Research in Ophthalmology 2012, 1(1): 1-5
DOI: 10.5923/j.ophthal.20120101.01
Humphrey Matrix Perimetry for Open-Angle Glaucoma
Screening in A High-Risk Population
Colin I Clement
1,2,3,*
, Ivan Goldberg
1,2,3
, Stuart L. Graham
2,4
,Paul R Healey
2,3,5
1
Glaucoma Unit, Sydney Eye Hospital, Sydney, 2000, Australia
2
Eye Associates, Sydney, 2000, Australia
3
The University of Sydney, Sydney, 2006 Australia
4
The Australian School of Advanced Medicine, Macquarie University, Sydney, 2109, Australia
5
Westmead Millenium Institute for Medical Research, Sydney, 2145, Australia
Abstract The aim of this study was to evaluate the role of Humphrey Matrix perimetry to better guide appropriate re-
ferral of at-risk individuals for glaucoma assessment. Fifty-two consecutive patients referred to a specialist glaucoma oph-
thalmology practice for an opinion about a new diagnosis of glaucoma were enrolled. Prior to assessment for presence of
open-angle glaucoma, patients performed Humphrey Matrix perimetry with results withheld from both patient and assess-
ing glaucoma specialist. Humphrey Matrix perimetry results were later compared with the outcome of clinical assessment
for each individual. Eleven of 52 participants were diagnosed with open-angle glaucoma (21.2%). Negative predictive
value of Humphrey Matrix perimetry in this population ranged from 87.1% to 93.6% depending on the indices used. Only
two individuals with normal Humphrey Matrix perimetry were diagnosed with open-angle glaucoma. These results suggest
Humphrey Matrix perimetry displays a high negative predictive value when screening for open-angle glaucoma in an
at-risk population. It may be useful in this context to better guide eye health professionals on appropriate referral of patients
for an opinion about glaucoma.
Keywords Glaucoma, Screening, Perimetry, Frequency Doubling Perimetry
1. Introduction
Glaucoma is an important cause of visual impairment with
an overall prevalence of 2%[1,2]. In most developed nations,
half the glaucoma in the community remains undiagnosed[1].
As glaucoma is a progressive disorder that may cause
blindness or significant visual field loss, earlier recognition
and intervention are important.
Examination by an eye health care professional for rea-
sons other than glaucoma provides an opportunity to detect
and appropriately refer individuals with undiagnosed glau-
coma. However, eye health care professionals may not ac-
curately detect such cases and may miss the diagnosis in a
significant number of individuals[3].
In addition to risk factor assessment and optic disc ex-
amination, perimetry may have a useful role in screening for
undiagnosed glaucoma. Problems with the use of perimetry
for screening include expense of equipment, long test times
and high false positive rates amongst naive subjects[4].
Mitchell et al.,[1] reported false positive rates of 20% with-
supra-threshold perimetry, which would overwhelm our
* Corresponding author:
colinc1@med.usyd.edu.au (Colin Clement)
Published online at http://journal.sapub.org/ ophthal
Copyright © 2012 Scientific & Academic Publishing. All Rights Reserved
ophthalmic services if such screening yielded referrals.
While supra-threshold perimetry in a reduced number of
stimulus locations may reduce the false positive rate, it does
so at the cost of sensitivity[1,5].
Frequency-doubling threshold (FDT) perimetry was de-
veloped in the hope of better detecting early glaucomatous
visual field loss and may be suitable for this purpose. It uses
a stimulus that non-linearly stimulates magnocellular retinal
ganglion cells[6]. Perhaps because of reduced redundancy,
damage to these cells can be detected relatively early in
glaucoma[7-9]. FDT perimetry is comparable with achro-
matic perimetry for detecting established mild, moderate and
severe field loss in glaucoma[10,11] and it identifies sig-
nificantly more patients with ocular hypertension that pro-
gress to glaucoma than does conventional achromatic pe-
rimetry[12,13].
A newer version of the FDT perimeter is the Humphrey
‘Matrix’, which uses a 5-degree rather than 10-degree square
stimulus. For the cental 24 degrees of field, the Humphrey
Matrix tests 55 zones verses 17 in the original machine. This
may yield improved performance.
Given the known high false positive rates for perimetric
screening, we wished to evaluate the ability of Matrix pe-
rimetry, performed after normal optometric assessment, to
identify patients without glaucoma, thereby reducing the
false positive rate of optometric referral for glaucoma. Thus