Steroids Early in Bacterial Keratitis May Improve Vision

Veronica Hackethal, MD

April 28, 2014

Adding topical corticosteroids 2 to 3 days after starting topical antibiotics for treatment of bacterial keratitis is linked to improved vision outcomes, according to a study published online April 24 in JAMA Ophthalmology.

Scarring as a result of bacterial keratitis is a leading cause of vision loss, and adjunctive treatment with corticosteroids may improve outcomes. However, the optimal timing and differential benefits of their use have been unclear, and depend on ulcer severity, according to Kathryn J. Ray, MA, from the F. I. Proctor Foundation and the Department of Epidemiology and Biostatistics, University of California, San Francisco, and colleagues.

Therefore, the investigators performed a secondary analysis of data from the Steroids for Corneal Ulcers Trial (SCUT).

"The primary outcome of the randomized controlled trial [SCUT] was unable to find a difference between steroid and placebo. This report is a secondary analysis that suggested that if patients presented early in their treatment, steroids do help," senior author Tom Lietman, MD, also from the F. I. Proctor Foundation and the Department of Epidemiology and Biostatistics and the Department of Ophthalmology, University of California, San Francisco, told Medscape Medical News. "One interpretation would be that the primary study showed that steroids were not dangerous, and this secondary analysis suggests that use early in the course of treatment with antibiotics could be beneficial."

The subanalysis looked at earlier addition (within 2 - 3 days) and later addition (after 4 or more days) of topical corticosteroids to topical moxifloxacin hydrochloride on best spectacle-corrected visual acuity (BSCVA) at 3 months' follow-up compared with placebo.

The study included 500 participants, of whom 8 were excluded because of missing data. Participants with earlier corticosteroid administration showed approximately 1-line better BSCVA (n = 311; −0.11 logMAR; 95% confidence interval, −0.20 to −0.02; P = .01) compared with those who received placebo early. In contrast, there was no significant difference between patients who started receiving corticosteroids after 4 or more days of antibiotic treatment compared with placebo (n = 139; 0.10 logMAR; 95% CI, −0.02 to 0.23; P = .14). Adjusting for sex, symptom duration, Nocardia vs non-Nocardia species, and scar size did not change these associations.

Ulcer severity also did not seem to affect the improvement associated with earlier corticosteroid administration. Patients with severe ulcers had a 3-line improvement in BSCVA with earlier treatment (n = 85; −0.27 logMAR; 95% CI, −0.50 to −0.40; P = .02) but also showed a nonsignificant 2-line improvement with later administration (n = 40; −0.24 logMAR; 95% CI, −0.59 to 0.10; P = 0.17). Those with moderately severe ulcers had a 1-line improvement with earlier treatment (n = 165, −0.09 logMAR; 95% CI, −0.20 to 0.01; P = .09), although BSCVA worsened by 2 lines with later administration (n = 72; 0.20 logMAR; 95% CI, 0.04 - 0.36; P = .01). Among those with mild ulcers, earlier treatment showed no significant advantage (n = 61; 0.02 logMAR; 95% CI, −0.09 to 0.13; P = .70), whereas those treated later had a 2-line decrease in BSCVA (n = 27; 0.19 logMAR; 95% CI, 0.02 - 0.36; P = .03).

Visual acuity also significantly improved with earlier corticosteroid addition in those with non-Nocardia keratitis (n = 289; 95% CI, −0.22 to −0.03 logMAR; P = .01) and in those without initial antibiotic treatment before enrolling in SCUT (n = 279; −0.12 logMAR; 95% CI, −0.21 to −0.04; P = .02).

This study may have been subject to methodological limitations inherent in subanalyses of larger trials. In addition, bacterial cultures were used as confirmation in all cases of bacterial keratitis, which could limit generalizability because mild to moderate corneal ulcers resulting from bacterial keratitis are often treated empirically with topical broad-spectrum antibiotics.

"Steroids are still contraindicated when we're not effectively killing the microbial agent; for example, with fungus or Nocardia," Dr. Lietman emphasized, "But if we know the cause is bacterial, and not Nocardia, steroids are safe and may be effective if given early."

Dr. Lietman also mentioned that many cornea specialists are already using this approach in clinical practice, a statement seconded by Guillermo Amescua, MD, when contacted by Medscape Medical News for comment. Dr. Amescua has done research on ocular immunology and is an assistant professor of clinical ophthalmology at the University of Miami Miller School of Medicine in Florida.

"I think this study confirms what we as clinicians have empirically observed after treating a large number of patients with moderate and severe bacterial keratitis," Dr. Amescua explained. "The early use of steroids in patients with moderate and severe bacterial keratitis helps to decrease inflammation, and this helps decrease pain and discomfort."

He added, "As a physician who takes care of multiple patients a month with moderate and severe corneal ulcers, the information provided by the SCUT trial and this study is very useful." He also emphasized a case-based approach and the importance of laboratory results to confirm bacterial infection and rule out fungi, parasites, mycobacteria, and acanthamoeba.

"The data presented by this paper [are] helpful, but the decision of starting steroids should always be made very carefully, and always by having a conversation with the patient about the pros and cons for doing this," Dr. Amescua said, "including risk for causing harm in patients with fungal keratitis."

This study was supported by the National Eye Institute and a Research to Prevent Blindness Award. Alcon provided moxifloxacin for the trial. The Department of Ophthalmology at the University of California, San Francisco, is supported by the National Eye Institute. The authors and Dr. Amescua have disclosed no relevant financial relationships.

JAMA Opthalmol. Published online April 24, 2014. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.