Acute postoperative bacterial endophthalmitis

Research output: Contribution to journalArticlepeer-review

Abstract

The EVS has provided important background data and has helped to guide treatment in patients with postoperative endophthalmitis. In addition to its two main conclusions that immediate vitrectomy is indicated in patients that present with LP vision, but not those that present with vision of hand movement or better, and that systemic amikacin and ceftazidime provide no benefit, the EVS has provided information leading to additional suggestions: (1) both vitreous and anterior chamber specimens should be obtained for cultures and Gram stain; (2) the current treatment approach does not eradicate infection in a substantial number of patients, and, therefore, new approaches should be sought; (3) even if a second set of intravitreous antibiotics are needed to eliminate the infection, good vision is still possible, and therefore an aggressive approach is justified; (4) presenting vision of LP or a positive gram stain portend a poor visual prognosis and a substantial chance of persistent infection after the first procedure and, therefore, the threshold for additional intervention should be lowered; and (5) signs of worsening inflammation after the first post-treatment day suggest persistent infection, and additional cultures should be done and intravitreous antibiotics should be injected. If the patient has not had vitrectomy, it should be performed. The EVS did not deal with the choice of antibiotics, but because of other information indicating that there is substantial risk associated with the use of amikacin, vancomycin and ceftazidime should be used for intravitreous injections.

Original languageEnglish (US)
Pages (from-to)83-88
Number of pages6
JournalOphthalmology Clinics of North America
Volume12
Issue number1
DOIs
StatePublished - 1999

ASJC Scopus subject areas

  • Ophthalmology

Fingerprint

Dive into the research topics of 'Acute postoperative bacterial endophthalmitis'. Together they form a unique fingerprint.

Cite this