Selective, Micropulse and Pattern Laser Trabeculoplasty in Medically Treated Open-angle Glaucoma.

Background and rationale Laser trabeculoplasty is an established first-line and adjunctive treatment in open-angle glaucoma. Three subvisible delivery protocols are in clinical use and differ substantially in the physics of energy delivery: selective laser trabeculoplasty (SLT) uses 3-nanosecond pulses titrated to microbubble formation; micropulse laser trabeculoplasty (MLT) uses a 577-nm envelope chopped at a low duty cycle; pattern laser trabeculoplasty (PLT) uses short-exposure, high-density automated patterns. Evidence comparing them is indirect: SLT has been compared with medication in randomized and registry studies, SLT and MLT have been compared in retrospective series and in meta-analyses concluding non-inferiority, and each modality has been compared separately against medical therapy. No trial had compared all three protocols head to head in the same cohort under a single protocol. Objectives The trial was designed to determine whether the three subvisible delivery protocols differ in IOP reduction and in treatment success at 6 months in eyes with open-angle glaucoma already receiving medical therapy. Design and setting Prospective, randomized, parallel-group, three-arm trial conducted at the Glaucoma Unit of the Department of Medicine - Ophthalmology, University of Udine, Italy. The protocol adhered to the tenets of the Declaration of Helsinki, was approved by the local institutional review board, and all participants gave written informed consent. The trial was not registered before enrolment of the first participant and is registered retrospectively. Participants Adults with primary or secondary open-angle glaucoma were eligible. Gonioscopy was performed in every eye and only gonioscopically open angles were included. Eyes were selected for subvisible laser trabeculoplasty according to standard practice at the unit (IOP inadequately controlled on medical therapy, intolerance of medical therapy, or both), so that all participants were receiving IOP-lowering medication at entry. Randomization and masking One eye per patient was enrolled; when both eyes were eligible, the study eye was the eye with the higher baseline IOP. Eyes were allocated 1:1:1 by a computer-generated random number sequence. Participants and the treating surgeon were not masked to allocation, which is not feasible when the three lasers differ in console and audible signature. All IOP measurements and anterior segment examinations were performed by a second ophthalmologist masked to allocation and to clinical history; outcome assessment was therefore single-masked. Medication changes were decided by the unmasked treating physician against a pre-set individual target, so masking did not extend to the co-intervention. Interventions Every eye received a single 360-degree session, performed by the same surgeon, after two drops of pilocarpine 2% and topical anaesthetic. SLT was delivered with a Tango laser (Ellex Medical Lasers, Adelaide, Australia) over 360 degrees using approximately 100 non-overlapping 400-micrometre spots and 3-nanosecond pulses. Pulse energy was titrated to produce visible microbubble formation in approximately 50% of applications, up to a maximum of 1.4 mJ per pulse. MLT was delivered with an IQ 577 laser (Iridex, Mountain View, California, USA): 577 nm, 1000 mW, 300-ms envelope, 300-micrometre spot, 15% duty cycle. PLT was delivered with a Pascal Streamline 577 (Topcon, Tokyo, Japan): power was titrated with 10-ms pulses until light blanching of the meshwork was seen, then exposure was shortened to 5 ms, rendering the treatment subvisible, with 100-micrometre spots delivered in 32 automatically rotating arc-shaped patterns of three rows of 13 spots, zero spacing, for approximately 1250 spots over 360 degrees. Follow-up and co-intervention Eyes were examined at baseline and at 1 day, 1 week, 1 month, 3 months, and 6 months. Each visit included anterior segment examination, best-corrected visual acuity, and Goldmann applanation tonometry performed at the same time of day (+/- 1 hour) by the masked assessor. At every visit three consecutive IOP readings were taken and their arithmetic mean was used in all analyses. IOP was additionally measured 1 hour after treatment to detect pressure spikes. An individual target IOP was set before treatment for every eye as the lower of a 25% reduction from presenting IOP (Early Manifest Glaucoma Trial criterion) or 18 mmHg (Advanced Glaucoma Intervention Study criterion). From month 1 onward medications were titrated against this target; the week-1 visit therefore precedes any protocol-driven medication change, and week-1 medication counts were used as the baseline medication burden. Sample size The sample-size calculation was based on the global treatment effect in an analysis of covariance model comparing three treatment groups. Assuming an omnibus effect size f = 0.35, a two-sided alpha of 0.05, and 80% power, 82 eyes were required; this was rounded to 84 eyes to allow equal allocation of 28 eyes per group. Ninety eyes were enrolled, providing approximately 84% power under the original assumptions. Statistical analysis All randomized eyes were analysed according to their assigned treatment group in the intention-to-treat population. The primary outcome was analysed using analysis of covariance with 6-month IOP as the dependent variable, treatment group as a fixed factor, and baseline IOP as a covariate; adjusted means and all three pairwise adjusted mean differences with 95% confidence intervals were derived from this model, with the Holm correction applied to the pairwise comparisons. Equivalence was assessed by two one-sided tests against the prespecified +/-2.0 mmHg margin. A linear mixed-effects model was used for the longitudinal IOP data. Cumulative treatment success was analysed by Kaplan-Meier estimation with the log-rank test and by Cox proportional-hazards regression.