Intrastromal Voriconazole and Amphotericin B for Fungal Keratitis: A Systematic Review and Meta-analysis
In brief
Intrastromal antifungals show no proven healing advantage for fungal keratitis
In a review of 36 studies, comparisons found no significant improvement in healing, need for corneal surgery, or vision with intrastromal voriconazole or amphotericin B versus other therapies. Uncontrolled studies reported healing in about 79% and 86% of eyes, respectively, but cannot show the drugs caused those outcomes; limited evidence leaves their role as add-on or rescue treatment uncertain.
- Journal
- The ocular surface (Q1)
- Published
- 28 October 2026
- Study design
- Systematic review / meta-analysis of RCTs
- Evidence level
- Level 1, High (CEBM 1a)
- Authors
- Karanvir Gill, Arvin Zaker, Aaryamaan Verma, Tajveer S Grewal, Valerie Mok, Sonia N Yeung, et al.
- PMID
- 42950985
- DOI
- 10.1016/j.jtos.2026.10.005
Why clinicians should know about it
- Picked for Ophthalmology (paper of the day, 2 November 2026): Systematic review evaluates therapies for sight-threatening fungal keratitis
Abstract
PURPOSE: To evaluate the efficacy and safety of intrastromal voriconazole (IS-VCZ) and intrastromal amphotericin B (IS-AMB) for fungal keratitis. METHODS: PubMed/MEDLINE, Embase, Scopus, and the Cochrane Library were searched from inception for randomized trials, comparative observational studies, and case series evaluating IS-VCZ or IS-AMB. Random-effects meta-analyses were performed for comparative and single-arm outcomes. RESULTS: Thirty-six studies were included: 33 evaluated IS-VCZ (732 eyes) and 7 evaluated IS-AMB (171 eyes), with four contributing to both groups. In uncontrolled single-arm analyses, pooled healing proportions were 79.1% (95% CI, 70.5%-85.7%) for IS-VCZ and 86.4% (95% CI, 71.9%-94.1%) for IS-AMB; these estimates were descriptive and do not establish treatment efficacy. Pooled therapeutic keratoplasty rates were 12.3% (95% CI, 6.7%-21.4%) and 13.2% (95% CI, 4.4%-33.4%), respectively. Compared with non-intrastromal therapies, IS-VCZ demonstrated no significant difference in clinical healing (OR 1.20, 95% CI 0.40-3.56), therapeutic keratoplasty (OR 1.29, 95% CI 0.59-2.82), or final visual acuity (MD 0.22, 95% CI -0.51 to 0.94). Comparative evidence remained limited, with wide confidence intervals across several outcomes. Comparative evidence for IS-AMB was limited and similarly demonstrated no significant differences in healing (OR 2.62, 95% CI 0.68-10.03), therapeutic keratoplasty (OR 0.45, 95% CI 0.02-8.23), or healing time (MD -5.64 days, 95% CI -25.05 to 13.77). Substantial heterogeneity was observed across most pooled outcomes. CONCLUSIONS: Current comparative evidence is insufficient to establish the efficacy of IS-VCZ or IS-AMB relative to existing antifungal therapies. Uncontrolled studies report favorable healing outcomes, but these findings cannot establish treatment efficacy. Intrastromal antifungal therapy may have a role as adjunctive or salvage treatment, but higher-quality comparative studies with standardized protocols are needed.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.