Purpose
To analyze reduction of intraocular pressure (IOP) by argon laser trabeculoplasty (ALT) in the Early Manifest Glaucoma Trial and factors influencing the effect of such treatment.
Design
Cohort study based on 127 patients from the treatment group of the Early Manifest Glaucoma Trial, a randomized clinical trial.
Methods
Patients randomized to the treatment arm of the Early Manifest Glaucoma Trial received a standard treatment protocol (topical betaxolol hydrochloride followed by 360-degree ALT) and then were followed up prospectively at 3-month intervals for up to 8 years. One eye per patient was included in the analyses. We investigated the relationship between IOP before ALT and subsequent IOP reduction and other factors that might have influenced the effect of ALT, including stage of the disease, trabecular pigmentation, presence of exfoliation syndrome, and treating surgeon.
Results
The mean ± standard deviation IOP before ALT and after betaxolol treatment was 18.1 ± 3.9 mm Hg, and the mean ± standard deviation short-term IOP reduction 3 months after ALT was 2.8 ± 3.9 mm Hg (12.6 ± 20.5%). The IOP before ALT strongly affected IOP reduction ( P < .001); each 3-mm Hg higher IOP before ALT value was associated with an additional mean IOP reduction of approximately 2 mm Hg. The treating surgeons also had a significant impact on IOP reduction ( P = 0.001), with mean values ranging from 5.8 to −1.3 mm Hg.
Conclusions
In this cohort, which included many patients with low IOP levels, IOP before ALT markedly influenced the IOP reduction induced by ALT, seen as a much larger decrease in eyes with higher IOP before ALT. The treating surgeon also had a significant impact on ALT outcome.
Argon laser trabeculoplasty (ALT) has been used to control open-angle glaucoma for more than 30 years, in some cases as primary treatment, but more often as adjunctive treatment. It seems clear that the size of the pressure-lowering effect achieved with ALT depends on the level of intraocular pressure (IOP) before laser treatment, as indicated by several studies showing that IOP reduction (expressed in millimeters of mercury) is larger in eyes with higher untreated pressures. Nevertheless, detailed knowledge is lacking with regard to the impact of baseline IOP on treatment effects, particularly concerning results obtained in patients with low IOP before laser treatment.
We realized that it should be possible to obtain new knowledge in this area by using data collected in the Early Manifest Glaucoma Trial (EMGT). In short, this randomized clinical trial aimed to evaluate the effectiveness of IOP-reducing treatment in patients with manifest open-angle glaucoma. The EMGT patients were randomized to a fixed treatment protocol (topical betaxolol plus ALT) or to no treatment. That approach differs from most other studies of ALT, which have administered laser treatment to patients with uncontrolled IOP. In the EMGT, IOP values before treatment varied over a large range, and the study included a large proportion of patients with normal or even low IOP. This feature offered us the opportunity to study the influence of IOP level before laser treatment on the pressure-lowering effect of ALT. Furthermore, treatment of the EMGT patients was not changed unless progression occurred, and thus data are available for several years after ALT with unchanged topical therapy. Also, the use of an untreated control arm in the EMGT allowed us to evaluate any long-term non–treatment-related effects, such as whether there was a tendency for IOP to increase over time. The purpose of the present study was to analyze the pressure-lowering effects of ALT in this group of prospectively observed patients, focusing on the role of the IOP before ALT level and other factors related to the IOP-reducing effects of ALT.
Methods
The EMGT study design has been described in detail previously. Briefly, this trial included patients with newly diagnosed and untreated primary open-angle glaucoma (POAG) or exfoliation glaucoma with early to moderate damage. Most patients were found in a large, population-based screening of 44 000 elderly citizens in the cities of Malmö and Helsingborg, Sweden. Eligible eyes had glaucoma with established visual field damage, repeatedly demonstrated by the Humphrey 30-2 full-threshold test and the glaucoma hemifield test being outside normal limits. Patients were randomized to treatment or no treatment and followed up prospectively by tonometry, perimetry, optic disc photography, and general ophthalmic examinations.
The current study included 127 of the 129 patients who initially were randomized to the EMGT treatment arm. Two patients were excluded for the following reasons: one declined ALT, and the other withdrew from the study directly after ALT. After randomization, patients received betaxolol eye drops 2 weeks before the laser procedure, which consisted of a full 360-degree ALT. One hour after pretreatment with 1% apraclonidine (Iopidine; Alcon, Fort Worth, Texas, USA), one hundred (range: 90–110) 50µm spots of 100 milliseconds’ duration were delivered to the pigmented part of the trabecular meshwork. The effect was 600 to 1400 mW, titrated to small bubble formation. The procedures were performed by 7 surgeons (A.H. and 6 independent surgeons) who were experienced glaucoma specialists or ophthalmologists with a special interest in glaucoma.
Treatment was unchanged as long as progression did not occur. Progression was defined as worsening of visual fields according to predefined criteria based on computerized analyses using glaucoma change probability maps, which have been shown to offer high sensitivity and specificity. Glaucoma progression also could be identified as worsening of the disc, which was assessed by masked readings at a disc photography reading center. The protocol included a safety hatch that entailed an exception to the principle of unchanged therapy; in short, if the IOP in treated eyes exceeded 25 mm Hg at 2 consecutive follow-up visits, additional pressure-lowering eye drops were prescribed. The extra drug was dorzolamide during the earlier phase of the trial, but later was changed to latanoprost. This exception resulted in only a few changes in therapy (see Results).
Patients were followed up at 3-month intervals. At each visit, tonometry was performed by certified, independent technicians using calibrated Goldmann applanation tonometers. Technicians were masked to the patient’s study group (treated or untreated) and to earlier IOP values, and follow-up measurements were performed at approximately the same time of day whenever possible.
Pressure-Reducing Effect
Analyses of the IOP-reducing effect of ALT included 1 eye of each patient; if both eyes were eligible, one was selected randomly. We used follow-up IOP measurements obtained every 3 months during the first year after ALT (n = 4) and every 6 months during the second year (n = 2) and all available 12-month measurements from the third year onward. Only IOP values recorded up to the time of progression were included, because treatment could have been changed after that point and thus could have affected IOP.
IOP before ALT was defined as the last IOP measurement before ALT, that is, the eyes then were treated with betaxolol. The reduction in IOP achieved by ALT was defined as the difference between the IOP before ALT value and the IOP value at the 3-month follow-up visit.
The mean and median IOP measurements for all patients remaining in the study at each time point are not reported here, because such values would be misleading. Inasmuch as higher IOP was associated with faster progression, patients with lower initial IOP levels gradually would have become more overrepresented among the nonprogressing group. Therefore, we instead calculated IOP slopes over time for each patient, which was carried out to assess changes that included only IOP values recorded while the patient was still in the study and had not reached progression or had not received treatment according to the safety hatch exception (see above). The patients subsequently were divided into those with at least 5 IOP measurements during follow-up and those with fewer than 5 such values, because errors in measurement changes over time can be particularly large when only a few observations are available. One patient with a single IOP measurement after ALT was excluded from these analyses.
Statistical significance of the difference between IOP before ALT and IOP 3 months after ALT was evaluated by 2-tailed t tests. Similarly, the t test was applied to compare the exfoliation and POAG patients with regard to the mean IOP reduction 3 months after ALT.
Factors Influencing Pressure Reduction
We also evaluated factors that might have influenced short-term (3-month) IOP reduction by ALT. The primary aim was to study the influence of IOP before ALT, but we also considered possible effects of the following: the treating surgeon, stage of the disease defined by the perimetric mean deviation (MD; average of 2 baseline fields), trabecular pigmentation (grades 0 through 3), and presence of exfoliations (dilated examination).
A stepwise multiple linear regression analysis of IOP reduction was performed using IOP before ALT, baseline MD, diagnosis (exfoliation glaucoma or POAG), and trabecular pigmentation grade as independent variables.
We used two-way ANOVA to analyze the influence of the treating surgeon on total IOP reduction and on percentage pressure reduction. IOP values before ALT were included in the analysis to discern any interaction between surgeon and IOP before ALT. Eyes were divided into 3 different pressure groups with equal IOP intervals (based on measurements before ALT): 16 mm Hg or less, 17 to 23 mm Hg, and 24 mm Hg or more. A 2-way analysis of variance using type III sums of squares was used to handle unbalanced data resulting from differences in the numbers of treated patients per surgeon and the pre-ALT group.
All statistical calculations were performed using SPSS software version 17.0 (SPSS, Inc, Chicago, Illinois, USA). Statistical significance was set at P < .05.
Results
Follow-up time ranged from at least 3 months to a maximum of 8 years (mean, 57 months; median, 60 months). Among the 127 patients, 3 (2.4%) were lost to follow-up during the study period. Sixty-four (50.4%) patients progressed, and 14 (11%) died. For 9 (7.1%) patients who either discontinued betaxolol or had a change of therapy, we included IOP values only as long as treatment remained unchanged.
Baseline Characteristics
Baseline characteristics of the 127 patients are shown in Table 1 . Eighty-one (64%) of the patients were female; 119 (94%) had POAG, and 8 (6%) had exfoliation glaucoma. Mean baseline IOP was 20.7 mm Hg. Almost all patients had slight to moderate trabecular pigmentation, and 92 (72.5%) had early visual field loss with a perimetric MD better than −6 dB.
| Characteristic | No. (%) | Mean ± SD (Median [Range]) |
|---|---|---|
| Age (years) | 127 (100) | 68.6 ± 4.8 (68.2 [58.2 to 78.9]) |
| Sex | ||
| Female | 81 (63.8) | — |
| Male | 46 (36.2) | |
| Diagnosis | ||
| Primary open-angle glaucoma | 119 (93.7) | — |
| Exfoliation glaucoma | 8 (6.3) | |
| Trabecular pigmentation | ||
| Grade 0 | 7 (5.5) | — |
| Grade 1 | 50 (39.4) | |
| Grade 2 | 62 (48.8) | |
| Grade 3 | 8 (6.3) | |
| IOP before any treatment (mm Hg) | 127 (100) | 20.7 ± 4.0 (20 [13 to 30]) |
| IOP immediately before argon laser trabeculoplasty (mm Hg) | ||
| All groups | 127 (100) | 18.1 ± 3.9 (18 [10 to 29]) |
| Group 1: 10 to 16 | 48 (37.8) | — |
| Group 2: 17 to 23 | 67 (52.8) | |
| Group 3: 24 to 29 | 12 (9.4) | |
| Baseline mean deviation (dB) | 127 (100) | −4.8 ± 3.7 (−4.2 [−14.7 to −1.3]) |
Short-term and Long-term Pressure Reduction
The mean IOP before ALT and after betaxolol treatment was low, 18.1 mm Hg. Mean short-term IOP reduction (3 months after ALT) was 2.8 mm Hg ( P < .001), or 12.6%. The pressure-lowering effect was larger in patients with exfoliation glaucoma than in those with POAG: 6.6 versus 2.5 mm Hg ( P = .003). However, this difference was not significant in the multivariate analysis (see below and Table 2 ). The mean of all IOP slopes over time was slightly positive (0.19 mm Hg/year ± 0.75) for the 113 patients (89.7%) with 5 or more IOP measurements after ALT ( Figure 1 ). Compared with those subjects, the 13 patients (10.3%) with fewer than 5 such measurements had greater mean IOP slopes, although those results might have been sensitive to measurement errors and likely were influenced by the selected characteristics and short follow-up of these 13 cases ( Figure 1 ).
| Variables | No. | Stepwise Multiple Linear Regression β ( P Value) | Univariate 2-Way ANOVA P Value a ( P Value) b |
|---|---|---|---|
| IOP immediately before ALT (mm Hg) | 127 | 0.683 (<.001) | <.001 (<.001) |
| Baseline perimetric deviation | 127 | −0.092 (.182) | — |
| Trabecular pigmentation grade | |||
| 0 | 7 | 0.002 (.981) | — |
| 1 | 50 | ||
| 2 | 62 | ||
| 3 | 8 | ||
| Exfoliations | |||
| Yes | 8 | 0.072 (.297) | — |
| No | 118 | ||
| Surgeon | |||
| 1 | 23 | — | .001 (.01) |
| 2 | 8 | ||
| 3 | 54 | ||
| 4 | 11 | ||
| 5 | 9 | ||
| 6 | 9 | ||
| 7 | 13 |
a Total IOP reduction achieved by ALT as dependent variable.
b Percentage IOP reduction achieved by ALT as dependent variable.
Factors Influencing Intraocular Pressure Reduction
IOP before ALT was a significant factor influencing the magnitude of IOP reduction ( Table 2 and Figures 2 and 3 ). Every 3 mm Hg of higher IOP before ALT was associated with an additional IOP reduction of approximately 2 mm Hg ( P < .001; r 2 = 0.47). Patients with an IOP before ALT of less than 15 mm Hg had no mean pressure lowering ( Figures 2 and 3 ).
