Performance of frequency-doubling technology perimetry in a population-based prevalence survey of glaucoma - The Tajimi study

Objective: To evaluate the performance of frequency-doubling technology (FDT) perimetry in a population-based glaucoma prevalence survey. Design: Population-based cross-sectional study. Participants: Participants older than 40 years randomly selected from the population of Tajimi City. Methods: Each participant underwent screening ophthalmic examinations including a visual field test using FDT with the C-20-1 screening protocol. A diagnosis of glaucoma was determined by glaucoma specialists with another detailed visual field test using Humphrey Field Analyzer (HFA; Humphrey Instruments, San Leandro, CA) with the 30-2 Swedish interactive threshold algorithm standard protocol and stereoscopic disc photographs. Main Outcome Measures: The ratios of reliable FDT results and the sensitivity and specificity for detecting glaucoma in a general population. Results: Of 5784 eyes of 2892 participants (age range, 40-92 years; refractive error, -23 to 11 diopters) in whom FDT was performed in both eyes, reliable results (<= 33% fixation loss and <= 33% false-positive errors) were obtained in 5707 eyes (98.7%), including 2871 right eyes (99.3%) and 2836 left eyes (98.1%) with a significant bilateral difference (P < 0.001, chi-square test). The rate of reliable FDT results did not differ between men and women (P = 0.81) but decreased with age. In 5582 eyes with reliable FDT results, FDT showed 1 or more abnormal points in the visual field in 502 eyes (9.0%), including 388 (7.3%) of 5295 normal eyes, 19 (16.4%) of 116 eyes of glaucoma suspects, and 95 (55.6%) of 171 eyes with definite glaucoma. The sensitivity and specificity values for detecting definite glaucoma were 55.6% and 92.7%, respectively. The positive and negative predictive values were 18.9% and 98.5%, respectively. In further analyses stratified with the mean deviation (MD) of the HFA, the sensitivities were 32.1%, 48.4%, 73.7%, and 96.6% for detecting definite glaucoma with an MD of more than -2 dB, an MD of -2 dB or less and more than -5 dB, an IVID of -5 dB or less and more than -8 dB, and an MD of -8 dB or less, respectively. Conclusions: In a population-based glaucoma screening study, FDT perimetry with the C-20-1 screening protocol was reliably performed in more than 98% of participants. The sensitivity for detecting glaucomatous visual field damages, especially early damage, was not sufficiently high, whereas the specificity was high.