Twenty-one RCTs (26 pairwise comparisons, range 26 to 690 patients) were included in the meta-analysis. Seven trials were double-blinded, 11 were single blinded, one was an open trial and information on blinding was not available for two trials. Reported follow-up ranged from three months to 24 months.
There was no statistically significant difference in mean IOP at three months between T2345 and polyquaternium-1-travoprost, BAK-bimatoprost 0.03%, BAK-bimatoprost 0.01%, BAK-travoprost, or BAK-latanoprost. There was a statistically significant difference in favour of T2345 when compared with BAK-tafluprost (WMD -0.9 mmHg, 95% CI -1.52 to -0.28).
Risk of hyperaemia was statistically significantly lower with T2345 than with polyquaternium-1-travoprost (OR 0.24, 95% CI 0.11 to 0.55), sofzia-travoprost (OR 0.37, 95% CI 0.16 to 0.84) BAK-bimatoprost 0.03% (OR 0.18, 95% CI 0.10 to 0.33), BAK-bimatoprost 0.01% (OR 0.27, 95% CI 0.13 to 0.56), BAK-tafluprost (OR 0.18, 95% CI 0.05 to 0.65), BAK-travoprost (OR 0.25, 95% CI 0.14 to 0.46) and BAK-latanoprost (OR 0.52, 95% CI 0.31 to 0.86).
The exploratory non-inferiority analysis results suggested that T2345 was also noninferior to all other prostaglandin analogues except polyquaternium-1-travoprost. In the indirect comparison of T2345 against polyquaternium-1-travoprost, the upper limit was 1.51 mmHg (which was above the widely used limit of noninferiority of 1.5 mmHg).
Publication bias was not assessed due to small number of the included trials for each comparison.