Endophthalmitis caused by Bacillus species.

PURPOSE: To investigate clinical settings, management, and visual outcomes of endophthalmitis caused by Bacillus species and to review in vitro effectiveness of antibiotics commonly used against Bacillus species.
DESIGN: Retrospective, consecutive case series.
METHODS: Record review of all patients with endophthalmitis caused by Bacillus species treated at Bascom Palmer Eye Institute between January 1, 1990 and July 1, 2007. Antibiotic sensitivities were conducted on 21 of 22 isolates.
RESULTS: Twenty-two eyes of 22 patients met study inclusion criteria. Median follow-up was 18 months. Clinical settings included open globe injury (18 eyes), endogenous (two eyes), delayed-onset bleb-associated (one eye), and acute-onset postoperative (one eye). Twelve (67%) of 18 patients with open globe injuries had intraocular foreign bodies. Presenting visual acuity (VA) was hand movements or better in 13 (59%) patients. Initial treatment included pars plana vitrectomy and injection of antibiotics in 14 eyes (64%), vitreous tap and injection of antibiotics in seven eyes (32%), and evisceration in one eye (5%). Four (18%) patients received additional doses of intravitreal antibiotics; 16 (73%) underwent secondary surgical procedures. Eight (36%) patients achieved a final VA of 20/400 or better and four (18%) achieved a final VA of 20/60 or better. All patients received intraocular vancomycin and a cephalosporin or aminoglycoside. Systemic antibiotics were used in 18 (82%) patients. Fifteen (68%) isolates were Bacillus cereus. All isolates tested were sensitive to vancomycin, gentamicin, and five fluoroquinolones. Only three of 21 isolates were susceptible to penicillin and cephalosporins.
CONCLUSIONS: Endophthalmitis caused by Bacillus species often results in poor visual outcomes. In vitro antibiotic sensitivities indicate that vancomycin, aminoglycosides, and fluoroquinolones were effective against Bacillus isolates, whereas cephalosporins were relatively ineffective.