Ординатура / Офтальмология / Английские материалы / Atlas of Glaucoma, Second Edition_Choplin, Lundy_2007
.pdf
82 Atlas of glaucoma
SUMMARY OF RESULTS FOR |
opportunity to objectively compare measurements |
|
GLAUCOMA DETECTION |
longitudinally in a given individual and thereby |
|
Historically, few studies have compared the perform- |
establish the progressive nature of the disease. The |
|
high spatial resolution of these devices may allow |
||
ance of multiple laser imaging techniques for identi- |
structural changes to be measured over time that are |
|
fying glaucoma in a single population. Comparisons |
too subtle to be detected by more conventional means |
|
within the same population are more meaningful than |
such as photography. However, for a true change to |
|
comparisons across populations (i.e. across studies) |
be detected, the change needs to exceed the vari- |
|
because the demographic characteristics, including |
ability of the measurement. Measurement variability |
|
glaucoma severity, of single-population studies are |
can be calculated by obtaining multiple scans at an |
|
controlled. Table 7.1 shows results from the only |
imaging session. Change that exceeds this measure- |
|
currently available study comparing new-generation |
ment variability can then be automatically meas- |
|
CSLO (HRT II), OCT (StratusOCT) and SLP (GDx |
ured and assessed for repeatability over successive |
|
VCC) in one population. No other published studies |
imaging sessions in order to confirm that a change has |
|
have compared two or more of these techniques, at |
truly occurred. Less sophisticated change analyses |
|
the time of writing. |
simply document the amount of change from |
|
Several studies investigating new-generation |
baseline measurements by subtraction. |
|
CSLO, OCT, SLP and RTA individually have shown |
|
|
that all techniques can discriminate between |
|
|
glaucomatous eyes (defined by abnormal standard |
Confocal scanning laser ophthalmoscopy |
|
automated perimetry) and healthy eyes. In addition, |
The HRT II offers two methods of measuring change |
|
studies investigating previous-generation tech- |
||
over time: Topographic Change Analysis and stereo- |
||
niques showed similar results for CSLO and OCT. |
||
metric parameter change. |
||
GDx VCC has been shown to discriminate better |
||
|
||
between glaucomatous and healthy eyes and to be |
|
|
better correlated with visual function measured with |
Topographic Change Analysis |
|
standard automated perimetry compared to the |
This analysis is performed automatically once suffi- |
|
previous-generation SLP (GDx with fixed corneal |
cient examinations have been acquired (at least three |
|
birefringence compensator). Significantly fewer |
consecutive examinations are recommended). The |
|
studies have investigated glaucoma detection using |
baseline and follow-up topographies are normalized |
|
RTA, compared to CSLO, OCT and SLP. |
to each other by correcting for magnification, displace- |
|
|
ment, rotation, and tilt between images. This process |
|
TECHNIQUES FOR DETECTING |
is independent of a reference plane or contour line. |
|
Clusters of 4 4 adjacent pixels are combined to |
||
CHANGE OVER TIME |
create super-pixels, which are colored if there is a |
|
An advantage of these imaging instruments is the |
significant local height change. The result is the |
|
change probability map, with red colors indicating |
||
plethora of quantitative and reproducible data that |
significantly more depression in the follow-up |
|
can be obtained. Given the large interindividual vari- |
examination – or glaucomatous progression (green |
|
ability in the characteristics of the optic nerve head |
colors indicate a significant elevation or improve- |
|
and RNFL which renders cross-sectional diagnosis |
ment). A cluster of at least 20 connected super- |
|
of glaucoma problematic, these instruments offer the |
pixels occurring in at least two or three (three is |
Table 7.1 ROC curve areas and sensitivities at fixed specificities for discriminating between glaucomatous and healthy eyes using the ‘best’ (largest ROC curve area) HRT II, StratusOCT and GDx VCC parameter. Population included 75 glaucoma eyes with average standard perimetry MD –4.87 dB (SD 3.9 dB) and
66 healthy eyes
Parameter |
Area under |
Sensitivity/specificity (%) |
|
ROC curve |
|
HRT II Vertical cup/disc ratio* |
0.83 |
59/95, 69/80 |
StratusOCT inferior RNFL thickness |
0.92 |
64/95, 89/80 |
GDx VCC Nerve Fiber Indicator |
0.91 |
61/97, 87/80 |
*HRT Moorfields Regression Analysis results performed similarly to vertical cup/disc ratio, but results are not reported because ROC curve areas and sensitivities at fixed specificities cannot be calculated for categorical variables.
|
Scanning laser imaging 83 |
||
|
|
|
|
suggested) consecutive follow-up topographies indi- |
can be displayed in a diagram along the time axis |
||
cates significant change (an error probability of 5% |
(a trend analysis) (Figure 7.11). |
||
compared to variability in the baseline topography) |
|
|
|
(Figure 7.10). |
Optical coherence tomography |
||
|
|||
Stereometric parameter change |
For assessing RNFL thickness change, the StratusOCT |
||
The stereometric parameter values can be exam- |
incorporates both an ‘RNFL Thickness Change |
||
ined for change over time. This requires a contour |
Analysis’ and an ‘RNFL Thickness Serial Analysis’. |
||
line to be placed around the optic nerve head on the |
The RNFL Thickness Change Analysis graphically |
||
baseline examination (see above). The parameter |
presents the change in RNFL thickness (x-axis) |
||
values inside that contour are then computed and |
around the optic disc (shown in TSNIT format along |
||
the contour line automatically transferred to follow- |
the y-axis) between two examinations. In addition, |
||
up examinations that are normalized to the baseline. |
changes (in micrometers) for each quadrant and for |
||
Stereometric parameter values are re-computed for |
12, 30 sectors are shown as pie charts. The ‘RNFL |
||
follow-up examinations and compared to the base- |
Thickness Serial Analysis’ displays the RNFL thick- |
||
line values. The result is the amount of change for |
ness (x-axis) around the optic disc (y-axis) for up to |
||
each parameter. The significance of a detected change |
four examinations on the same axes. Both reports |
||
can be estimated from the typical variability of the |
can be used to compare RNFL thickness measure- |
||
parameter values of about 5%. The parameter values |
ments between tests over time (Figure 7.12). |
||
Figure 7.10 HRT II Topographic Change Analysis. HRT II Topographic Change Analysis of an optic disc of a glaucoma patient over a period of 8 years, using three consecutive follow-up topographies to confirm change. The topographical map on the left is the baseline topography, with the follow-up topographies arranged chronologically to the right. Areas of significant depression in local height measurements (colored red) denote areas of the neuroretinal rim and peripapillary retina that are progressively enlarging with time. A progressively enlarging area of depression of height is seen in the temporal-inferior peripapillary area.
84 Atlas of glaucoma
Figure 7.11 HRT II Trend Analysis. HRT II Trend Analysis report of the same patient illustrated in Figure 7.10. The global, temporal-inferior and temporal-superior regions are shown for neuroretinal rim volume. A progressive reduction of rim volume in the temporal-inferior region is evident and consistent with the temporal-inferior peripapillary depression shown in Figure 7.10.
Figure 7.12 RNFL thickness change analysis. A composite figure illustrating available StratusOCT information for detecting RNFL thickness change over time. For the comparison graph, the purple curve is the first scan and the blue curve is the second. This glaucoma patient (different from that shown in Figures 7.10 and 7.11) shows some RNFL thinning superiorly, inferiorly and inferotemporally evident on the thickness change (top) and comparison thickness (bottom) graphs. The change is shown in micrometers by quadrant and 30 sector on the RNFL thickness charts. Some RNFL thickness improvement is also shown.
Scanning laser imaging 85
No change detection analyses currently are available for StratusOCT evaluation of optic disc topography or macular thickness.
line, allowing comparison of the RNFL profile over time (Figure 7.13).
Scanning laser polarimetry
The GDx VCC provides an ‘Advanced Serial Analysis’ report that displays a selection of up to four scans of the same eye chronologically, depicting the changes in RNFL with time (Figure 7.13). The report presents a Nerve Fiber Layer map, Deviation from Normal map, Difference from Baseline map, and TSNIT Parameter Table with trend analysis. The Deviation from Normal and Difference from Baseline maps are color coded to show probability of measurement compared to the normative database, and micrometer change from baseline, respectively. Additionally, a combination TSNIT graph shows the TSNIT RNFL thickness measurements at baseline and at each follow-up as a different-colored
Retinal thickness analysis
The Retinal Thickness Analyzer (RTA) system allows the clinician to generate ‘follow-up reports’ for both retinal thickness and optic disc topography. These reports are produced when a patient is scanned at least twice using the same analysis type. For optic disc topography, a cup/disc area map is shown for each consecutive scan, a deviation map illustrates areas of thickness deviation compared to the baseline thickness map, and stereometric parameters, with deviation from baseline values, are shown for each follow-up examination. Regression analysis results also are shown for all images, and changes in these results may signal glaucomatous progression (Figure 7.14).
Figure 7.13 GDx VCC Scanning Laser Polarimeter. GDx VCC Scanning Laser Polarimeter ‘Advanced Serial Analysis’ report from the same patient illustrated in Figure 7.12. The examination at the top of the report is considered the reference (baseline) measurement for the selected serial analysis. Compared to baseline, the follow-up examination shows a slight decrease in RNFL thickness, based on the Nerve Fiber Layer map (left). This decrease is reflected as an increase in the area of the image that is outside normal limits based on the Deviation from Normal map, and a corresponding area of the image showing RNFL thinning compared to the reference image (arrow on Difference from Baseline map). In addition, one more TSNIT parameter is outside normal limits (inferior average RNFL thickness), the NFI has increased from 33 to 43, the Trend Analysis indicates a slight downward slope in all parameters, and the TSNIT graph shows a slight decrease in RNFL thickness superiorly and inferiorly.
86 Atlas of glaucoma
Figure 7.14 Retinal Thickness Analyzer. A RTA glaucoma follow-up report from a healthy eye is shown for illustrative purposes. The upper figure includes the baseline disc area image, with overlying rim/cup area, and the lower figure includes the follow-up image. Potential changes in cup size would be shown on the deviation map. Deviation from baseline values for all parameters are shown in the lower right table.
SUMMARY OF RESULTS FOR
CHANGE DETECTION
Because glaucoma is a slowly progressing disease, and because the newest-generation imaging technologies described in this chapter have been in use for only a few years, no studies have yet reported the ability of HRT II, StratusOCT, GDx VCC, and RTA optic disc topography measurement to detect change over time.
One relevant study showed change over time in retinal height measurements in glaucomatous eyes using a previous version of the currently available confocal scanning laser ophthalmoscopy topographic change analysis software for the HRT. This study showed significant change in retinal height (i.e. tissue thinning) in 69% of 77 eyes followed for approximately 5 years. Significant change in visual function accompanied 27% of these eyes and tissue thinning was corroborated by assessment of stereoscopic optic disc photographs in 80% of those eyes for which stereophotographs were available.
CONCLUSIONS
Computer-based imaging techniques show considerable promise for detecting abnormalities in eyes with glaucoma. The fact that they provide objective, reproducible, quantitative measurements ondemand makes them clinically attractive. While the new-generation instruments discussed in this chapter have not been well tested for glaucoma detection, they appear to perform as well as, or better than, their predecessors. Although it is not recommended that clinical diagnoses be made based solely on measurements from these instruments, it appears obvious that the measurements obtained can supplement other available clinical information. However, clinicians need to be aware of the strengths and limitations of these techniques, including understanding the importance of good image quality, before utilizing imaging data for clinical decision-making. Finally, although these instruments appear promising for the detection of glaucomatous change over time, available scientific
Scanning laser imaging 87
data neither sufficiently support nor reject this claim. Further research is necessary to provide a
strong, evidence-based recommendation for their use in the monitoring of this disease.
Further reading
Chauhan BC, McCormick TA, Nicolela MT, LeBlanc RP. Optic disc and visual field changes in a prospective longitudinal study of patients with glaucoma: comparison of scanning laser tomography with conventional perimetry and optic disc photography. Arch Ophthalmol 2001; 119: 1492–9
Greenfield DS. Optic nerve and retinal nerve fiber layer analyzers in glaucoma. Curr Opin Ophthalmol 2002; 13: 68–76
Medeiros FA, Zangwill LM, Bowd C, Weinreb RN. Comparison of the GDx VCC scanning laser polarimeter, HRT II confocal scanning laser ophthalmoscope, and stratus OCT optical coherence tomograph for the detection of glaucoma. Arch Ophthalmol 2004; 122: 827–37
Medeiros FA, Zangwill LM, Bowd C, et al. Evaluation of retinal nerve fiber layer, optic nerve head, and macular thickness
measurements for glaucoma detection using optical coherence tomography. Am J Ophthalmol 2005; 139: 44–55
Reus NJ, Lemij HG. Diagnostic accuracy of the GDx VCC for glaucoma. Ophthalmology 2004; 111: 1860–5
Wollstein G, Garway-Heath DF, Hitchings RA. Identification of early glaucoma cases with the scanning laser ophthalmoscope. Ophthalmology 1998; 105: 1557–63
Zangwill LM, Medeiros FA, Bowd C, Weinreb RN. Optic nerve imaging: recent advances. In: Grehn F, Stamper R, eds. Glaucoma. Berlin: Springer-Verlag, 2004; 63–88
Zeimer R, Asrani S, Zou S, et al. Quantitative detection of glaucomatous damage at the posterior pole by retinal thickness mapping. A pilot study. Ophthalmology 1998; 105: 224–31
8 Psychophysical and electrophysiological testing in glaucoma: visual fields and other functional tests
Neil T Choplin
INTRODUCTION
The detection of nerve fiber loss and prevention of its development and progression is the ultimate goal of the ophthalmologist in the diagnosis and management of patients with glaucoma. At the present time, there is no proven reliable and consistent way to ‘count’ optic nerve fibers, compare the ‘count’ to known normals to determine the presence or absence of glaucomatous disease, and accurately determine whether the ‘count’ is changing over time. As axons are lost through the disease process, visual function declines in relation to the loss of fibers serving the region of the loss. Therefore, tests of optic nerve function are integral to the management of glaucoma patients as an indirect measure of the number of axons remaining.
Tests of optic nerve function can be objective or subjective. A variety of different test modalities have been investigated, looking for the ideal test that would be: (1) easy to administer; (2) mostly objective (requiring minimal decision-making or other efforts on the part of the patient); (3) highly sensitive to early loss or minimal change in the disease state; and (4) specific for glaucoma.
No objective test meeting these criteria has been found. Clinical functional testing for glaucoma consists mostly of ‘psychophysical’ tests, which by their very nature are highly subjective and susceptible to the shortcomings of human subjects. These tests require the patients to perceive something, intellectually interpret what was perceived,
remember the instructions regarding what is supposed to be said or done if the correct ‘something’ was perceived, and effect some type of verbal or motor response. Perception, memory, interpretation and action thus characterize such psychophysical tests, and can all (individually, collectively or in some combination) lead to variability in the results. This can sometimes lead to false conclusions regarding the state of the disease process and inappropriate treatment.
FUNCTIONAL TESTS IN GLAUCOMA
There is good evidence that chronic glaucoma selectively damages large optic nerve fibers, although fibers of all sizes are damaged. Functional testing, therefore, should be directed at the visual tasks subserved by these axons in an attempt to identify early damage (Table 8.1). Some investigators have shown associated functional changes attributable to the loss of large optic nerve fibers, including declines in contrast sensitivity at high spatial frequencies, loss of pattern-evoked electroretinographic responses (PERG) after death of ganglion cells, and attenuation of PERG responses for high temporal frequencies of stimulation, supporting the idea that some part of glaucoma damage involves loss of the function of these large-diameter nerve fibers. Other investigators have suggested that the PERG may be helpful in identifying patients with ocular hypertension who will develop signs of glaucoma.
89
90 Atlas of glaucoma
Table 8.1 Functional tests other than standard visual fields that have been used for glaucoma. Many different electrophysiological and psychophysical tests have been investigated in glaucoma, particularly in an attempt to identify early damage or risk factors for development or progression of damage. This table lists the various modalities that have been investigated, how they are performed, what they measure and what the findings are in glaucoma patients
Test name |
Objective or |
How performed |
What it measures |
Findings in |
Reference |
|
subjective |
|
|
glaucoma |
|
Central contrast |
Subjective |
Low-contrast |
Minimum differ- |
Mean thresholds |
Atkin et al., |
sensitivity |
|
flickering stimuli |
ence in luminance |
were lower than |
1979; 1980 |
|
|
presented to the |
between stimulus |
normal controls. |
|
|
|
central four |
and background of |
Some ocular hyper- |
|
|
|
degrees |
a flickering |
tensive patients |
|
|
|
|
stimulus |
also show reduction |
|
|
|
|
|
in threshold |
|
Temporal contrast |
Subjective |
Stimulus of |
Intensity of stimu- |
Frequency-specific |
Breton |
sensitivity |
|
increasing |
lus required to |
loss at 15 Hz, non- |
et al., 1991 |
|
|
frequency of |
perceive flicker at |
frequency specific |
|
|
|
flicker presented |
each presented |
mean sensitivity |
|
|
|
to the fovea |
flicker frequency |
loss greater in |
|
|
|
|
|
glaucoma patients |
|
|
|
|
|
than suspects |
|
Pattern-evoked |
Objective |
Reversing |
Waveform gener- |
Second negative |
Trick, 1985; |
electro-retinographic |
|
checkerboard |
ated, latencies of |
wave selectively |
Weinstein |
responses (PERG) |
|
pattern presented |
responses and |
depressed in |
et al., 1988 |
|
|
on a television |
amplitudes for |
patients with glau- |
|
|
|
screen at varying |
each wave compo- |
coma. Some ocular |
|
|
|
contrast while |
nent |
hypertensives |
|
|
|
standard ERG |
|
showed similar |
|
|
|
measurements |
|
responses |
|
|
|
are obtained |
|
|
|
Contrast sensitivity |
Subjective |
Patient views |
Ability to detect |
Decreased contrast |
Nadler |
|
|
figures of decreas- |
correct orientation |
acuity in glaucoma |
et al., 1990 |
|
|
ing contrast at |
of stripe pattern as |
patients, particu- |
|
|
|
multiple spatial |
contrast |
larly at high spatial |
|
|
|
frequencies and |
decreases |
frequencies |
|
|
|
identifies orienta- |
|
|
|
|
|
tion of stripe |
|
|
|
|
|
pattern |
|
|
|
Color vision |
Subjective |
Farnsworth– |
Defects in color |
Blue-yellow defects |
Sample |
|
|
Munsell 100 Hue |
vision |
may occur early in |
et al., 1986 |
|
|
Test, Farnsworth |
|
glaucoma, red- |
|
|
|
D-15 Test, Nagel or |
|
green defects |
|
|
|
Pickford–Nicholson |
|
appear with |
|
|
|
anomaloscope |
|
advanced optic |
|
|
|
|
|
neuropathy |
|
Color contrast sen- |
Subjective |
Computer-driven |
Color contrast |
Sensitivity to blue |
Gunduz |
sitivity |
|
color television |
threshold as a |
and red lights (rela- |
et al., 1988 |
|
|
system |
fraction of the |
tive to green) sig- |
|
|
|
|
maximum color |
nificantly less than |
|
|
|
|
available for color |
controls |
|
|
|
|
combinations on |
|
|
|
|
|
each color axis |
|
|
|
|
|
|
|
|
Psychophysical and electrophysiological testing in glaucoma 91
Test name |
Objective or |
How performed |
|
subjective |
|
Scotopic retinal |
Subjective |
Patient dark |
sensitivity |
|
adapted then pre- |
|
|
sented with flicker- |
|
|
ing stimulus |
|
|
increasing in lumi- |
|
|
nance until seen |
Flicker perimetry |
Subjective |
Static visual field |
|
|
testing performed |
|
|
with target flicker- |
|
|
ing at 25 |
|
|
flashes/second |
Flicker visual- |
Objective |
Standard VEP |
evoked potential |
|
recording while |
|
|
viewing a flickering |
|
|
stimulus of varying |
|
|
frequency |
Color pattern- |
Objective |
Standard VEP |
reversal visual- |
|
recording while |
evoked potential |
|
viewing a reversing |
|
|
checkerboard of |
|
|
black-white, black- |
|
|
red or black-blue |
Visual-evoked |
Objective |
Standard VEP |
potentials after |
|
measurement |
photostress |
|
before and follow- |
|
|
ing 30 seconds of |
|
|
photostress |
High-pass reso- |
Subjective |
Rings of varying |
lution perimetry |
|
size presented to |
|
|
peripheral retina, |
|
|
patient indicates |
|
|
seen or not seen |
What it measures |
Findings in |
Reference |
|
glaucoma |
|
Sensitivity of dark- |
Reduction in absolute |
Glovinsky |
adapted retina to |
whole-field retinal |
et al., 1992 |
a flashing light |
sensitivity |
|
Threshold for tar- |
Flicker threshold ele- |
Feghali |
get luminance |
vated in glaucoma |
et al., 1991 |
against constant |
patients compared to |
|
background |
normals but not com- |
|
|
pared to nonflickering |
|
|
stimulus |
|
Amplitude and |
Glaucoma patients |
Schmeisser |
phase responses |
showed amplitude loss |
et al., 1992 |
of VEP |
at high flicker frequen- |
|
|
cies, correlating with |
|
|
visual field damage |
|
P1-wave peak |
Glaucoma and ocular |
Shih et al., |
time and ampli- |
hypertensives showed |
1991 |
tude for each pat- |
significant decreases |
|
tern |
in the measured |
|
|
parameters compared |
|
|
to normal, especially |
|
|
to the black-red and |
|
|
black-blue checker- |
|
|
boards |
|
Time for VEP |
Longer VEP recovery |
Parisi and |
recording to return |
time required for glau- |
Bucci, 1992 |
to prestress |
coma patients with |
|
baseline |
intermediate values in |
|
|
ocular hypertensives |
|
|
compared to normals |
|
'Ring' threshold, |
Similar to standard |
Sample |
peripheral visual |
perimetry except that |
et al., 1992 |
acuity |
target size, rather |
|
|
than luminance, is the |
|
|
variable |
|
Blue-on-yellow |
Subjective |
Modified auto- |
Visual field |
Abnormalities to |
Johnson |
perimetry (short- |
|
mated static |
thresholds to blue |
blue-on-yellow may |
et al., |
wavelength auto- |
|
threshold perimeter |
stimuli on a yellow |
precede those to |
1993a; |
mated perimetry |
|
with yellow back- |
background |
standard |
1993b |
or SWAP) |
|
ground and blue |
|
white-on-white and |
|
|
|
projected stimuli |
|
may predict which |
|
|
|
|
|
patients will develop |
|
|
|
|
|
loss or progress |
|
Multifocal visual- |
Objective |
Standard VEP |
P1-wave peak |
Decreased amplitude |
Goldberg |
evoked |
|
recording while |
time and ampli- |
in a cluster of 3 test |
et al., 2002 |
potentials |
|
viewing a reversing |
tude for each |
locations |
|
|
|
white-black |
location |
|
|
|
|
checkerboard |
|
|
|
|
|
which tests 60 |
|
|
|
|
|
locations simulta- |
|
|
|
|
|
neously |
|
|
|
