方法：计算机检索PubMed，Medline，CNKI及中国生物医学文献数据库收录的，并辅以手工检索、因特网搜索的有关拉坦前列素与噻吗心安治疗原发性开角型青光眼和高眼压症的随机对照试验(RCT)。按照纳入和排除标准限定研究对象，通过Jadad评分量表进行文献质量评估后，针对眼压下降比例、药物不良反应2项内容，使用Cochrane协作网提供的RevMan 5.0软件进行Meta分析。

结果：共纳入9项RCT，合计555例患者。Meta分析结果显示：(1)拉坦前列素滴眼液与噻吗心安滴眼液降眼压效果，在 2，6，12wk时差异均有统计学意义(P<0.01)，加权平均差(WMD)分别为：在2wk\，在6wk\和12wk\。(2)随访结束时，结膜充血、异物感为拉坦前列素的两种较为常见的不良反应， 但其发生率拉坦前列素组与噻吗心安组比较， 结膜充血的发生率\〖OR=2.25，95% CI(0.99，5.08)\〗，异物感的发生率\〖OR=2.48，95% CI(1.02，6.03)\〗，显示二者差异均无统计学意义。

结论：治疗原发性开角型青光眼和高眼压症，拉坦前列素降眼压效果在用药12wk内较噻吗心安好； 两者在12wk内引起结膜充血、异物感、虹膜色素加深、视野损害等的不良反应方面，差异不明显。由于纳入研究的样本量偏小，且方法学质量中等，致使本系统评价结果论证强度不高，因此还需要开展更多的高质量的临床随机对照研究，以便更客观、准确、全面地评价其疗效和安全性。

METHODS: This was an evidence-based medicine science study. Pertinent studies of randomized controlled trial(RCT)were identified through searches of PubMed, Medline, CNKI and China Biology Medicine disc. The intensive searching by hand and internet was also designed. According to inclusion and exclusion criteria, the study object was limited. Quality of literatures were evaluated by experienced researchers according to the Jadad Score. RevMan 5.0 offered by Cochrane was used for Meta-analysis aimed at the ratio of low intraocular tension and adverse drug reaction.

RESULTS: A total of 9 RCT sadded up to 555 patients were involved. The results of meta-analysis showed that, a)the difference at 2, 6, 12wk both had statistical significance, latanoprost showed greater intraocular pressure(IOP)lowering efficacy compared with timolol. The weighted mean difference(WMD)at 2, 6, 12wk was respectively \〖-0.76, 95% CI(-1.32 to -0.20)\〗, \〖-1.15, 95% CI(-1.68 to 0.63)\〗 and \〖-1.01, 95% CI(-1.42 to -0.61)\〗. b)The difference in conjunctival congestion(OR=2.25, 95% CI 0.99 to 5.08)and foreign body sensation(OR=2.48, 95% CI 1.02 to 6.03 )between latanoprost and timolol group was not statistically significant.

CONCLUSION: Latanoprost showed greater IOP lowering efficacy at 12wk compared with timolol for OAG and OH patients. The conjunctival hyperemia, foreign body sensation, iris pigmentation deepened, vision damage of latanoprost group at 12wk compared with timolol, the difference was not significant. This conclusion is not powerful enough in proof due to the medium methodology quality of the included studies, so a large number of high-quality RCTs with large sample are needed for objectively, precisely and entirely evaluating the efficacy.