Table 1

The role of topical corticosteroids in management of microbial keratitis

Author, yearNoStudy designControl groupResultsComments
Aronson,17 196917Retrospective case seriesNoDegree of therapeutic success depended on severity of original lesion.Sixteen cases of bacterial keratitis, one case of Candida keratitis. Hourly corticosteroid drops used for 3–6 weeks. No statistical analyses reported.
Coster,18 198777Retrospective case seriesNo‘Successful outcome’ was achieved in 69% of cases, where topical corticosteroids used, compared with 78% of cases, where no corticosteroid used.Successful outcome not defined. Findings biased by more severe cases more likely to receive corticosteroids.
Carmichael,14 199040Prospective randomised trialYesNo difference between groups in visual outcome, healing rate or complications.Antibiotics only (controls) vs antibiotics and topical dexamethasone (cases). Dexamethasone commenced as early as 24 h postantibiotic therapy.
Srinivasan,15 200942Prospective double-blind randomised trialYesNo difference between groups in visual outcome or complications. Delayed healing in corticosteroid group noted.Antibiotics only (controls) vs antibiotics and topical prednisolone (cases). Prednisolone commenced after a minimum of 48 h of antibiotics.
Blair,16 201130Prospective randomised trialYesNo difference between groups in residual ulcer size at 10 weeks, final visual acuity or healing time.Antibiotics only (controls) vs antibiotics and topical dexamethasone (cases). Dexamethasone commenced after 48 h of antibiotics.
Srinivasan,19 2012500Prospective double-blind randomised trialYesNo difference between groups in final visual outcome, healing rate or complications. Better final visual acuity in subgroup with central ulcers or worst initial vision that received corticosteroids.Antibiotics only (controls) vs antibiotics and topical prednisolone (cases). Prednisolone commenced after a minimum of 48 h of antibiotics.