Br J Ophthalmol: first published as 10.1136/bjo.72.12.892 on 1 December 1988. Downloaded from British Journal of Ophthalmology, 1988, 72, 892-896 Levobunolol compared with timolol: a four-year study ORNA GEYER,' MOSHE LAZAR,' GARY D NOVACK,23 DAVID SHEN,3 AND CHERYL Y ETO3 From the 'Department ofOphthalmology, Ichilove Medical Center, Sackler School ofMedicine, University of Tel Aviv, Israel; the 2University of California, Irvine, California, USA; and the 3Department of Clinical Research, Allergan Pharmaceuticals, Irvine, California, USA SUMMARY Fifty-one patients with raised intraocular pressure (IOP) were treated for up to four years with one of three ophthalmic solutions: 0 5 % levobunolol, 1% levobunolol, or 0-5 % timolol. The study was conducted as a double-masked, randomised trial in which medications were administered twice daily to both eyes. Levobunolol and timolol were equally effective in reducing overall mean IOP; reductions were greater than 8 8 mmHg in all three treatment groups. The study showed levobunolol to be as safe and effective as timolol in the long-term control of raised TOP. copyright. Glaucoma is a chronic disease, generally requiring a then instructed to administer a single drop of the lifetime of treatment. Most evaluations of new study medication into each eye twice daily (9 am and agents, however, are of relatively limited durations. 9 pm) for the duration ofthe study. The cup-disc ratio In the present report we describe our findings in a was measured subjectively with a direct ophthal- four-year, double-masked comparison of levo- moscope and visual fields by Goldmann kinetic bunolol, a new 131432-adrenoceptor antagonist, and perimetry. This is a to our IOP values for right and left eyes were averaged for timolol. follow-up previous one-year http://bjo.bmj.com/ report of this comparison.' analysis. Owing to differences at baseline an analysis of covariance model (ANCOVA) was used to Subjects and methods analyse mean changes of IOP from baseline.2 Within- group changes from baseline were tested for signific- As detailed previously,' this was a randomised, ance with the paired t test. Overall mean changes double-masked comparison of three treatments: from baseline were tested by analysis of variance 0.5% levobunolol, 1% levobunolol, or 0-5% timolol. (ANOVA) for repeated measures design.3 Kaplan- All were administered twice daily into both eyes. Meier survival curves were used to evaluate time to on September 30, 2021 by guest. Protected Patients with chronic open-angle glaucoma or ocular treatment failure due to uncontrolled IOP.4 Survival hypertension characterised by untreated IOP values curves were compared among the groups by the log- of 23 mmHg or higher in both eyes were considered rank method.5 Results were considered statistically for the study. Excluded from participation were significant if p values were -0.05. patients with contraindications to the topical or systemic use of beta blockers, those with secondary Results or narrow-angle glaucoma or aphakia, and those who had used systemic beta blockers within the three Fifty-one patients participated in the study. One months immediately preceding the study. patient was diagnosed as having ocular hypertension; The study consisted of 35 visits including a baseline the rest had open-angle glaucoma. Other demo- examination. After the baseline examination and as graphic characteristics are presented in Table 1. close as possible to 9 am the patients received a single As shown in Fig. 1, the cumulative probability of drop of the test medication into each eye. They were successful control of IOP (life-table analysis) for four Correspondence to Dr Gary D Novack, Nelson Research, 1001 years was approximately 40% for each treatment Health Sciences Road West, California 92715, USA. group. There was no significant difference among 892 Br J Ophthalmol: first published as 10.1136/bjo.72.12.892 on 1 December 1988. Downloaded from Levobunolol compared with timolol: afour-yearstudy 893 100 - ... o 0.5% Levobunolol 90 - * 1.0% Levobunolol - 80 * 0.5% Timolol J-Qc) 0~ 70 - 2 Fig. 1 Cumulative probability of a_- successful control ofIOP - c) 60 - approximately 40% in each > treatment group. 3 50- v.. 0- .0....-. 40 - JUin 30 60 90 120 150 180 210 Week groups. Most of the efficacy failures occurred within Table 2 Status ofpatients at end offour years the first two years of the study. The status of all the patients at the end of the study 0-5% 10% 0-5% Total is shown in Table 2. In addition to the patients whose Levobunolol Levobunolol Timolol treatment was terminated owing to inadequate con- Successfully 4 (23%) 4 (24%) 6 (35%) 14 (27%) trol of IOP, approximately 8% had their treatment completed copyright. Terminated: terminated for drug-related adverse effects. These Inadequate 7 (41%) 7 (41%) 9 (53%) 23 (45%) were in the two levobunolol groups and included control of three reports of topical allergies to levobunolol and lOp one report of dyspnoea. All adverse effects occurred Terminated: Drug-related 3 (18%) 1 (6%) 0 (0%) 4 (8%) after approximately one year of treatment. adverse As shown in Fig. 2, there was a significant decrease experience of mean IOP from baseline in all three treatment Discontinued: Not drug 3 (18%) 5 (29%) 2 (12%) 10 (20%) groups throughout the four years of the study. http://bjo.bmj.com/ related Overall, the mean decrease in IOP was 9-5 mmHg, Total 17 17 16 50 9*6 mmHg, and 8-8 mmHg in the 0*5% levobunolol, 1% levobunolol and 0-5% timolol groups, respect- ively. There were no significant differences among the groups. As shown in Table 3, there were reports of clinically significant increases in cup/disc ratio (¢0-2) and in progressive visual field loss in approxi- Table 1 Study population* mately one-half of the patients. The treatment on September 30, 2021 by guest. Protected Variable 0-5% 1% 0-5% groups were similar in this respect. Levobunolol Levobunolol Timolol Overall, the mean changes in pupil size were less (n=17) (n=17) (n=17) than +0-25 mm. Mean decreases in Schirmer tests Age (years) were less than 2-5 mm. No changes were seen in the Mean±SD 70-9±6-9 70-8±9-1 69-1±6-7 Sex Male 10 10 12 Table 3 Cup-disc ratio increases and visualfield losses Female 6 7 5 Not reported 1 0 0 0.5% 1% 0.5% Iris colour Levobunolol Levobunolol Timolol Blue 0 1 3 Green 6 7 6 Cup-disc ratios 7 (44%0) 9 (53%) 11 (65%o) Brown 11 9 8 increase-<02 Glaucoma diagnosis Glaucomatous visual 6 (38%) 3 (18%) 10 (59%) Open-angle glaucoma 16 (94%) 17 (100%) 17 (100%) field losses* Ocular hypertension 1 (6%o) 0 (0%) 0 (0%) *Defined as Bjerrum's scotoma, nasal step, generalised constriction, *No significant differences among the groups were noted. or enlarged blind spot not present at baseline examination. Br J Ophthalmol: first published as 10.1136/bjo.72.12.892 on 1 December 1988. Downloaded from 894 Orna Geyer, Moshe Lazar, Gary D Novack, David Shen, and Cheryl YEto 34- o 0.5% Levobunolol I 30- * 1% Levobunolol_ E * 0.5% Timolol II.- u 26- cn C'Q) 0O 22- 0 o 18 - U) IA 14.- .II 1 0 30 60 90 120 150 180 210 Week copyright. Fig. 2 Significant decrease ofmean IOP in each treatment group. 95 - http://bjo.bmj.com/ 85 ~ _ 0- 75- on September 30, 2021 by guest. Protected 4-, I--, 0 65- 0 U) o 0.5% Levobunolol 55- * 1% Levobunolol _ * 0.5% Timolol 41,1 I I 0 30 60 90 120 150 180 210 Week Fig. 3 Mean heart rate decreased in first week oftherapy, but decreases did not subsequently change. Br J Ophthalmol: first published as 10.1136/bjo.72.12.892 on 1 December 1988. Downloaded from Levobunolol compared with timolol: afour-year study 895 175 - 165 - 155 - 145 - E "I 135 - 125 - 0.5% Levobunolol 115 - co 1% Levobunolol _ co 105 - * 0.5% Timolol m 95 - 85 - :3 75- 65 - IJFN 9;1. I 0 90 30 60 120 150 180 210 copyright. Week Fig. 4 Decreases in mean systolic and diastolic blood pressure in each treatment group. blink test. None of these changes was of clinical or resolved within one month. The third, who used 1% statistical significance. levobunolol, reported dizziness at three visits. While decreases in mean heart rate of 5-10 bpm with the first week Discussion were observed of therapy (Fig. 3), http://bjo.bmj.com/ these decreases did not change significantly over time. There were decreases in mean systolic and In this study levobunolol was as safe and effective as diastolic blood pressure in all treatment groups (Fig. timolol for the long-term treatment of raised IOP. 4). There were no significant differences among The long-term efficacy failure rate of approximately groups in overall mean changes in heart rate or blood 15% per year was not only similar for levobunolol pressure (Table 4). In general, the incidence of and timolol in this study but similar to that in a two- clinically significant biomicroscopic and ophthalmo- year study comparing the same treatments.6 Interest- scopic findings were similar among the groups. ingly, most of the efficacy failures occurred in the first on September 30, 2021 by guest. Protected Patients' complaints were few, numbering only two years. Outside the treatment protocol we found three. One patient using 0.5% timolol had epiphora that these patients could be controlled with the at one visit. Another using 0-5% levobunolol addition of pilocarpine or dipivefrin.