Purchase this article with an account.
C. Laurent; Bacterial identification using universal PCR in endophthalmitis. . Invest. Ophthalmol. Vis. Sci. 2004;45(13):4004.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose:To improve identification and speciation of bacteria involved in acute endophthalmitis, using polymerase chain reaction (PCR) technique. Methods:This prospective study (2002–2003) included 28 patients (mean age 67.5 years, 22–96) with endophthalmitis after cataract surgery (n = 17), glaucoma surgery (n = 5), secondary implantation (n = 4) or radial keratotomy (n = 1). One patient had an endogenous endophthalmitis. In emergency, an anterior chamber puncture was performed under aseptic conditions (associated with intravitreal injection of antibiotics). Aqueous humor samples were collected in every patients: 100 microL for standard culture and 150 microL for PCR. In 11 patient (40%) a posterior vitrectomy was performed 2–5 days after the first intravitreal injection of antibiotics and vitreous samples were analysed using the same techniques. Results:In aqueous humor, microbiological diagnosis was made on 31% of the samples using cultures (1 Staphylococcus aureus, 2 Enterococcus faecalis, 2 Staphylococcus epidermidis, 2 Streptococcus pneumonaie, 1 Moraxella) and on 60 % using PCR (2 Enteroccus faecalis, 3 Staphylococcus epidermidis, 3 Streptococcus pneumoniae, 5 Moraxella, 1 Streptococcus sanguis, 1 Streptococcus deficiens, 1 Serratia marcescens). In two cases (7%), PCR was negative whereas cultures were positive (1 S. aureus, 1 S. epidermidis). The association of PCR and cultures allowed the identification of the bacteria in 65 % of the cases. The correlation between both techniques was 100%. For the vitreous samples (11/28), the microbiological diagnosis was made in 18% using cultures (1 Staphylococcus epidermidis, 1 Streptococcus sanguis) and in 72% using PCR (1 Enterococcus faecalis, 4 Staphylococcus epidermidis, 2 Streptococcus pneumoniae, 1 Streptococcus sanguis). The culture was negative in cases with negative PCR. In two cases (18%, 2 Staphylococcus epidermidis), the analysis of the vitreous samples allowed the diagnosis whereas the analysis on aqueous humor was negative. In one case (9%, 1 Moraxella) the analysis of the vitreous sample did not detect the bacteria detected in aqueous humor. Considering aqueous humor and vitreous samples, the infectious agent was identified in 75% of the cases. Conclusions:These results confirm that the use of PCR offers great advantage compared to conventional microbiologic testing and demonstrates that the association of culture and PCR improves the identification of the causative pathogens in endophthalmitis. A rapid and exact diagnosis is still critically important for a good functional outcome. In this context PCR is helpful for an appropriate treatment.
This PDF is available to Subscribers Only