Investigative Ophthalmology & Visual Science, Vol. 29, No. 2, February 1988
Copyright © Association for Research in Vision and Ophthalmology
Effect of Long-Term Contact Lens Wear on Corneal
Endothelial Cell Morphology and Function
Keith H. Carlson, William M. Bourne, and Richard F. Drubaker
Anterior segment fluorophotometry (topical) and central endothelial cell photography were performed
on 40 long-term (2-23 years) contact lens wearers (four groups of ten each: hard, soft, gas permeable,
and gas permeable plus prior lens usage) and 40 non-contact lens wearers of similar ages. Morphologically, the endothelial cells of contact lens wearers showed greater variability in size and shape compared to controls. The mean endothelial cell size in contact lens wearers (307 ± 35 ixm2) was smaller
than that of controls (329 ± 38 Mm2, P < 0.01). There was an increase in the coefficient of variation of
cell size of the contact lens group (0.35 ± 0.06 versus 0.25 ± 0.04 for controls, P < 0.0001). The
endothelial cell mosaic contained a smaller percentage of hexagonal cells in contact lens wearers (66
± 8) compared to controls (71 ± 7, P < 0.01). There was a compensatory increase in five-sided cells.
Functionally, there was no difference in corneal clarity, central corneal thickness or endothelial
permeability to fluorescein (3.78 ± 0.57 X 10"4 cm/min versus 3.85 ± 0.55 X 10"4 cm/min for
controls) between the two groups. Aqueous humor flow was increased 7% in contact lens wearers. We
found no correlation between oxygen transmissibility, estimated underlying oxygen tension, or duration of wear of the contact lenses and any morphologic or functional variable. We also found no
differences between the four groups of contact lens wearers except that the gas permeable lens wearers
had more hexagonal and less pentagonal cells. Long-term contact lens wear induces morphologic
changes in the corneal endothelium. These changes, however, do not result in functional abnormalities
detectable by the methods of this study. Invest Ophthalmol Vis Sci 29:185-193,1988
Contact lens usage for the correction of refractive
error is widespread. These lenses, in their various
forms, have been available for over 30 years. In 1981,
Schoessler and Woloschak described morphologic
changes in the corneal endothelium which they attributed to chronic contact lens usage1; since that time,
numerous reports have confirmed and extended their
findings.2""
The corneal endothelium is a single cell layer of
neural crest origin12 that forms the most posterior
layer of the cornea. The integrity of this layer is crucial for maintenance of normal corneal transparency.
Its main function is to keep the cornea in a relatively
dehydrated state for optimum clarity. This is accomplished by means of an energy-dependent metabolic
pump and a relative barrier to fluid flow between the
endothelial cells. Loss of endothelial function results
in prompt corneal swelling.
The endothelium appears to have limited regenerative potential. If human or primate endothelial cells
are damaged, the remaining cells in the area respond
by enlargement and migration to cover the defect.13"19 Mitotic activity is minimal.
Several studies have shown that with increasing age
there is a decrease in endothelial cell density and a
concurrent increase in the variability of cell size.20"24
Numerous insults including infection, trauma or intraocular surgery can damage and destroy endothelial
cells and thereby hasten the "aging" process.14 Given
this limited endothelial regenerative potential, as cell
loss occurs a point may be reached where the remaining cells are not in sufficient number to enlarge
and cover the entire posterior corneal surface, producing corneal decompensation.25
A number of studies have shown statistically significant alterations in the endothelium of young, longterm contact lens wearers.1"1' Although the mean cell
size is normal, there is a wide variability of individual
cell sizes which is out of proportion to age-matched
controls. The cellular variability seen in these young
contact lens wearers is similar to that seen in much
From the Department of Ophthalmology, Mayo Clinic, Rochester, Minnesota.
Supported in part by NIH grants EY-02037 and EY-00634, Research to Prevent Blindness, Inc., New York, New York, and the
Mayo Clinic and Foundation, Rochester, Minnesota.
Presented at the annual meeting of the Association for Research
in Vision and Ophthalmology, Sarasota, Florida, April 30, 1986.
Submitted for publication: December 23, 1986; accepted September 24, 1987.
Reprint requests: William M. Bourne, MD, Department of Ophthalmology, Mayo Clinic, Rochester, MN 55905.
185
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933369/ on 07/28/2017
186
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / Februory 1988
Table 1. Characteristics of contact lens subgroups
Group 1:
n = 10
Standard hard contact lenses
—polymethylmethacrylate (10)
—five males, five females
Group 2:
n = 10
Daily wear soft contact lenses
—Hydron® (2), Hydron 04® (2), Soflens® (4), Ciba
Clear® (l),Ciba aqua® (1)
—five males, five females
Group 3:
n = 10
Gas permeable hard contact lenses
—Polycon I® (2), Polycon II® (4), Boston II® (2),
Paraperm® (2)
—three males, seven females
Group 4:
n = 10
Gas permeable hard contact lenses plus prior hard
or soft lens wear
—Polycon I® (1), Polycon II® (6), Paraperm® (1),
Optacryl 95® (1), Optacryl K® (1)
—five males, five females
Manufacturers:
Hydron and Hydron 04: American Hydron, Woodbury, NY.
Soflens: Bausch and Lomb, Rochester, NY.
Ciba Clear and Ciba aqua: CIBA Vision Care, Atlanta GA.
Polycon I and II: Sola-Syntex Ophthalmics, Phoenix, AZ.
Boston II: Polymer Technology Corporation, Wilmington, MA.
Paraperm: Paragon Optical Inc., Mesa, AZ.
Optacryl 95 and K: Optacryl Inc., Inglewood, CO.
older non-contact lens wearing populations. This
gives the impression of an "accelerated" aging process.
Several interesting questions arise. Do these morphologic changes in the endothelium translate into
functional changes? Are endothelial permeability and
central corneal thickness affected in contact lens
wearers? Do changes in endothelial cell morphology
or function vary with the type of contact lens worn?
Are such changes transient or permanent and progressive? In this study we addressed the first three
questions.
Materials and Methods
Forty long-term (at least 2 years) contact lens
wearers age 17 to 42 years (28 ± 6 years, mean
± standard deviation) were recruited from the Mayo
Clinic contact lens practice. All subjects had a negative ophthalmic history and examination, ie normal
visual acuity, intraocular pressure and slit lamp appearance. The corneas of all subjects were clear and
without edema. There were four subject groups often
subjects each (Table 1). Forty subjects (20 of each sex)
with normal eye examinations who had never worn
contact lenses served as the control group. The age
range of the control group was 12 to 48 years (mean
= 29 ±11 years). We obtained written informed consent for the study from all subjects.
The study consisted of two parts: endothelial cell
Vol. 29
photography and anterior segment fluorophotometry. All of the subjects had five to ten photographs
taken of the central 3-4 mm of each cornea with a
contact specular microscope (Heyer-Schulte Medical
Optics Center, Irvine, CA). The microscope had a
rectangular field of view of approximately 0.04 mm2
of endothelial area. The central corneal thickness of
each eye was also measured with the specular microscope.26 Photogrammetric measurements of the volumes of the anterior chambers were recorded.27
The photographic negatives were projected at a
magnification of X500, and the areas of the individual cells were measured with an electronic planimeter
(Numonics, Inc., Lansdale, PA). For each eye, 100
cells were traced from several negatives by a masked
observer. This technique of measurement has a precision of 2% and a reproducibility (difFerent photographs of the same cornea) of 7% if only 50 cells are
traced.22 The mean and standard deviation of endothelial cell size, as well as the coefficient of variation
of cell size (standard deviation/mean), were recorded
for the 100 cells traced from each eye. In addition, the
cell shape or polygonality was determined by counting the number of vertices (or sides) of each traced
cell. Polygonality was expressed as the percentage of
cells with n sides for n = 5, 6, 7, etc.
On a separate day, 30 hr after contact lens removal,
the subject instilled Fluress® (0.25% fluorescein and
0.4% benoxinate; Barnes-Hind Inc., Sunnyvale, CA)
into both eyes every 5 min for 20 min starting at 2:00
AM. Three fluorophotometric measurements were
performed at 4 hr intervals with a scanning ocular
fluorophotometer28 beginning at 8:00 AM. The measurements were made through the central 5 mm of
the cornea and anterior chamber. The lids and lashes
were carefully rinsed to remove any dried deposits of
fluorescein prior to the first measurement.
The concentration of fluorescein in the mid-horizontal plane of the anterior segment was measured
each time. The autofluorescence of the corneal
stroma, measured before application of fluorescein,
was subtracted from the totalfluorescence.The mass
of fluorescein in the corneal stroma (Mc) was calculated as the product of the mean concentration of
fluorescein in the stroma (Cc) and the geometric volume of the stroma (Vc) (assumed to be 70 fd for all
eyes). The mass of fluorescein in the anterior
chamber (Ma) was calculated as the product of the
mean concentration of fluorescein in the anterior
chamber (Ca) and the geometric volume of the anterior chamber (Va) measured for each eye photogrammetrically.27
The rate of clearance of fluorescein from the anterior chamber by diffusion andflowwas calculated for
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933369/ on 07/28/2017
No. 2
EFFECT OF CONTACT LENS WEAR ON ENDOTHELIUM / Carlson er ol.
each temporally spaced pair of measurements of fluorescence in the eye:
rate of clearance = (AMC + AMa)/CaAT
(1)
Ca is the "average" concentration of fluorescein in
the anterior chamber during the time interval, calculated from the formula:
ln[Ca(l)/Ca(2)]
(2)
Ten percent of the clearance of fluorescein was assumed to be due to diffusional losses and 90% due to
the flow of aqueous humor through the anterior
chamber: rate of flow of aqueous humor = rate of
clearance X 0.9.
We calculated the cornea-to-anterior chamber
transfer coefficient (kc.ca) from the same measurements of fluorescein concentration using the relationships denned by Jones and Maurice.29 The endothelial permeability to fluorescein was calculated
from the transfer coefficient as follows: Permeability
= kc.ca X CT/0.64 in which CT is the central corneal
thickness as measured with the specular microscope,
and 0.64 is the anterior chamber-to-cornea equilibrium distribution ratio determined by Ota et al.30 v
Polarization of fluorescence in the corneal stroma
and anterior chamber was measured fluorophotometrically using modifications in the scanning fluorophotometer as described by Herman et al.31 The polarization is related to the proportion of the emitted
light that is linearly polarized parallel to the same
plane as the polarized excitatory light. The polarization offluorescencegives an indication of binding of
the fluorescein molecule in tissue or solution.
For each subject, the following contact lens information was recorded: brand name, number of years
worn, hours of daily wear, water content, oxygen permeability (Dk), central lens thickness (L) and estimated oxygen tension under the lens.
The manufacturers provided the oxygen permeability (Dk) for each specific lens, except for the polymethylmethacrylate lenses (group 1), for which we
used the permeability value reported by O'Neal et
al.32 The central lens thickness (L) was determined by
direct micrometer measurement for the hard and gas
permeable lenses and from the manufacturers' specifications for the soft lenses. The oxygen transmissibility (Dk/L) was calculated for both lenses of each subject using the central lens thickness for L. The oxygen
tension under each lens was estimated by the method
of Fatt and Lin33 using a blink period of 5 seconds, a
tear film thickness of 32 nm, and a tear mixing effi-
187
ciency of 0.05 for the soft lenses and 0.15 for the hard
and gas permeable lenses.
Analysis of Data
We recorded the following values for both eyes of
each subject: (1) Endothelial permeability to fluorescein; (2) Central corneal thickness; (3) Aqueous
humorflow;(4) Mean endothelial cell size; (5) Coefficient of variation of cell size; (6) Relative frequency of
cell shapes; (7) Polarization of fluorescence in the
cornea; and (8) Polarization of fluorescence in the
anterior chamber.
First, we compared the data (values 1-8) of all
contact lens wearers (n = 40) with that of the control
group (n = 40). Second, we compared each contact
lens subgroup separately with the control group.
Third, we looked for significant differences between
the four contact lens subgroups. Fourth, we looked
for sex differences in the contact lens wearers. Fifth,
we looked for correlations with oxygen transmissibility of the lenses, estimated oxygen tension under the
lenses, and duration of lens wear. A two-sample t-test
and Wilcoxon test were performed for each variable
analyzed. Correlations were tested with a Spearman
coefficient of correlation. A probability of 0.05 was
considered statistically significant.
Results
Table 2 summarizes the combined data from all
four contact lens-wearing groups and the control
group. The data listed in Tables 2 and 3 are mean
values from both eyes. The data from the right and
left eyes were also analyzed separately, and the results
were similar. Endothelial permeability to topically
applied fluorescein was not statistically different between the two groups. We found average values of
3.78 ± 0.57 X 10"4 cm/min for the contact lens group
and 3.85 ± 0.55 X 10~4 cm/min for the control
group. In a study such as this, a statistically significant difference (P <, 0.05, two-sided) would be found
90% of the time (90% power) if the true difference in
permeability were at least 4.1 X 10"5 cm/min (11%)
and the population from which the samples were
drawn were distributed normally. In other words, we
can be 90% confident that long-term contact lens
wearers, as a group, differ by less than 11% in endothelial permeability from non-contact lens wearers.
Central corneal thickness was similar for both groups.
A slight increase in aqueous humor flow was found
in contact lens wearers, 2.84 ± 0.51 /ul/min compared
to 2.65 ± 0.52 ^l/min for the controls. This was of
borderline statistical significance (Wilcoxon: P
<0.05, t-test: P< 0.10).
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933369/ on 07/28/2017
188
INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / February 1988
Vol. 29
Table 2. Summary data: all contact lens wearers versus controls
40
Contact lens wearers
(mean ± SD)
•40
Controls
(mean ± SD)
Endothelial permeability tofluorescein(X10~4 cm/min)
Central corneal thickness (mm)
Aqueous humor flow (/zl/min)
3.78 ± 0.57
0.54 ± 0.03
2.84 ±0.51
3.85 ±0.55
0.54 ± 0.03
2.65 ± 0.52
Endothelial cell size (jtni2)
Coefficient of variation of cell size
Relative frequency of cell sides (%)
5 sides
6 sides
7 sides
8 sides
Polarization offluorescence(cornea)
Polarization offluorescence(anterior chamber)
Fluorescein concentration (ng/ml)f
Age (yrs)
307 ± 35
0.35 ± 0.06
329 ± 38
0.25 ± 0.04
20 ± 4
66 ± 8
12 ± 3
2± 1
0.19(n= 18) ±0.02
0.03(n= 18) ±0.03
752 ± 400
28.1 ± 6
16 ± 4
71 ± 7
11 ± 3
1±1
0.18(n = 38)±0.02
0.03(n = 38) ± 0.03
646 ± 402
29.1 ± 11
*P<0.10—t-test.
Morphologic analysis of the endothelial cells
showed a decrease in mean cell size in contact lens
wearers compared to the controls, 307 ±35 /tm2 and
329 ± 38 nm2 respectively. In addition, there was an
increase in the coefficient of variation of cell size
among contact lens wearers (0.35) compared to controls (0.25).
Contact lens wearers showed a decrease in the relative frequency of endothelial cells with six sides, 66%
compared to 71% for controls, and a corresponding
increase infive-sidedcells.
Polarization of fluorescence, an indication of protein binding, was measured in 18 of the 40 contact
lens wearers (Table 2). No significant difference was
found between the contact lens and control groups
with respect to cornea and anterior chamber polarization offluorescence.The initial concentration of fluorescein in the corneal stroma was not statistically
different between the two groups (Tables 2 and 3).
Finally, we found no differences by sex in any of the
variables analyzed in either the contact lens or control groups.
Table 3 summarizes the data of each subgroup of
contact lens wearers compared to the control group.
Endothelial permeability and central corneal thickness were relatively constant across all four contact
lens groups as well as the control group. Wilcoxon
rank sum analysis showed no differences between the
groups.
The borderline increase in aqueous humor flow
described above is again seen in Table 3. Hard, soft
and gas permeable contact lens wearers all showed a
numerical increase in aqueous humor flow, but only
in soft contact lens wearers was this difference statis-
Probability
(Wilcoxon)
>0.58
>0.78
<0.05
(<0.l0)*
<0.01
<0.0001
<0.001
<0.01
>0.61
<0.002
>0.15
>0.38
>0.15
>0.10
t Initial concentration offluoresceinin corneal stroma, 6 hr after topical
application.
tically significant. Those subjects who currently wear
gas permeable lenses but had previously worn hard or
soft lenses actually had a small insignificant decrease
in aqueousflow.Wilcoxon rank sum analysis showed
no differences between the groups.
Endothelial cell size was decreased in all contact
lens groups compared to the control group. The
change was greatest for gas permeable contact lens
wearers (299 ^m2 versus 329 Mm2), and least for soft
contact lens wearers (321 jum2 versus 329 fim2). The
coefficient of variation of cell size uniformly increased across all contact lens groups. The greatest
increase was in hard contact lens wearers (0.38) and
the least increase was in gas permeable contact lens
wearers (0.32). Wilcoxon rank sum analysis found no
differences within the four contact lens groups with
respect to endothelial cell size and coefficient of variation.
The relative frequency of five and six-sided endothelial cells was found to differ across the four contact
lens groups. Gas permeable lens wearers had a higher
proportion of six-sided cells and a lower proportion
of five-sided cells (72% and 16% respectively) than
the other three groups (P < 0.01 and P < 0.001).
The oxygen transmissibility (Dk/L) and estimated
oxygen tension of each lens were compared to the
mean endothelial cell size, the coefficient of variation
of cell size and the percentage of hexagonal cells associated with it. No statistically significant correlations were found. The preceding correlations were
also not significant when each contact lens group was
tested separately. Aqueous humor flow, central corneal thickness, polarization offluorescence,endothelial permeability to fluorescein and the percentage of
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933369/ on 07/28/2017
able 3. Subgroups of contact lens wearers vs control group*
Control^
dothelial permeability
(X10~4cm/min)
ntral corneal thickness
(mm)
ueous humor flow
(Ml/min)
dothelial cell size (fim2)
efficient of variation
lative frequency of cell
sides (%)
5 sides
6 sides
7 sides
8 sides
uorescein concentration
(ng/ml)§
e (years)
Mean ± standard deviation.
From Table 2.
Hard
Soft contacts
Probability^
(n = 10)
Probability^
Gas permeable
(n = 10)
Probability^
Gas perm
prior
(n = 10)
(n = 40)
(n = 10)
3.85 ± 0.55
3.92 ± 0.58
>0.72
3.67 ± 0.52
>0.37
3.80 ± 0.78
>0.85
3.72 ± 0.40
0.54 ± 0.03
0.54 ± 0.03
>0.96
0.53 ± 0.03
>0.55
0.55 ± 0.03
>0.21
0.54 ± 0.02
2.65 ± 0.52
329 ± 38
0.25 ± 0.04
2.89 ± 0.48
305 ± 34
0.38 ± 0.08
>0.19
<0.08
<0.00001
2.97 ± 0.56
321 ± 19
0.33 ± 0.04
<0.06
>0.52
<0.0001
2.89 ± 0.50
299 ± 32
0.32 ± 0.06
>0.19
<0.02
<0.001
2.61 ±0.51
305 ± 49
0.34 ± 0.05
16 ± 4
71 ± 7
11 ±3
1±1
23 ± 4
61 ± 7
12 ± 4
2± 1
<0.0001
<0.003
>0.84
>0.09
21 ± 2
64 ± 6
12 ± 4
1± 1 ,
<0.0005
<0.02
>0.78
>0.45
16 ± 3
72 ± 4
10 ± 2
2± 1
>0.48
>0.83
>0.17
<0.02
20 ± 5
66 ± 9
12 ± 3
2± 1
646 ± 402
29.1 ± 11
903 ± 465
32.5 ± 6.8
755 ± 433
24.4 ± 5.4
>0.39
702 ± 436
29.2 ± 5.4
>0.07
650 ± 245
26.3 ± 5.2
>0.57
% Contact lens subgroup compared to control group (Wilcoxon).
§ Initial concentration offluoresceinin corneal stroma, 6 hr after topical application.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933369/ on 07/28/2017
190
INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / February 1988
Vol. 29
Table 4. Contact lens data
Water content (%)
Mean
Range
Years worn
Mean
Range
Hours per <jay worn
Mean
Range
Oxygen transmissibility (Dk/L) X 10~9
cm ml 02/sec ml mm Hg
Mean
Range
Estimated oxygen tension under
contact lens (mm Hg)
Mean
Range
Hard
Soft
Gas permeable
1.5
1.5
39.0
37.5-45.0
<l-<2
14.1
5-23
2-11
13.4
8-16
14.2
12-16
4.9
<1
3.2
2-7
14.5
5-17
Gas permeable*
plus other
<1
<l-<2
3.2
2-8
14.7
12-16
0.17
0.1-0.3
12.1
5.4-16.5
11.0
4.6-17.3
10.9
0.6-21.3
15.0
—
37.3
16.0-54.0
49.9
26.0-76.0
50.6
17.0-95.0
* Data are for gas permeable hard contact lenses currently worn. Subjects had previous hard or soft contact lens usage.
cells with sides other than six also showed no significant correlations with DK/L or estimated oxygen
tension.
Table 4 summarizes additional contact lens data.
Percent water content of lens, number of years worn
and hours worn per day are listed for each contact
lens group. We found no statistically significant correlations between wearing time (either years or
hours/day) and any morphologic or functional variable. This was true across all subgroups of contact
lens wearers.
Discussion
This study agrees with the results of other investigators in showing that contact lens wear affects the
corneal endothelium. There is an increased variation
of endothelial cell size and shape in the eyes of contact lens wearers. The ratio of the standard deviation
of cell size divided by the mean cell size, termed the
coefficient of variation of cell size, is a useful measure
of the variability of endothelial cell size in a given
subject. Numerous studies,3'5'6-8"11 including this one
(Table 2), show an increased coefficient of variation
in contact lens wearers. In addition, the percentage of
hexagonal cells making up the endothelial mosaic
decreases in contact lens wearers, and there is a compensatory increase in cells of other than six sides
(Table 2). Contact lens-induced hypoxia is thought to
be responsible for these endothelial changes.1"6 It is
noteworthy that the gas permeable lens wearers had a
statistically significantly higher proportion of sixsided cells (P < 0.01) and lower proportion of fivesided cells (P < 0.001) than the other three groups
(Table 3). It has been reported that cellular hexago-
nality is a more sensitive indicator of endothelial
damage or instability than is the coefficient of variation of cell size.34
If corneal hypoxia is responsible for,these morphologic changes, then a contact lens which provides the
greatest amount of oxygen to the cornea should produce the least amount of endothelial change.
Schoessler et al7 have shown that lenses of extremely
high oxygen transmissibility (Silicone Elastomer—
Silsoft® (Bausch and Lomb, Rochester, NY): Dk/L
= 340 X 10"9 cm ml O2/sec ml mm Hg for a lens 0.1
mm thick) produce no morphologic changes in the
corneal endothelium. O'Neil et al32 and Holden and
Mertz35 found that a minimum lens oxygen transmissibility (Dk/L) of 75-87 X 10"9 cm mlO2/sec ml mm
Hg is needed to prevent contact lens-induced corneal
edema with overnight wear. The contact lenses used
in the present study had substantially lower oxygen
transmissibilities (Table 4). Because we used central
lens thickness rather than average thickness36 to calculate Dk/L, the true oxygen transmissibilities of the
lenses may be somewhat different, but this should not
affect our conclusions.
We found no significant correlation between oxygen transmissibility, estimated oxygen tension or duration of lens wear and any morphologic or functional variable, even when each contact lens group
was considered separately. It is possible that these
values do not mirror corneal oxygenation closely because of other factors such as lens movement and tear
exchange and our inability to measure them accurately.
Whereas others1"11 have shown either no change or
an increase in the mean endothelial cell size of contact lens wearers, we recorded a decrease in size that
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933369/ on 07/28/2017
No. 2
EFFECT OF CONTACT LENS WEAR ON ENDOTHELIUM / Carlson er ol.
was present across all subgroups (Tables 2, 3). Because we analyzed 100 different cells from several
small-field photographs, it is possible that proportionally more large cells than small cells were untraceable (and therefore not measured) because they
crossed the margins of the photographs. If this error
in sampling had occurred, however, we probably
would not have observed a much larger coefficient of
variation of cell size in the contact lens group. Nevertheless, our results need to be confirmed by studies
with wide-field photography. In our contact lens
group, despite the large variation in size and shape, a
substantial proportion of endothelial cells were much
smaller than the mean cell size of the control group.
Why a larger proportion of small endothelial cells
should be present in contact lens wearers is puzzling.
It has been shown that mean endothelial cell size
increases with age.20"24 In our study, however, there
were no age differences between the contact lens and
control groups (Tables 2, 3). One explanation is the
possible faulty sampling mentioned above, which we
feel is unlikely. A second unsupported explanation is
that endothelial cells are dividing in response to metabolic stress. Hirst et al,6 observing clusters of small
cells in endothelial cell photographs, suggested the
possibility of active endothelial mitosis despite evidence to the contrary after other types of injury.13"19
Aqueous humor flow was slightly increased in contact lens wearers (Tables 2, 3). The change inflowwas
quite small (7%) and probably is unimportant. We
hestitate to speculate about its cause (increased lactate in the cornea37 or aqueous humor,38 for example)
until the finding is confirmed.
We looked for differences between the groups that
could account for the changes in flow observed. Contact lens wearers removed their lenses 30 hr prior to
the study to allow the corneal epithelium to return to
a baseline state. Accordingly, we found no difference
between the initial fluorescein concentration within
the corneal stroma of contact lens wearers and controls (Tables 2, 3). There was no evidence of increased
binding of fluorescein within the stroma of either
group as shown by the similar values of the polarization offluorescence(Table 2). Had increased binding
been present, it could have quenched the fluorescent
signal from the corneal stroma and, thereby, influenced the flow calculation. The anterior chambers
of myopic contact lens wearers may have been larger
than the controls, but no correlation was found between the anterior chamber volume and aqueous
flow.
With these morphologic changes in the corneal endothelium, is endothelial function impaired? Corneal
clarity and central corneal thickness were the same in
both groups. Likewise, endothelial permeability to
191
fluorescein was not statistically different; with the
techniques used in this study, we can state with 90%
confidence that the true difference in endothelial permeability between contact lens wearers and controls
is not greater than 11%. By the techniques of this
study, it appears that the endothelium of contact lens
wearers functions as well as that of non-contact lens
wearers. However, as Holden et al10 have shown, epithelial and stromal thickness decrease following contact lens removal. They found that stromal thickness
was 2.5% greater in the lens-wearing eye than in the
control eye immediately following lens removal, but
decreased to become thinner than the control eye by
the second day. Stromal thickness continued to decrease and reached a constant level in 1 week. Our
subjects removed their lenses 30 hr prior to the study.
Perhaps our corneal thickness measurements were
made during the period of transition from increased
stromal thickness to stromal thinning, and thus we
were unable to record a difference in corneal thickness between controls and contact lens wearers.
The changes in the endothelial cell morphology
seem to develop early and regress slowly, if at all, after
abandoning contact lens wear. Zantos and Holden39
and others40"45 have shown that the earliest endothelial cell changes (endothelial bleb response) appear
minutes after initial exposure to contact lenses. Holden et al1046 observed that increased variation in endothelial cell size is apparent within 2 weeks of commencing extended soft lens wear. In addition, they
observed patients 1 month after discontinuing contact lenses and found only a slight reduction in the
variability of endothelial cell size that was not statistically significant. MacRae et al" studied former hard
contact lens wearers who discontinued lens wear for
an average of 4.3 years. These individuals showed
persistent endothelial alterations compared to controls. From our data, patients with prior hard or soft
contact lens wear and subsequent gas permeable lens
wear showed no improvement in endothelial cell
morphology. However, the oxygen transmissibility in
the gas permeable lenses was not substantially greater
than that of the soft lenses (Table 4). Moreover, subjects with gas permeable lenses and no prior lens wear
showed endothelial changes similar to that of soft and
hard contact lens wearers.
Stocker and Schoessler8 noted a significant increase
in polymegathism (coefficient of variation of cell size)
with increased wearing time in 15 hard contact lens
wearers. We did not demonstrate a significant correlation between these factors, although we studied
only ten subjects in each group. The range of ages of
our subjects was less than that in their study. The
effect of age, which is also correlated with polymegathism,20"24 on the regression between wearing time
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933369/ on 07/28/2017
192
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / February 1988
and polymegathism was not evaluated by Stocker and
Schoessler.8 The average age of the control subjects
and contact lens wearing subjects in our study was
young (29 ± 11 years and 28 ± 6 years, respectively).
We have recently observed that endothelial permeability to fluorescein increases by a factor of 23%
from age 5 years to 79 years.47 At the same time,
endothelial cell morphology becomes more variable
with advancing age. The increased variability in endothelial cell size and shape in older individuals is
similar to that seen in much younger contact lens
wearers. Time will tell if contact lens wearing plus
advancing age have additive effects on the corneal
endothelium and thereby lessen its functional reserve
in times of stress.
In summary, the use of daily wear soft, standard
hard or gas permeable hard contact lenses induces
morphological changes in the corneal endothelium.
Contact lens wearers show greater variation in the
size and shape of endothelial cells compared to noncontact lens wearing controls. By the methods used in
this study, however, no differences in endothelial
function were found between these contact lens
wearers and controls. Endothelial permeability to fluorescein, central corneal thickness and corneal clarity
were the same for both groups.
Key words: contact lens wear, specular microscopy, fluorophotometry, endothelial morphology, endothelial function
References
1. Schoessler JP and Woloschak MJ: Corneal endothelium in
veteran PMMA contact lens wearers. International Contact
Lens Clinic 8:19, 1981.
2. Caldwell DR, Kastl PR, Dabezies OH, Miller KR, and Hawk
TJ: The effect of long-term hard lens wear on corneal endothelium. Contact and Intraocular Lens Medical Journal 8:87,
1982.
3. Woloschak MJ: The corneal endothelial appearance in unilateral contact lens wearers. Masters thesis, Ohio State University,
1983.
4. Caillau S and Cochet P: Corneal endothelium and contact
lenses. Contactologia 5:32, 1983.
5. Schoessler JP: Corneal endothelial polymegathism associated
with extended wear. Int Contact Lens Clinic 10:148, 1983.
6. Hirst LW, Auer C, Cohn J, Tseng SCG, and Khodadoust AA:
Specular microscopy of hard contact lens wearers. Ophthalmology 91:1147, 1984.
7. Schoessler JP, Barr JT, and Freson DR: Corneal endothelial
observations of silicone elastomer contact lens wearers. Int
Contact Lens Clinic 11:337, 1984.
8. Stocker EG and Schoessler JP: Corneal endothelial polymegathism induced by PMMA contact lens wear. Invest Ophthalmol Vis Sci 26:857, 1985.
9. MacRae SM, Matsuda M, and Yee R: The effect of long-term
hard contact lens wear on the corneal endothelium. CLAO J
11:322, 1985.
10. Holden BA, Sweeney DF, Vannas A, Nilsson KT, and Efron
N: Effects of long-term extended contact lens wear on the
human cornea. Invest Ophthalmol Vis Sci 26:1489, 1985.
Vol. 29
11. MacRae SM, Matsuda M, Shellans S, and Rich LF: The effects
of hard and soft contact lenses on the corneal endothelium.
Am J Ophthalmol 102:50, 1986.
12. Johnston MC, Noden DM, Hazelton RD, and Coulombre JL:
Origins of avian ocular and periocular tissues. Exp Eye Res
29:27, 1979.
13. Van Horn DL and Hyndiuk RA: Endothelial wound repair in
primate cornea. Exp Eye Res 21:113, 1975.
14. Bourne WM, McCarey BE, and Kaufman HE: Clinical specular microscopy. Trans Am Acad Ophthalmol Otolaryngol
81:743, 1976.
15. Doughman DJ, Van Horn D, Rodman WP, Byrnes P, and
Lindstrom RL: Human corneal endotheliaJ layer repair during
organ culture. Arch Ophthalmol 94:1791, 1976.
16. Treffers WF: Human corneal endothelial wound repair: In
vitro and in vivo. Ophthalmology 89:605, 1982.
17. Matsubara M and Tanishima T: Wound-healing of the corneal
endothelium in the monkey: A morphometric study. Jpn J
Ophthalmol 26:264, 1982.
18. Tsuru T, Araie M, Matsubara M, and Tanishima T: Endothelial wound-healing of monkey cornea: Fluorophotometric and
specular microscopic studies. Jpn J Ophthalmol 28:105, 1984.
19. Steinhorst U, Bohnke M, Singh G, and Draeger J: Electron
microscopic studies on endothelial wound healing in vitro.
Fortschr Ophthalmol 81:288, 1984.
20. Laing RA, Sandstrom MM, Berrospi AR, and Leibowitz HM:
Changes in the corneal endothelium as a function of age. Exp
Eye Res 22:587, 1976.
21. Stefansson A, Miiller O, and Sundmacher R: Non-contact
specular microscopy of the normal corneal endothelium: A
statistical evaluation of morphometric parameters. Graefes
Arch Clin Exp Ophthalmol 218:200, 1982.
22. Bourne WM: Morphologic and functional evaluation of the
endothelium of transplanted human corneas. Tr Am Ophthalmol Soc 81:403, 1983.
23. Suda T: Mosaic pattern changes in human corneal endothelium with age. Jpn J Ophthalmol 28:331, 1984.
24. Schimmelpfennig B: Topography of age-related changes in
corneal endothelial cell size: Some preliminary microscopic
observations. Klin Monatsbl Augenheilkd 184:353, 1984.
25. Mishima S: Clinical investigations on the corneal endothelium: XXXVIII Edward Jackson memorial lecture. Am J Ophthalmol 93:1, 1982.
26. Bourne WM and Enoch JM: Some optical principles of the
clinical specular microscope. Invest Ophthalmol 15:29, 1976.
27. Johnson S, Coakes RL, and Brubaker RF: A simple photogrammetric method of measuring anterior chamber volume.
Am J Ophthalmol 85:469, 1978.
28. McLaren JW and Brubaker RF: A two-dimensional scanning
ocular fluorophotometer. Invest Ophthalmol Vis Sci 26:144,
1985.
29. Jones RF and Maurice DM: New methods of measuring the
rate of aqueous flow in man with fluorescein. Exp Eye Res
5:208, 1966.
30. Ota Y, Mishima S, and Maurice DM: Endothelial permeability
of the living cornea tofluorescein.Invest Ophthalmol 13:945,
1974.
31. Herman DC, McLaren JW, and Brubaker RF: A method of
determining concentration of albumin in the living eye. Invest
Ophthalmol Vis Sci 1988, in press.
32. O'Neil MR, Poise KA, and Sarver MD: Corneal response to
rigid and hydrogel lenses during eye closure. Invest Ophthalmol Vis Sci 25:837, 1984.
33. Fatt I and Lin D: Oxygen tension under a soft or hard, gas-permeable contact lens in the presence of tear pumping. J Optom
Physiol Opt 53:104, 1976.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933369/ on 07/28/2017
Jo. 2
EFFECT OF CONTACT LENS WEAR ON ENDOTHEUUM / Corlson er ol.
34. Matsuda M and Bourne WM: Long-term morphologic
changes in the endothelium of transplanted corneas. Arch
Ophthalmol 103:1343, 1985.
35. Holden BA and Mertz GW: Critical oxygen levels to avoid
corneal edema for daily and extended wear contact lenses.
Invest Ophthalmol Vis Sci 25:1161, 1984.
36. Brennan NA: Average thickness of a hydrogel lens for gas
transmissibility calculations. Am J Optom Physiol Opt 61:627,
1984.
37. Bergmanson JPG, Johnsson J, Soderberg PG, and Philipson
BT: Lactate levels in the rabbit cornea and aqueous humor
subsequent to non-gas permeable contact lens wear. Cornea
4:173, 1985/1986.
38. Hamano H, Hori M, Hamano T, Kawabe H, Mikami M,
Mitsunaga S, and Hamano T: Effects of contact lens wear on
mitosis of corneal epithelium and lactate content in aqueous
humor of rabbit. Jpn J Ophthalmol 27:451, 1983.
39. Zantos SG and Holden BA: Transient endothelial changes
soon after wearing soft contact lenses. Am J Optom Physiol
Opt 54:856, 1977.
40. Barr JT and Schoessler JP: Corneal endothelial response to
rigid contact lenses. Am J Optom Physiol Opt 57:267, 1980.
41. Vannas A, Jukka M, Jukka S, Reijo A, and Esko J: Contact
42.
43.
44.
45.
46.
47.
193
lens induced transient changes in corneal endothelium. Acta
Ophthalmol 59:552, 1981.
Schoessler JP, Woloschak MJ, and Mauger TF: Transient endothelial changes produced by hydrophilic contact lenses. Am
J Optom Physiol Opt 59:764, 1982.
Buckley RJ: Transitory endothelial changes induced by soft
contact lenses. In Acta: XXIV International Congress of Ophthalmology, Henkind P, editor. Philadelphia, J.B. Lippincott
Co., 1982, pp. 1149-1153.
Vannas A, Holden BA, and Makitie J: The ultrastructure of
contact lens induced changes. Acta Ophthalmologica 62:320,
1984.
Holden BA, Williams L, and Zantos SG: The etiology of transient endothelial changes in the human cornea. Invest Ophthalmol Vis Sci 26:1354, 1985.
Holden BA, Sweeney DF, Vannas A, Kotow M, La Hood D,
Grant T, Swarbrick H, and Efron N: Contact lens induced
endothelial polymegathism. ARVO Abstracts. Invest Ophthalmol Vis Sci 26(Suppl):275, 1985.
Carlson KH, Bourne WM, and Brubaker RF: Variations in
human corneal endothelial cell morphology and permeability
to fluorescein with age. ARVO Abstracts. Invest Ophthalmol
Vis Sci 28(Suppl):325, 1987.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933369/ on 07/28/2017