Clinical profile and outcome in Bacillus endophthalmitis

Objective: To study the clinical presentation, microscopic and organismal culture correlation of vitreous, and species-specific outcome in Bacillus endophthalmitis Design: Retrospective noncomparative case series. Intervention: Thirty-one culture proven Bacillus endophthalmitis patients between January 1991 and February 1998 underwent vitrectomy and intravitreal antibiotic injection. Lensectomy was combined when considered essential, and on a few occasions intravitreal dexamethasone was added. The patients also received topical and systemic antibiotics. The undiluted vitreous biopsy was the source for microbiologic evaluation (microscopy and culture sensitivity). Main Outcome Measures: The duration of symptoms, the presenting visual acuity, and influence of intravitreal dexamethasone with intravitreal antibiotics were examined for any statistical correlation with the final visual acuity. Results: Trauma was the major cause of infection. Vitreous biopsy microscopy demonstrated gram-positive bacillus in 28 of 31 cases, and polymicrobial infection was seen in 12 instances. All Bacillus species were sensitive to gentamicin, followed by vancomycin and ciprofloxacin. Clinical treatment within 7 days of symptoms, use of intravitreal vancomycin, and absence of polymicrobial infection were associated with better visual outcome. Conclusions: With appropriate treatment that essentially consists of vitrectomy and intravitreal antibiotics, patients with Bacillus endophthalmitis are likely to benefit in many instances. Gram-positive bacilli detected on vitreous microscopy should be empirically treated as Bacillus species unless otherwise proved. Ophthalmology 2001;108:1819-1825 (C) 2001 by the American Academy of Ophthalmology.