Bactericidal effect of intravitreal levofloxacin in an experimental model of endophthalmitis.

AIM

To evaluate the efficacy of intravitreal 1.5% levofloxacin versus combined treatment with vancomycin and ceftazidime in an experimental model of bacterial endophthalmitis.

METHODS

The right eye of 75 New Zealand White rabbits was inoculated in the vitreous cavity with strains of Staphylococcus epidermidis, Staphylococcus aureus or Pseudomonas aeruginosa. The study evaluated treatment with 1.5% levofloxacin compared with 1% vancomycin plus 2.2% ceftazidime, and a control group. At 24 h after treatment, 100 microl vitreous and aqueous humor were harvested, and the number of colony-forming units (CFU) per ml was determined. The microbiological status of the vitreous and aqueous humor samples was tested at day 2, 3, 5 and 8. The Peyman classification was used to assess the severity of endophthalmitis.

RESULTS

For the S. epidermidis endophthalmitis model, levofloxacin treatment produced an approximately 3 log decrease in CFU/ml of vitreous relative to that in untreated eyes (p = 0.023), whereas vancomycin plus ceftazidime produced a 5 log reduction in CFU/ml (p = 0.023). However, in the S. aureus model, levofloxacin treatment produced an approximately 4 log reduction in CFU/ml, whereas vancomycin plus ceftazidime produced an approximately 2 log reduction in CFU/ml relative to that of the control (p = 0.018 and p = 0.019, respectively). Treatment of P. aeruginosa-infected rabbit eyes with levofloxacin produced a 5 log reduction in CFU/ml of vitreous, whereas vancomycin plus ceftazidime produced an approximately 3-4 log reduction in CFU/ml relative to that of the control (p = 0.001). However, there was no significant difference between levofloxacin and vancomycin plus ceftazidime on comparing the reduction of CFU/ml of vitreous for any of the bacterial strains studied.

CONCLUSIONS

Levofloxacin appears to be effective in treating experimental endophthalmitis in the rabbit model. Further studies, especially on the toxicity of intravitreal levofloxacin, are required before a clinical role for the drug in postoperative therapy can be determined.