Purpose
To evaluate control of intraocular pressure (IOP) and IOP fluctuation in patients with ocular hypertension or glaucoma treated with fixed-combination brimonidine–timolol compared with brimonidine or timolol monotherapy.
Design
Post hoc analysis of data from 2 identical, 12-month, randomized, double-masked, multicenter trials.
Methods
Patients were treated bilaterally with fixed brimonidine–timolol twice a day (n = 385), brimonidine tartrate 0.2% 3 times a day (n = 382), or timolol 0.5% twice a day (n = 392). Diurnal IOP was measured at follow-up visits at weeks 2 and 6 and months 3, 6, 9, and 12. IOP fluctuation was defined as the standard deviation of IOP measurements.
Results
The percentage of patients with mean diurnal IOP <18 mm Hg and short-term (daily) IOP fluctuation ≤2 mm Hg was statistically significantly higher in the brimonidine–timolol group than in the brimonidine or timolol group at each follow-up visit (at month 12, brimonidine–timolol 43.0%; brimonidine 18.9%, timolol 33.5%, P ≤ .017). At each hour (8 AM, 10 AM, 3 PM, and 5 PM), the percentage of patients with mean IOP <18 mm Hg and long-term (intervisit) IOP fluctuation ≤2 mm Hg was statistically significantly higher with brimonidine–timolol than with brimonidine or timolol alone (at 8 AM, brimonidine–timolol 41.0%, brimonidine 11.3%, timolol 23.7%, P < .001).
Conclusions
Patients treated with fixed-combination brimonidine–timolol were more likely than patients treated with either brimonidine or timolol alone to achieve a combination of low mean IOP and low short-term (daily) or long-term (intervisit) IOP fluctuation.
Intraocular pressure (IOP) is a well-established risk factor for the development and progression of glaucoma. Clinical studies have demonstrated that IOP lowering is effective in reducing the development of glaucoma in ocular hypertension (OHT) patients and in slowing the progression of glaucomatous visual field (VF) loss in patients with normal or elevated IOP. Evidence from these studies has suggested that consistent achievement of a low target pressure may minimize VF deterioration.
The effect of IOP fluctuation on the risk of VF progression in glaucoma is currently under debate. Several studies have suggested that long-term fluctuation in IOP is related to the rate of VF loss in glaucoma. In the Advanced Glaucoma Intervention Study (AGIS), intervisit fluctuation in IOP was a strong risk factor for visual field progression independent of the level of IOP. An independent association between long-term IOP fluctuation and the risk of glaucoma progression was also reported in a cohort study of 151 patients treated at 12 independent practices. Finally, in a study of 408 glaucoma patients with postoperative pressures consistently <18 mm Hg following combined trabeculectomy, cataract extraction, and posterior chamber lens implantation, those patients with smaller or larger long-term IOP fluctuation had a similar low mean IOP (approximately 11 mm Hg), but the patients with larger long-term IOP fluctuation (defined as a standard deviation [SD] of IOP measurements >2 mm Hg) had increased VF deterioration. In contrast, long-term fluctuation in IOP was not related to glaucomatous progression in the Early Manifest Glaucoma Trial, in which only mean follow-up IOP was found to be a significant risk factor for progression. Similarly, studies in OHT patients have reported no association between long-term IOP fluctuation and the risk of conversion to glaucoma. These disparate findings may be explained by differences in the patient populations studied and the treatments used. In a recent analysis of data from the AGIS, intervisit fluctuation in IOP was a predictive factor for progression in the tercile of patients with the lowest mean IOP but not in the tercile with the highest mean IOP. These results are consistent with previous findings and suggest that long-term IOP fluctuation may be an important risk factor for VF progression in patients with lower IOP, whereas the level of IOP may be the predominant risk factor for progression in patients with higher IOP.
A fixed combination of brimonidine tartrate 0.2% and timolol maleate 0.5% (brimonidine–timolol) has recently become available for lowering IOP in glaucoma and OHT. Fixed brimonidine–timolol has been shown to reduce IOP as effectively as concomitant treatment with separate bottles of brimonidine and timolol. Further, the fixed combination has been demonstrated to reduce IOP more effectively than either brimonidine or timolol monotherapy and to have an improved tolerability profile compared with brimonidine monotherapy. The incidence of ocular allergy, a common side effect of brimonidine treatment, has been shown to be reduced with twice-daily fixed brimonidine–timolol compared with brimonidine monotherapy, regardless of whether brimonidine monotherapy is dosed 2 or 3 times daily.
Because both IOP fluctuation and mean IOP may affect the risk of glaucomatous progression, a post hoc analysis was undertaken to evaluate control of short-term and long-term IOP fluctuation as well as mean IOP in glaucoma and OHT patients treated with fixed brimonidine–timolol, brimonidine, or timolol in a previously reported 12-month comparison study.
Methods
Two 12-month, randomized, phase 3, double-masked, parallel-group clinical trials compared fixed-combination brimonidine tartrate 0.2%–timolol maleate 0.5% ophthalmic solution (Combigan) with its component medications used separately as monotherapy in patients with glaucoma or ocular hypertension. The trials had the same design and data were pooled for analysis. All drugs were from Allergan, Inc (Irvine, California, USA). Details of the study methods and results have been previously published. Briefly, following an appropriate washout period of any previous ocular hypotensive medication, a total of 1159 patients were treated with fixed-combination brimonidine tartrate 0.2%–timolol maleate 0.5% (brimonidine–timolol) twice daily, brimonidine tartrate 0.2% (Alphagan) 3 times daily, or timolol maleate 0.5% twice daily in each eye for 12 months. IOP was measured at 8 AM (before instillation of study drug), 10 AM, 3 PM (before instillation of study drug or vehicle), and 5 PM at baseline, weeks 2 and 6, and months 3, 6, and 12. At month 9, IOP was measured only at 8 AM and 10 AM.
Analyses of IOP fluctuation were based on the eye with the larger baseline IOP fluctuation and used observed values in the intent-to-treat population, consisting of all randomized patients. To avoid biasing the results toward lower values for IOP fluctuation, there was no imputation for missing values. Each analysis included all patients for whom complete data were available at all time points. In the primary analyses, IOP fluctuation was defined as the SD of IOP measurements. In supplementary analyses (not shown), IOP fluctuation was defined as the range of IOP measurements to confirm that the conclusions drawn were not dependent on the use of the SD as the measure of IOP fluctuation.
For each patient, at each study visit, the mean diurnal IOP (the mean of the IOP measurements at all hours) was calculated. Short-term, daily IOP fluctuation was defined as the SD associated with the mean diurnal IOP. The mean IOP over the 6 follow-up visits was calculated at each hour (8 AM, 10 AM, 3 PM, and 5 PM). Long-term, intervisit IOP fluctuation at a particular hour was defined as the SD associated with this mean IOP. The percentage of patients in each group demonstrating short-term or long-term IOP fluctuation of ≤2 mm Hg was evaluated. The primary outcome measure was composite endpoints for control of both IOP fluctuation (short-term or long-term) and IOP, defined as IOP fluctuation of ≤2 mm Hg plus attainment of a target mean IOP of <18 mm Hg. These cutoffs were selected based on a paper by Hong and associates in which patients with IOP <18 mm Hg after a triple procedure who also had IOP fluctuation of ≤2 mm Hg experienced a significant reduction in VF deterioration. Secondary sensitivity analyses used IOP fluctuation cutoffs of ≤3 mm Hg and ≤4 mm Hg for the composite endpoints.
Statistical analyses were performed using SAS version 9.1 (SAS Institute Inc, Cary, North Carolina, USA) and a 2-sided alpha level of 0.05. The proportions of patients demonstrating IOP fluctuation ≤2 mm Hg, or achieving a composite endpoint, were compared between groups using χ 2 tests or, when at least 25% of the expected cell counts were less than 5, Fisher exact tests. Baseline mean diurnal IOP and short-term IOP fluctuation were analyzed using a 2-way analysis of variance model with treatment and site as factors. Comparisons between the treatments were based on the least squares means from this model. Pairwise comparisons were performed for fixed brimonidine–timolol vs brimonidine and fixed brimonidine–timolol vs timolol, as these were the differences of interest.
Power estimates for the primary outcomes were calculated using procedure PTTOU in nQuery Advisor (Statistical Solutions, Saugus, Massachusetts, USA). The estimated power of the analysis of the composite endpoint of short-term IOP fluctuation ≤2 mm Hg and mean diurnal IOP <18 mm Hg ranged from 52% to 83% across visits for fixed brimonidine–timolol vs brimonidine and was >99% at each visit for fixed brimonidine–timolol vs timolol. The estimated power of the analysis of the composite endpoint of long-term IOP fluctuation ≤2 mm Hg and mean IOP <18 mm Hg at a particular hour across visits ranged from 78% to >99% across hours for fixed brimonidine–timolol vs brimonidine and from 83% to >99% across hours for fixed brimonidine–timolol vs timolol.
Results
Baseline Intraocular Pressure and Fluctuation
At untreated baseline, the average mean diurnal IOP in the eye with the worst fluctuation was 23.0 mm Hg in the fixed brimonidine–timolol group, 23.3 mm Hg in the brimonidine group, and 23.3 mm Hg in the timolol group. The differences in mean diurnal IOP between the fixed brimonidine–timolol group and the brimonidine or timolol groups were not statistically significant ( P ≥ .128). The mean daily IOP fluctuation at untreated baseline was also similar between the fixed brimonidine–timolol group (2.3 mm Hg) and the brimonidine (2.2 mm Hg) and timolol (2.3 mm Hg) groups ( P ≥ .273).
Efficacy in Intraocular Pressure Lowering
The previously published analysis of IOP-lowering efficacy in this study used mean values from both eyes as the IOP value for the individual patient. In the present analysis based on the (worse) eye with the larger baseline IOP fluctuation, the percentage of patients that achieved a target mean diurnal IOP of <18 mm Hg was statistically significantly larger in the fixed brimonidine–timolol group than in the brimonidine or timolol groups at each follow-up visit ( Figure 1 ).
Short-Term (Daily) Intraocular Pressure Fluctuation During Treatment
Complete data from at least 278 of 385 patients (72.2%) in the fixed brimonidine–timolol group, 204 of 382 patients (53.4%) in the brimonidine group, and 315 of 392 patients (80.4%) in the timolol group were available for each of the analyses of short-term and long-term IOP fluctuation. Early discontinuations of patients from the study accounted for most of the missing data. As reported previously, the most common reason for early discontinuation from the study was adverse events (14.3% of brimonidine–timolol patients, 30.6% of brimonidine patients, and 5.1% of timolol patients).
By chance, at baseline there was a trend for a lower percentage of patients in the fixed brimonidine–timolol group than in the brimonidine group or timolol group having short-term (daily) IOP fluctuation of ≤2 mm Hg ( P ≤ .088, Table 1 ). Nonetheless, a statistically significantly higher percentage of patients in the fixed brimonidine–timolol group than in the brimonidine group had daily IOP fluctuation of ≤2 mm Hg at each follow-up visit ( P ≤ .002, Table 1 ). The differences in daily IOP fluctuation between fixed brimonidine–timolol and timolol were not statistically significant at weeks 2 and 6 or at month 6. At months 3, 9, and 12, patients in the timolol group were statistically significantly more likely than patients in the fixed brimonidine–timolol group to have daily IOP fluctuation ≤2 mm Hg ( P ≤ .020, Table 1 ).
| Visit | Treatment Group | P Value | |||
|---|---|---|---|---|---|
| Brimonidine–Timolol N = 384 | Brimonidine N = 381 | Timolol N = 392 | Brimonidine–Timolol vs Brimonidine | Brimonidine–Timolol vs Timolol | |
| Baseline | 41.4% (157/379) | 47.7% (180/377) | 47.5% (184/387) | .080 | .088 |
| Week 2 | 64.5% (240/372) | 38.3% (137/358) | 70.1% (263/375) | <.001 | .102 |
| Week 6 | 67.7% (247/365) | 40.4% (135/334) | 73.7% (269/365) | <.001 | .074 |
| Month 3 | 65.2% (234/359) | 39.9% (124/311) | 73.2% (260/355) | <.001 | .020 |
| Month 6 | 66.3% (218/329) | 34.8% (93/267) | 72.3% (248/343) | <.001 | .089 |
| Month 9 b | 60.3% (184/305) | 47.0% (110/234) | 75.1% (253/337) | .002 | <.001 |
| Month 12 | 61.3% (174/284) | 39.6% (84/212) | 71.4% (232/325) | <.001 | .008 |
a Short-term daily fluctuation in intraocular pressure is the standard deviation of the diurnal intraocular pressure measurements taken at a particular visit.
b At month 9 diurnal intraocular pressure measurements were taken only at 8 AM and 10 AM; at all other visits measurements were taken at 8 AM, 10 AM, 3 PM, and 5 PM.
Long-Term (Intervisit) Intraocular Pressure Fluctuation During Treatment
As shown in Table 2 , the percentage of patients with long-term (intervisit) IOP fluctuation during follow-up of ≤2 mm Hg was statistically significantly higher in the fixed brimonidine–timolol group than in the brimonidine group at each hour (8 AM, 10 AM, 3 PM, and 5 PM; P ≤ .023). The differences in long-term IOP fluctuation between fixed brimonidine–timolol and timolol were not statistically significant at 10 AM, 3 PM, or 5 PM. However, a statistically significantly higher percentage of patients in the fixed brimonidine–timolol group than in the timolol group had long-term IOP fluctuation of ≤2 mm Hg at the 8 AM measurements ( P = .014).
| Hour | Treatment Group | P Value | |||
|---|---|---|---|---|---|
| Brimonidine–Timolol N = 288 | Brimonidine N = 213 | Timolol N = 327 | Brimonidine–Timolol vs Brimonidine | Brimonidine–Timolol vs Timolol | |
| 8 AM | 69.8% (201/288) | 55.4% (118/213) | 60.3% (196/325) | <.001 | .014 |
| 10 AM | 67.4% (188/279) | 57.3% (121/211) | 68.5% (220/321) | .023 | .763 |
| 3 PM | 67.3% (189/281) | 53.1% (110/207) | 68.3% (220/322) | .002 | .780 |
| 5 PM | 70.1% (195/278) | 56.4% (115/204) | 69.8% (220/315) | .002 | .936 |
a Long-term intervisit fluctuation in intraocular pressure is the standard deviation of the intraocular pressure measurements at that hour across all follow-up visits.
Short-Term Control of Both Intraocular Pressure and Fluctuation
At each follow-up visit, a statistically significantly higher percentage of patients in the fixed brimonidine–timolol group than in the brimonidine group or the timolol group achieved both a mean diurnal IOP <18 mm Hg and daily IOP fluctuation ≤2 mm Hg ( P ≤ .044; Figure 2 ). Sensitivity analyses using the composite endpoints of a mean diurnal IOP <18 mm Hg and daily IOP fluctuation ≤3 mm Hg or ≤4 mm Hg also showed better short-term control of both IOP and IOP fluctuation with fixed brimonidine–timolol ( Supplemental Figures 1 and 2 , available online at AJO.com ). A statistically significantly higher percentage of patients in the fixed brimonidine–timolol group than in either the brimonidine group or the timolol group demonstrated both mean diurnal IOP <18 mm Hg and short-term IOP fluctuation ≤3 mm Hg or ≤4 mm Hg at each follow-up visit.
Long-Term Control of Both Intraocular Pressure and Fluctuation
At each hour (8 AM, 10 AM, 3 PM, and 5 PM), a statistically significantly higher percentage of patients in the fixed brimonidine–timolol group than in the timolol group achieved both mean IOP <18 mm Hg at that hour across visits and long-term IOP fluctuation ≤2 mm Hg ( P ≤ .006; Figure 3 ). Patients in the fixed-combination group were also statistically significantly more likely than those in the brimonidine group to achieve both the target pressure of <18 mm Hg and long-term IOP fluctuation ≤2 mm Hg at 8 AM, 10 AM, 3 PM, and 5 PM ( P ≤ .003; Figure 2 ). Results of the sensitivity analyses using different cutoffs for IOP fluctuation led to similar conclusions ( Supplemental Figures 3 and 4 , available online at AJO.com ). The percentage of patients achieving mean IOP <18 mm Hg and long-term IOP fluctuation ≤3 mm Hg or ≤4 mm Hg at a particular hour across follow-up was statistically significantly higher in the fixed brimonidine–timolol group than in the timolol group at all hours. The percentage of patients achieving mean IOP <18 mm Hg and long-term IOP fluctuation ≤3 mm Hg or ≤4 mm Hg at a particular hour across follow-up was statistically significantly higher in the fixed brimonidine–timolol group than in the brimonidine group at all hours except 5 PM. At 5 PM, the peak effect of the additional afternoon brimonidine dose was compared with the near-trough effect of fixed brimonidine–timolol.
