Download The relationship between pore density and outflow facility in

Investigative Ophthalmology & Visual Science, Vol. 33, No. 5, April 1992
Copyright © Association for Research in Vision and Ophthalmology
The Relationship Between Pore Density
and Outflow Facility in Human Eyes
R. Rand Allingham, Annelies W. de Kater, C. Ross Ethier, P. John Anderson, Ellen Hertzmark, David L. Epstein
The inner wall (IW) endothelial lining of Schlemm's canal was examined in six normal human eyes and
four eyes with primary open angle glaucoma (POAG). Outflow facility was measured using constant
pressure perfusion at 15 mmHg, eyes were fixed at 15 mmHg, and the IW endothelial lining was
isolated and examined by scanning electron microscopy. Pore density, pore diameter, and bulge density
were recorded by quadrant, and pore size and density were used to estimate IW endothelial facility,
resistance, and hydraulic conductivity (facility per unit area). In POAG eyes, pores were less common
(489 ± 172 vs 1437 ± 423 pores/mm2; P < .005) and appeared to be more unevenly distributed than in
normal eyes. A regional analysis of pore density (by quadrant) failed to detect a significant difference
between quadrants of normal or POAG eyes. Pore density was correlated with measured outflow
facility in normal eyes alone (P < .02) and when normal eyes were pooled with POAG eyes (P < .001).
The percentage of total resistance attributed to the IW endothelium was 5.8% in normals compared to
9.5% in POAG eyes. This indicates there is a greater pressure drop across the IW endothelium in
POAG eyes, suggesting that an intrinsic difference in IW endothelial function exists between normal
and glaucomatous eyes. However, this difference alone does not account for the decreased outflow
facility in POAG eyes. IW endothelial hydraulic conductivity is markedly higher than that of other
vascular endothelia. We hypothesize that this may protect the IW endothelial lining of Schlemm's
canal from mechanical stress induced by the relatively high rate of transcellular fluid flow. Invest
Ophthalmol Vis Sci 33:1661-1669,1992
The majority of aqueous humor exits the anterior
chamber via the conventional outflow system, ie, the
trabecular meshwork (TM). Perfusion studies have
suggested that pathways exist for aqueous flow
through the TM and into the lumen of Schlemm's
canal.1"4 Garron et al reported the presence of "giant
vacuoles" within the inner wall (IW) endothelial lining of Schlemm's canal.5 Holmberg first described
transcellular pathways that ranged from 0.5-1.5 nm
in diameter. These were later named pores.6 Tripathi
subsequently proposed a mechanism for aqueous
flow through the pores of the giant vacuoles into
Schlemm's canal.7-8
From the Howe Laboratory of Ophthalmology, Massachusetts
Eye and Ear Infirmary and Harvard Medical School, Boston, Massachusetts; and Biomedical Institute and Mechanical Engineering,
University of Toronto, Toronto, Ontario, Canada.
This work is supported in part by NEIEY01894; by a grant from
National Glaucoma Research, a program of the American Health
Assistance Foundation, Rockville, Maryland; and by a grant from
the Massachusetts Lion's Eye Research Fund, Boston, Massachusetts.
Submitted for publication: July 15, 1991; accepted October 31,
1991.
Reprint Requests: R. Rand Allingham, MD, University of Texas
Southwestern Medical Center at Dallas, 5161 Harry Hines Blvd.,
Dallas, TX 75235-8895.
In a series of elegant perfusion experiments, Grant
determined that the site of greatest resistance to
aqueous outflow in normal and glaucomatous human
eyes resided within the outer TM, made up of the
corneoscleral meshwork, juxtacanalicular tissue, and
inner endothelial lining of Schlemm's canal.910 In
normal eyes, the outer TM accounted for approximately 75% of total resistance to aqueous outflow.
Recently, using a microcanulation technique,
Maepea and Bill determined that the outer TM contributes 90% of total resistance in nonhuman primate
eyes.11 To localize the main sites of resistance within
the TM, Bill and Svedbergh examined the IW endothelial lining of Schlemm's canal of normal human
eyes with scanning electron microscopy (SEM).12 By
analyzing pore density and pore size, these investigators estimated that the contribution of the endothelial
lining of Schlemm's canal accounted for no more
than 10% of total outflow resistance in the normal
human eye. In a similar investigation in nonhuman
primate eyes, Moseley et al estimated the IW endothelial resistance contributed 10-24% of total resistance.13
There has not been a similar investigation of the
endothelial lining of Schlemm's canal in eyes with
primary open angle glaucoma (POAG). In addition,
in previous studies, aqueous outflow facility was not
1661
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933167/ on 05/10/2017
1662
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / April 1992
measured. Therefore, we have performed a morphometric analysis of the IW endothelial lining of
Schlemm's canal in normal and glaucomatous human eyes by SEM, and have correlated measured outflow facility with morphologic appearance and calculated resistance of the IW endothelial lining of
Schlemm's canal.
Material and Methods
Six normal human autopsy eyes from three donors
(donor ages 62-84) and four POAG eyes from two
donors (donor ages 73-88) were obtained from the
National Disease Research Interchange, Philadelphia,
PA, or the Foundation for Glaucoma Research, San
Francisco, CA. The glaucoma eyes had a well documented history of POAG, including elevated intraocular pressure, glaucomatous visual field loss, and
optic atrophy. Specific glaucoma treatment in all
four POAG eyes consisted of a combination of pilocarpine, timolol, and dipivefrin. There was no history
of ocular surgery, laser treatment, or ocular disease
other than POAG in the eyes studies. All donor eyes
were enucleated within 3 hr of death, stored in a
moist chamber at 4° C, and perfused within 18 hr
postmortem.
Perfusion Technique
Eyes were placed in a bath of Sorensen's phosphate
buffered solution at 24°C. A 23 G inflow needle was
inserted through the limbus, threaded through the pupil, and inserted into the posterior chamber, avoiding
contact with the lens or ciliary body. The inflow needle was connected to a reservoir containing Dulbecco's phosphate buffered saline with 5.5 mM glucose at a height equivalent to 15 mmHg. A second 23
G outflow needle was inserted through the limbus, left
in the anterior chamber, and attached to a reservoir
containing the same solution at a height corresponding to 12 mmHg. The anterior chamber was exchanged with 0.5 cc of Dulbecco's phosphate buffered
saline and the outflow needle was clamped. Outflow
facility was measured using constant pressure perfusion at 15 mmHg.9 The eyes were perfused for 60 min
or until the outflow facility was constant for at least 30
min, whichever was longer. The anterior chamber
was exchanged with 0.5 cc of a fixative solution containing 2.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M Sorensen's phosphate buffered solution
(pH 7.3), after which the outflow needle was clamped.
The eyes then were further perfused at 15 mmHg with
thisfixativefor an additional 60 min, after which they
were partially bisected at the equator and immersion
fixed in the same fixative solution overnight.
Vol. 33
Tissue Preparation and Examination
After fixation, the eyes were bisected and the posterior segment was examined for evidence of glaucomatous optic nerve atrophy. The lens was dissected from
the anterior segment and two meridional wedges were
taken from each quadrant. The IW of Schlemm's
canal was isolated in the following manner. Meridional wedges were placed on edge and held in position by pins placed through the cornea anteriorly and
the sclera posteriorly. Using an operating microscope,
an incision was made into the anterior portion of
Schlemm's canal, which was opened by the application of gentle traction on the ciliary body, allowing
access to the posterior margin of the canal without
damaging the inner endothelial surface. The resulting
specimen thus consisted of iris, ciliary body, and the
outer TM with the IW endothelial lining exposed.
Specimens were dehydrated through graded alcohols,
critical point dried, mounted on stubs, sputter coated,
and examined using a JEOL JSM 35 or JEOL 6400
SEM (JEOL USA, Inc. Applications Laboratory, Peabody, MA).
The IW endothelium of Schlemm's canal was photographed at X1800 using Polaroid (Cambridge, MA)
55 film. Ten micrographs were taken of each quadrant (40 per eye). Each micrograph was examined in a
masked fashion for numbers of pores, pore diameter,
and "bulges" (representing giant vacuoles or endothelial cell nuclei). Each micrograph was analyzed independently by two of us (RRA and AWdK) until interobserver variability was less than 10%. Openings in
the endothelial lining were considered pores rather
than artifactual tears if the edges were smooth and
regular. Pore diameters were measured using a X10
measuring magnifier (Peak, Japan) with a reticle calibrated to 0.1 mm. The largest diameter was recorded
for oval pores. Counts of bulges were repeated in a
masked fashion until interobserver comparisons were
within 10%. Pore density, bulge density, and calculated IW endothelial facility data were obtained for
each quadrant to analyze regional variability. Whole
eye pore density was calculated by averaging all quadrant pore densities for a given eye. (This method implicitly assumes that pore density is uniform in each
quadrant.) Bulge density was calculated in the same
manner. Pore density distributions were used to calculate IW endothelial facility for that quadrant, as described below.
Quantitative Analysis
The resistance to aqueous outflow contributed by
the endothelial lining of Schlemm's canal was calculated in the manner described by Bill and Sved-
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933167/ on 05/10/2017
No. 5
PORE DENSITY IN NORMAL AND GLAUCOMATOUS HUMAN EYES / Allinghom er ol
bergh.l2 Briefly, the resistance toflowthrough a "pore
invagination unit" is estimated as a function of pore
diameter using elementary fluid mechanics. Pores are
then grouped according to diameter, and all pores
within a given size grouping are assigned a resistance
based on the mean diameter within that group. The
exception to this procedure are pores with diameters
^2 jum, which are assigned resistances corresponding
to a diameter of 2 jum. This was considered to be a
reasonable upper limit of the diameter of the abluminal opening to the giant vacuole. We used the same
pore size groupings as Bill and Svedbergh.
IW endothelial facility is calculated by multiplying
the mean pore density in a given size range by facility
per pore (defined as the reciprocal of the resistance per
pore invagination unit), summing over all pore size
ranges, and multiplying by total IW area considered.
For whole eye calculations, the IW endothelial surface
was taken to be 11 mm2. For quadrant analysis, it was
a quarter of this area.14
In addition to IW facility, we also calculated the
hydraulic conductivity of the IW endothelium, defined as IW facility per unit area (units: fi\/
(min • mmHg • cm2). This standard quantity measures
the intrinsic ease with which fluid crosses a barrier—
in this case, the IW endothelium.
Statistical Analysis
Data, for whole eyes and quadrants, were analyzed
using a regression analysis program that took into account intrapatient correlations as described by Lipsitz
1663
and Harrington.15 Values are given as means ± one
standard deviation.
Results
Morphology
The endothelial surface of the IW of Schlemm's
canal consisted of a continuous lining of elongated
cells generally oriented circumferentially within
Schlemm's canal, ie, parallel to the longitudinal axis
of the canal (Fig. 1). Cells were 100-150 /*m or more
long and typically 4-7 ^m wide. Commonly, one or
more discrete bulging structures, presumably giant
vacuoles or cell nuclei, were seen protruding at
various intervals from a single endothelial cell. Many
of these structures were fusiform in shape, measuring
5-7 fim long and 4-5 fim wide. Some bulges had small
discrete openings, or pores, typically ranging from
0.3-2 /xm in diameter (Fig. 2). Pores appeared to be
randomly distributed in normal eyes. Pores were typically located on bulges, but were occasionally seen on
flatter regions of the inner endothelial lining.
Although the IW endothelium in eyes of patients
with POAG appeared similar to normals in certain
areas of the IW, differences were noted. The circumferential orientation of the endothelial cells was less
apparent. Although the appearance of the pores was
similar to those in normal eyes, they were frequently
observed to be clustered, rather than randomly distributed (Fig. 3). Bulging structures seemed to vary more
in shape and size in POAG eyes. In addition, regions
Fig. 1. Overview SEM of the inner wall endothelial lining in a normal eye perfused at 15 mm Hg. Elongated endothelial cells contain one or
more bulging structures, representing nuclei or vacuoles. A pore (arrow) is seen in a flat section of the inner wall endothelium. Original
magnification X650.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933167/ on 05/10/2017
1664
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / April 1992
Vol. 33
Figs. 2-4. Fig. 2. Higher magnification
SEM of a normal eye. A pore (arrow) is observed at the base of a bulging structure. Original magnification X35OO. Fig. 3. SEM of the
inner wall endothelium in a POAG eye. The
appearance of the bulges in the inner wall lining is less uniform. A cluster of pores (arrows)
is present. Original magnification X35OO. Fig.
4. SEM of the inner wall endothelial lining in
a glaucomatous eye depicting an area devoid
of bulges and pores. Suchflatareas were more
commonly seen in POAG eyes than in normal
eyes. Original magnification X3100.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933167/ on 05/10/2017
PORE DENSITY IN NORMAL AND GLAUCOMATOUS HUMAN EYES / Allingham er al
No. 5
1665
Table 1. Data from individual eyes
No.
8864
8867
8901
8872
8928
1
OD
OS
OD
OS
OD
OS
OD
OS
OD
OS
Age
Sex
Nl vs. POAG
Measured
facility
(fil/min X mm Hg)
84
M
62
F
70
F
73
M
88
F
Nl
Nl
Nl
Nl
Nl
Nl
POAG
POAG
POAG
POAG
0.15
0.21
0.25
0.27
0.33
0.29
0.13
0.12
0.05
0.13
Calculated IWE*
facility
(\d/min X mm Hg)
Area IW
analyzed
2.89
3.09
6.83
4.64
5.06
4.21
1.62
1.11
0.54
1.32
1.02 X 10s
9.7 X 10"
7.4 X 10"
5.6 X 10"
9.1 X 10"
9.2 X 10"
8.7 X 10"
8.6 X 10"
1.03 X 105
1.04 X 10s
(nm2)
IWE = inner wall endothelium of Schlemm's canal.
where the IW wasflatwere noted more commonly in
POAG eyes (Fig. 4). Few pores were seen in the flattened regions.
Quantitative Analysis
Whole eye analysis: Age, sex, measured outflow facility, calculated IW endothelial facility, and area of
IW endothelium analyzed by SEM were tabulated for
normal and POAG eyes (Table 1), and a comparison
between normal eyes and POAG eyes was made (Table 2). Measured outflow facility was 0.25 ± 0.06 /A/
min • mmHg in normal eyes compared to 0.11 ± 0.04
/Lil/min • mmHg in POAG eyes (P < .004). Mean pore
density was 1437 ± 423 pores/mm2 in normal eyes
and was 489 ± 172 pores/mm2 in POAG eyes (P
< .004). Calculated IW endothelial facility was 4.45
± 1.44 and 1.15 ± 0.46 /il/min • mmHg, in normal
and POAG eyes, respectively (P < .02). A histogram
of pore counts in each pore diameter group for normal and POAG eyes was made (Fig. 5). The relative
distribution of pores per pore diameter group was similar between normal and POAG eyes. Mean pore size,
bulge density, and area of IW endothelium analyzed
per eye were not significantly different between normal and POAG eyes.
Regression analysis was performed examining the
relationship between pore density and measured outflow facility in normal eyes, POAG eyes, and in a
combined analysis that included both groups (Fig. 6).
Pore density was related to measured outflow facility
in normal eyes (P < .02) and in the combined group
(P < .001). However, this relationship did not reach
statistical significance in the POAG group when analyzed alone (P < .27). No statistically significant relationship was found between pore diameter or bulge
density and measured outflow facility in these groups.
The relationship between measured outflow resistance in whole eyes and calculated IW endothelial resistance (reciprocal of calculated IW facility) was examined. As measured total resistance increased, calculated IW resistance increased for normal and
POAG eyes (Fig. 7). However, the rate of increase (as
measured by the slope of the regression lines in Fig. 7)
was different for POAG and normal eyes. Specifically,
the percentage resistance attributed to the IW endothelium in normal eyes was 5.8 ± 1.2%, while in POAG
eyes it was 9.5 ± 1.2% (P < .02). Because normal and
POAG eyes were perfused at the same constant pressure, this implies that the pressure drop across the IW
endothelium was different in normal and POAG eyes
Table 2. Comparison normal vs. POAG (Whole eyes)
POAG
(± STD)
Normal
(± STD)
Mean measured outflow facility
(/nl/min X mm Hg)
Mean pore density (per mm2)
Calculated IW facility
(jtl/min X mm Hg)
Calculated IW resistance
(% of measured total resistance)
Mean pore diameter (nm)
Mean area of IW Analyzed (/xm2)
Mean bulge density (per mm2)
P
0.25 ± 0.06
1437 ± 423
0.11 ± 0.04
489 ± 172
0.004
0.004
4.45 ± 1.44
1.15 ±0.46
0.02
5.8 ± 1.2
0.91 ±0.12
8.5 ± 1.7 X 10"
7.3 ± 1.0 X 103
9.5
0.85
9.5
6.7
0.02
0.46
0.42
0.45
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933167/ on 05/10/2017
± 1.2
±0.12
± 0.98 X 10"
± 1.5 X 103
Vol. 33
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / April 1992
1666
IW Endothelial Resistance vs. Measured Outflow Resistance
Mean Pore Count By Pore Diameter
.1-.3
.4-.7
.8-1.1
12-1.6
1.6-1.9
2.0-2.3
2.4-2.7
2.8-3.1
3.2-3.5
>3.5
Pore Diameter (urn)
Fig. 5. Histogram of the number of pores per eye within each
pore diameter grouping in normal and POAG eyes. Vertical bars
are ± STD.
(0.87 mmHg vs. 1.43 mmHg for normal and POAG
eyes, respectively).
There was a statistically significant difference between the calculated hydraulic conductivity of the IW
endothelium in normal eyes compared to POAG
eyes. The hydraulic conductivity of the IW endothelium was 40.5 ±13.1 /il/(min • mmHg • cm2) in normal eyes and 10.4 ± 4.1 /il/(min • mmHg • cm2) in
POAG eyes (P < .02).
Regional analysis: Quadrant pore densities were
calculated by pooling data from corresponding quadrants from individual eyes. Mean quadrant pore densities in normal eyes ranged from 1262 ± 464 pores/
mm2 in the superonasal quadrant to 1523 ± 759
pores/mm2 in the inferotemporal quadrant (Fig. 8).
In POAG eyes, mean quadrant pore densities ranged
Outflow Resistance (mmHg x min/ul)
Fig. 7. Graph of calculated IW endothelial resistance versus total
measured outflow resistance in normal and POAG eyes. The slope
was greater in POAG eyes than in normals (P < 0.02).
from 373 ± 127 pores/mm2 in the inferotemporal
quadrant to 630 ± 564 pores/mm2 in the superonasal
quadrant. There were no significant differences in regional mean pore density among quadrants in normal
eyes, POAG eyes, or when combined. The pore density was significantly lower in three of four quadrants
of POAG eyes when compared to the same region in
normal eyes (P < .05), the exception being the superonasal quadrant (P < .11).
Comparisons of quadrant-calculated IW endothelial facility were made in normal and POAG eyes. In
normal eyes, calculated IW endothelial facility per
Regional Pore Density By Quadrant
Pore Density vs. Measured Outflow Facility
Q
Normal
•
POAG
2 D
.8 -
D
D
1.6 Q
1.4 -
I
t '*
1.2 -
<§ 0.8 S
D
1 -
S.
oe
-
0.4 D
0.8 -
02 -
•
0.8 -
IT
0.4 -
SN
ST
Total
IN
IT
SN
ST
o Normal
•
0.2 -
POAG
0 Outflow Facility (ul/mln/mmHg)
Fig. 6. Graph of the relationship between pore density and measured outflow facility. The relationship was statistically significant
in normal eyes alone (P < 0.02) and when data were pooled with
POAG eyes (P< 0.001).
Fig. 8. Bar graph of regional pore density per quadrant comparing normal with POAG eyes. There were no significant differences
in regional pore density among quadrants of normal or POAG eyes
or when data were combined. Pore density was significantly lower
in POAG eyes when compared with the corresponding region in
normal eyes, with one exception, the superonasal quadrant (P
< 0.11). Total is the overall mean pore density. IN, IT, SN, and ST
are inferonasal, inferotemporal, superonasal, and superotemporal,
respectively. Vertical bars are ± STD.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933167/ on 05/10/2017
No. 5
PORE DENSITY IN NORMAL AND GLAUCOMATOUS HUMAN EYES / Allingham er ol
quadrant ranged from 1.5 ± 0.8 ^l/min«mmHg in
the superonasal quadrant to 0.8 ± 0.5 /ul/min • mmHg
in the superotemporal quadrant. In POAG eyes, the
calculated IW endothelial facility per quadrant
ranged from 0.2 ± 0.1 jtl/min • mmHg in the superonasal quadrant to 0.5 ±0.3 ^1/min • mmHg in the inferonasal quadrant. There were no significant differences between the calculated IW endothelial facility
per quadrant in normal eyes and POAG eyes. The IW
endothelial facility per quadrant was significantly
lower in POAG eyes than in the respective normal
eyes except for the inferotemporal quadrant (P < .06).
Discussion
Pore/Facility Correlation
The endothelial lining of Schlemm's canal previously has been estimated to contribute up to 24% of
total outflow resistance in normal human and primate eyes.1213 In addition, it has been shown that
pore and giant vacuole density increase with increased perfusion pressure.16"21 In this study, we have
demonstrated that a relationship exists between measured outflow facility and pore density in normal eyes
alone and when combined with POAG eyes. This relationship did not reach statistical significance in
POAG eyes when analyzed alone, possibly because of
the small number, the relatively lower outflow facility
of POAG eyes, or pathological changes in the IW endothelium.
Although the postmortem changes may have influenced the results of this study—and may influence
the results of any study that uses donor tissue—we
made every effort to prevent artifactual changes. Donor eye tissue was harvested soon after death. Measurements of outflow facility and perfusion fixation
were performed within 18 hr. Furthermore, the difference observed in pore density between normal and
glaucomatous eyes, and the noted relationship between pore density and outflow facility, cannot be explained by postmortem time differences. The possibility that pore formation could be occurring postmortem also was considered. We examined specimens
from emersion fixed eyes that were otherwise identically prepared and found pore density to be near zero
(R. Allingham, A. de Kater, D. Epstein, Howe Laboratory of Ophthalmology, Boston, MA, unpublished
data). Therefore, we believe that fluid flow through
the trabecular meshwork was responsible for the formation of endothelial pores observed in this study.
There are at least three possible reasons for the positive correlation between outflow facility and pore
density. (1) The endothelial lining itself is responsible
for a majority of total resistance in the human eye. (2)
1667
The endothelial lining of Schlemm's canal is responsible for "modulating" total resistance in the eye. This
implies that the endothelial lining could alter fluid
flow resistance "upstream" within the TM, perhaps
by altering the resistance of flow pathways within the
juxtacanalicular region. (3) The IW endothelium is a
passive resistor to aqueous outflow, and pore density
is modulated by a variable such as fluid flow or pressure drop across the IW endothelial lining.
In our study, the mean calculated resistance of the
inner endothelial lining of Schlemm's canal was 5.8%
of total measured resistance in normal eyes and 9.5%
in POAG eyes. These values are consistent with those
of Bill and Svedbergh.12 Therefore, it appears unlikely
that the IW endothelium by itself produces a majority
of total aqueous outflow resistance and explanation
(1) can be eliminated.
Explanation (2) is similar to (1) in that it attributes
a controlling role on facility to the endothelial lining,
although in this case by interaction with other tissues
"upstream." The present study does not allow us to
rule out this mechanism. Even if it eventually proves
not to be a dominant factor, hydraulic interactions of
the IW endothelium with the juxtacanalicular region
still may contribute significantly to overall resistance.
Explanation (3) rests on the widely accepted idea
that the pores in the IW endothelium open in response to transmural pressure. The reduction in pore
number in POAG eyes can be seen as a consequence
of lower facility causing a lower fluid flow rate. This
could produce a lower pressure drop across the IW
endothelium (with perfusion pressure held constant).
However, as noted above, not only was the overall
flow resistance higher in POAG eyes but the IW resistance was proportionately greater. The calculated IW
endothelial transmural pressure drop is 64% higher
for POAG eyes (0.87 mmHg for normal eyes and 1.43
mmHg for POAG eyes). The higher transmural pressure in the presence of lower flow clearly implies that
the reduction in pore number was more than could be
accounted for by lower fluid flow alone. Therefore,
the IW endothelium in POAG eyes appears to be
more resistant to pore formation. There is not enough
data in this study to characterize the pressure/pore
formation process. A similar study that analyzes pore
density after perfusion at different levels of constant
flow (rather than constant pressure) may clarify this
and other questions.
Evidence that supports an intrinsic difference between the IW endothelium of Schlemm's canal in
normal and POAG eyes comes from tracer studies.
De Kater et al perfused normal and POAG autopsy
eyes with cationic ferritin (CF).22 In normal eyes, the
abluminal cell surface of giant vacuoles was consis-
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933167/ on 05/10/2017
1668
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / April 1992
tently decorated. However, in POAG eyes, CF staining was invariably absent at this location. These investigators concluded that an alteration of the electrical
charge was present in this region or CF entry was
blocked from entering the vacuoles in POAG eyes. In
a study that examined the distribution of sialic acid
moieties associated with the endothelial lining of normal and glaucomatous eyes, Tripathi et al found an
increased density of sialic acid on luminal and abluminal surfaces in POAG specimens compared to normal eyes.23 Increased levels of sialic acid, which is believed to stabilize cell membranes, could induce abnormally high cell rigidity, ultimately affecting the
process of vacuolization as well as pore formation. In
addition, sialic acid is known to interact with a variety
of extracellular matrix molecules that could directly
or indirectly affect the IW endothelium. This study
suggests that there also may be a functional alteration
of the IW endothelium in POAG. The effect of medical therapy on the function of the IW endothelium of
glaucoma eyes is not known.
Regional Variability
Regional differences in aqueous outflow have been
suggested by some investigators.24'25 We were unable
to detect a statistically significant difference in regional pore density or calculated IW endothelial facility in normal or POAG eyes. However, this does not
rule out the possibility that regional variation exists.
In fact, we observed substantial variation in pore and
bulge distribution within segments of normal and
glaucomatous eyes, although this was not localized to
a specific quadrant. Interestingly, de Kater et al found
evidence for regional variation in flow in POAG
eyes.22 Qualitatively, we found that glaucomatous
eyes appear to have greater variability in pore distribution than normal eyes. Pores were more clustered
in POAG, suggesting the presence of localized areas of
increased aqueous outflow. In addition, there appeared to be a greater number of areas where the IW
endothelium was flat with few pores. Flattened regions may represent areas where Schlemm's canal was
collapsed. If so, pore formation (and related aqueous
flow) through these areas appears to have been diminished. These findings lend further support to the impression that there is greater regional variability in
aqueous outflow in POAG eyes.
Preliminary evidence suggests there may be more
giant vacuoles in normal eyes than in POAG eyes.26
In the present study, we observed no apparent relationship between "bulges" and outflow facility. However, SEM is not well suited to analyzing giant vacuole formation because it is not possible to clearly dis-
Vol. 33
tinguish between giant vacuoles and IW endothelial
cell nuclei.
Comparison With Other Endothelia
In this study, the calculated hydraulic conductivity
of the IW endothelium of normal eyes was 40.5
±13.1 jul/(min • mmHg • cm2), while in POAG eyes it
was 10.4 ± 4.1 /il/(min • mmHg • cm2). In the circulatory system, the primary region for fluid flow across
the endothelial lining is within the capillaries. The
hydraulic conductivity of several different vascular
endothelia has been calculated (Table 3).27 Interestingly, the hydraulic conductivity of the IW endothelium of Schlemm's canal in normal eyes is 30 times
greater than that of fenestrated renal capillary endothelium and far greater than other types of vascular
endothelia. The hydraulic conductivity of POAG IW
endothelium is lower than that of normals, but is still
nearly an order of magnitude greater than fenestrated
renal capillary endothelium. We speculate that high
hydraulic conductivity may be essential to IW endothelial cell survival. In the absence of low resistance
transcellular endothelial pores, stresses on the IW endothelial lining, induced by aqueous flow, might
prove excessive, causing disruption of cell-to-cell and
cell-to-matrix attachments. Ultimately, the underlying IW endothelial basement membrane could be exposed to the contents of Schlemm's canal. If blood
were to reflux into Schlemm's canal, platelet aggregation could occur on exposed basement membrane
surfaces, resulting in obstruction of aqueous pathways
within the juxtacanalicular tissues or occlusion of
Schlemm's canal itself.28 Of course, if blood were to
reflux through these areas into the anterior chamber,
visual acuity also might be adversely affected.
Table 3. Comparison between hydraulic
conductivity of IW endothelium of Schlemm's canal
and capillary endothelium
Endothelium
IW endothelium, Schlemm's canal
Normal
POAG
Fenestrated capillaries*
Renal (frog, mammal)
Intestinal mucosa (mammal)
Nonfenestrated capillaries*
Frog mesentery
Dog lung
Cat hind leg
Tight junction*
Rabbit brain
Lp (fil/(min x mm Hg X cm2)
40.5 ± 13.1 (STD)
10.4 ±4.1 (STD)
1.2
1.0 X 10"'
4.0 X 10"3
2.7 X 10"3
2.0 X 10"3
2.4 X 10"5
* Values for hydraulic conductivity derived from Renkin.28
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933167/ on 05/10/2017
No. 5
PORE DENSITY IN NORMAL AND GLAUCOMATOUS HUMAN EYES / Allinghom er ol
Key words: Schlemm's canal, outflow facility, endothelial
pores, scanning electron microscopy, flow resistance, hydraulic conductivity, primary open angle glaucoma, trabecular meshwork, aqueous humor outflow system.
Acknowledgments
The authors would like to thank Victor Merritt and
Aliakbar Shahsafaei for excellent technical assistance, Pat
Basler for secretarial assistance, Mark Johnson, for stimulating discussions on the dynamics of fluid flow, and Robert
Glynn, for suggestions regarding the statistical analysis. The
authors are grateful to JEOL USA, Inc. Applications Laboratory, Peabody, MA, for the use of the JEOL 6400.
14.
15.
16.
17.
18.
References
1. Karg SJ, Garron KL, Feeney L, and McEwen WK: Perfusion
of human eyes with latex microspheres. Arch Ophthalmol
61:68, 1959.
2. Feeney L and Wessig S: Outflow studies using an electron
dense tracer. Trans Am Acad Ophthalmol Otolaryngol 70:791,
1966.
3. Grierson I, Lee WR, and Abraham S: The appearance of the
outflow apparatus of the eye after staining with ruthenium red.
Acta Ophthalmol 55:827, 1977.
4. Barany E: Pore size and passage of paniculate matter through
the trabecular meshwork. Doc Ophthalmol 13:41, 1959.
5. Garron L, Feeney ML, Hogan MJ, and McEwen WK: Electron
microscopic studies of the human eye. Am J Ophthalmol
46:27, 1958.
6. Holmberg A: Thefinestructure of the inner wall of Schlemm's
canal. Arch Ophthalmol 62:956, 1047, 1959.
7. Tripathi RC: Ultrastructure of Schlemm's canal in relation to
aqueous outflow. Exp Eye Res 7:335, 1968.
8. Tripathi RC: Mechanisms of the aqueous outflow across the
trabecular wall of Schlemm's canal. Exp Eye Res 11:116,1971.
9. Grant WM: Further studies on facility offlowthrough the trabecular meshwork. Arch Ophthalmol 60:523, 1958.
10. Grant WM: Experimental aqueous perfusion in enucleated human eyes. Arch Ophthalmol 69:783, 1963.
11. Maepea O and 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 49:645,
1989.
12. Bill A and Svedbergh B: Scanning electron microscopic studies
of the trabecular meshwork and the canal of Schlemm—an
attempt to localize the main resistance to outflow of aqueous
humor in man. Acta Ophthalmol 50:295, 1972.
13. Moseley H, Grierson I, and Lee WR: Mathematical modelling
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
1669
of aqueous humour outflow from the eye through the pores in
the lining endothelium of Schlemm's canal. Clin Phys Physiol
Meas4:47, 1983.
McEwen WK: Application of Poiseuille's law to aqueous flow.
Arch Ophthalmol 60:290, 1958.
Lipsitz SR and Harrington D: Analyzing correlated binary data
using SAS. Comput Biomed Res 23:268, 1990.
Grierson I and 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 39:505, 1984.
Grierson I and 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 19:21, 1974.
Grierson I and Lee WR: Light microscopic quantitation of the
endothelial vacuoles in Schlemm's canal. Am J Ophthalmol
84:234, 1977.
Grierson I and Lee WR: Pressure-induced changes in the ultrastructure of the endothelium lining Schlemm's canal. Am J
Ophthalmol 80:863, 1975.
Ainesworth JR and Lee WR: Effects of age and rapid high-pressure fixation on the morphology of Schlemm's canal. Invest
Ophthalmol Vis Sci 31:745, 1990.
Johnstone MA and Grant WM: Pressure-dependent changes in
the structures of the aqueous outflow system of human and
monkey eyes. Am J Ophthalmol 75:365, 1973.
de Kater AW, Melamed S, and Epstein DL: Patterns of
aqueous humor outflow in glaucomatous and nonglaucomatous human eyes. A tracer study using cationized ferritin. Arch
Ophthalmol 107:572, 1989.
Allingham RR, de Kater AW, Shahsafaei A, and Epstein DL:
Relationship between outflow facility and giant vacuoles of
Schlemm's canal in normal and glaucomatous human eyes,
abstract. Invest Ophthalmol Vis Sci 31(Suppl):338, 1990.
Ellingsen BA and Grant WM: Trabeculotomy and sinusotomy
in enucleated human eyes. Invest Ophthalmol Vis Sci 11:21,
1972.
Campbell DG, Boys-Smith JW, and Woods WD: Variation of
pigmentation and segmentation of pigmentation in primary
open angle glaucoma, abstract. Invest Ophthalmol Vis Sci
25(Suppl):122, 1984.
Raviola G and Raviola E: Paracellular route of aqueous outflow in the trabecular meshwork and canal of Schlemm. A
freeze-fracture study of the endothelial junctions in the sclerocorneal angle of the macaque monkey eye. Invest Ophthalmol
Vis Sci 21:52, 1981.
Renkin EM: Multiple pathways of capillary permeability. Circ
Res 41:735, 1977.
Hamanaka T and Bill A: Role of blood in Schlemm's canal,
abstract. Invest Ophthalmol Vis Sci 31(Suppl):337, 1990.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933167/ on 05/10/2017