Interdevice agreement in the measurement of physiognomy parameters and frame angles to prescribe progressive addition lenses

Background The aim of this study is to describe interdevice agreement between different methods and traditional methods (frame ruler). Methods The agreement of the interpupillary distance, nasopupillary distance and fitting point height measured with four devices (PD-5, OptiCenter, Visioffice and a frame ruler) and of pantoscopic and frame wrap angles measured with three devices (OptiCenter, Visioffice and Essilor standard ruler) was assessed in 21 healthy volunteers, by a Bland-Altman analysis; mean difference and limits of agreement (LoA) were calculated. Results Frame ruler nasopupillary distance measurements showed limited agreement with PD-5 [-0.38 +/- 1.03 (LoA -2.40 to 1.64) and 0.44 +/- 1.02 (LoA -1.72 to 2.61)] and Visioffice [0.62 +/- 1.24 (LoA -1.81 to 3.05) and -0.16 +/- 1.72 (LoA -3.54 to 3.22)] measurements for the right and left eyes, respectively. Poor agreement was found for interpupillary distances (PD-5 [0.21 +/- 1.47 (LoA -2.67 to 3.09)], OptiCenter [-0.05 +/- 1.16 (LoA -2.32 to 2.22)] and Visioffice [0.46 +/- 1.95 (LoA -3.36 to 4.28)]), fitting point height (OptiCenter [-1.27 +/- 2.56 (LoA -6.27 to 3.75) and -0.92 +/- 2.77 (LoA -6.35 to 4.51)] and Visioffice [-5.88 +/- 6.21 (LoA -18.05 to 6.29) and -5.98 +/- 6.12 (LoA -17.98 to 6.02)] for the right and left eyes, respectively) and pantoscopic and frame wrap angles (OptiCenter [-4.13 +/- 3.75 (LoA -11.48 to 3.22) and -1.09 +/- 0.60 (LoA -2.27 to 0.09)] and Visioffice [-6.18 +/- 3.53 (LoA -13.10 to 0.74) and -1.93 +/- 3.49 (LoA -8.77 to 4.91)], respectively). Conclusions These results suggest that measurements of physiognomy and frame angles are not interchangeable between assessed devices and that these differences could induce lens centration errors with a large impact on progressive addition lens prescriptions.