Purpose
Goldmann perimetry is the current gold-standard kinetic visual field (VF) test in paediatric practice, but is no longer commercially available, so there is a pressing need to investigate the value of other perimeters. From a prospective, observational study, we describe normative values in children for kinetic perimetry undertaken using Goldmann and Octopus 900 perimeters and compare outputs to inform future policy/practice.
Methods
154 children aged 5-15 years (patients and siblings attending Moorfields Eye Hospital, London), without ocular pathology that could cause a visual field defect, underwent kinetic perimetry with both Goldmann and Octopus perimeters using standardised protocols, plotting two isopters (randomised between isopters III4e, I4e and I2e) and a blind spot. Bland-Altman plots were used to compare isopter area between perimeters and linear quantile mixed-effects regression models (lqmm) of raw data were used to describe normative isopter shape.
Results
Linear regression of visual field area and age showed model coefficients of 176.67 (III4e, 95% CI: [93.8, 259.5]), 167.32 (I4e, [86.4, 248.2]) and 141.7 (I2e, [80.9, 202.5]) for Goldmann isopters. These represent the change in area (deg2) for each additional year of age from 5 to 15 years. For Octopus perimetry, coefficients of 184.4 (III4e, [98.9, 269.8]), 129.1 (I4e, [39.6, 218.5]) and 26.7 (I2e, [-40.9, 94.4]) were found i.e. a significant increase in VF area with age for Goldmann isopters III4e, I4e and I2e and Octopus isopters III4e and I4e.<br /> No significant difference between Goldmann and Octopus VF area was found when comparing isopters III4e (p=0.224), I4e (p=0.205) and I2e (p=0.376). However, lqmm (graphical) outputs demonstrated small, clinically apparent differences in isopter shape and normative confidence estimates between Goldmann and Octopus fields for all plotted isopters e.g. a larger nasal field corresponded with a smaller temporal field, maintaining total VF area.
Conclusions
Kinetic perimetry using the Octopus produces findings consistent with, but not identical to, the Goldmann perimeter in children without VF defects. Thus, interpretation of kinetic fields should be based upon age-specific normative data developed for specific perimeter models. Children undergoing serial VF tests should be assessed using the same model of perimeter at each visit.