方法：以“维生素C”“角膜溃疡”为关键词，检索Embase、Cochrane library、Pubmed、中国知网数据库、万方数据库中结膜下注射维生素C治疗角膜溃疡的随机对照研究，检索时间为数据库建库至2021-02； 采用Cochrane偏倚风险评估工具对纳入研究进行质量评估，对纳入文献的发表偏倚运用Stata软件进行Egger检验，使用RevMan 5.3统计软件进行Meta分析。

结果：共纳入符合条件的32篇，全部为中文文献，合计4 514例患者，对照组采用常规治疗，试验组采用结膜下注射维生素C联合常规治疗； Meta分析结果表明，试验组的疗效优于对照组(OR=4.61，95%CI：3.70，5.74)； 复发率低于对照组(OR=0.24，95%CI：0.18，0.31)； 同时治愈时间短于对照组(MD=-6.29，95%CI：-7.08，-5.51)； 32项研究对药物不良反应报告不充分。对总有效率、复发率以及治愈时间进行Egger检验，结果显示总有效率和治愈时间存在发表偏倚，但通过剪补法分析得知存在的发表偏倚未对结果造成影响。

结论：结膜下注射维生素C治疗角膜溃疡具有一定可行性，临床疗效、治愈时间以及复发率均优于常规治疗组； 但由于纳入的研究证据强度较低，需要更多的多中心、随机双盲临床试验，从而提高维生素C治疗角膜溃疡疗效的证据强度。

METHODS: Using “vitamin C” and “corneal ulcer” as keywords, the randomized controlled trials(RCT)of subconjunctival injection of vitamin C in Embase, Cochrane library, Pubmed, CNKI and Wanfang database in the treatment of corneal ulcer was searched. Relevant reference published before February 2021. The risk bias tool provided by the Cochrane cooperation scale was used to assess the risk bias of included studies of RCTs. The publication bias of the included studies were assessed by the Egger's test with Stata software. Meta-analysis was performed with RevMan 5.3.

RESULTS: A total of 32 eligible articles were included, all of which were Chinese literatures, with a total of 4 514 patients. The control group was treated with routine treatment, and the experimental group was treated with subconjunctival injection of vitamin C combined with routine treatment. The Meta-analysis showed that the experimental group was better than control group(OR=4.61, 95%CI:3.70, 5.74). The recurrence rate was lower than control group(OR=0.24, 95%CI: 0.18, 0.31). The cure time in experimental group was shorter than control group(MD= -6.29, 95%CI: -7.08, -5.51), 32 studies do not adequately report adverse drug reactions(ADR). Egger's test was performed on the total effective rate, recurrence rate and cure time. The results showed that there was publication bias in the total effective rate and cure time. Trim and fill method showed that the publication bias did not affect the results.

CONCLUSION: Subconjunctival injection of vitamin C is feasible in the treatment of corneal ulcer. The clinical effect, cure time and recurrence rate were better than routine treatment. However, due to the low intensity of the included research evidence, more multicenter, randomized double-blind clinical trials are needed to improve the evidence intensity of vitamin C in the treatment of corneal ulcer.