Background: Fungal keratitis, one of the most common causes of ocular mycosis, is the second most common cause of blindness in the world, after cataracts. The aim of the study was to compare the efficacy of conventional topical, systemic medications and intracameral voriconazole injection in visual and structural outcomes in keratomycosis.

Material and methods: We conducted a hospital-based observational study of 45 patients of 45 eyes with smear-positive fungal keratitis. Patients were categorized into three groups: Group I received systemic topical with oral ketoconazole 200 mg, Group II — topical medications with intracameral voriconazole 50 μgm/0.1 mL, Group III — topical medications with both oral ketoconazole 200 mg and intracameral voriconazole 50 μgm/0.1 mL.

Results: The common fungal organism is identified as Fusarium. The mean final visual acuity (VA) was 1.25 ± 0.32, 1.47 ± 1.05, and 1.22 ± 0.37 logMAR in Group I, group II, and Group III, respectively. The mean improvement in VA was 0.33 ± 0.07, 0.01 ± 0.71, and –0.19 ± 0.02 logMAR without significant change (p = 0.9). There was a significant difference in VA between the final postoperative follow-up period and baseline in Group I cases (p = 0.0019). Whereas no difference in VA between the final postoperative follow-up period and baseline in either Group II (p = 0.0671) or Group III (p = 0.1505) cases. The difference in time between the disappearance of hypopyon and the mean time to infection healing was not statistically significant (p = 0.1). Three cases in each group were perforated, and keratoplasty was performed. These perforated cases did not show culture positive. Histopathology identified the isolated organisms as Aspergillus species (n = 3) and Fusarium species (n = 2) in the corneal buttons.

Conclusion: The differences in VA between the three methods were not statistically significant, indicating no treatment method superior to others (inter-group). However, in Group I, there was a significant difference in VA between the final postoperative follow-up period and baseline (p = 0.0019). There was no difference in VA between these time intervals in either Group II (p = 0.0671) or Group III (p = 0.1505). Within-group or intra-group analysis reveals that the Group I method is more effective for VA. The success rate of the method depended cumulatively on the duration of intracameral voriconazole in the anterior chamber, non-drainage of hypopyon, and individual clinical response.