Therapeutic interventions for herpes simplex virus epithelial keratitis

  • Review
  • Intervention


  • KR Wilhelmus

Prof Kirk Wilhelmus, Professor of Ophthalmology, Cullen Eye Institute, Department of Ophthalmology, Baylor College of Medicine, 6565 Fannin Street, NC-205, Houston, TX 77030, USA.



Many clinical trials have been performed on the acute treatment of dendritic epithelial keratitis. Surveys of ocular antiviral pharmacology and of herpes simplex virus (HSV) eye disease have evaluated different interventions, but a systematic review of all comparative clinical studies has not previously been undertaken.


The objective of this review was to compare the effects of various therapeutic interventions for dendritic or geographic HSV epithelial keratitis.

Search strategy

We searched the Cochrane Central Register of Controlled Trials - CENTRAL (which contains the Cochrane Eyes and Vision Group Trials Register) in The Cochrane Library (Issue 3, 2006), MEDLINE (1966 to July 2006, week 3), EMBASE (1980 to 2006, week 30), LILACS (up to August 2006), SIGLE (1980 to March 2005), ZETOC (1 August 2006), BIOSIS (up to 2005), JICT-EPlus (up to 2005), Index Medicus (1960 to 1965), Excerpta Medica Ophthalmology (1960 to 1973), reference lists of primary reports and review articles, and conference proceedings pertaining to ocular virology.

Selection criteria

This review included comparative clinical trials that assessed one-week or two-week healing rates of topical ophthalmic or oral antiviral agents and or physical or chemical debridement in people with active epithelial keratitis.

Data collection and analysis

The review author extracted data and assessed trial quality. Interventions were compared by the proportions of participants healed at seven days and at fourteen days after trial enrolment.

Main results

This review included data from 98 trials that randomised a total of 5211 participants. Compared to idoxuridine, the topical application of vidarabine, trifluridine, or acyclovir resulted in a significantly greater proportion of participants healing within one week of treatment. Among these latter three antiviral agents, no treatment emerged as significantly better for the therapy of dendritic epithelial keratitis. Insufficient placebo-controlled studies were available to assess debridement and other physical or physicochemical methods of treatment. Interferon monotherapy had a slight beneficial effect on dendritic epithelial keratitis but was not better than other antiviral agents. Interferon was very effective when combined with another antiviral agent such as trifluridine.

Authors' conclusions

Currently available antiviral agents are effective and nearly equivalent. The combination of a nucleoside antiviral with either debridement or with interferon seems to speed healing. Future trials of the acute treatment of HSV epithelial keratitis must aim to achieve adequate statistical power for assessing the primary outcome of epithelial healing and should consider the effect of lesion size and other characteristics on treatment response.

Plain language summary

Plain language summary

Antiviral, interferon, and debridement treatments for herpes simplex eye disease

Herpes simplex is the most common virus acquired by humans, and ocular herpes is a prevalent and recurrent infection. Several treatments, ranging from eye medications to wiping or scraping, aim to shorten the course of herpetic eye disease when infection of the corneal surface, known as a dendrite, occurs. To examine the effectiveness of treatment options, the review author included 98 clinical trials from Europe, North America, Asia, Australia, and Africa that involved 5,211 participants who had a corneal dendrite. Healing rates at one week and two weeks were compared using different treatments, including placebo, antivirals, interferon, and ocular surface removal known as debridement. Placebo was not very effective since only 25% of corneal dendrites healed spontaneously within one week and slightly less than half by two weeks. During the 1960s, the first nucleoside antiviral drug idoxuridine (IDU) was discovered and was shown to lead to faster healing of herpetic corneal infection compared to placebo. Randomized trials subsequently found that antiviral drugs such as trifluridine and acyclovir were better than IDU; about 5 times as many patients healed within one week with trifluridine or acyclovir as with IDU. Further trials showed that trifluridine and acyclovir were equivalently effective and led to healing in two thirds of treated patients by one week and in approximately 90% by two weeks. Small trials with bromovinyldeoxyuridine (BVDU), ganciclovir, or foscarnet suggested similar effectiveness as trifluridine or acyclovir. In one trial, acyclovir taken by mouth was as good as acyclovir applied to the eye. Interferon was better than placebo and as effective as other antiviral drugs. The combination of interferon-alfa eye drops and either trifluridine or acyclovir resulted in faster healing of dendritic keratitis than treatment with trifluridine or acyclovir alone; 90% of eyes healed within one week with combined interferon-antiviral therapy. Debridement by heat, chemicals, swabbing, or abrasion was commonly used before the development of antiviral drugs; the joint use of debridement and antiviral therapy seemed to speed healing and reduced recrudescent infection. Unfortunately, undue inconsistency among too few studies precluded knowing whether patients who had their infected corneal surface wiped off and then received antiviral treatment healed faster than those treated only with an antiviral medication. This review was limited by the many different therapies that have been studied and could not include 63 other trials that used inadequate methods or reported insufficient information