Evaluation and comparison of commercial vitreous fluorophotometers.

The Metricon vitreous fluorophotometer and two models by Gamma Scientific were evaluated. The lower limit of detection of the Metricon and the 2900 Gamma Scientific were found to be satisfactory. Because of the DR2 Gamma Scientific high dark noise, an external electronic filter had to be added to achieve a lower limit of detection, which was borderline for vitreous fluorophotometry. The error of the systems was found to be acceptable except for the DR2 Gamma Scientific at low concentrations. However, it was demonstrated that the determining factor for reliability is the level of artifactual readings. It was shown that the level of artifact is acceptable for the Gamma Scientific models but high for the Metricon, in which significant error could be induced in patients with borderline fluorescein leakage. Other technical aspects of the systems are discussed.