We examined 19 eyes of 19 patients with a mean age of 42.3 ±
11.1 (SD) years (range, 26–65 years). The male-to-female ratio was
10:9. Included were 12 eyes with inflammatory (n = 5) or
diabetic (n = 7) macular edema, and 7 eyes with uveitis
without edema. The latter category of eyes with subnormal visual acuity
without evidence of edema on ophthalmoscopy or angiography consisted of
four eyes with a history of macular edema, one eye in which macular
edema developed approximately 1 year after SLO measurement, and two
eyes without macular edema in the past or after a follow-up of 2 and 3
years. Patients with substantial media opacities were excluded.
Corrected visual acuity was measured with the Snellen letter projector
on the day of SLO measurement and 6 months thereafter. Diagnosis of
macular edema was based on both clinical and angiographic criteria. A
recent (less than 6 months previously) fluorescein angiogram was
available for 17 patients. In two patients without edema in recent
ophthalmoscopy (and no need for angiography), an angiogram was
available from approximately 2 years before SLO measurement. Six months
after SLO measurement, a follow-up fluorescein angiogram was made in
all but three patients. Macular edema as shown on late-phase angiogram
(approximately 10 minutes after injection of the dye), was graded by
two masked observers: grade 0 when no edema was found (n = 7), grade 1 when less than 25% of the macular area was affected by
edema (n = 4), grade 2 when the affected area was between
25% and 66% (2/3, n = 7), and grade 3 when more
than 66% of the macular area was affected (n = 1). In case
of discrepancy of grading, the two observers discussed the angiograms
until they agreed on a final angiogram grade. Mean duration of macular
edema was 3.3 years ± 31.4 months (SD), ranging from 7
months to 9 years.
Measurements were performed with a custom-built SLO (the apparatus and
clinical use have been described earlier).
7 8 Normal
values for photoreceptor directionality and density were obtained by
measuring 25 healthy volunteers (average age, 30 ± 11.3 years),
19 of whom were available for cross-section analysis. One or two drops
of tropicamide (0.5%) were instilled into the eye to induce mydriasis.
Accurate positioning and steady fixation of the eye were ensured by two
forehead rests and a bite board with a dental compound. Subjects were
asked to look at a fixation cross. Refraction was corrected if
necessary. The SLO images covered a retinal area of 23° × 18° with
the fovea as center. Directionality and visual pigment density of
photoreceptor cells in the central 2° × 2° area were determined,
because the optical density in this area is almost entirely due to
photopigments of cones that primarily determine central visual
acuity.
9 The maximum reflectance was determined by finding
the position at which the reflectance of the parafoveal cones was at
its maximum.
As mentioned before, the optical SCE is due to the directionality of
the photoreceptor cells. By the joint movement of entrance and exit
beam across the pupil plane in healthy eyes, a change in reflectance
was recorded, and a gaussian-like reflectance function with a peak
close to the middle of the pupil plane was obtained.
10 The
peakedness of the curve, and thus the directional sensitivity, was
defined as ρ. We used the parameter ρ as an indicator of
directional sensitivity for a group of photoreceptors. The average
value for directional sensitivity (ρ) of a normal subject was
0.24 ± 0.13 (2 SD). The SCE was recorded by finding the pupil
position resulting in maximum parafoveal reflection, followed by
acquiring a series of images (approximately every 0.25 mm) from the
nasal to the temporal pupil edge.
For measurement of visual pigment density, the eye was adapted to a
strong light that bleached away more than 95% of the visual pigments.
Next, dark adaptation was performed during which visual pigments fully
regenerated. For imaging and 2 minutes of bleaching at 514 nm (5.9 log
troland [td]) was used. Dark adaptation was performed for 8 minutes
(2.2 log td). A retinal aperture of 0.43° was used. The density of
the visual pigments is estimated by taking the logarithm of the ratio
of light reflection between the light-adapted and dark-adapted
fundus.
7 The implicit assumption was that the difference
in the two images is caused by the appearance of visual pigment. The
average value of a normal subject’s pigment density was 0.37 ±
0.11 (2 SD).
Strong reflections—for instance, from sources anterior to the
photoreceptors—may lead to high levels of stray light. We determined
these levels (average 32 ± 10 arbitrary units [AU]; normal
range, 12–52 AU) by fitting a model of Gorrand and
Delori
10 through the data points, and we also calculated
the reflectance centrally in the fovea.
10 11 12 Cross-section images through the fovea (a line profile of the
reflection along a horizontal line through the fovea) were obtained in
all patients with macular edema and in 19 normal eyes.
The study conformed with the tenets of the Declaration of
Helsinki. Informed consent was obtained from all participating
patients, and the study protocol was approved by the Committee for
Scientific Research in Humans of our hospital. Statistical analysis
included Student’s t-test, χ2 test,
Fisher’s exact test, correlation coefficients, and linear regression
analysis. P ≤ 0.05 was considered statistically
significant.