874. Does minimum inhibitory concentration mediate the relationship between causative organism and clinical outcome in bacterial keratitis?
Session: Oral Abstract Session: Antimicrobial Resistance and Susceptibility
Saturday, October 22, 2011: 11:15 AM
Room: 151AB

Different organisms are associated with severity in clinical outcomes in systemic and ocular bacterial infections.  Antimicrobial susceptibility likely plays a role in this relationship, but it has been difficult to assess in prospective setting.  If clinical outcomes and minimum inhibitory concentration (MIC) differ significantly across organisms and there is a significant relationship between MIC and clinical outcome, a mediation analysis can be conducted to assess the role MIC plays in the effect of organism on outcome.  Ocular infections offer an ideal environment to study these questions, as we can control for specific baseline measurements for the corresponding outcome and analyze the relationship during the course of treatment.


The Steroids for Corneal Ulcers Trial (NEI U10EY015114) was a randomized, placebo-controlled, double-masked clinical trial assessing benefit with topical corticosteroids as adjunctive therapy for culture-positive bacterial keratitis.  Enrolled patients received topical moxifloxacin for at least three weeks.  Antibiotic susceptibility testing was performed using Etest.  Visual acuity was measured at presentation and 3 weeks.  The relationship of organism and both MIC and 3-week visual acuity was assessed by analysis of variance (ANOVA).  The relationship of MIC and 3-week visual acuity was assessed by linear regression.  A mediation analysis was performed using seemingly unrelated regression.


Of 500 patients enrolled in the trial, MIC was available for 480.  The most commonly-isolated organisms were Streptococcus pneumoniae (248, 52%) and Pseudomonas aeruginosa (109, 23%).  Causative organism was associated with 3-week visual acuity (P=0.008) and MIC (P<0.001).  There was a significant relationship between MIC and 3 week visual acuity (0.019, 95% CI 0.004 to 0.033, P=0.01).  Mediation analysis estimated that approximately 13% (95% CI 3 to 24%, P=0.015) of the effect of organism on 3 week visual acuity was indirect via MIC. 


MIC is a significant mediator of the relationship between organism and outcome, but the effect is likely small.  Determination of causative organism and in vitro susceptibility is important for prognostic value and initiation of appropriate therapy.

Subject Category: A. Antimicrobial agents and Resistance

Catherine Oldenburg, MPH1, Lalitha Prajna, MD2, Muthiah Srinivasan, MD3, Jeena Mascarenhas, MD3, Kathryn Ray, MA1, David Glidden, PhD4, Nisha Acharya, MD, MS1 and Thomas Lietman, MD1, (1)Francis I. Proctor Foundation, University of California, San Francisco, San Francisco, CA, (2)Ocular Microbiology, Aravind Eye Care System, Madurai, India, (3)Cornea Clinic, Aravind Eye Care System, Madurai, India, (4)Department of Epidemiology and Biostatistics, University of California, San Francisco, San Francisco, CA


C. Oldenburg, None

L. Prajna, None

M. Srinivasan, None

J. Mascarenhas, None

K. Ray, None

D. Glidden, None

N. Acharya, None

T. Lietman, None

Findings in the abstracts are embargoed until 12:01 a.m. EST Thursday, Oct. 20 with the exception of research findings presented at IDSA press conferences.