Travoprost compared with other prostaglandin analogues or timolol in patients with open angle glaucoma or ocular hypertension: a meta-analysis of randomised controlled trials

Methodological quality of the review: Medium confidence

Author: Li N, Chen XM, Zhou Y, Wei ML, Yao X

Geographical coverage: United States of America (USA), Brazil, Australia and Italy

Sector: Open angle glaucoma, ocular hypertension

Sub-sector: Side effects

Equity focus: None specified

Review type: Effectiveness review

Quantitative synthesis method: Meta-analysis

Qualitative synthesis method: Not applicable

Background: The effects and side effects of travoprost are still uncertain compared with other prostaglandin analogues or timolol in patients with open-angle glaucoma or ocular hypertension.

Objectives: The objective of this meta-analysis of randomised controlled trials (RCTs) was to evaluate the incidence of reported side-effects and intra-ocular pressure (IOP)-lowering effect of travoprost versus other prostaglandin analogues (latanaprost, bimatoprost, unoprostone) or timolol.

Main findings: In this meta-analysis, 12 articles including 3048 patients with open angle glaucoma (OAG) or ocular hypertension (OH) were included. Studies included were conducted in the USA, Italy, Brazil and Australia.

The combined results showed that travoprost 0.004% was more effective than timolol or travoprost 0.0015% in lowering IOP, but not more effective than bimatoprost or latanoprost. Travoprost 0.004% caused a higher percentage of hyperemia than timolol, latanoprost or travoprost 0.0015%.

Authors noted that there was an increased incidence of pigmentation with travoprost than timolol. Travoprost 0.004% caused a higher percentage of eyelash changes than timolol, latanoprost or travoprost 0.0015%. Authors concluded that travoprost is more effective than timolol in lowering IOP in patients with OAG or OH. Compared with other prostaglandin analogues, travoprost appears to be equivalent to bimatoprost and latanoprost. In addition, authors stated that although a limited number of local side-effects were reported, no serious treatment-related side-effects were reported.

Methodology: PUBMED, EMBASE, Chinese Bio-medicine Database and Cochrane Controlled Trials Register were used to identify the potentially relevant RCTs. The search terms used were travatan or travoprost. Literature reference proceedings were hand-searched at the same time. Titles and abstracts title were screened for potentially relevant articles. Two reviewers conducted searches independently, and the results were combined.

The outcome measures were the incidence of reported side-effects (hyperemia, iris pigmentation, eyelash changes) and mean IOP pooled over treatment visits. Inclusion and exclusion criteria included:

• RCTs
• Patients with OAG (including primary and secondary OAG) or OH
• Intervention – travoprost versus other prostaglandin analogues or timolol, initiated at the same time and with the same other treatment
• At least one of the following primary outcome variables: IOP, side effects

Only articles written in English or Chinese were included. Quality assessment followed Cochrane Eyes and Vision Protocol Development Guidelines. Two authors independently extracted data of included studies. Heterogeneity was examined, however authors did not assess publication bias due to the small number of studies included.

Applicability/external validity: Authors did not discuss the applicability/external validity of results.

Geographic focus: Most studies were from the USA, two from Italy, one from Australia and one from Brazil.

Summary of quality assessment: Overall, there is medium confidence in the conclusions about the effects of this study. Authors used appropriate methods to screen studies for inclusion, extract data and quality-assess included studies. In addition, authors used appropriate methods to analyse findings of included studies. However, authors did not conduct a thorough search of the literature: authors did not avoid language bias, did not search for grey literature and did not contact authors/experts as part of the search strategy. This indicates the presence of publication bias within the review which may impact of the validity of the overall findings.