Effect of Fixation Pressure on Juxtacanalicular Tissue and
Schlemm's Canal
Richard D. Ten Hulzen and Douglas H. Johnson
Purpose. A quantitative study was performed to compare the effect of two commonly used
fixation pressures, 0 mm Hg (immersion) and 10 mm Hg (perfusion), on the porosity of the
juxtacanalicular tissue and the size of Schlemm's canal.
Methods. Twelve pairs of human eyes were studied by fixing one eye with perfusion fixation
and the fellow eye with immersion fixation: Morphometric analysis of the juxtacanalicular
tissue and Schlemm's canal was performed. Outflow resistance was calculated from these
measurements and compared with the measured outflow resistance obtained in six eyes.
Results. Schlemm's canal was narrowed in perfusion-fixed eyes, with a 47% smaller crosssectional area than in immersion-fixed eyes (P = 0.04). Juxtacanalicular tissue of perfusionfixed eyes had a 13.4% increase in the relative amount of empty space when compared with
immersion-fixed fellow eyes (P = 0.04). Solid tissue components were almost equally divided
among amorphous basement membrane, tendon and sheath material, and cytoplasm. No
obvious washout of extracellular material was noted in perfused tissue. Measured outflow
resistance was 100 times larger than outflow resistance of the juxtacanalicular tissue calculated
from histologic measurements.
Conclusions. Perfusion fixation at physiologic intraocular pressure caused a 47% decrease in
the area of Schlemm's canal and a mean increase of 13.4% in the relative amount of empty
space in the juxtacanalicular tissue compared with immersion-fixed fellow eyes. Perfusion of
fixative did not appear to cause washout of extracellular material. Perfusion-fixed tissue appears preferable for studies of Schlemm's canal and for ultrastructural studies of the aqueous
outflow pathways within the juxtacanalicular tissue. Invest Ophthalmol Vis Sci. 1996; 37:114124.
X ressure-induced changes in the conformation of the
trabecular meshwork were first described by Johnstone and Grant1 more than 20 years ago. Subsequent
studies at various levels of intraocular pressure in human and monkey eyes confirmed their findings.2"7 At
an intraocular pressure of 0 mm Hg, Schlemm's canal
appears widely dilated, few giant vacuoles are found,
the juxtacanalicular region appears relatively compact, and the outer trabecular lamellae are closely
spaced.1"6 This conformation prevents blood reflux
from Schlemm's canal to the anterior chamber,
From the Department of Ophthalmology, Mayo Clinic and Mayo Foundation,
Rochester, Minnesota.
Supported in part by National Institutes of Health research grant EY07065, by an
unrestricted grant from Research to Prevent Blindness, Inc., (New York, New York),
and by the Mayo Foundation (Rochester, Minnesota).
Submitted for publication May 15, 1995; revised July 26, 1995; accepted September
14, 1995.
Proprietary interest category: N.
Reprint requests: Douglas H. Johnson, Mayo Clinic, 200 First Street SW, Rochester,
MN 55905.
114
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
allowing the meshwork to function as a one-way valve.1
As the level of pressure within the anterior chamber
increases, Schlemm's canal narrows, the number and
size of giant vacuoles increase, the juxtacanalicular
region expands, and the trabecular lamellae separate.1"7 Johnstone and Grant1 found increasingly
greater differences in the conformation of the meshwork and Schlemm's canal at pressures of 15 mm Hg
and higher. The lumen of the canal was of "moderate" size at 0 and 5 mm Hg, narrowed at 15 mm Hg,
and virtually collapsed at 30 mm Hg. Juxtacanalicular
tissue appeared to expand progressively to fill in the
canal at these higher pressures.1 These studies were
performed in enucleated human eyes, with episcleral
venous pressures of 0 mm Hg.
Despite the more physiologic appearance of the
trabecular meshwork with perfusion fixation (fixative
introduced into the anterior chamber, mimicking the
flow of aqueous), many studies have continued to use
Investigative Ophthalmology & Visual Science, January 1996, Vol. 37, No. 1
Copyright © Association for Research in Vision and Ophthalmology
Immersion Versus Perfusion Fixation
115
1^* ^ ^ tO i-H CM
tO CO CM f^» O^ "^
© I-H I-H I-H © O
© © © © q ©
©'©©©©'©
immersion fixation (anterior chamber pressure, 0 mm
Hg).8"16 In addition, trabeculectomy specimens, the
most readily available tissue from patients with glaucoma, are fixed at 0 mm Hg.16"20 To understand aqueous flow through this region and to compare the results of studies performed with either method of fixation, it is necessary to know the magnitude of
difference that fixation pressure may have on the architecture of the juxtacanalicular tissue, particularly
the optically empty spaces.
This study was performed to quantitate differences in the juxtacanalicular tissue and in Schlemm's
canal with perfusion and immersion fixation. Six pairs
of eyes were studied by fixing one eye with perfusion
fixation and the fellow eye with immersion fixation.
Biologic variability between fellow eyes was a potential
confounding problem for this comparison; a previous
study21 found a mean difference of 23% between fellow eyes in the relative amount of empty space in
the juxtacanalicular tissue. Thus, we studied a second
group of eyes, in which it was possible to compare the
effect of immersion and perfusion fixation within a
single eye. In this group, trabecular biopsy was performed at the start of the experiment, while the eye
was at 0 mm Hg, and was followed by watertight closure of the scleral incision before fixation by perfusion.
CM t O
© o
q ©
© ©
i n oo ooCT>t o i—i
co m oo oo m m
f J ( N ( N ( N
© q q © © p
©'©'©'©©©
O5 tO
I-H CM
qq
© ©
r^ t^ tom
-^ I-H CM© CM
I-H
m
CM
CO CM
m 00 CM CM tO <*
tO ^f •"'f© CO in
CO
I-H I-H
m
00
I-H
I-H
I-H
00
l-H I-H
m m
t-H
r-H
CM ©
00
to
CM to ©
I-H
r— !"">
m eo
O> 00 »n
I-H
I-H I-H
CO © CO © CO CM CO <* CO 00 CO © m
I-H CM
to <y>
eo
CO
CO CO
CO CO CO CM CM CO
CO CO
00 CO
© I-H
m
I-H
© to CM
to >o t.O
© CO © CM to 00 ^
CO I-H CM CO CM I-H CM
I-H
00
CO to
©
O
f © t i n
©
to I-H CM
CO CM
CM
m
m
CM
tO 00
METHODS
TtoococMTtoo
I-HI-HCMI-H
8|S
I-H
I—( T f I-H t O I-H I-H
Group 1
CO 00
cocMto©CMina»eo©CMCMOini-H
tOOOOOt^tOOOcOi-HOi-HCMCM-^00
|
o o o o n ^
0 © © r t ( a ^
r- CM
mto-<ti'^CMini>toinx>coi-Hi>o6
i>CMr^©oototOT^m©a>CMini>
CMCMCMeOCMCMCMCOCMCMi-HCMCMCM
Cfiv
| u ^
a
Ji
Normal eyes from six donors were obtained at autopsy
and processed within 24 hours of death. Mean age of
the donors was 74 ± 6 years (range, 69 to 83 years
± SD; Table 1). No eyes had undergone intraocular
surgery or had a history of disease. One eye of each
pair was fixed by perfusion offixativeinto the anterior
chamber at either 10 or 15 mm Hg; eyes from three
of the donors in this group have been reported21 (Table 1). Fellow eyes were fixed by opening them at the
equator and immersing them in fixative.
Consent for use of all autopsy eyes was obtained
from next of kin. The protocol was approved by the
Institutional Review Board on the use of human tissue,
and the study met the recommendations of the Declaration of Helsinki.
Group 2
©
as
eo
x^ oo
to
gI c
•^
co
•^*
m
co
^
-*
o>
o
m
^
co
CM
in
co
co
m
co
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
J3
u
^
8g e
Normal eyes from six donors were obtained at autopsy
and processed within 18 hours of death. Mean age of
the donors was 79 ± 12 years (range, 67 to 96 years
± SD; Table 2). Two eyes had undergone cataract
surgery. No other history of ocular surgery or disease
was present.
Characteristics of Eyes From Experiment 2 (Trabecular Biopsy)
rs)
SC
Length
Fixation
Sex Eye
F
L
P
Trab B X2
M
R
R
IM
P
Trab B X2
F
L
R
IM
P
Trab B X2
F
L
R
IM
P
Trab B X2
F
L
R
Im
P
Trab B Xl
L
P
Trab B Xl
R&L
P
Trab
F
L
P
R
Trab B Xl
Not useable
P
Trab
IM
(fim)
SD
205
209
185
258
242
281
245
158
253
114
159
268
170
181
137
219
154
200
277
227
100
9
48
46
6
47
71
14
36
34
18
34
44
—
17
—
19
27
92
—
208.8
199.2
246.8
63.9
34.7
42.7
SC
Area
(fim2)
2050
2452
1956
2310
1909
7496
1230
848
2736
1798
TA
4435
725
1955
925
2643
8925
2299
2757
2492
jcr
SD
Empty
Space (%)
8.6
11.6
8.9
6.7
9.3
7.6
8.3
10.5
9.0
9.9
8.0
11.3
9.9
5.3
10.0
4.5
9.9
4.9
14.2
9.0
3.6
4.4
2.5
2.5
0.7
2.1
1.1
0.9
1.9
4.3
2.6
3.5
2.5
—
3.7
—
2.8
0.6
3.1
—
41.1
43.5
33.8
29.2
31.4
31.2
35.9
35.8
30.7
41.5
29.2
34.1
34.3
37.6
42.9
14.6
38.6
26.1
37.4
19.9
9.6
87.9
9.2
2.5
2.3
1.5
37.3
31.0
32.5
Thickness
SD
1289
1213
1032
758
TA
4888
626
TA
678
774
TA
821
272
—
4899
—
153
486
1407
—
1828.3 708.4
2000.0 684.0
4155.8 2455.6
(fim)
ESSC
SD
4.5
TA
4.0
3.9
2.2
8.3
0.8
11.6
9.1
13.8
10.3
6.5
12.9
—
11.7
—
0.8
16.3
2.0
—
4.5
8.1
1.7
(%)
8.0
7.5
8.3
8.3
1.7
3.1
11.1
2.8
1.0
9.3
2.5
1.2
21.3
10.9
11.7
SD
0.0120
—
3.3
3.1
3.4
0.0363
0.0609
0.0444
0.0153
0.0303
0.0101
TA
6.40
5.10
0.90
0.80
6.70
2.30
0.30
0.40
0.10
1.00
16.40
11.30
—
16.5
6.80
2.40
4.00
10.2
5.0
3.4
SD
0.0184
0.0145
0.0353
0.0420
0.0499
0.0391
0.0243
0.0436
0.0584
0.0303
0.0844
0.0449
0.1033
0.0125
0.0184
0.1181
0.0609
0.0653
0.0421
0.1074
6.70
7.5
9.2
8.1
6.5
R (calculated)
(mm Hg/ftl per minute)
R (measured)
(mm Hg/ftl per
5.26
TA
0.0123
0.0195
0.0061
0.0100
0.0070
0.0190
0.0437
0.0200
0.0786
0.0404
0.1320
5.00
3.85
1.82
6.67
—
0.0224
7.69
—
0.0600
0.0747
0.0155
4.55
—
4.98 ± 1.
fixation at 10 mm Hg; Trab B = trabecular biopsy; IM = immersion fixation; TA = technical artifact prevented analysis of this component in both biopsies; SC =
nal; SD = standard deviation; JCT = juxtacanalicular tissue; E = optically empty space; ES-SC = optically empty space adjacent to Schlemm's canal; R = right; L = le
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
Immersion Versus Perfusion Fixation
Trabecular Biopsy
Trabecular biopsies were performed while the eyes
were at 0 mm Hg intraocular pressure, before perfusion, and were considered to represent immersion
fixation. The trabecular biopsy technique was similar
to standard surgical trabeculectomy except for watertight closure of the scleral flap. A one-half thickness
scleral flap was raised (4X5 mm), and an inner block
of meshwork was excised. Each meshwork sample was
complete and included a portion of the peripheral
cornea anteriorly and a portion of the ciliary muscle
tips posteriorly. The scleral flap was closed tightly with
multiple 10-0 nylon sutures, and the edges of the flap
were sealed with cyanoacrylate adhesive. After the trabecular biopsy, eyes were perfused with Dulbecco's
saline at 10 mm Hg for approximately 90 minutes,
and the facility of outflow was measured. Facility was
determined by measuring the volume of fluid required to maintain intraocular pressure at 10 mm Hg
using a computerized system with a servo-controlled
stepper motor. Facility was calculated with the equation C = F/P.22 Eyes were perfused with fixative at 10
mm Hg.
In 4 of 6 pairs of eyes, trabecular biopsy was performed in two quadrants, 180° apart, on one eye of
each pair, and the fellow eye was fixed by immersion
(group 2a). In 2 of 6 pairs of eyes, trabecular biopsy
was performed on both eyes of each pair. In these
eyes, only one biopsy was obtained from each eye to
allow examination of meshwork 180° away from the
biopsy site (group 2b). In one of these four eyes, the
biopsy was technically unsatisfactory, and the eye was
omitted from the study.
Effect of Trabecular Biopsy on Remaining
Meshwork
Removal of a piece of meshwork may decrease the
outflow resistance of the eye by creating openings into
Schlemm's canal at each edge of the biopsy site,
allowing fluid to bypass the meshwork. The effect of
trabecular biopsy on the remaining meshwork was
studied three ways:
1. Outflow resistance was determined in a preliminary experiment measuring the facility of outflow before and after trabecular biopsy in 10
eyes. At a perfusion pressure of 10 mm Hg, outflow facility increased by 4% after the biopsy
(mean facility before biopsy, C = 0.23 ± 0.06 //I/
minute per mm Hg; mean facility after biopsy,
scleral flap sealed, 0.24 ± 0.04).
2. Facility of outflow was measured after trabecular
biopsy in seven eyes reported in Table 2 (0.24
± 0.05 /xl/minute per mm Hg).
3. Comparison of the biopsy specimen with tissue
from the same quadrant as the biopsy site and
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
117
with meshwork from 90° or 180° distant. If a
significant degree of bypass of the meshwork
through the cut ends of the canal occurred at
the biopsy site, tissue from that quadrant would
resemble immersion-fixed tissue—with a larger
canal area and less empty space in the juxtacanalicular tissue—more than it would resemble perfusion-fixed tissue in quadrants distant from the
biopsy site.
Tissue Processing
Fixative was 1% paraformaldehyde-2% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.4. Wedges of
the limbal region, including the trabecular meshwork,
were taken from at least three quadrants, dehydrated
in ascending alcohols, and embedded in Epon-Araldite. Semithin sections (1 jum) were stained with toluidine blue and examined by light microscopy. Grids
were stained with uranyl acetate and lead citrate and
examined with aJEOL (Tokyo, Japan) 1200 transmission electron microscope.
Morphologic Analysis
Schlemm's Canal. Light microscopy was used to
measure the anterior-to-posterior length of Schlemm's
canal, perimeter of the canal, and area of the canal.
Measurements were made at X400 magnification, using the Zeiss (Thornwood, NY) IBAS 2000 image analysis system.21 Reproducibility was determined by reanalysis of at least one quadrant from each eye and
gave values within 8%. IBAS analysis of the canal
length in eyes previously measured with a graticule
gave results within 3% of the previously reported values.21
Juxtacanalicular Region. Juxtacanalicular tissue was
defined as the tissue underlying Schlemm's canal, extending from Schlemm's canal endothelial cells to die
empty space adjacent to the first trabecular lamella18'21
(Fig. la). In regions that were less well defined, the
first trabecular lamella was defined as having either a
central "core" area of elastic-like tissue with some
surrounding cortex, or as having a horizontally elongated trabecular cell overlying such tissue (Fig. lb).
Schlemm's canal endothelial cells and giant vacuoles
were not included in the juxtacanalicular tissue area.
Initial examination involved low magnification
(X600) overview photographs on the electron microscope for orientation and aid in the determination
of the boundaries of the juxtacanalicular tissue. Four
micrographs of the juxtacanalicular tissue were then
made at X2500 using a sampling protocol. These areas
included the region underlying die anterior end of
the canal, two areas from mid-canal, and the region
underlying the posterior end of the canal. From previous work, this sampling regimen was found to provide
a reasonable representation of the juxtacanalicular tis-
Investigative Ophthalmology & Visual Science, January 1996, Vol. 37, No. 1
118
sc
sessed from micrographs of each sample region, measured perpendicularly to the long axis of the canal
from four predetermined spots on each photograph.
Empty space (ES) was measured by densitometry
and was considered to be "optically empty space"
when viewed with electron microscopy; no tissue components of any kind were included in this measurement.21
Empty space touching Schtemm's canal (ES-SC) was
traced by hand. This parameter was found by LutjenDrecoll23 to correlate with outflow facility in monkey
eyes.
Solid tissue was determined by subtracting the
amount of empty space from the total juxtacanalicular
area in the eyes in group 1. In the eyes in group 2,
each of the various solid tissue components of the
juxtacanalicular tissue was determined by densitometry, including amorphous basement membrane (type
I plaque), tendon and sheath material (type II and III
plaque), and cytoplasm.21'24
Analysis of the electron micrographs was repeated
on two samples on each of three separate days and
gave results differing by an average of 9%. Calibration
of the magnification on the electron microscope, micrographs, and of the IBAS system was determined
using a 15,240 lines/inch calibration grating (Ladd,
Burlington, VT).
Calculations
Measurements from each of the four micrographs per
FIGURE 1. Overview of juxtacanalicular tissue (JCT) and cortissue section were combined to determine the value
neoscleral meshwork. (A) The JCT has a characteristic loose
appearance of cells and extracellular matrix. The JCT
for each section. Data from the different quadrants
boundary is indicated by the line (arrows). Asterisk denotes were combined to give a mean value per eye. For the
tissue that resembles a trabecular lamella but does not have
two eyes fixed in the same fashion from one donor,
a central core of elastic-like material and, dius, is included
results from the eyes were combined to yield the
in the JCT. (B) JCT boundary is less distinct: Outermost
mean. Similarly, when there were two trabecular bioplamella-like tissue has central core of elastic material (el)
sies from one eye, the results were combined to yield
with some surrounding basement membrane and is not ina mean value for that eye. Values of all data are mean
cluded in the JCT (both micrographs are from same histo± standard deviation.
logic section of eye 435, immersion fixation. Original magniTwo-tailed paired Rests were used for comparison
fication, X2500.
between trabeculectomy and perfused samples from
the same eye and for comparison between fellow eyes.
Pearson's correlation coefficient was used to test for
sue.21 The photographic technique was standardized,
correlations among measured facility and histologic
using the same accelerating voltage (60 kV) and phomeasurements.
tometer readings for all pictures.
Calculations of permeability and resistance of the
The IBAS image analysis system was used to determine the relative areas of the various juxtacanalicular juxtacanalicular tissue were performed using the
method of Ethier and colleagues,25 which considers
components.21 This system allowed determination of
the juxtacanalicular tissue to be a porous nitration bed
area either by densitometry or by hand tracing. Meaand uses the Carmen-Kozeny equation to calculate its
surements were expressed relative to the total area of
specific hydraulic conductivity. Flow resistance is then
the juxtacanalicular tissue, as measured in the four
calculated using Darcy's law, which describes flow
micrographs from each quadrant. Images from microthrough porous materials.26 The measured thickness
graph negatives were fed into the IBAS with a video
of the juxtacanalicular tissue was used, as was the mean
camera, and the image was enlarged to a final magnilength of Schlemm's canal. The circumference of the
fication of X 6,250.
Thickness of the juxtacanalicular tissue was as- meshwork was taken as 36 mm, and the viscosity of
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
119
Immersion Versus Perfusion Fixation
1O(Jrt>
TBx
FIGURE 2. Light microscopy of specimens from same donor (donor 23,1-fjm sections). Toluidine blue. Original magnification, X670. (A) Perfused eye. Schlemm's canal narrow; giant
vacuoles present (small arrows). Uveal meshwork cells appear intact {arrowheads). (B) Trabecular biopsy from same eye as in A. Uveal meshwork cells appear edematous, with rounded
cytoplasm and nuclei (arrowheads). (C) Immersion-fixed fellow eye. Schlemm's canal appears
large. Uveal meshwork cells appear intact (arrowheads).
aqueous humor at 37°C was assumed to be 0.72 centipoise. 25
RESULTS
Overall appearance of immersion-fixed and perfusionfixed eyes fit the pattern reported by Johnstone and
Grant. 1 Immersion-fixed eyes tended to have a dilated
Schlemm's canal, with lamellae closely spaced. Perfusion-fixed eyes tended to have a smaller canal area,
with lamellae more widely spaced apart (Fig. 2).
Schlemm's Canal
Structure. Schlemm's canal was variable in appearance and size. The canal was a single, large channel
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
in 56% (45/80) of quadrants; in 44% of quadrants,
the canal contained at least two or more channels (two
channels, 23%; three channels, 14%; four or more
channels, 7%).
Morphometric measurements. The mean length of
Schlemm's canal was 248 ± 53 /xm for all eyes (mean
± SD; Tables 1, 2). Variability in the length of the
canal among quadrants of an individual eye was approximately 21% for perfusion and immersion groups
combined (Table 3). The difference between fellow
eyes for canal length ranged from 6 to 154 jj,m, with
a mean difference of 37 //m (14%), and the mean of
the coefficient of variation between fellow eyes was
16%.
The area of Schlemm's canal was 47% smaller in
120
Investigative Ophthalmology & Visual Science, January 1996, Vol. 37, No. 1
3. Circumferential Variability Within
the Same Eye
TABLE
Perfusion Fixed
Schlemm's canal
Length
Area
Perimeter
Immersion Fixed
CV(%)
Range
(%)
CV(%)
Range
(%)
26
46
28
37
56
40
16
33
16
27
45
27
29
23
59
53
41
14
37
47
59
54
23
JCT
Empty space
ES-SC
Amorphous BM
Tendon-sheath
Cells
19
62
20
39
21
78
30
52
36
C V = mean of coefficients of variation of all individual eyes
(SD/mean); range = mean of ranges of all individual eyes
[(highest quadrant - lowest quadrant)/highest quadrant]; JCT =
juxtacanalicular tissue; ES-SC = empty space adjacent to
Schlemm's canal; BM = basement membrane.
ration because the region of origin of the septae in
the juxtacanalicular tissue was composed of densely
arranged collagen (Fig. 3a). In another variation, the
juxtacanalicular tissue appeared to resemble the wall
of a vascular channel rather than a loose, permeable
filtration tissue. Compact layers of basement membrane and collagen appeared underlying the canal,
with no evidence of giant vacuole formation (Fig. 3b).
Several of these variations could be present within the
same eye, potentially misleading the interpretation of
the configuration of the juxtacanalicular tissue if only
one region was sampled.
Washout of extracellular matrix was not seen histologically in perfusion-fixed eyes. This was also evident on morphometric measurement because the relative proportions of the various solid tissue components (amorphous basement membrane, sheath and
perfusion-fixed eyes than in immersion-fixed fellow
eyes (P — 0.04, groups 1 and 2 combined; Tables 1,
2). The canal was 34% smaller in perfusion-fixed eyes
in experiment 1 and 56% smaller in the perfusionfixed eyes in experiment 2. Variability in the canal
area around the circumference of an individual eye
was approximately 40% (Table 3).
The area of the canal in trabecular biopsy specimens was closer to the eye from which the biopsy was
taken than to the fellow eye. Despite the difference
in fixation pressures between the biopsy (immersion)
and the remaining perfused tissue, the area of the
canal in the biopsy specimen was only 9% larger than
in the perfused tissue from the same eye (Table 2).
Similarly, comparison of trabecular biopsy specimens
with individual quadrants from the same eye, whether
the same quadrant as the biopsy site or with quadrants
90° or 180° away, revealed no significant differences
in canal area.
The perimeter of the canal tended to be smaller
in perfusion-fixed eyes than in immersion-fixed eyes
(593 /xm ± 98 /xm versus 668 ± 140 /im), although
this did not reach statistical significance (groups 1 and
2 combined; data not shown).
Juxtacanalicular Tissue
Structure. Juxtacanalicular tissue appearance was
variable. Juxtacanalicular tissue, composed of electron-lucent spaces, extracellular materials, and trabecular cells, usually appeared loose and open (Fig. 1).
When underlying a collector channel, juxtacanalicular
tissue appeared to have more open space and a less
distinct transition to the corneoscleral meshwork. The
presence of septae often changed this open configu-
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
FIGURE 3. Configurations of juxtacanalicular tissue (JCT).
(A) Collector channel entering with Schlemm's canal. Juxtacanalicular tissue region is a combination of loosely arranged material (asterisk) and more densely layered collagen
that appears to be the origin of a septa (S) (eye 23R, immersion fixed). Magnification, X600. (B) Posterior region of
Schlemm's canal and meshwork. JCT (arrows) appears layered and moderately dense, reminiscent of vascular channel. No giant vacuoles are seen in this region, despite perfusion fixation (eye 23L). Original magnification, X2500; (inset), X600.
Immersion Versus Perfusion Fixation
40-
D Perfused (n = 6)
13 Trabec(n = 6)
13 Immersed (n = 4)
X±SE
30-
O
121
X
20-
x
10-
Empty space
ES-SC
Basement
membrane
Tendon &
sheath
Cells
FIGURE 4. Morphometric analysis of juxtacanalicular tissue.
Eyes are from group 2.
tendon material, and cytoplasm) did not differ significantly between the different methods of fixation
(Fig. 4). Giant vacuoles, more common in perfusionfixed tissue, rarely contained debris or other material.
Of interest, the occasional random cell seen entering
Schlemm's canal appeared to pass between the inner
wall cells, not through a giant vacuole.
Morphometric Measurements. Thickness. The juxtacanalicular tissue of perfusion-fixed eyes was thinner
than that of immersion-fixed eyes (7.1 ± 1.6 jum versus
8.7 ± 1 . 7 fxm; P = 0.006; paired eyes; Tables 1, 2).
This was found whether the first intertrabecular space
was excluded (Tables 1, 2) or included (data not
shown). Juxtacanalicular tissue appeared wider in die
posterior region than in the anterior region in perfusion-fixed eyes (8.9 ±1.7 fim versus 6.5 ± 2.2 /J,m; P
= 0.05) but not in immersion-fixed eyes (8.7 ± 2.3
//m versus 8.3 ± 2.9 /mi).
Comparison of juxtacanalicular thickness in trabecular biopsy specimens with the remaining tissue
from the same quadrant, or quadrants 90° and 180°
away, revealed no significant differences.
Empty space. The amount of empty space relative
to the juxtacanalicular area was 13.4% greater in perfusion-fixed eyes than in fellow eyes fixed by immersion (P = 0.04; groups 1 and 2 combined). Values
were similar in both experimental groups: 13% more
empty space in perfusion-fixed eyes in group 1 and
13.5% more empty space in perfusion-fixed eyes in
group 2 (Tables 1, 2). Variability among quadrants
within a single eye was similar between perfusion- and
immersion-fixed eyes (Table 3).
Comparison of the trabecular biopsy specimens
(considered immersion fixation) with the remainder
of the perfusion-fixed tissue from within the same eye
revealed the perfusion-fixed tissue to have 20% more
empty space than the biopsy specimens (P = 0.14).
No significant differences were noted among quadrants near the biopsy site or regions 90° or 180° away.
The amount of empty space touching Schlemm's
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
canal was larger by 60% in perfusion-fixed eyes than
in fellow eyes fixed by immersion (8.4% ± 2.5% versus
3.4% ± 2.7%; P = 0.01; Tables 1, 2). Note the large
variation of this measurement, with a coefficient of
variation of approximately 60% (Table 3). A comparison of biopsy specimens with perfusion-fixed tissue
from the same eye (group 2) also revealed more empty
space touching the canal in the perfusion-fixed tissue
(10.2% ± 3.3% versus 5.0% ± 3.1%; P = 0.02; Table 2).
Solid tissue. The areas of the solid tissue components relative to the total juxtacanalicular area were
approximately equally divided among amorphous
basement membrane, tendon and sheath material,
and cytoplasm (Fig. 4). Fixation method did not appear to make a significant difference in the relative
proportions of these components. Circumferential
variability among quadrants within the same eye was
similar between perfusion- and immersion-fixed eyes
(Table 3).
Effect of obtaining biopsy on meshwork sample. All tra-
becular biopsy specimens were satisfactory in that they
contained the entire trabecular meshwork and
Schlemm's canal. No obvious qualitative differences
were noted between the histologic appearance of the
juxtacanalicular region from biopsy specimens and tissue from the other quadrants from the same eye; differences in morphometric measurements have been
described previously. In die meshwork itself, cells in
the inner meshwork region, especially uveal meshwork
cells, were edematous and rounded in 8 of 11 biopsy
specimens from the seven eyes with trabecular biopsy
(Fig. 2). Uveal meshwork cells in the other quadrants
examined from these eyes were rounded or swollen
in only 5 of 23 perfused quadrants from these seven
eyes. Damage to uveal meshwork cells from excision
of the specimen seemed the likely cause.
Permeability and Resistance Calculations
Perfusion-fixed eyes had a 37% lower calculated flow
resistance than immersion-fixed eyes: 0.0247 ± 0.0100
mm Hg//il per minute versus 0.0393 ± 0.0134 mm
Hg///1 per minute (P = 0.04; Tables 1, 2). Perfusionfixed meshwork also had a lower calculated flow resistance than trabecular biopsy samples from the same
eye (Table 2). Note the calculated values are two orders of magnitude smaller than die measured outflow
resistance in these same eyes of 4.98 ± 1.90 mm Hg/
fil per minute (Table 2). This difference of two orders
of magnitude is similar to that reported by Ethier et
al.25
DISCUSSION
In perfusion-fixed eyes, the trabecular meshwork had
a larger area of empty space in the juxtacanalicular
122
Investigative Ophthalmology & Visual Science, January 1996, Vol. 37, No. 1
tissue and a smaller area of Schlemm's canal than in
immersion-fixed fellow eyes. The most striking difference was in the area of Schlemm's canal, which was
smaller in perfusion-fixed eyes by 47% (P = 0.04).
Although narrowed by the pressure within the anterior chamber, the canal area does not contribute to
the outflow resistance unless the canal collapses,
which did not occur at the physiologic pressure used
in this study. At higher pressures (30+ mm Hg), the
canal can collapse and may then contribute to outflow
resistance.1-472728
Composition and configuration of juxtacanalicular tissue are probably more important in affecting
outflow resistance than the area of Schlemm's canal.22-24'29'30 The amount of empty space in the juxtacanalicular region was increased in perfusion-fixed
eyes by 13.4% (P = 0.04). This increase was relatively
small because of the physiologic perfusion pressures
used in the study. At higher intraocular pressures, the
juxtacanalicular tissue would be more expanded, and
the relative amount of empty space would be expected
to be increased.'"4l27 The pressure differential found
across the meshwork in living eyes is the difference
between intraocular pressure and episcleral venous
pressure: 16 - 9 = 7 mm Hg.31"33 This is close to
the pressures used in the current study, in which the
episcleral venous pressure was 0 mm Hg.
The difference in the amount of empty space between the two fixation pressures cannot be attributed
to mere biologic variation between fellow eyes. Because our previous study21 found a mean difference
of 23% in the amount of empty space between fellow
eyes, a comparison of values from within the same eye
seemed to be the best possible method to eliminate
this question. Eyes from experiment 2, in which a
trabecular biopsy was performed before the remaining
meshwork was fixed by perfusion, revealed values similar to those of experiment 1, in which no biopsy was
performed.
Washout of extracellular material was not observed in the perfusion-fixed eyes. Giant vacuoles usually appeared empty and rarely contained debris or
underlying basement membrane. Amorphous basement membrane and tendon and sheath material appeared qualitatively similar in perfusion- and immersion-fixed tissue. Quantitation of the amount of solid
tissue revealed no significant difference in the relative
amounts of amorphous basement membrane or tendon and sheath material between perfused and immersed tissue—further evidence against perfusion
fixation causing artifactual washout of extracellular
components (Fig. 4). The amorphous basement membrane appears to be interconnected, material that
serves as an anchor for the endothelial lining cells of
the canal rather than a collection of material washed
downstream and held in place by the endothelial cells.
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
Washout of glycosaminoglycans or gel filling the
empty spaces is possible, although impossible to detect
using conventional electron microscopy processing.
Indirect evidence of washout, however, could be
the apparently contradictory finding that the juxtacanalicular tissue was thinner in perfusion-fixed tissue
and yet was more porous. This difference in thickness
could be the result of variability between eyes: The
juxtacanalicular tissue was thinner in perfused meshwork in group 1 but not in group 2 eyes. Previous
reports of washout of extracellular material have concerned monkeys perfused at higher pressures.3"527 No
other reports have analyzed the juxtacanalicular tissue
of human eyes at higher pressures.
The increase in empty space in the perfusion-fixed
meshwork resulted in a lower calculated outflow resistance in perfusion-fixed eyes by 37% (P = 0.04). Calculated resistance values, however, were two orders of
magnitude smaller than the measured outflow resistance, a difference first reported by Ethier.25 Explanation of this difference between measured and calculated flow resistance values seems key to understanding outflow physiology. If the site of resistance does
lie within the juxtacanalicular tissue, calculations using the Carmen-Kozeny-Darcy's law model predict
that a change in the amount of empty space (porosity)
or a change in the length of Schlemm's canal must
occur. If the amount of empty space is key, it would
have to decrease to approximately 7% to match the
measured outflow resistance. It is of interest that the
amount of empty space touching Schlemm's canal was
8.4% in perfusion-fixed meshwork; however, changes
in the pathlength to fluid flow and other parameters
prevent a simple application of this data to the model.
If a shortening of Schlemm's canal were to occur,
calculations indicate that the canal would have to decrease to 3 /im in length to account for the measured
outflow resistance.
Other theories for aqueous outflow resistance
within the juxtacanalicular tissue have been suggested,
including resistance caused by extracellular glycosaminoglycans lining or even filling the outflow pathways,2534 entrapment of proteins in the extracellular
matrix in animal eyes,35 and interaction between the
juxtacanalicular tissue and the inner wall of
Schelmm's canal.36
The cellular lining of the inner wall of Schlemm's
canal, which was not included in the flow resistance
calculations, may contribute significant resistance to
flow. Indeed, the exact mechanism of fluid flow across
this lining is unknown; flow through giant vacuole
formation and flow through intercellular junctions
have been proposed.37'38
Some researchers have looked outside the juxtacanalicular region for the source of outflow resistance,
including resistance caused by the collector channels
123
Immersion Versus Perfusion Fixation
and aqueous veins39 42 and resistance caused by the
cellular lining of channels in the outer corneoscleral
meshwork adjacent to the juxtacanalicular region (cul
de sacs).20
No correlation was found among measured outflow resistance and ultrastructural components of the
juxtacanalicular tissue. Unlike Lutjen-Drecoll, 23 no
relationship was found between the measured outflow
resistance and the amount of empty space touching
Schlemm's canal. She analyzed young monkey eyes
given pilocarpine intracamerally before enucleation
and subsequent fixation by immersion. Our eyes, in
which outflow facility was measured, were from older
humans and were fixed by perfusion. Measurements
of Schlemm's canal and the juxtacanalicular region
were similar to those reported by others. 7 ' 43 Canal
length decreases with age, from approximately 350
fim at 20 years of age to approximately 208 fim by 80
years of age. Our value of 248 /im is in keeping with
these numbers. Other reports 18 ' 212544 on the amount
of empty space in the juxtacanalicular region in human eyes range from 25% to 70%, measured on immersion-fixed tissue.
The biopsy itself did not appear to affect the remaining meshwork. Facility of outflow increased by
only 4% after trabecular biopsy in the preliminary
study of 10 eyes. This finding is similar to that reported
by Grant, who commented that removal of a tiny block
of trabecular meshwork and Schlemm's canal (termed
trabeculo-canalectomy by Grant) "would produce a
relatively slight increase in facility of outflow."39 This
small effect was thought to be caused by a limited
circumferential flow in Schlemm's canal, as determined from trabeculotomy experiments at pressures
of 25 mm Hg.39'40 Using a lower perfusion pressure,
Rosenquist and colleagues41 found that a trabeculotomy (incising meshwork without removal of meshwork) had a larger effect on outflow. Opening the
meshwork only 1 clock hour (30°) decreased outflow
resistance by 32% when performed at a perfusion pressure of 7 mm Hg. Such a change in resistance implies
circumferential flow within the canal, which could
lessen the pressure differential across the meshwork
in the quadrants near the biopsy. We did not find
evidence of this; morphometric measurements of canal area, juxtacanalicular thickness, and relative
amount of empty space within the juxtacanalicular
tissue did not show major differences among different
quadrants, whether from the same quadrant as the
biopsy site or regions 90° or 180° away.
Is perfusion fixation preferable to immersion fixation? Although large differences in the appearance
of the juxtacanalicular tissue and the amount of empty
space were not present, neither was obvious washout
of extracellular material in perfusion-fixed eyes. The
consistent finding of increased empty space in the
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
juxtacanalicular tissue with this physiologic perfusion
pressure, as well as the increased amount of empty
space touching Schlemm's canal in perfusion-fixed tissue, indicate that perfusion may open collapsed flow
pathways not evident at 0 mm Hg. 2 This concept is
supported by the findings of Ainsworth and Lee, 7 who
reported perfusion fixation to cause a more uniform
distribution of giant vacuoles around the circumference of the eye. Additional evidence for this is the
increase in the number of giant vacuoles with increasing intraocular pressure. 1 " 7 Thus, although relatively
small differences in the juxtacanalicular tissue occur
between the two methods of fixation at physiologic
pressures, the potential of finding aqueous flow pathways that would be closed at 0 mm Hg weighs in favor
of perfusion fixation.
Key Wards
electron microscopy, image analysis, morphometry, porous
media, Schlemm's canal, trabecular meshwork
References
1. Johnstone MA, Grant WM. Pressure-dependent
changes in structures of the aqueous outflow system
of human and monkey eyes. Am J Ophthalmol.
1973; 75:365-383.
2. Grierson I, Lee WR. Changes in the monkey outflow
apparatus at graded levels of intraocular pressure: A
qualitative analysis by light microscopy and scanning
electron microscopy. Exp Eye Res. 1974; 19:21-33.
3. Grierson I, Lee WR. The fine structure of the trabecular meshwork at graded levels of intraocular pressure:
1: Pressure effects within the near-physiological range
(8-30 mmHg). Exp Eye Res. 1975; 20:505-521.
4. Grierson I, Lee WR. The fine structure of the trabecular meshwork at graded levels of intraocular pressure:
2: Pressures outside the physiological range (0 and 50
mmHg). Exp Eye Res. 1975;20:523-530.
5. Grierson I, Lee WR. Pressure-induced changes in the
ultrastructure of the endothelium lining Schlemm's
canal. Am J Ophthalmol. 1975; 80:863-884.
6. Grierson I, Lee WR. Light microscopic quantitation
of the endothelial vacuoles in Schlemm's canal. Am]
Ophthalmol. 1977; 84:234-246.
7. Ainsworth JR, Lee WR. Effects of age and rapid highpressure fixation on the morphology of Schlemm's
canal. Invest Ophthalmol Vis Sci. 1990;31:745-750.
8. Grierson I, Lee WR, Moseley H, Abraham S. The trabecular wall of Schlemm's canal: A study of the effects
of pilocarpine by scanning electron microscopy. BrJ
Ophthalmol. 1979; 63:9-16.
9. McMenamin PG, Lee WR. Age related changes in extracellular materials in the inner wall of Schlemm's
canal. Graefe'sArch ClinExp Ophthalmol. 1980;212:159172.
10. McMenamin PG, Lee WR, Grierson I, Grindle FCJ.
Giant vacuoles in the lining endothelium of the human Schlemm's canal after topical timolol maleate.
Invest Ophthalmol Vis Sci. 1983; 24:339-342.
124
Investigative Ophthalmology & Visual Science, January 1996, Vol. 37, No. 1
11. Grierson I, Howes RC, Wang Q. Age-related changes
in the canal of Schlemm. Exp Eye Res. 1984; 39:505512.
12. McMenamin PG, Lee WR, Aitken DAN. Age-related
changes in the human outflow apparatus. Ophthalmology. 1986;93:194-209.
13. Liitjen-Drecoll E, Kaufman PL, Eichhorn M. Longterm timolol and epinephrine in monkeys: I: Functional morphology of the ciliary processes. Trans Ophthalmol Soc UK 1986; 105:180-195.
14. Liitjen-Drecoll E, Kaufman PL. Biomechanics of
echothiophate-induced anatomic changes in monkey
aqueous outflow system. Graefe's Arch Clin Exp Ophthalmol. 1986; 224:564-575.
15. AlvaradoJ, Murphy C, PolanskyJ.Juster R. Age-related
changes in trabecular meshwork cellularity. Invest Ophthalmol Vis Sci. 1981;21:714-727.
16. AlvaradoJ, Murphy C, Juster R. Trabecular meshwork
cellularity in primary open-angle glaucoma and nonglaucomatous normals. Ophthalmology. 1984; 91:564579.
17. Rodrigues MM, Spaeth GL, Sivalingam E, Weinreb S.
Histopathology of 150 trabeculectomy specimens in
glaucoma. Trans Ophthalmol Soc UK 1976;96:245-255.
18. Alvarado JA, Yun AJ, Murphy CG. Juxtacanalicular tissue in primary open angle glaucoma and in nonglaucomatous normals. Arch Ophthalmol. 1986; 104:15171528.
19. Murphy CG, Johnson M, Alvarado JA. Juxtacanalicular
tissue in pigmentary and primary open angle glaucoma: The hydrodynamic role of pigment and other
constituents. Arch Ophthalmol. 1992; 110:1779-1785.
20. Alvarado JA, Murphy CG. Outflow obstruction in pigmentary and primary open angle glaucoma. Arch Ophthalmol. 1992; 110:1769-1778.
21. Buller CR, Johnson DH. Segmental variability of the
trabecular meshwork in normal and glaucomatous
eyes. Invest Ophthalmol Vis Sci. 1994; 35:3841-3851.
22. Grant WM. Facility of flow through the trabecular
meshwork. Arch Ophthalmol. 1955;54:245-248.
23. Liitjen-Drecoll E. Structural factors influencing outflow facility and its changeability under drugs: A study
in Macaca arctoides. Invest Ophthalmol. 1973; 12:280294.
24. Rohen JW. Why is intraocular pressure elevated in
chronic simple glaucoma? Anatomical considerations.
Ophthalmology. 1983;90:758-765.
25. Ethier CR, Kamm RD, Palaszewski BA, Johnson MC,
Richardson TM. Calculations of flow resistance in the
juxtacanalicular meshwork. Invest Ophthalmol Vis Sci.
1986;27:1741-175O.
26. LevickJR. Flow through interstitium and other fibrous
matrices. Quart JExp Physiol. 1987;72:403-438.
27. Svedbergh B. Effects of artificial intraocular pressure
elevation on the outflow facility and the ultrastructure
of the chamber angle in the vervet monkey (Cercopithecus ethiops). Ada Ophthalmol. 1974; 52:829-845.
28. Johnson MC, Kamm RD. The role of Schlemm's canal
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
in aqueous outflow from the human eye. Invest Ophthalmol Vis Sci. 1983;24:320-325.
Liitjen-Drecoll E, Shimizu T, Rohrbach M, Rohen
JW. Quantitative analysis of 'plaque material' in the
inner and outer wall of Schlemm's canal in normal
and glaucomatous eyes. Exp Eye Res. 1986; 42:443-455.
Rohen JW, Lutjen-Drecoll E, Flugel C, Meyer M,
Grierson I. Ultrastructure of the trabecular meshwork
in untreated cases of primary open-angle glaucoma
(POAG). Exp Eye Res. 1993;56:683-692.
Leydhecker W, Akiyama K, Neumann HG. Der intraokulare Druck gesunder menschlicher Augen. Klin
Monatsbl Augenheilkd. 1958; 133:662-670. In German.
Phelps CD, Armaly MF. Measurement of episcleral
venous pressure. Am J Ophthalmol. 1978; 85:35-42.
Maepea O, Bill A. The pressures in the episcleral veins,
Schlemm's canal and the trabecular meshwork in
monkeys: Effects of changes in intraocular pressure.
Exp Eye Res. 1989;49:645-663.
Barany EH. In vitro studies on the resistance to flow
through the chamber angle. Ada Soc Med Upsal.
1953; 59:260-276.
Johnson M, Gong H, Freddo TF, Ritter N, Kamm R.
Serum proteins and aqueous outflow resistance in bovine eyes. Invest Ophthalmol Vis Sci. 1993; 34:35493557.
Johnson M, Shapiro A, Ethier CR, Kamm RD. Modulation of outflow resistance by the pores of the inner was
endothelium. Invest Ophthalmol Vis Sci. 1992; 33:16701675.
Tripathi RC. Mechanism of the aqueous outflow
across the trabecular wall of Schlemm's canal. Exp Eye
Res. 1971;11:116-121.
Epstein DL, Rohen JW. Morphology of the trabecular
meshwork and inner-wall endothelium after cationized ferritin perfusion in the monkey eye. Invest
Ophthalmol Vis Sci. 1991; 32:160-171.
Grant WM. Symposium: Microsurgery of the outflow
channels. Lab Res Trans. Am Acad Ophthalmol Otoloaryngol. 1972; 76:398-404.
Ellingson BA, Grant WM. Trabeculotomy and sinusotomy in enucleated human eyes. Invest Ophthalmol.
1972;ll:21-28.
Rosenquist R, Epstein D, Melamed S, Johnson M,
Grant WM. Outflow resistance of enucleated human
eyes at two different perfusion pressures and different
extents of trabeculotomy. CurrEyeRes. 1989; 12:12331240.
Jocson VL, Grant WM. Interconnections of blood vessels and aqueous vessels in human eyes. Arch Ophthalmol. 1965; 73:707-720.
Allingham RR, de Kater AW, Ethier CR. Morphometric analysis of Schlemm's canal in normal and glaucomatous human eyes. ARVO Abstracts. Invest Ophthalmol Vis Sci. 1991; 32:788.
Miyazaki M, Segawa K, Urakawa Y. Age-related
changes in the trabecular meshwork of the normal
human eye. JpnJ Ophthalmol. 1987;31:558-569.