wrong randomization (n = 1)
Analyzed (n = 66)
postoperative visit; anterior chamber inflammation at every post-
operative visit graded as 0 to 4 based on standard criteria as noted
above; Snellen visual acuity at every postoperative visit in loga-
rithm of the minimum angle of resolution units; TM pigmentation
recorded as an ordinal variable from 0 to 4þ, where 0 represents no
pigment and 4þ represents dense homogenous pigment; number of
glaucoma medications needed per patient in each group; and pro-
gression to surgical therapy in each group.
Covariates
Those considered to be confounders or effect modifiers were used
in the subgroup analysis. These covariates were built into the
exposure-outcome analysis using multivariate generalized linear
models and included the following: baseline IOP; baseline TM
pigmentation; age; gender; glaucoma risk factors by patient history,
including hypertension, diabetes, myopia, family history of glau-
coma (a binary variable), and central cornea thickness; and number
of glaucoma medications at baseline.
Sensitivity Analyses
The primary outcome analysis was repeated for the per-protocol
population. Patients who had increased the number of
 Hutnik et al 
SLT vs. ALT in Patients with Previous SLT

Table 2. Baseline Characteristics of Patients Table 2. (Continued.)

Selective Laser Argon Laser Selective Laser Argon Laser

Trabeculoplasty Trabeculoplasty Trabeculoplasty Trabeculoplasty
(n [ 64) (n [ 66) (n [ 64) (n [ 66)

Center PDS 4 3
Western University 16 17 PXF 7 12
University of Toronto 10 11 OHT 6 4
University of Calgary 17 16 PXF and OHT* 1 0
University of Alberta 2 2
Dalhousie University 13 15 ACI ¼ anterior chamber inflammation; ALT ¼ argon laser trabeculoplasty;
McMaster University 5 5 BCVA ¼ best-corrected visual acuity; CCT ¼ central corneal thickness;
McGill University 1 0 IOP ¼ intraocular pressure; logMAR ¼ logarithm of the minimum angle of
Study eye was right eye 28 43 resolution; OHT ¼ ocular hypertension; PAS ¼ peripheral anterior
Male gender 34 40 synechiae; PDS ¼ pigmentary dispersion syndrome; POAG ¼ primary
Age (yrs) open-angle glaucoma; PXF ¼ pseudoexfoliation syndrome; SD ¼ standard
Mean (SD) 64.95 (10.60) 67.07 (9.77) deviation; TMP ¼ trabecular meshwork pigmentation.
Range 35e89 42e89 *Defined as IOP >21 mmHg, open drainage angles observed on gonioscopy
Ethnicity without glaucomatous optic disc damage, detectable nerve fiber layer
White 54 57 defect, or visual field loss.
Black 3 3
Other 7 6
BCVA (logMAR)
Mean (SD) 0.11 (0.12) 0.11 (0.26) medications, had undergone a glaucoma surgery, or had received
Range e0.12 to 0.48 e1 to 1 another glaucoma laser treatment during the 12-month follow-up
IOP (mmHg) were removed from this analysis. A sensitivity test for the pri-
Mean (SD) 21.57 (3.14) 21.52 (3.30)
mary end point also was conducted with extreme case analysis (the
Range 16.5e29.5 15.5e32.5
best and worst case assessment), in which the missing IOP mea-
CCT (mm)
Mean (SD) 552.92 (38.52) 559.74 (38.25)
surements at 12 months were imputed with the minimum IOP of all
Range 452e647 484e682 the participants in the SLT group and the maximum IOP for those
Modified Schaffer (0e4) in the ALT group, and reverse.
2 2 3
3 35 35
4 27 28 Trial Management
TMP (0e4)
0 5 7
A steering committee composed of trial members oversaw the
1 30 31 day-to-day management of the study. The study database creation,
2 23 24 maintenance, and data analysis were performed at the Western
3 5 3 University Clinical Trial Data Center in London, Canada. All data
4 1 1 security standards set out by the Western University Data Safety
PAS (present) 2 1 Monitoring Board were met by the data center. Quality control
Cup-to-disc ratio was undertaken by both the steering committee and the data
Mean (SD) 0.64 (0.20) 0.66 (0.17) management center. An independent 3-member Data Safety
Range 0.2e0.9 0.2e0.9 Monitoring Committee was struck before the start of the study
Risk factors and oversaw the study, performed interim analyses, and audited
Family history of POAG 23 23 all adverse outcomes. The Data Safety Monitoring Committee met
Age (>60 yrs) 47 53 twice yearly.
Myopia 14 19
Elevated IOP (>21 mmHg) 40 42
Ethnic background 10 9
(“yes” if not white) Results
Concomitant medical 22 19
conditions (hypertension, Table 2 provides the baseline characteristics of the patients based
diabetes, hypothyroidism) on demographics, descriptive ocular variables, baseline IOP,
Other 5 0 glaucoma risk factors, and prelaser treatment. Of note, the
No. of glaucoma medication baseline IOP was almost identical in the 2 groups.
used at baseline Table 3 demonstrates the results of the primary outcome,
0 28 35 namely, the IOP change at 1 year compared with baseline for
1 5 3
SLT vs. ALT. This difference was 0.33 mmHg (3.16 mmHg for
2 16 16
SLT and 2.83 mmHg for ALT) and was not statistically
3 13 11
4 2 1
significant (P ¼ 0.71) in the intention-to-treat analysis. Table 4
Diagnosis demonstrates the same analysis as in Table 3 but at all the
POAG 46 47 different follow-up time points. Selective laser trabeculoplasty
had a significantly lower IOP reduction at early time points, 1 week
(Continued) and 1 month, but this effect was lost by 3 months.

5
 Ophthalmology Volume -, Number -, Month 2018

Table 3. Intraocular Pressure Change from Baseline to the 12-Month Visit

Difference in the
Selective Laser Argon Laser Intraocular Pressure Change
Trabeculoplasty Trabeculoplasty (95% Confidence Interval) P Value
Complete case analysis 0.33 (e1.40 to 2.05) 0.71
No. 64 66
Mean (SD) 3.16 (4.96) 2.83 (5.06)
Per-protocol analysis 0.56 (e0.91 to 2.03) 0.45
No. 49 41
Mean (SD) 3.37 (3.79) 2.81 (3.12)

SD ¼ standard deviation.
Intraocular pressure measured in millimeters of mercury.

Survival Analysis Adverse Events

We defined laser failure in 3 different ways, first, when the IOP
failed to be reduced by more than 3 mmHg (absolute IOP failure).
With this definition, the hazard ratio was 0.70 (SLT compared with
ALT failure, P ¼ 0.07). The KaplaneMeier curve for this defini-
tion of failure is shown in Figure 2, with log-rank test results that
were not statistically significant (P ¼ 0.06). However, with both
analyses, there was a strong trend (although not a statistically
significant one) for ALT to fail more quickly than SLT.
Our second definition of failure was IOP reduction of less
than 20% (relative IOP failure). As with the more than 3-mmHg
definition, the hazard ratio was 0.70 (P ¼ 0.06) with respect to
SLT failure compared with ALT. The KaplaneMeier curve is
shown in Figure 3, with SLT again showing a longer time to
failure than ALT, although this time the result was statistically
significant (P ¼ 0.0499). Hence, with a failure definition of
20% reduction of IOP, ALT failure was quicker than SLT by
both the multivariate Cox model and the KaplaneMeier sur-
vival curve (Fig 3), but only the survival curve reached statistical
significance.
The final definition of failure included adding any medication,
needing repeat laser treatment, or needing incisional glaucoma
surgery (intervention failure). In this definition, both the multi-
variate proportions hazard model and the univariate log-rank test
showed a statistically significant better time to failure for SLT
over ALT. In the Cox model, the hazard ratio was 0.52
comparing SLT failure with ALT failure (P ¼ 0.047), and the
KaplaneMeier curve shown in Figure 4 also demonstrated
significantly early failure with ALT over SLT (P ¼ 0.04, log-
rank test).
At all time points, there was no difference between the 2 groups
with respect to visual acuity, TM pigmentation, or anterior cham-
ber inflammation. The only exception was anterior segment
inflammation at the 1-week time point, when a significantly higher
proportion of SLT patients (22%) demonstrated inflammation
compared with the ALT group (9%; P ¼ 0.047). Furthermore,
there were no IOP spikes at 1 hour in either group.
Number of Glaucoma Medications at the
12-Month Follow-up Visit
During the 12-month follow-up period, 14 patients (22%) in the
SLT group and 22 patients (33%) in the ALT group added at least 1
medication. The risk difference was e0.11 (95% CI, e0.27 to
0.04; P ¼ 0.15). The relative risk was 0.66 (95% CI, 0.37e1.17;
P ¼ 0.15). The number of glaucoma medications used at 12
months was comparable between the 2 laser treatment groups (1.64
in the SLT group and 1.52 in the ALT group; P ¼ 0.59).
Subgroup Analyses
The subgroup analyses based on diagnosis of POAG, baseline IOP,
age, gender, glaucoma risk factors, central corneal thickness,
number of glaucoma medications used at baseline, prostaglandin
analog users, carbonic anhydrase inhibitor users, and time of pre-
vious SLT showed an equivalent IOP reduction effect or a
nonsignificant IOP change difference at 12 months between the
SLT and ALT groups (Table 5).

Table 4. Intraocular Pressure Change from Baseline to Different Time Points

Selective Laser Argon Laser Difference

Time Point Trabeculoplasty (n [ 64) Trabeculoplasty (n [ 66) (95% Confidence Interval) P Value
1 wk 3.20 (4.43) 1.55 (3.79) 1.65 (0.22e3.08) 0.02*
1 mo 4.66 (3.67) 2.83 (3.54) 1.83 (0.58e3.08) 0.005*
3 mos 4.17 (3.01) 3.32 (4.64) 0.85 (e0.51 to 2.21) 0.22
6 mos 3.26 (3.60) 3.27 (4.02) e0.01 (e1.34 to 1.31) 0.99

Data were analyzed based on complete cases and are mean (standard deviation) unless otherwise indicated. Intraocular pressure measured in millimeters of
mercury.
*P < 0.05.

6
 Hutnik et al 
SLT vs. ALT in Patients with Previous SLT
Figure 2. KaplaneMeier curve showing survival estimates (success defined
as intraocular pressure [IOP] change of more than 3 mmHg from baseline
IOP). ALT ¼ argon laser trabeculoplasty; SLT ¼ selective laser
trabeculoplasty.
Discussion
The primary aim of this study was to determine the 1-year
efficacy of SLT vs. ALT after failure of previous 360

SLT. Our results indicated that they are equal, both
achieving approximately 3-mmHg lowering of IOP after 12
months of treatment. This was true with both the intention-
to-treat analysis and the per-protocol analysis. Although this
was the primary question, we believe the design of this
study answers more than this question alone.
The first issue is that although the IOP decreases were
equal at 1 year, essentially all of the time-to-failure analyses
consistently favored SLT over ALT in the analysis. We
cannot exclude that this is because the effect of SLT in this
repeat study is superior to ALT at early time points, but then
Figure 3. KaplaneMeier curve showing survival estimates (success defined
as intraocular pressure [IOP] change of more than 20% of baseline IOP).
ALT ¼ argon laser trabeculoplasty; SLT ¼ selective laser trabeculoplasty.
Figure 4. KaplaneMeier curve showing survival estimates (success as no
additional interventions). ALT ¼ argon laser trabeculoplasty; SLT ¼ se-
lective laser trabeculoplasty.
dissipates with time. However, it is more likely that SLT
does show modest superiority to ALT in terms of consis-
tently keeping IOP lower over the course of 1 year, even
though they are equal at 1 year.
Also interesting is that the SLT (and ALT) effect at 1 year is
a decrease of only 3 mmHg. This is less than our RCTs that
have looked at primary SLT vs. ALT have shown by
approximately 50%.5,6 It implies strongly that although initial
SLT may produce no histologic changes to the TM, there is
likely some change nevertheless because the second SLT is
not as effective as the first. It is difficult to know if this
decreased efficacy is because of direct physiologic or
biochemical changes to the meshwork or some other more
complex reason. Some authors have examined mechanistic
changes in the angle after SLT that are not histologic.22 Our
SLT patients continue to be enrolled in an ongoing open-
label study of further SLT that will determine if even further
SLT laser procedures produce the same results as the second
or whether further dissipation of the effect occurs.
Table 5 shows the SLT vs. ALT results stratified by
several variables. The ALT vs. SLT pressure drop is not
different at 1 year stratified by any of these variables.
However, the effect of both lasers is more effective in
patients with higher initial IOP, and this is consistent with
primary SLT results.7 The effect of repeat SLT itself did
not seem to depend on the use of previous glaucoma
medications, including prostaglandins. There was a
slightly better effect of repeat SLT if topical carbonic
anhydrase medications were not used; however, this effect
was small and not statistically significant. Furthermore, the
timing of the first SLT, whether recent or further in the
past (more or less than 3 years), did not matter in terms of
repeat SLT effect. Both glaucoma diagnosis and patient
age were not predictors of the outcome, making it likely
that duration of disease also was not an important factor.
Our study has the advantage of being a multicenter RCT
whose participants were all experienced SLT users and SLT
trial researchers. We believe that this increases both the
7
 Ophthalmology Volume -, Number -, Month 2018

Table 5. Intraocular Pressure Change from Baseline to the 12-Month Visit in Subgroups

Selective Laser Argon Laser Mean Difference

Subgroup Trabeculoplasty Trabeculoplasty (95% Confidence Interval) P Value
POAG e0.22 (e1.90 to 1.46) 0.80
No. 48 48
Mean (SD) 2.68 (3.58) 2.90 (4.64)
Baseline IOP (mmHg)
<22 0.80 (e0.64 to 2.24) 0.27
No. 38 39
Mean (SD) 2.58 (2.73) 1.78 (3.54)
22 e0.35 (e4.03 to 3.33) 0.85
No. 26 27
Mean (SD) 4.01 (6.65) 4.36 (6.68)
Age (yrs)
<66 1.51 (e0.61 to 3.63) 0.16
No. 33 30
Mean (SD) 3.01 (3.23) 1.50 (5.05)
66 e0.62 (e3.33 to 2.08) 0.65
No. 31 36
Mean (SD) 3.32 (6.00) 3.94 (5.09)
Gender
Female 0.13 (e2.64 to 2.89) 0.93
No. 30 26
Mean (SD) 2.63 (5.60) 2.50 (4.57)
Male 0.58 (e1.68 to 2.84) 0.61
No. 34 40
Mean (SD) 3.63 (3.84) 3.05 (5.59)
Glaucoma risk factors
2 0.53
No. 36 33
Mean (SD) 3.72 (5.25) 2.87 (5.91) 0.85 (e1.83 to 3.53)
>2 0.74
No. 28 33
Mean (SD) 2.44 (3.97) 2.79 (4.42) e0.35 (e2.52 to 1.81)
CCT (mm)
<556 e0.35 (e3.04 to 2.34) 0.79
No. 33 31
Mean (SD) 3.02 (4.98) 3.37 (5.77)
556 0.96 (e1.31 to 3.22) 0.40
No. 31 35
Mean (SD) 3.32 (4.54) 2.36 (4.63)
No. of glaucoma medications at baseline
0 e0.23 (e2.53 to 2.07) 0.84
No. 28 35
Mean (SD) 3.03 (5.14) 3.25 (3.99)
1 0.91 (e1.73 to 3.54) 0.50
No. 36 31
Mean (SD) 3.27 (4.47) 2.36 (6.30)
PGAs at baseline
No e0.14 (e2.16 to 1.89) 0.89
No. 33 42
Mean (SD) 2.95 (4.86) 3.09 (3.95)
Yes 1.00 (e2.14 to 4.14) 0.53
No. 31 24
Mean (SD) 3.38 (4.68) 2.38 (6.91)
CAIs at baseline
No 0.28 (e1.55 to 2.11) 0.76
No. 51 50
Mean (SD) 3.31 (4.80) 3.03 (4.47) 0.37 (e4.33 to 5.06) 0.87
Yes
No. 13 16
Mean (SD) 2.58 (4.63) 2.22 (7.10)

8
 Hutnik et al 
SLT vs. ALT in Patients with Previous SLT
Table 5. (Continued.)
Selective Laser
Subgroup Trabeculoplasty
Previous SLT (yrs)
<3
No. 35
Mean (SD) 3.23 (5.02)
3
No. 29
Mean (SD) 3.07 (4.66)
CAI ¼ carbonic anhydrase inhibitors; CCT ¼ central corneal thickness; IOP ¼ intraocular pressure; PGA ¼ prostaglandin analog; POAG ¼ primary open-
angle glaucoma; SD ¼ standard deviation; SLT ¼ selective laser trabeculoplasty.
Data were analyzed based on complete cases. Intraocular pressure measured in millimeters of mercury.
validity and precision of our results as well as the generaliz-
ability. We followed the World Glaucoma Society Random-
ized Clinical Trial Recommendations whenever possible,
including when measuring IOP. Figure 1 demonstrates that
the randomization method was very robust.
The drawbacks of our study included a non-washout
phase and the fairly narrow patient demographic profile.
A washout phase has the advantage of studying the laser
free of medications. However, there are significant prob-
lems with washouts. First, the ethics of stopping a medi-
cation(s) needed for a blinding disease are very significant
and may not be accepted by some physician recruiters and
some patients. This is the biggest drawback of the washout
concept. The time needed for washout often is poorly un-
derstood, making the practicality of performing washouts
difficult. Finally, in the real-world scenario, patients use
glaucoma medications, and so a washout decreases
generalizability considerably. The effect of medications not
being washed out was mitigated by several strategies. First,
we tried to keep medications constant at every visit, and
the average preoperative medication number and 1-year
medication number was minimally different in both
groups. We also carried out both univariate and multivar-
iate analyses (controlling for medication use among other
covariates) that produced identical results. Our patients
were demographically quite homogenous, with a high
percentage of white patients. This may reduce generaliz-
ability to some populations. Finally, patient eye drop
compliance was monitored by patient history only.
Although not a disadvantage, our protocol included
steroids as a postoperative medication. However, we
believe that it is unlikely to have had an effect on IOP,
given the short duration of use. Furthermore, the steroid
was given to both groups.
In summary, in patients who have undergone previous
360
SLT, both SLT and ALT are equal in lowering IOP at
1 year, although the effect of SLT is slightly superior to that
of ALT at earlier time points throughout the year. The SLT
effect seems to be approximately 50% of the primary SLT
treatment. Repeat SLT, like first SLT, is more effective in
patients with higher initial IOP. The effect of third SLT
treatments and beyond is now being evaluated by us in an
open-label study.
Argon Laser Mean Difference
Trabeculoplasty (95% Confidence Interval) P Value
e0.23 (e2.67 to 2.21) 0.85
31
3.47 (4.88)
0.80 (e1.72 to 3.32) 0.53
35
2.27 (5.44)
References
1. Latina MA, Park C. Selective targeting of trabecular mesh-
work cells: in vitro studies of pulsed and CW laser in-
teractions. Exp Eye Res. 1995;60:359e372.
2. Kennedy JB, SooHoo JR, Kahook MY, Seibold LK. Selective
laser trabeculoplasty: an update. Asia-Pacific J Ophthalmol.
2016;5:63e69.
3. Kramer TR, Noecker RJ. Comparison of the morphologic
changes after selective laser trabeculoplasty and argon laser
trabeculoplasty in human eye bank eyes. Ophthalmology.
2001;108:773e779.
4. Latina MA, Sibayan SA, Shin DH, et al. Q-switched 532-nm
Nd:YAG laser trabeculoplasty (selective laser trabeculo-
plasty). Ophthalmology. 1998;105:2082e2090.
5. Damji KF, Shah KC, Rock WJ, et al. Selective laser tra-
beculoplasty v argon laser trabeculoplasty: a prospective
randomized clinical trial. Br J Ophthalmol. 1999;83:
718e722.
6. Damji KF, Bovell AM, Hodge WG, et al. Selective laser
trabeculoplasty vs argon laser trabeculoplasty: results
from a 1 year RCT. Br J Ophthalmol. 2006;90(12):
1490e1494.
7. Hodge W, Damji KF, Rock W, et al. Baseline IOP predicts
selective laser trabeculoplasty success at 1 year post-treatment:
results from a randomised clinical trial. Br J Ophthalmol.
2005;89(9):1157e1160.
8. Birt CM. Selective laser trabeculoplasty retreatment after prior
argon laser trabeculoplasty: 1-year results. Can J Ophthalmol.
2007;42(5):715e719.
9. Mao AJ, Pan XJ, McIlraith I, et al. Development of a pre-
diction rule to estimate the probability of acceptable intraoc-
ular pressure reduction after selective laser trabeculoplasty in
open-angle glaucoma and ocular hypertension. J Glaucoma.
2008;17(6):449e454.
10. Martow E, Mao A, Hutnik CML. SLT and adjuvant medical
therapy. A prediction rule analysis. J Glaucoma. 2011;20(4):
266e270.
11. Kent SS, Hutnik CM, Birt CM, et al. A randomized clinical
trial of selective laser trabeculoplasty versus argon laser tra-
beculoplasty in patients with pseudoexfoliation. J Glaucoma.
2015;24(5):344e347.
12. Hong BK, Winer JC, Martone JF, et al. Repeat selective laser
trabeculoplasty. J Glaucoma. 2009;18:180e183.
13. Avery N, Ang GS, Nicholas S, Wells A. Repeatability of
primary selective laser trabeculoplasty in patients with
9
 Ophthalmology Volume -, Number -, Month 2018
primary open-angle glaucoma. Int Ophthalmol. 2013;33:
501e506.
14. Khouri AS, Lari HB, Berezina TL, et al. Long term efficacy of
repeat selective laser trabeculoplasty. J Ophthalmic Vis Res.
2014;9:444e448.
15. Ayala M. Intraocular pressure reduction after initial failure of
selective laser trabeculoplasty (SLT). Graefes Arch Clin Exp
Ophthalmol. 2014;252:315e320.
16. Polat J, Grantham L, Mitchell K, Realini T. Repeatability of
selective laser trabeculoplasty. Br J Ophthalmol. 2016;100:
1437e1441.
17. Durr GM, Harasymowycz P. The effect of repeat 360-degree
selective laser trabeculoplasty on intraocular pressure control
in open-angle glaucoma. J Fr Ophtalmol. 2016;39:261e264.
18. Shaarawy TM, Sherwood MB, Grehn F, eds. Guideline on
Design and Reporting of Glaucoma Surgical Trials. Amster-
dam: Kuegler Publications; 2009.
Footnotes and Financial Disclosures
Originally received: June 12, 2018.
Final revision: September 6, 2018.
Accepted: September 25, 2018.
Available online: ---. Manuscript no. 2018-1355.
1
Department of Ophthalmology, Western University, Ontario, Canada.
2
Department of Ophthalmology, University of Calgary, Calgary, Canada.
3
Department of Ophthalmology, Dalhousie University, Halifax, Canada.
4
Department of Ophthalmology, University of Toronto, Toronto, Canada.
5
Department of Ophthalmology, McMaster University, Hamilton, Canada.
6
Department of Ophthalmology, University of Alberta, Alberta, Canada.
7
Department of Ophthalmology, McGill University, Montreal, Canada.
Presented at: Association for Research and Ophthalmology Annual
Meeting, AprileMay 2018, Honolulu, Hawaii.
Financial Disclosure(s):
The author(s) have made the following disclosure(s): E.S.: Consultant e
Bausch & Lomb, Novartis, Alcon; Financial support e Allergan.
M.D.: Financial support e Allergan, Ivantis.
H.S.: Financial support e Alcon/Novartis, Allergan, Bausch & Lomb,
Glaukos, Aerie Pharmaceuticals, Johnson & Johnson, Ivantis, Zeiss.
Supported by the Canadian Institute of Health Research (2012 operating
grant no.: 123233). The sponsor had no role in the design or conduct of this
research.
HUMAN SUBJECTS: Human subjects were included in this study. The
human ethics committees at Western University Research Ethics Board
approved the study. This study was conducted in accordance with the
ethical standards of the Western University Research Ethics Board and the
1964 Helsinki declaration. All participants provided informed consent.
10
19. Canadian Ophthalmological Society Glaucoma Clinical Prac-
tice Guideline Expert Committee; Canadian Ophthalmological
Society evidence-based clinical practice guidelines for the
management of glaucoma in the adult eye. Can J Ophthalmol.
2009;44(Suppl 1):S27.
20. American Academy of Ophthalmology. Primary open-angle
glaucoma. Preferred Practice Pattern. San Francisco, CA:
American Academy of Ophthalmology; 2005. Available at:
http://one.aao.org/asset.axd?id¼2e86ca2e-9db0e43c9-b605- c004
a82b6ea5. Accessed August 21, 2018.
21. Asrani S, Zeimer R, Wilensky J, et al. Large diurnal fluctua-
tions in IOP are an independent risk factor in patients with
glaucoma. J Glaucoma. 2000;9(2):134e142.
22. Lee J, Kagan D, Roumeliotis G, et al. Secretion of MMP-3 by
co-cultured pigmented and non-pigmented human trabecular
meshwork cells following selective laser trabeculoplasty. Clin
Exp Ophthalmol. 2016;44:33e42.
No animal subjects were included in this study.
Author Contributions:
Conception and design: Hutnik, Crichton, Ford, Nicolela, Shuba, Birt,
Sogbesan, Damji, Dorey, Saheb, Hodge
Analysis and interpretation: Hutnik, Crichton, Ford, Nicolela, Shuba, Birt,
Sogbesan, Damji, Dorey, Saheb, Klar, Guo, Hodge
Data collection: Hutnik, Crichton, Ford, Nicolela, Shuba, Birt, Sogbesan,
Damji, Dorey, Saheb, Hodge
Obtained funding: Hodge
Overall responsibility: Hutnik, Crichton, Ford, Nicolela, Shuba, Birt,
Sogbesan, Damji, Dorey, Saheb, Klar, Guo, Hodge
Abbreviations and Acronyms:
ACI ¼ anterior chamber inflammation; ALT ¼ argon laser trabeculoplasty;
BCVA ¼ best corrected visual acuity; BJO ¼ British Journal of
Ophthalmology; CCT ¼ central corneal thickness; CI ¼ confidence in-
terval; IOP ¼ intraocular pressure; logMAR ¼ logarithm of minimum
angle of resolution; OHT ¼ ocular hypertension; PAS ¼ peripheral anterior
synechiae; PDS ¼ pigmentary dispersion syndrome; POAG ¼ primary
open-angle glaucoma; PXF ¼ pseudoexfoliation syndrome;
RCT ¼ randomized clinical trial; SD ¼ standard deviation;
SLT ¼ selective laser trabeculoplasty; TM ¼ trabecular meshwork;
TMP ¼ trabecular meshwork pigmentation.
Correspondence:
William Hodge, MD, PhD, St. Josephs Healthcare London, 268 Grosvenor
Street, London, Ontario N6A 4V2, Canada. E-mail: William.Hodge@sjhc.
london.on.ca.