Mycotic Antimicrobial Localized Injection: A Randomized Clinical Trial Evaluating Intrastromal Injection of Voriconazole.

PURPOSE

To determine if there is a benefit to adjuvant intrastromal voriconazole (ISV) injections for primary treatment of filamentous fungal keratitis.

DESIGN

Outcome-masked, randomized controlled clinical trial.

PARTICIPANTS

Patients with moderate vision loss resulting from a smear-positive fungal ulcer.

METHODS

Study eyes were randomized to topical natamycin plus ISV injection versus topical natamycin alone.

MAIN OUTCOME MEASURES

The primary outcome of the trial was microbiological cure on 3-day repeat culture analysis. Secondary outcomes included microbiological cure on 7-day repeat culture analysis; 3-week and 3-month best spectacle-corrected visual acuity; infiltrate or scar size or both; rate of perforation; therapeutic penetrating keratoplasty (TPK); and other adverse events.

RESULTS

A total of 151 patients with smear-positive ulcers were screened and 70 were enrolled at Aravind Eye Hospital, Pondicherry, India. Baseline cultures grew Fusarium in 19 samples (27%), Aspergillus in 17 samples (24%), and other filamentous fungi in 19 samples (27%) and showed negative results in 13 samples (19%). Those randomized to ISV injection had 1.82 times the odds of 3-day culture positivity after controlling for baseline culture status (95% confidence interval [CI], 0.65-5.23; P = 0.26, bias-corrected logistic regression) and 1.98 times the odds of positive 7-day culture results, after controlling for baseline culture status (95% CI, 0.69-5.91; P = 0.20, bias-corrected logistic regression). Those randomized to ISV injection showed 0.5 logMAR lines (approximately 0.5 Snellen lines) of decreased visual acuity (95% CI, -2.6 to 3.6 lines; P = 0.75) and 0.55 mm worse infiltrate or scar size or both at 3 months after controlling for baseline values (95% CI, -0.13 to 1.25; P = 0.11). Intrastromal voriconazole injections showed a 2.85-fold increased hazard of perforation after controlling for baseline infiltrate depth (95% CI, 0.76-10.75; P = 0.12) but no difference in the rate of TPK (hazard ratio, 0.95; 95% CI, 0.44-2.04; P = 0.90).

CONCLUSIONS

There seems to be no benefit to adding ISV injections to topical natamycin in the primary treatment of moderate to severe filamentous fungal ulcers. Studies consistently suggest that voriconazole has a limited role in the treatment of filamentous fungal ulcers.