Polymerase chain reaction analysis of corneal epithelial and tear samples in the diagnosis of Acanthamoeba keratitis.

PURPOSE Acanthamoeba is an uncommon cause of corneal infection in which the best visual outcome follows prompt diagnosis and a long course of appropriate antimicrobial therapy. Because conventional detection techniques for Acanthamoeba have certain limitations, we investigated the ability of the polymerase chain reaction (PCR) to confirm the clinical diagnosis of Acanthamoeba keratitis, with the ultimate aim of achieving early diagnosis. METHODS Using two different pairs of primers, PCR was performed on representative cultured Acanthamoeba isolates to confirm the assay's ability to amplify Acanthamoeba DNA from a wide range of acanthamoebae. Subsequently, corneal epithelial samples from 19 patients and tear samples from 12 patients with Acanthamoeba keratitis were analyzed by PCR for the presence of Acanthamoeba DNA. RESULTS Acanthamoeba DNA was amplified by PCR from 16 (84%) of 19 corneal epithelial samples, whereas Acanthamoeba was cultured from 10 samples (53%), all of which were PCR positive. Tear samples from 8 (66%) of 12 patients were positive on PCR testing, and one tear sample was PCR positive, whereas the corresponding epithelial biopsy had yielded a negative PCR result. Samples from culture-positive patients were positive on PCR testing more frequently than those from culture-negative patients (10/10 culture-positive corneal epithelial and 5/7 [71%] culture-positive initial tear samples versus 6/9 [66%] culture-negative corneal epithelial and 2/5 [40%] culture-negative tear samples). All control epithelial (n = 15) and tear (n = 15) samples yielded negative results. CONCLUSIONS PCR was a more sensitive diagnostic test than a culture for Acanthamoeba keratitis, and the use of two different primers achieved better sensitivity than a single set. A PCR of a tear sample also may be a useful complementary test and, in combination with PCR of epithelial samples, would prove particularly helpful in confirming the clinical diagnosis in culture-negative cases.