Background Pythium insidiosum is a parafungus that causes keratitis resembling fungal keratitis. This study compares outcome in a large cohort of patients with P insidiosum keratitis treated with antifungal drugs, to a pilot group treated with antibacterial antibiotics.

Methods Between January 2014 and December 2016, 114 patients with culture positive P insidiosum keratitis were included in the study. A subset of culture isolates was tested in vitro for response to nine antibacterial antibiotics by disc diffusion and E test. Patients were treated with topical natamycin in 2014, 2015 and up until mid 2016. Thereafter, the patients received a combination of topical linezolid and topical and oral azithromycin. Therapeutic penetrating keratoplasty (TPK) was done for patients not responding to medical therapy.

Results In vitro disc diffusion assay showed linezolid to be most effective. The rate of TPK was significantly higher in 2015 compared with 2016 (43/45, 95.6% vs 22/32, 68.8%; p=0.002). Eighteen patients were treated with antibacterial and 14 were treated with antifungal antibiotic in 2016. One patient was lost to follow-up in each group. The rate of TPK was higher and proportion of healed ulcers was lower (p=0.21, Fisher’s exact test) in the group on antifungal therapy (TPK—11/13, 84.6%; Healed—2/13, 15.3%) compared with the group on antibacterial therapy (TPK—11/17, 64.7%; Healed—6/17, 35.2%).

Conclusions We report favourable but not statistically significant response of P insidiosum keratitis to antibacterial agents in a pilot series of patients. Further evaluation of this strategy in larger number of patients is recommended.