Diagnostic utility of polymerase chain reaction on intraocular specimens to establish the etiology of infectious endophthalmitis.

PURPOSE

To evaluate the utility of polymerase chain reaction (PCR) on intraocular clinical specimens (aqueous humor [AH] and vitreous fluid [VF]) as an etiologic diagnostic tool relative to microbiological culture methods in infectious endophthalmitis.

METHODS

Conventional bacterial and mycologic cultures and PCR for eubacterial and panfungal genomes were applied for etiologic diagnosis on pairs of AH and VF obtained from 72 patients with clinically established infectious endophthalmitis.

RESULTS

Based on cultures, an infectious etiology was established in 27 (37.5%) of 72 patients. PCR detected infectious etiology in all 72 patients. PCR increased the clinical sensitivity over culture by 62.5% (p<0.0001, McNemar test). The frequency of culture positivity, single infections, and polymicrobial infection varied significantly among the types of endophthalmitis (p<0.0001, chi-square test). PCR detected an infectious etiology in 48 patients and polymicrobial infection in 24 patients. An etiology was established by PCR on 56 (77.8%) AH and 65 (90.3%) VF of the 72 patients and this difference had no statistical significance.

CONCLUSIONS

PCR on intraocular specimens as an etiologic diagnostic tool has been shown to be specific and severalfold more sensitive than cultures and clinically useful. Therefore, PCR may be considered the gold standard to establish the etiology of infectious endophthalmitis. As there is no statistically significant difference in the results of PCR on AH and VF, PCR on AH could be the method of choice considering safety and simplicity of the procedure of its collection.