What is epidemic keratoconjunctivitis?
Epidemic keratoconjunctivitis is inflammation of conjunctiva, the membrane covering the sclera (white outer coating of the eye) and inside of the eyelids, usually caused by specific strains of a group of common viruses known as adenoviruses. The infection can spread easily within households, healthcare settings, and the community. In some individuals, the inflammation leads to scarring of the cornea ('infiltrates') and conjunctiva, which causes persistent discomfort and poor vision.
How is it treated?
Treatment is usually supportive with cool compresses, artificial tears, and sometimes steroids.
What did we want to find out?
We wanted to know whether any existing topical medication can relieve symptoms or signs and prevent complications, and if these medications were well-tolerated.
What we did
We reviewed randomized controlled trials (a type of study where participants are randomly assigned to one of two or more treatment groups) of children and adults with epidemic keratoconjunctivitis. We summarized the results of these studies and rated our confidence in the evidence according to study sizes and methods.
What we found
We found 10 studies that involved 892 people with epidemic keratoconjunctivitis (9 to 82 years old); studies lasted from 7 days to 18 months. When compared with artificial tears, antiviral agents appeared to shorten the duration of symptoms or signs. Povidone-iodine alone led to more disease recovery within the first seven days of treatment. We found no evidence that any treatment prevented corneal scars more often than artificial tears. The immunosuppressant cyclosporin A was no more effective than steroids in treating corneal scars. Cyclosporin A and tacrolimus eye drops often caused eye discomfort but did not increase people's intraocular pressure as compared to steroids.
What are the limitations of the evidence?
Nearly all of the included studies had flawed study methods and were reported poorly. These weaknesses raised our concerns about the study findings and reduced our confidence in the overall evidence summarized in the review.
How up-to-date is the evidence?
The evidence is current to April 2021.
The evidence for the seven specified outcomes was of low or very low certainty due to imprecision and high risk of bias. The evidence that antiviral agents shorten the duration of symptoms or signs when compared with artificial tears was inconclusive. Low certainty evidence suggests that PVP-I alone resolves signs and symptoms by seven days relative to artificial tears. PVP-I or PVA-I, alone or with steroid, is associated with lower risks of SEI development than artificial tears or steroid (very low certainty evidence). The currently available evidence is insufficient to determine whether any of the evaluated interventions confers an advantage over steroids or artificial tears with respect to virus eradication or its spread to initially uninvolved fellow eyes. Future updates of this review should provide evidence of high-level certainty from trials with larger sample sizes, enrollment of participants with similar durations of signs and symptoms, and validated methods to assess short- and long-term outcomes.
Viruses cause about 80% of all cases of acute conjunctivitis. Human adenoviruses are believed to account for 65% to 90% of cases of viral conjunctivitis, or 20% to 75% of all causes of infectious keratoconjunctivitis worldwide. Epidemic keratoconjunctivitis (EKC) is a highly contagious subset of adenoviral conjunctivitis that has been associated with large outbreaks at military installations and at medical facilities. It is accompanied by severe conjunctival inflammation, watery discharge, and light sensitivity, and can lead to chronic complications such as corneal and conjunctival scarring with discomfort and poor quality of vision. Due to a lack of consensus on the efficacy of any pharmacotherapy to alter the clinical course of EKC, no standard of care exists, therefore many clinicians offer only supportive care.
To assess the efficacy and safety of topical pharmacological therapies versus placebo, an active control, or no treatment for adults with EKC.
We searched the Cochrane Central Register of Controlled Trials (CENTRAL, which contains the Cochrane Eyes and Vision Trials Register; 2021, Issue 4); Ovid MEDLINE; Ovid Embase; Latin American and Caribbean Health Sciences database (LILACS); ClinicalTrials.gov; and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP), with no restrictions on language or year of publication. The date of the last search was 27 April 2021.
We included randomized controlled trials in which antiseptic agents, virustatic agents, or topical immune-modulating therapy was compared with placebo, an active control, or no treatment.
We used standard Cochrane methodology.
We identified 10 studies conducted in Asia, Europe, the Middle East, and North Africa with a total of 892 participants who were treated for 7 days to 6 months and followed for 7 days up to 1.5 years.
Study characteristics and risk of bias
In most studies participants were predominantly men (range: 44% to 90%), with an age range from 9 to 82 years. Three studies reported information on trial registration, but we found no published study protocol. The majority of trials had small sample sizes, ranging from 18 to 90 participants enrolled per study; the only exception was a trial that enrolled 350 participants. We judged most studies to be at high or unclear risk of bias across risk of bias domains.
We included 10 studies of 892 EKC participants and estimated combined intervention effects in analyses stratified by steroid-containing control treatment or artificial tears. Six trials contributed to the comparisons of topical interventions (povidone-iodine [PVP-I], trifluridine, ganciclovir, dexamethasone plus neomycin) with artificial tears (or saline). Very low certainty evidence from two trials comparing trifluridine or ganciclovir with artificial tears showed inconsistent effects on shortening the mean duration of cardinal symptoms or signs of EKC. Low certainty evidence based on two studies (409 participants) indicated that participants treated with PVP-I alone more often experienced resolution of symptoms (risk ratio (RR) 1.15, 95% confidence interval (CI) 1.07 to 1.24) and signs (RR 3.19, 95% CI 2.29 to 4.45) during the first week of treatment compared with those treated with artificial tears. Very low certainty evidence from two studies (77 participants) suggested that PVP-I or ganciclovir prevented the development of subepithelial infiltrates (SEI) when compared with artificial tears within 30 days of treatment (RR 0.24, 95% CI 0.10 to 0.56).
Four studies compared topical interventions (tacrolimus, cyclosporin A [CsA], trifluridine, PVP-I + dexamethasone) with topical steroids, and one trial compared fluorometholone (FML) plus polyvinyl alcohol iodine (PVA-I) with FML plus levofloxacin. Evidence from one trial showed that more eyes receiving PVP-I 1.0% plus dexamethasone 0.1% had symptoms resolved by day seven compared with those receiving dexamethasone alone (RR 9.00, 95% CI 1.23 to 66.05; 52 eyes). In two trials, fewer eyes treated with PVP-I or PVA-I plus steroid developed SEI within 15 days of treatment compared with steroid alone or steroid plus levofloxacin (RR 0.08, 95% CI 0.01 to 0.55; 69 eyes). One study found that CsA was no more effective than steroid for resolving SEI within four weeks of treatment (RR 0.84, 95% CI 0.67 to 1.06; N = 88). The evidence from trials comparing topical interventions with steroids was overall of very low level certainty.
Antiviral or antimicrobial agents plus steroid did not differ from artificial tears in terms of ocular discomfort upon instillation (RR 9.23, 95% CI 0.61 to 140.67; N = 19). CsA and tacrolimus eye drops were associated with more cases of severe ocular discomfort, and sometimes intolerance, when compared with steroids (RR 4.64, 95% CI 1.15 to 18.71; 2 studies; N = 141). Compared with steroids, tacrolimus did not increase the risk of elevated intraocular pressure (RR 0.07, 95% CI 0 to 1.13; 1 study; N = 80), while trifluridine conferred no additional risk compared to tear substitute (RR 5.50, 95% CI 0.31 to 96.49; 1 study; N = 97). Overall, bacterial superinfection was rare (one in 23 CsA users) and not associated with use of the intervention steroid (RR 3.63, 95% CI 0.15 to 84.98; N = 51). The evidence for all estimates was of low or very low certainty.