Aim of this study was to evaluate the long-term reproducibility of frequency doubling technology (FDT) screening procedures.
Longitudinal data of 433 eyes of 294 patients with no progression of glaucomatous optic disc atrophy were retrospectively analyzed: 62 control eyes, 184 ocular hypertensive eyes, 104 preperimetric, and 83 perimetric open-angle glaucoma eyes. All subjects had annual tests with the FDT perimeter and a standardized ophthalmologic examination (ie, conventional perimetry, optic disc inspection, tonometry, lens opacity measurement for exclusion of cataract). The present analysis used a published overall screening score with case-wise recalculation of missed localized probability levels. We analyzed long-term variability by correlation analysis, sign tests, and limits of agreement (LoA) as introduced by Altman and Bland. All subjects had at least 2 annual tests. Three hundred twenty-six eyes had 2 annual tests with the C-20 procedure and at least 1 test with the N-30 protocol another year later. One hundred thirty-five eyes had 1 C-20 and 2 annual tests with the N-30 protocol.
Analyses of repeated measurements revealed a significant learning effect (P<0.001, LoA: −4, 17) between the first and second examination but no significant difference between the second and following tests with the C-20 protocol (P>0.6, LoA: ±9). In addition, there was no significant difference between second C-20 and N-30 tests (P>0.5, LoA: −12, 6).
The study demonstrates the variability of FDT tests over several years. Longitudinal FDT-results in a clinical study showed a higher reproducibility if the first test was discarded. Reproducibility of screening with the N-30 protocol is comparable to the C-20 procedure if an overall score is considered.
*Department of Ophthalmology and University Eye Hospital, Friedrich-Alexander University Erlangen-Nürnberg at Erlangen
†Department of Biostatistics and Clinical Epidemiology, Charité University Medicine Berlin, Hindenburgdamm, Berlin, Germany
Supported by: Deutsche Forschungsgemeinschaft, Bonn, Germany (SFB 539).
Proprietary interest: None.
Reprints: Folkert K. Horn, PhD, Universitäts-Augenklinik, Schwabachanlage 6, D-91054 Erlangen, Germany (e-mail: firstname.lastname@example.org).
Received for publication September 28, 2006; accepted March 2, 2007