Article Text

Download PDFPDF

Susceptibility testing and clinical outcome in fungal keratitis
  1. Brett L Shapiro1,
  2. Prajna Lalitha2,
  3. Allison R Loh1,3,
  4. Annette W Fothergill4,
  5. Namperumalsamy V Prajna2,
  6. Muthiah Srinivasan2,
  7. Amit Kabra2,
  8. Jaya Chidambaram1,
  9. Nisha R Acharya1,
  10. Thomas M Lietman1
  1. 1F.I. Proctor Foundation, University of California, San Francisco, California, USA
  2. 2Aravind Eye Hospitals, Madurai, India
  3. 3University of Pennsylvania School of Medicine, Philadelphia, USA
  4. 4University of Texas Health Sciences Center at San Antonio, Texas, USA
  1. Correspondence to Dr Thomas M Lietman, F.I. Proctor Foundation, Room S309, 513 Parnassus Avenue, University of California, San Francisco, San Francisco, CA 94143, USA; tom.lietman{at}

Statistics from

Fungal keratitis causes significant morbidity, especially in tropical climates, and is notoriously difficult to manage. The choice of antifungal agent for fungal keratitis remains largely empirical, with no consensus on the role of susceptibility testing in guiding therapy. Studies suggest that susceptibility and outcome may be associated in systemic fungal infections with some dimorphic fungi,1 but this correlation may not exist for filamentous fungi or in ocular disease because of frequent topical dosing and high drug concentration. Given the availability of new topical medications, tailoring antifungal therapy based on microbial sensitivity is important.2 Here, we assess whether fungal susceptibility testing correlates with clinical outcomes in cases of fungal keratitis.


This study was a retrospective case review of consecutive patients with culture proven fungal keratitis presenting to the Aravind Eye Hospital cornea clinic between March and July 2004. Of 98 consecutive patients, minimum inhibitory concentration (MIC) data was available for 90.3 Eighty-one corresponding charts were available, and 54 charts had follow-up of at least 3 weeks to allow determination of healing. All cases were cultured, and treated with natamycin 5% suspension every hour for 48 h while awake, then every 2 h until re-epithelialisation. Patients received cycloplegic drops at the discretion of their physician. Patients were hospitalised until re-epithelialisation, which ensured delivery of study medication. Stored fungal isolates were sent to the University of Texas Health Sciences Center in San Antonio, Texas, USA, for MIC testing for natamycin using macrobroth dilution. Since patients were treated with natamycin solely, only MIC testing for natamycin was analysed. The natamycin used was pharmaceutical grade powder obtained from Alcon Labs, Dallas, Texas, USA. Susceptibility was measured using standardised methods.3–5

For each organism group, the MIC50 and MIC90 were estimated as the median MIC and 90th percentile, respectively, using the tab command in STATA (STATACorp LP, 4905 Lakeway Drive, College Station, Texas, USA). An ulcer was considered healed when the epithelial defect was found to be <1 mm in maximum diameter with slit-lamp biomicroscopy. A healing time of <3 weeks from presentation was considered a good result, and longer healing times a poor result. Logistic regression was used to predict a good/poor result, the primary outcome, using log2-transformed MIC as a covariate. The regression model was analysed using Pearson's goodness of fit. A Fisher's exact test was used to correlate genus (Aspergillus vs Fusarium) to good/poor result. All analyses were performed using STATA 9.2.


The baseline characteristics and MICs of 54 patients with fungal corneal ulcers are described in table 1.

Table 1

Fungal organisms isolated from retrospective review of corneal ulcers from March–June 2004 (n=54)*

A lower MIC was significantly associated with a good outcome, as was Fusarium species (as opposed to Aspergillus species) (table 2).

Table 2

Univariate analysis predicting healing at 3 weeks in fungal corneal ulcers (n=54)

When restricted to a subgroup of organisms, for example Fusarium cases, the relationship between MIC and clinical outcome was similar in magnitude, but no longer statistically significant (OR=0.51, 95% CI 0.12 to 2.05, p=0.34). In addition, when restricted to only Aspergillus cases, the relationship was neither significant nor similar in magnitude (OR=1.37, 95% CI 0.39 to 4.88, p=0.63).


Many factors contribute to the success or failure of fungal keratitis management, including ulcer size, ulcer location, organism, penetration of antimicrobial agent and susceptibility of the organism to treatment. In bacterial keratitis, studies suggest that susceptibility of the organism to the agent in vitro correlates with outcome.6 7 It remains unclear whether susceptibility correlates with outcome in fungal keratitis.8 9 In systemic fungal disease, researchers suggest that the role of susceptibility testing may be similar to that of bacterial susceptibility testing, where approximately 90% of susceptible cases and 60% of resistant cases respond to therapy.1 Antifungal susceptibility testing is associated with outcome in mucosal candidiasis and candidaemia, and antifungal susceptibilities influence treatment recommendations.10

In fungal keratitis, in vitro susceptibility did correlate with outcome. A twofold increase in MIC was associated with a 47% reduction in the odds of healing. In addition, the organism is associated with outcome. Since only 54 of 90 cases with completed susceptibility testing had clinical data available, the study had limited generalisability. Further prospective studies would be necessary to assess whether MIC provides information useful to the clinician once the organism species has been identified, as well as the effect of other covariates such as toxicity, prior medications, age and sex.


The Department of Ophthalmology at UCSF is supported by a core grant from the National Eye Institute, EY02162, That Man May See and the South Asia Research Fund. In addition, this work was supported in part by a grant from the Doris Duke Charitable Foundation to UCSF to fund Clinical Research Fellow Allison Loh.


View Abstract


  • Funding The funding for this research was provided by the National Eye Institute, EY02162, That Man May See, the South Asia Research Fund and the Doris Duke Charitable Foundation. The funding organisations played no role in the design, collection, analysis and/or interpretation of data, writing or submission of the paper.

  • Competing interests None.

  • Ethics approval This study was conducted with the approval of the institutional review boards at Aravind Medical Research Foundation and University of California San Francisco (CHR #H9332-21899-01).

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.