Abstract
Ocular toxoplasmosis is the major cause of posterior uvetis in European populations. The clinical diagnosis of toxoplasmic chorioretinitis is based upon ophthalmoscopic findings, which are often but not always typical. Laboratory testing is therefore important to confirm the etiology of the disease. In the present 2-year prospective study, the relative diagnostic sensitivities of the three analytical techniques (enzyme-linked immunosorbent assay [ELISA], immunoblotting, and PCR) were compared by using a group of patients (n = 19) with suspected ocular toxoplasmosis. The relative specificities of the three techniques were assessed by including two control groups of patients: one with nontoxoplasmic and noninflammatory ocular disease (n = 48) and the other with nontoxoplasmic and inflammatory ocular disease (n = 20). All 19 of the clinically suspect patients had serological evidence of exposure to Toxoplasma gondii: 17 had been previously infected, and 2 had current infection. The analysis of paired aqueous humor and serum samples by ELISA and immunoblotting revealed the local production of specific antibodies of the immunoglobulin G type in 63% (12 of 19) and 53% (10 of 19) of patients, respectively. PCR analysis of aqueous humor samples confirmed the presence of T. gondii DNA in 28% (5 of 18) of cases. When combined, ELISA, immunoblotting, and PCR findings confirmed the toxoplasmic origin of retinal lesions in 83% (15 of 18) of patients. The relative specificities of the three techniques were 89% for ELISA and immunoblotting and 100% for PCR.