High- versus low-density multifocal pupillographic objective perimetry in glaucoma


  • Conflict/competing interest: T Maddess receives royalty income from Carl Zeiss Meditec arising from the sale of the Matrix/FDT perimeters. He and Dr AC James are the inventors of several patents and patent applications currently under licence by the Australian National University to Seeing Machines Pty, the builder of the TrueField prototype used here. That IP relates mainly to the methods used in the TrueField.
  • Funding sources: NHMRC Development grant 410201 and Seeing Machines Pty.

Correspondence: Professor Ted Maddess, ARC Vision Centre, John Curtin School of Medical Research, Australian National University, Canberra, ACT 0200, Australia. Email: ted.maddess@anu.edu.au



Multifocal pupillographic objective perimetry was compared using 24 and 44 regions per visual field.


Experimental design in a university setting.


Twenty-seven normal control and 36 age-matched glaucoma patients.


The four test variants differed in the mean interval between stimuli: 4, 1 or 0.25 s; and the number of visual field regions tested within the central 60 degrees: 24 or 44. All subjects had their diagnostic status confirmed by optical coherence tomography, two forms of perimetry and slit-lamp biomicroscopy. Both eyes were measured concurrently in 2.73 ± 0.45 min/eye (mean ± standard deviation), and tests were repeated about 2 weeks apart.

Main Outcome Measures

The main outcome measures were: (i) mean change in light sensitivity due glaucoma; and (ii) areas under Receiver Operator Characteristic plots for detecting glaucoma.


For all four variants, consensual responses, female gender and age produced small but significant sensitivity differences, and sensitivity declined with age by ≤−0.27 dB/decade (all P < 0.0003). The best diagnostic accuracy (area under curve 93.2 ± 3.89%) was produced by the one-presentation/s 44-region protocol. Across the four protocols, the effect of repeat testing was small (all methods ≤0.15 dB).


Presentation rate had little effect, but increasing the tested density from 24 to 44 regions/field improved diagnostic power. Given that multifocal pupillographic objective perimetry also provides information on response delay and afferent versus efferent defects at every visual field region, it may be a useful adjunct to perimetry.