Abstract
Purpose :
Endophthalmitis is a vision-threatening condition that requires immediate diagnosis and treatment. The aim of our study was to compare different methods of microbiological examination of intraocular fluids and to evaluate the microbiological spectrum and antibiotic resistances.
Methods :
Data from endophthalmitis patients from three large ophthalmic institutions were retrospectively analyzed. One center used Freiburg Endophthalmitis Set (FRES) for germ detection, while the other two centers sent native specimens to a microbiology laboratory for processing. The FRES contains all materials to apply the samples directly in the operating room onto suitable culture media and to prepare für PCR. The following data were evaluated: microbial detection rate, microbial spectrum, and antibiotic resistance.
Results :
A total of 421 cases were identified. Cause of endophthalmitis was surgery in 88%, endogenous in 9% and posttraumatic in 3%. The most common procedures leading to endophthalmitis were cataract surgery and intravitreal drug administration. Germ detection was successful in 54% overall. With FRES the detection rate was 75%, without FRES the rate was 41% and 30%, respectively. Most of the bacteria detected were gram positive (96%), especially coagulase negative staphylococci (64%). While there was no resistance to vancomycin, distinct resistance to 1st/2nd generation cephalosporins, oxacillin, amicoglycosides, tetracyline, clindamycin, and erythromycin existed.
Conclusions :
With the FRES, the germ detection rate can be significantly increased by direct processing of the samples in the operating theater. Vancomycin remains the antibiotic with the best susceptibility to the germs detected. Other antibiotics have considerable resistances, so that it is very useful to determine their susceptibility by means of culture methods. In consequence, empiric therapy should include vancomycin (intravitreal/intravenous) and should be adjusted during the course depending on the resistance pattern.
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.