Kirk R Wilhelmus

Eye disease due to herpes simplex virus (HSV) commonly presents as epithelial keratitis which, though usually self-limiting, may persist or progress without treatment. To compare the relative effectiveness of antiviral agents, interferon, and corneal debridement in the treatment of HSV epithelial keratitis. We searched CENTRAL (which contains the Cochrane Eyes and Vision Group Trials Register) (2014, Issue 12), PubMed (January 1946 to 31 December 2014), EMBASE (January 1980 to 31 December 2014), Latin American and Caribbean Health Sciences Literature Database (LILACS) (January 1982 to 31 December 2014), System for Information on Grey Literature in Europe (OpenGrey) (January 1995 to 31 December 2014), BIOSIS (January 1926 to 5 May 2014), Scopus (January 1966 to 31 December 2014), Japan Science and Technology Institute (J-Global) (January 1975 to 31 December 2014), China National Knowledge Infrastructure (CNKI) (January 1979 to 31 December 2014), British Library's Electronic Table of Contents (Zetoc) (January 1993 to 7 May 2014). We looked for trials listed on the the metaRegister of Controlled Trials (www.controlled-trials.com), ClinicalTrials.gov (www.clinicaltrials.gov), the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (www.who.int/ictrp/search/en), Chinese Clinical Trial Registry, the U.S. Food and Drug Administration (FDA) (www.fda.gov/), National Institute for Health and Clinical Excellence (NICE) (www. nhs.uk) and the European Medicines Agency (EMA) (www.ema.europa.eu/ema/) as of 31 December 2014. There were no language or date restrictions in the search for trials. We also culled literature digests and conference proceedings as of 15 April 2014. There were no language or date restrictions in the search for trials. Randomised and quasi-randomised trials of HSV dendritic or geographic epithelial keratitis were included that reported the proportion of eyes healed at one week, two weeks, or both after enrolment. We tabulated data on study characteristics, risk of bias, and outcomes and used direct comparisons to estimate a risk ratio (RR) and, when feasible, a hazard ratio (HR) with a 95% confidence interval (CI). Heterogeneity was assessed by an inconsistency index. A multiple treatment comparison meta-analysis consolidated direct and indirect comparisons of relative healing at 14 days. One hundred thirty-seven studies involving 8333 eyes met the inclusion criteria. Placebo-controlled studies were heterogeneous in comparison with idoxuridine (RR 1.74; 95% CI 1.03 to 2.91) and few in number for vidarabine (RR 1.81; 95% CI 1.09 to 3.01), interferon (RR 1.32; 95% CI 1.06 to 1.64), and debridement. Vidarabine (RR 1.13; 95% CI 1.02 to 1.25), trifluridine (RR 1.30; 95% CI 1.18 to 1.43), acyclovir (RR 1.23; 95% CI 1.14 to 1.34), and brivudine (RR 1.34; 95% CI 1.18 to 1.51) were more effective than idoxuridine. Trifluridine (RR 1.17; 95% CI 1.03 to 1.32) and acyclovir (RR 1.11; 95% CI 1.03 to 1.19) were more effective than vidarabine. No significant differences in healing emerged among trifluridine, acyclovir, brivudine, and foscarnet although few studies compared brivudine or foscarnet with other antivirals. Any potential advantage of ganciclovir compared to acyclovir was mitigated by study heterogeneity and possible publication bias. Only one study evaluated the joint use of two topical antivirals. In a limited number of studies, oral acyclovir (RR 0.92; 95% CI 0.79 to 1.07) or the combination of oral acyclovir with a topical antiviral (RR 1.36; 95% CI 0.68 to 2.74) appeared as effective as a single topical antiviral agent. Compared to topical antiviral monotherapy, the combination of an antiviral with either interferon or debridement had inconsistent effects on expediting healing and improving outcome. Placebo-controlled studies of HSV epithelial keratitis are limited to superseded interventions. Trifluridine and acyclovir are more effective than idoxuridine or vidarabine and similar in therapeutic effectiveness. Brivudine and foscarnet do not substantially differ in effectiveness from trifluridine or acyclovir. Ganciclovir is at least as effective as acyclovir. The addition of interferon to a nucleoside antiviral agent and the combination of debridement with antiviral treatment need to be further assessed to substantiate any possible advantage in healing.