This review concluded that bimatoprost is more effective in reducing morning elevated intraocular pressure (IOP) than latanoprost, and at achieving the target IOP at 3 months. Both treatments were generally well tolerated. Given several considerations with the included studies, such as clinical differences between the studies and limitations with the data synthesis, the authors’ conclusions may not be reliable.

To compare the efficacy and tolerability of bimatoprost with latanoprost in treating patients with elevated intraocular pressure (IOP).

PubMed, EMBASE, the Chinese Biomedical Database, and the Cochrane Controlled Trials Register were searched up to February 2008 for publications in English; the search terms were reported. In addition, the references of original articles were screened, and abstracts of society meetings and databases of relevant pharmaceutical manufacturers were also searched.

Randomised controlled trials (RCTs) comparing bimatoprost and latanoprost monotherapy in patients with glaucoma (excluding those with normal tension glaucoma) or ocular hypertension were eligible for inclusion. Patients with glaucoma were required to have an elevated IOP (>21 mmHg) without medication and a glaucomatous visual field, optic disc changes, or retinal nerve fibre layer (RNFL) defects. Patients with ocular hypertension were required to have an IOP >21 mmHg without medication and a normal visual field, optic disc and RNFL. Eligible studies were required to include a washout period, with a post washout baseline IOP measuring between 22 and 38 mmHg. The included studies compared bimatoprost 0.03% eyedrops with latanoprost 0.005% eyedrops, administered as one drop in the evening in one or both eyes, and were carried out in Brazil, China, Greece, Israel, Italy, Turkey and the USA. The majority of patients were female and mean ages ranged from 52 to 71 years. The most prevalent diagnosis was primary open-angle glaucoma, but patients with exfoliative glaucoma, other types of glaucoma (not including pigmentary glaucoma, chronic angle-closure glaucoma, or pseudo phakic/aphakic glaucoma), and ocular hypertension were also included.

Studies reporting the percentage reduction in morning mean IOP (measured between 8 am and 10 am) or diurnal mean IOP as the primary outcome, or the proportion of patients achieving the target IOP (≥17 mmHg) as the secondary outcome, were eligible for inclusion. The included studies also reported drug-related adverse events (defined as those appearing immediately after drug administration and subsequently resolved) such as iris pigmentation, dry eye, eye irritation, eye pain, pruritis and visual disturbance, and ocular adverse events such as conjunctival hyperemia, ocular inflammation and cystoid macular oedema.

Two reviewers screened studies for relevance.

Validity was assessed according to the Jadad et al. criteria (randomisation, methods of concealment, study design) and the number of withdrawals. The included studies received a score out of 5, with a score of 3 or more denoting good quality.

Two reviewers independently performed the validity assessment.

For continuous data on morning and diurnal IOP, percentage changes in baseline and end point IOP were calculated, and for target IOP, the proportion of patients achieving target were extracted, ultimately to calculate the mean reduction in IOP (IOPR) with 95% confidence intervals (CIs). Data were extracted on an intention-to-treat basis, using a customised data extraction form, and were stratified by duration of treatment: 1 month (including observations obtained between 30 days and 7 weeks), 3 months (including observations between 12 weeks and 3 months) and 6 months. Dichotomous data on adverse events were converted into rate differences (RDs) with 95% CIs.

Two reviewers extracted the data and any disagreements were resolved by discussion.

The RDs were pooled using the DerSimonian and Laird random-effects model. Continuous variables were pooled using weighted mean differences (WMDs). Heterogeneity was assessed using Cochran’s Q statistic and the I² test. Sensitivity analyses were undertaken by removing industry-sponsored studies.

Thirteen RCTs (n=1,302 in the meta-analysis; 760 patients received bimatoprost and 757 received latanoprost) were included in the review: three double-blind parallel, two double-blind crossover, seven single-blind parallel and one single-blind crossover. Sample sizes ranged from 30 to 272 patients and withdrawal rates were between 0 and 12.9%.

Four studies scored 5 points on the Jadad scale and 9 studies scored 4.

Morning IOPR (12 studies): overall, IOPRs were lower with bimatoprost than with latanoprost at each time point; the WMD was 2.59% (95% CI: 0.81, 4.37, p=0.004) at 1 month, 2.41% (95% CI: 0.58, 4.25, p=0.01) at 3 months and 5.60% (95% CI: 2.95, 8.26, p<0.001) at 6 months. Sensitivity analyses to remove industry-sponsored trials did not significantly alter the results at 6 months (2 studies), but at 1 month (3 studies) and 3 months (1 study) the differences were no longer significant.

Diurnal IOPR (6 studies): no significant differences were reported at 1 month (3 studies) or 3 months (3 studies).

IOP target (5 studies): 2 studies reported a significant difference between the two treatment groups, with more bimatoprost patients achieving IOP target at 3 months than latanoprost patients (RD 12%, 95% CI: 4.0, 21.0, p=0.004). No other time points showed significant differences between the two groups.

Adverse events: conjunctival hyperemia was the most prevalent adverse event, with the bimatoprost group reporting greater incidence rates than the latanoprost group (RD 20%, 95% CI: 15.0, 24.0, p<0.001). No other significant differences in adverse events were reported between the two groups.

There was significant heterogeneity among studies reporting diurnal IOP at 1 month (p=0.004) and among studies reporting pruritis (p<0.001).

Bimatoprost is significantly more effective in reducing morning IOP than latanoprost at all time points and at achieving the target IOP at 3 months. Both treatments were well tolerated, although there was a significantly greater incidence of conjunctival hyperemia when using bimatoprost compared with latanoprost.

The review question was clear and was supported by appropriate inclusion criteria for the participants, interventions, outcomes and study design. A relevant literature search was undertaken using four electronic databases and other appropriate sources. However, studies were restricted by language, which means that language bias might have been introduced, and were also restricted to those published. Validity was assessed using published criteria and the included studies were reported to be of good quality. Attempts were made to minimise reviewer error and bias at each stage of the review. Appropriate methods were used to investigate heterogeneity but, as the authors reported significant clinical heterogeneity, it may not have been appropriate to pool the results. The authors also reported potential selection bias with the included studies, with patient selection based on previous treatment responses. Furthermore, sample sizes were small, CIs appeared wide, and follow-up was only short term. Given these considerations and limitations with the data synthesis, the authors' conclusions may not be reliable.

Practice: The authors did not state any implications for clinical practice.

Research: The authors stated that longer term, non-industry-sponsored RCTs are required to assess the efficacy of bimatoprost in patients with glaucoma and ocular hypertension, particularly in patients who have not responded adequately to treatment with latanoprost. Further long-term research is also required to examine the risk of rare but serious adverse effects in patients using topical prostaglandin analogues.

Not stated.

This record is a structured abstract written by CRD reviewers. The original has met a set of quality criteria. Since September 1996 abstracts have been sent to authors for comment. Additional factual information is incorporated into the record. Noted as [A:....].