Prospective and randomized comparison of frequency doubling perimetry vs standard automated perimetry in patients with glaucoma

Aim: To compare frequency doubling perimetry (FDP) versus standard automated perimetry in glaucoma. To evaluate the reproducibility of both methods. Patients and Methods: All the patients were tested (full threshold test) twice with both FDP (N-30) and Humphrey Field Analyzer II Model 750 (HFA) (30-2 procedures) in random sequence, within one day. The parameters mean deviation (MD), pattern standard deviation (PSD) and measured thresholds per quadrant and center areas were evaluated for FDP/HFA comparison in 45 glaucoma patients. The same parameters were used for testing the reproducibility of FDP (n = 46) and HFA (N = 39). Additionally, the center and each quadrant area were checked for any deviation in the total deviation probability plots. For statistical analysis Kappa coefficients and Bland/Altman plots were used. Results: An average MD (FDP vs HFA) of -7.3+/-5.8 dB vs -8.9 +/- 7.6 dB and PSD of 6.9 +/- 2.4 dB vs 6.5 +/- 3.6 dB were found. Kappa coefficients denote marginal accordance (kappa: 0.11-0.38) for area deviations. In a total of 225 areas HFA detected 191 deviations and FDP 165. HFA measured more negative deviation compared with FDP in the case of MD < -15 dB. The reproducibility was 0.98 (MD) and 0.92 (PSD) for FDP and 0.98 (MD) and 0.95 (PSD) for HFA (95% confidence interval). Conclusions: There was great conformity between FDP and HFA in glaucoma patients. HFA detected more deviations in the total deviation probability plots than FDP. There was a high reproducibility of both methods. FDP is an appropriate tool for detecting visual field loss in glaucoma patients.