|
Figure 47. Quantitative Cornea Endothelium Damage by SEM
vs. N-VP Monomer Concentration for 0.25 Mrad
Dose y-Radiation Graft Coatings on PMMA.
Wavenumber (cm'1) .
4000 3000 2500 2000 1500 1300 1100 1000 900 800 700 650 625
Figure 38.- Infrared Spectra for PMMA and PVP (Literature, Reference 62).
63
The hydrophilic polymer graft coatings were
characterized to determine the nature of the coating, most
particularly the properties which might influence the
biophysical behavior. The different tests which were
performed are discussed in the following sections.
Measurement of the intrinsic viscosity of the external
polymer formed in solution in y-radiation graft coating was
made to understand the relative amounts of polymer formed
under the different conditions in the study of radiation
dose and monomer concentration. The molecular weight of the
external polymer determined by these measurements would be a
relative measure of the extent of polymerization for the
graft coatings themselves and could be correlated to the
biophysical behavior and other properties of the graft
coating.
The contact angle of water was a surface property
measured to show the water wettability of the coatings and
their hydrophilicity, relative to PMMA. Contact angle was
measured for hydrated samples. Closely related to the
measurement of contact angle for water, the critical surface
tension for the graft coated surfaces was determined. It
was felt that the surface energy of the material might have
a significant influence on the contact adhesion resulting
from contact to an endothelium surface.
Scanning electron microscopy was used to examine the
surfaces and detect and record any changes in the surface
161
The Yradiatin graft coatings may form a diffuse,
boundary layer which not only modified the surface by its
presence as a coating, but also by acting as a thin hydrated
gel, which when compressed will release water to act as a
lubricant and prevent adhesion. RF plasma coatings,
particularly those formed at high energy levels are very
heavily crosslinked and may not be able to act as a
lubricating layer, with a high water content as well as the
Y-radiation graft coatings.
12. The reduction of adhesive force and prevention of
tissue damage by the permanent graft coatings of hydrophilic
polymers prepared and characterized here show that safer
IOLs are possible. These thin and adherent coatings seem to
provide a barrier to adhesion with tissue surfaces, without
interfering with IOL function.
13. Preliminary studies conducted on peritoneal
adhesions from abdominal surgery and desquamation from
endotrachea tube intubation have shown additional areas
where the use of hydrophilic polymer coatings is
advantageous in preventing tissue damage. The permanent
hydrophilic polymer graft coatings developed here may be
suitable for use in these applications.
% Transmission
Wavelength (nm)
Figure 41. Ultraviolet-Visible Transmission Spectra of y-Radiation Graft Coatings of
PVP on PMMA, 0.25 Mrad Dose Level.
131
12
blood and other tissues, issued by the U. S. Department of
Health and Human Services, extensive and exhaustive
protocols for the testing of toxicity and thrombogenicity
are detailed (26,27), but no testing of cell adhesion to
materials (other than blood cell adhesion) is recognized or
recommended. The adhesion of cells and tissue to materials
has* been relatively ignored.
However, some research has been devoted to the study of
cell adhesion to surfaces and reviews are available (28,29).
This research has been relatively limited dealing only with
cell adhesion pertinent to tissue culture (30), studies of
surgical adhesives and dental restoratives (31-33),
bacterial adhesion to otic implants (34), marine bacteria
and barnacle adhesion (35), and changes in cell surfaces
accompanying tumor cell metastasis (21). Few studies have
attempted to measure the actual strength of cell adhesion to
surfaces (36).
In these studies, tissue cultures were grown on
different substrates and then were subjected to shear
forces. Although a number of methods have been used, the
most extensive studies have used the shearing force
generated by a metal disc immersed in the culture medium and
lowered to a measured distance from the cells and then
rotated at a known velocity (36) The percentage of cells
of the tissue culture detached by this constant shear force
for a set period of time was measured. For a shear force of
2
7.4 dynes/cm applied for twenty seconds, 27.8% of the
8
Although other materials have been suggested, PMMA is the
material which is currently used for nearly all IOLs.* The
style or shape of lens, its placement in either the anterior
or posterior chamber (Figure 1-B & C), the surgical
technique for insertion and the method of fixation within
the eye have been varied extensively since the first
implantation, but the optical portion of the lens has
remained as originally fabricated from PMMA.
Studies have shown that after IOL insertion that there
was an average loss of almost half of the central corneal
endothelial cells (3,4); these cells form a monolayer on the
posterior surface of the cornea (Figure 1-D). These cells
serve an essential function in maintaining fluid balance
within the cornea. Damage to the endothelium will result in
leakage of fluid into the corneal layers and will cause
swelling and blurred vision. This monolayer is nonregen-
erative in man so that the damage is irreversible.
Our initial studies (5,6) revealed that this cell loss
was due to contact with.the acrylic IOL surface a't the time
of the insertion surgery (Figures 2 & 3). Others have
confirmed that such contact causes endothelium cell loss
(7,8) and efforts to prevent the contact and damage were
undertaken.
The only exceptions are poly(propylene) support loops and a
commercially available IOL made of glass with poly(imide)
supports (Lunell, NYC).
143
Figure 46.3. 3Q% Cornea Endothelium Damage after Contact
with 2Q% N-VP/0.25 Mrad Graft Coating.
50% Cornea Endothelium Damage after Contact
with 1Q% N-VP/Q.25 Mrad Graft Coating.
65
The tissue adhesion properties of several common
biomedical polymers were also evaluated for comparison to
the graft coated acrylic system. The contact angle for
water of these materials was also measured for the same
purpose.
4.2 Intrinsic Viscosity Molecular Weight of PVP
The molecular weight of the external polymer formed in
solution during Y-radiation graft coating gave an indication
of the relative molecular weight of the polymer (PVP)
actually grafted to the PMMA substrate; the molecular
weights of PVP formed under Y-radiation graft coating
conditions are presented in Table 3. The linearity of the
molecular weight versus initial N-VP concentration plot,
shown in Figure 16, coincides with previously reported data
(58), as did results for 0.1 and 0.25 Mrad dose levels. The
molecular weights for a given initial monomer concentration
were consistently higher for the higher dose level. The
difference became larger with increasing concentration.
In addition, the insolubility of the precipitated
homopolymer of the 5% and 10% N-VP concentrations at the
0.5 Mrad dose level in methanol, the solvent used for the
viscometry, indicated, along with the gelation of the
external homopolymer at 20% and 30% monomer concentrations
for that radiation dose level, the tendency for increased
crosslinking at the 0.5 Mrad dose level. Gelation of PVP in
solution has previously been shown to be a function of both
61
damage quantitatively by optical microscopy with NBT
staining were unsuccessful; counting of damaged cells at
high magnification was unable to distinguish between
materials of widely different tissue adhesion properties.
The. failure of these experiments demonstrated the need for
an in vitro method for evaluation and accurate measurement
of the tissue-material adhesion phenomenon.
4.1.4 Overview of Thesis Research
As described above, surface modification by solutions
of hydrophilic polymers was shown effective in reducing or
eliminating tissue-material adhesion in several different
areas of surgery. However, a clinical need for preventing
IOL-cornea endothelium adhesion and its damaging conse
quences existed and efforts were focused on this problem.
Although solutions of PVP used clinically showed promise
(5), a permanent modification of the IOL surface eliminates
the need to apply coating by the surgeon in the operating
room. Greater safety would also be assured by permanent
modification if the IOL became detached following
implantation, when damage to the endothelium from contact,
just as produced during the original insertion, might occur.
Cast films of a hydrophilic polymer would eliminate the
application of the coating by the surgeon, but would not be
a permanent means to prevent any post-surgical occurrence of
cornea endothelium-IOL adhesion.
110
100
90
80
70
60
50
40
30
20
10
PMMA
30%/0.25 104
1
_L
0
_J 1 _J L_
5 10 20 30
% N-VP (Initial Concentration in Polymerization Media)
¡ummary of Zisman Plots for Critical Surface Tensions (y ) of PVP on PMMA
rs. % N-VP for y-Radiation Graft Coatings of 0.25 Mrad Dose.
00
107
Figure 33.25. 30% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
Af<; 0 ** ry .
V r .
*T?v ' /
Jr '
f y
ff? *
/ /--'V
*
i t f
* /
r
'
(
^ *
* *
' ¥.> *
15KU X2000 3025
* ,
10.0U UFMSE
Figure 33.26. 30% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
134
in the literature (41). This spectral change is attributed
to radiolysis of the PMMA. The effects of the y-radiation
at the dose levels used in the production of the coatings in
this study are relatively insignificant, but higher
radiation doses, as for sterilization, may have detrimental
effects in some uses of the PMMA.
The UV-Vis spectra of the RF plasma coatings show that
these coatings also have no effect on the transmission of
the material over this spectral region (Figure 44).
4.9 PVP-Iodine Interaction
PVP forms a stable chemical complex with iodine, the
exact nature of which remains to be completely understood.
Our attempts to detect and measure the complexing of iodine
by the Y-radiation graft coatings and the RF plasma coatings
were unsuccessful by both UV-Vis and ATR-IR spectroscopy.
The spectra for the coatings were unaltered by soaking the
samples in a saturated solution of iodine in water
overnight.
Iodine has been shown to alter both spectra for
solutions of PVP. However, since the PVP alone was
undetected by both methods of spectroscopy, it was unlikely
that the PVP-Iodine complex would be detected. The spectra
obtained confirmed this, as the spectra were unchanged.
Other research has shown that the addition of iodine
does alter the spectra for PVP in solution (64). The IR
spectral changes are slight and the presence of the highly
105
Figure 33.21. 10% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
Figure 33.22. 20% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
% Transmission
Wavelength (nm)
Figure 42. Ultraviolet-Visible Transmission Spectra of y-Radiation Graft Coatings of
PVP on PMMA, 0.5 Mrad Dose Level.
132
% Transmission
Wavelength (ntn)
Figure 40. Ultraviolet-Visible Transmission Spectra of YRadiation Graft Coatings of
PVP on PMMA, 0.1 Mrad Dose Level.
130
To my Mother and Father,
who made it all possible.
Figure 13. Plexiglass Tank (filled with Saline) with
Tissue Sample.
140
was easily within the standard deviation of the measurements
and shows no significant difference. However, as a plot of
N-VP concentration versus the adhesive force for each of the
three radiation doses shows, the adhesive force drops
significantly with increasing monomer concentration; the
decline in adhesive force is greatest for the 0.25 Mrad
coatings, as the average adhesive force decreases to about
one-fourth of the adhesive force for PMMA (Figure 45).
The decrease in adhesive force with the change in
coating conditions parallels the change in contact angle for
water for the same change in coating conditions. The change
in contact angle is more linear, but the decrease in
adhesive force with increasing hydrophilicity is a
correlation which the initial hypothesis of the usefulness
in developing a hydrophilic coating to prevent the contact
adhesion phenomenon proposed.
Quantitative cornea endothelium damage measurement by
SEM for the contact with these same Y-radiation graft
coatings of PVP on the PMMA confirm the ability of the
coatings to reduce the severity of the contact adhesion.
Figure 46 shows representative SEM photos of damaged
endothelium. As seen in Table 11, PMMA produced an average
destruction of 60% of the cells within the area of contact;
this number does not immediately decrease with the coatings
produced at only 5% N-VP concentration and 0.25 Mrad, but as
monomer concentrations increase past this level, the extent
of endothelium damage shows a dramatic decrease. Although
42
3.10 ESCA (Electron Scattering for Chemical Analysis)
Samples for ESCA were sent to Dr. Christopher Batich*
of Dupont Central Research (Wilmington, Delaware) for
analysis. These samples were mounted to copper wire to
eliminate any background interference in the spectra. A
Kratos ES-300 (Kratos, Manchester, England) ESCA
spectrometer was used, with aluminum (Kot) as the x-ray
source. Data for each peak were collected for twenty
minutes.
Analysis of the raw data was performed by Dr. Batich.
Peak heights and areas were determined for C, N, O, Cl, Cu
and Si. The ratio relative to carbon was then calculated
for each element.
3.11 Infrared Spectroscopy
Surface analysis of the coated surfaces was made by
infrared spectroscopy using a Perkin-Elmer Model 283B
spectrometer (Perkin-Elmer, Stamford, CN). The analysis was
made on the sample surfaces by use of an attentuated total
reflectance (ATR) accessory (No. 185-0382, Perkin-Elmer).
Maximum sensitivity was obtained using KRS-5 crystals as the
prism elements at angles of 60 and 45. Samples, in the
form of coated acrylic "slabs," were placed on both sides of
the element to maximize the spectra's strength. The
greatest slit width available was used, along with a
*Present address, Dept, of Materials Science, University of
Florida.
LIST OF FIGURES
Figure Page
1* Cross Section of Eye and Cornea: (A) Lense,
(B) Anterior Chamber, (C) Posterior Chamber and
(D) Endothelium . 7
2. Scanning Electron Micrograph of Cornea Endothel
ium (X900) 9
3. Scanning Electron Micrograph of Cornea
Endothelium Damaged by Contact to Acrylic Intra
ocular Lens (X2000) 9
4. Predominant Reaction Scheme for y-Radiation
Graft Coatings of PVP on PMMA 15
5. y-Radiation Effects in Solution 16
6. Plasma Polymerization Processes 19
7. Infrared Spectrum of N-VP Monomer (As Distilled). 30
8. Infrared Spectrum of N-VP Monomer (Literature,
Reference 53) 31
9. Schematic Drawing of Co-60 Source, Department
of Radiation Biology, University of Florida ... 33
10. Schematic Drawing of KF Plasma Apparatus .... 37
11. Tissue-Polymer Adhesion Measurement Instrument 45
12. Acrylic Sample in Holder 46
13. Plexiglass Tank (Filled with Saline) with
Tissue Sample 48
14. Loading of Acrylic-Endothelium Interface .... 51
15. Deflection of Glass Fiber due to Acrylic-
Endothelium Adhesion with Lowering of Micrometer
Stage 52
ix
BIOGRAPHICAL SKETCH
\ John Wesley Sheets, Jr., was born on September 17,
1953, in Jacksonville, Florida. He was educated in
Jacksonville and graduated in 1971 from S.W. Wolfson High
School.
Following graduation, the author attended Vanderbilt
University in Nashville, Tennessee, for two years. He then
spent the summer of 1973 at Jacksonville University,
Jacksonville, Florida, before entering the University of
Florida in September of 1973. He graduated with the degree
of Bachelor of Science in zoology in June of 1975. After
one year of post-baccalaurate studies in chemistry, he
entered graduate studies in the Department of Materials
Science and Engineering, University of Florida, in the
Spring of 1976.
While pursuing M.S. and Ph.D. studies at the University
of Florida, the author has served as a graduate research
associate. He is a member of the Society of the Sigma Xi,
Tau Beta Pi, Alpha Sigma Mu and the Society for Plastics
Engineers.
172
23
To obtain the tissue samples, adult New Zealand albino
rabbits were sacrificed by an intravenous overdose of sodium
pentobarbital and the eyes were enucleated immediately upon
death. The corneas were removed with a 3 mm rim of sclera
and the lense-iris diaphragm was peeled from the cornea.
The corneas were placed endothelium-side up on a teflon
blo'ck and the central areas were punched out with a 7 mm
trephine. An identical procedure was employed in isolating
7 mm cornea buttons from fresh human eyes. Only paired (two
eyes from the same donor) human eyes were used, so that one
could serve as a control for the other. The corneas were
then placed on the acrylic surfaces, described as follows.
Endothelium damage was determined by either optical
microscopy with nitro-blue tetrazolium (NBT) staining or
scanning electron microscopy (SEM).
The cornea buttons were contacted to acrylic hemi
spheres which had been dipped in either a balanced salt
solution or a 40 weight % solution of poly(vinyl
pyrrolidone) (PVP) (GAF Corporation, New York; PVP K 29-32,
40,000 MW) in a balanced salt solution. The acrylic
hemisphere was held upright following dipping and the cornea
button was placed, endothelium side down, on the acrylic
surface. Contact time was varied from 1 to 60 seconds.
NBT staining (52) was performed by incubating the
corneas, following contact to an acrylic surface, at 37C
for 15 minutes with a drop of NBT stock solution and a drop
of reduced diphosphopyridine (DPNH) (0.3 mg/ml in water) on
Figure 33.1. Untreated PMMA.
Figure 33.2. Untreated PMMA.
Figure 33. Scanning Electron Micrographs of Scraped
y-Radiation Graft Coatings of PVP on PMMA.
77
uncoated PMMA and therefore furnish information about
materials with surfaces much different than the hydrophilic
coatings prepared by Y-radiation graft and RF plasma
polymerizations. These materials are all polymers which are
important or potentially important for biomedical use in
applications in which the contact adhesion phenomena could
play a significant role.
4.4 Critical Surface Tension
Critical surface tension measurements were made to
additionally characterize the 0.25 Mrad Y-radiation graft
coating series and the RF plasma coatings for correlation to
the biophysical measurements and the other surface
characterizations. However, the high surface energies of
these surfaces made the results of these measurements
difficult to interpret.
In order to measure the contact angle over a range of
different surface tension liquids, liquids which showed a
finite contact angle and would not spread (contact angle of
zero) were required. These liquids were all of relatively
high surface tension (greater than 50 dynes/cm) and the
majority of which are polar in nature. The result is polar
interaction with the hydrophilic PVP graft coated surfaces.
The Zisman plot for PMMA, tested at ambient conditions,
showed a critical surface tension of 38 dynes/cm, which is
dextran were applied to tissue surfaces, surgical gloves and
sponges prior to contact with internal organs and tissue.
Female mongrel dogs (12-20 kg) were anesthesized
initially with thiamylal sodium (Surital, Parke-Davis,
Morris Plains, NJ) (20 mg; 0.5 cc/kg of dogs weight) and
maintained on halothane with controlled ventilation
following endotracheal tube insertion. Abdomens were
prepped with a PVP-iodine (Betadine, Purdue Frederick,
Norwalk, CT) solution and draped for a sterile
intra-abdominal procedure. A midline incision was made with
a cold knife and bleeding vessels were electrocoagulated.
Dogs were divided into the following 4 groups of 4 dogs
each, and all were subjected to a complete exteriorized
bowel exploration as follows:
I)Organs manipulated with dry gloves and dry sponges.
II)Manipulation with dry gloves but using saline-wet
sponges and organs wet with sterile saline.
Ill)Gloves and sponges wet with 25 weight % soluble
dextran (Sigma, St. Louis; 200,000-300,000 MW) in
saline solution. Immediately after opening, 50 ml
dextran solution was sprayed on exposed organs
before manipulation. Just before closure 50 ml
dextran solution was again sprayed on organs.
IV)Same as III using 25 weight % PVP (GAF, New York;
40,000 MW) in saline.
A standardized procedure for organ manipulation was
used for each dog involving palpation of all organs,
rinses in distilled water. Examination of the polished
surfaces by SEM was performed on random samples to assure
smooth surfaces.
The slabs were cut from medical grade acrylic sheet and
were rinsed with distilled water before use.
' 3.3 Purification of Monomer (N-VP)
The N-vinyl pyrrolidone (N-VP) monomer was obtained
from Eastman Kodak Chemicals (Rochester, NY) and contained
0.1% NaOH as an inhibitor to prevent polymerization. The
monomer was purified by distillation under vacuum prior to
use to remove this additive and any other impurities.
Vacuum distillation was performed in a batch process; a
1000 ml Erlenmeyer flask with a teflon-coated magnetic
stir-bar was charged with approximately 500 ml of the
as-received N-VP. The flask was heated by an electric
mantle and supported by a magnetic stirrer. The flask was
connected by a Claisen tube with a 6" high sidearm (which
held a thermometer to register distillation temperature) to
a water cooled Leibig condenser (40 cm jacket length). This
condenser was lined by a connection joint to a 500 ml
Erlenmeyer receiving flask in an ice bath and to the vacuum
source. Vacuum was supplied by a mechanical pump (Model
D-150, Precision Scientific, Chicago) via a liquid nitrogen
cold trap. The vacuum attained was approximately
100 micron.
Uncleanable gel formed at monomer
concentrations of 20% and greater.
Figure 19. Contact Angle for Water vs. % Monomer for
0.5 Mrad Dose y-Radiation Graft Coatings of
PVP on PMMA.
98
^ 4
15KU X500 1001 10.0U UFMSE
*,
Figure 33.7. 10% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
15KU X2000 1001 10.0U UFMSE
Figure 33.8. 10% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
171
62. Pouchert, C.J., Aldrich Library of Infared Spectra,
Aldrich Chemical Company, Milwaukee, Wisconsin, 1975.
63. Harrick, N.J., Internal Reflection Spectroscopy, John
Wiley and Sons, Inc., New York, 1967, pg. 284.
64. Kaneniwa, N., and Ikekawa, A., Chem. Pharm. Bull., 22,
2990 (1974).
65. Tanzawa, H., Nagaoka, S., Suzuki, J., Kobayashi, S.,
Masubuchi, Y., and Kikuchi, T., in Biomedical Polymers,
Goldberg, E.P. and Nakajima, A., eds., Academic Press,
New York, pg. 189.
% Transmission
Wavelength (nm)
Figure 43. Ultraviolet-Visible Transmission Spectra of y-Irradiated PMMA.
133
129
infrared spectroscopy, particularly in light of the high
absorption of the PMMA substrate.
In an attempt to corroborate this conclusion, films of
PVP cast from chloroform solution onto PMMA were made over a
range of thicknesses, and ATR-IR spectra were taken
(Figure 39). The thinest detectable film was made from a 5%
PVP' solution with a 1 mil doctor blade; its spectrum showed
a strongly absorbing peak at 6.0 micron, indicating the
detection of both PVP and the PMMA substrate. SEM
examination of this surface, abraded to remove only the PVP
film, showed that the film was on the order of a few microns
in thickness, considerably thicker than the y-radiation
graft and RF plasma coatings which had been examined under
the same conditions. The coating thickness must therefore
be below the limit of detection of ATR-IR, that is less than
one micron.
4.8 Ultraviolet-Visible Spectroscopy
The ultraviolet-visible (UV-Vis) spectrum of PMMA shows
a sharp loss in transmission at 300 nanometers. The UV-Vis
spectra of the radiation graft coated samples show no marked
changes in the spectrum, although the severity of the drop
in transmission is lessened (Figures 40-42). This change in
spectrum is independent of the N-VP concentration in the
polymerization media and is related only to the level of
Y-radiation dose. The effects of y-radiation are more
pronounced at higher doses as can be seen in Figure 43 and
1Q1
Figure 33.13. 30% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
Figure 33.14. 30% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
Internet Distribution Consent Agreement
In reference to the following dissertation:
AUTHOR: Sheets, John
TITLE: Hydrophilic polymer coatings to prevent tissue adhesion / (record
number: 473800)
PUBLICATION DATE: 1983
<^Ak\
, as copyright holder for the aforementioned dissertation,
hereby grant specific and limited archive and distribution rights to the Board of Trustees of the University of
Florida and its agents. I authorize the University of Florida to digitize and distribute the dissertation described
above for nonprofit, educational purposes via the Internet or successive technologies.
This is a non-exclusive grant of permissions for specific off-line and on-line uses for an indefinite term. Off-line
uses shall be limited to those specifically allowed by "Fair Use" as prescribed by the terms of United States
copyright legislation (cf, Title 17, U.S. Code) as well as to the maintenance and preservation of a digital archive
copy. Digitization allows the University of Florida to generate image- and text-based versions as appropriate
and to provide and enhance access using search software.
This/^rant of permissions prohibits use of the digitized versions for commercial use or profit.
Signature-of Copyright Holder
Printed draped Name of Copyright Holder/Licensee
Personal information blurred
iff 'll- ot
Date of Signature
Please print, sign and return to:
Cathleen Martyniak
UF Dissertation Project
Preservation Department
University of Florida Libraries
P.O. Box 117007
Gainesville, FL 32611-7007
3
(down to 25%) had only a small reduction in contact angle
for water, to 60.
Because there was no systematic variation of parameters
involved in the RF plasma coating process, it is impossible
to determine, at this point, the reasons for the anonamlies
to the previous correlations. The reduction in the
endothelium damage without a similar reduction in the
adhesive force does not seem logical, and will require
further research to provide an explanation. Still, the RF
plasma coatings, produced from both N-VP and HEMA monomers,
are effective in the reduction of endothelium damage;
whether or the correlation exist between surface properties
and the biophysical behavior of these RF plasma coatings
will require a more exhaustive study.
The testing of the various biomedical polymers by these
biophysical tests is shown in Tables 14 and 15. The
adhesive force for urethane is down within the range of the
values found for the Y-radiation graft coatings, and its
contact angle is quite low (37). However, the damage to
the cornea endothelium produced by contact with urethane is
60% which matches that for PMMA. The rest of the materials
tested are hydrophobic, with contact angles as high as 105,
for teflon. The adhesive force values are also high, on the
same level as the PMMA, as are the extent of damage to the
endothelium, measured by SEM. The only exception to this is
the 40% damage assessment for teflon, which is the result of
96
Figure 33.4. 5% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.'
Figure 9. Schematic Drawing of Co-60 Source, Department of Radiation Biology,
University of Florida.
142
Figure 46.2. 10% Cornea Endothelium Damage after Contact
with 30% N-VP/Q.25 Mrad Graft Coating.
Figure 46. Representative Scanning Electron Micrographs
of Cornea Endothelium Damage.
Figure 46.1. 0% Cornea Endothelium Damage after Contact
with 30% N-VP/Q.25 Mrad Graft Coating.
Table 10. Summary of Adhesive Force Measurements for Cornea Endothelium Contact
with Y-Radiation Graft Coatings of PVP on PMMA.
Polymer Surface
(% N-VP Monomer/
Y-Radiation Dose (Mrad))
Average Adhesive
Force (mg/crn )
Standard _
Deviation (mg/cin )
Number of
tests
PMMA
405
162
93
5%/0.1
362
80
21
10%/0.1
179
48
21
20%/0.1
174
118
29
30%/0.1
108
34
29
0%/0.25
490
212
21
5%/0.25
156
40
20
10%/0.25
70
83
24
20%/0.25
105
29
22
30%/0.25
111
19
25
5%/0.5
209
29
26
10%/0.5
146
27
28
139
ACKNOWLEDGEMENTS
The author would like to acknowledge the following
individuals who were invaluable to him and this research:
Dr. Eugene P. Goldberg, special thanks for his patience and
guidance as my mentor; Drs. Moshe Levy and Shimon Reich, for
their paternal and scholarly assistance; Dr. Moshe Yalon,
for his friendship and rare talents; the faculty and staff,
especially Drs. Stanley R. Bates, Christopher D. Batich,
Charles L. Beatty and John J. Hren, for showing how much fun
science can be; Dr. Larry L. Hench and June Wilson, for
their encouragement and advice; and Drs. Mutaz B. Habal,
Randall Olson, Jeffery Katz and Herbert E. Kaufman, for
their collaboration in the preliminary research.
Also, the author wishes to express his sincere
appreciation of his friends: Thomas C. Saitta (counselor),
William E. Longo (confidant), Walter J. McCracken
(compatriot), Cindy L. Flenniken (colleague), Alan D. Walsh
(crony) and Ronald A. Palmer (coach).
The author is also indebted to the National Institutes
of Health-Eye-Institute and Intermedies Intraocular, Inc.
for grants which provided partial financial support for this
study.
iv
APPENDIX
INTRINSIC VISCOSITY MEASUREMENT
I. Sample Preparation
A. Concentration
C=0.5-1.0g/100ml
B. Enough sample to prepare, a concentration in above
range is weighted out exactly. The sample is then
dissolved in an appropriate solvent in a
volumetric flask. Because at least 8 ml is needed
for the measurement/ the volume of solution should
be greater than 10 ml (20-25 ml is recommended).
C. The solution should be filtered with a solvent-
resistant filter prior to measurement.
II. Thermostat for Constant Temperature Bath
A. The temperature should be set at the appropriate
temperature. Fluctuation of temperature should be
within 0.01C.
III. Viscometer
A.
It
should be kept clean and free from dust.
B.
Cleaning Procedure
1.
Chromerge cleaning solution can be used for
cleaning
2.
Rinse
3. -Dry with vacuum aspirator. Do Not Heat in
drying oven.
IV. Viscosity Measurement
A. Viscometer must be set exactly perpendicular.
B. Solvent Viscosity:
1. 10 ml of the chosen solvent is added to the
day viscometer
164
103
Figure 33.17.
5% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
Figure 33.18. 5% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
Table
Pase
14. Summary of Adhesive Force Measurements for
Cornea Endothelium Contact with Various
Biomedical Polymers 151
15. Quantitative Cornea Endothelium Damage by SEM
for Various Biomedical Polymers 152
16. Summary of Measurements for y-Radiation
Graft Coatings of PVP on PMMA 156
viii
99
Figure 33.9. 10% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
Figure 33.10. 2Q% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
s J (. \
Eugene P. Goldberg, Chairman
Professor of Materials Science
and Engineering
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
( ^.^'i.afry
Professor/of Materials Science
and Engineering
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
102
Figure 33.15. 30% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
Figure 33.16. 5% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
Figure 10. Schematic Drawing of RE Plasma Apparatus.
To Lina
100 Watt RF
Generator
Cos#
Surface Tension (Dynes/cm)
Figure 24. Zisman Plot of Critical Surface Tension (y ) of y-Radiation
Graft Coating: 10% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
00
to
(1) HOMOPOLYMERIZATION
(2) BRANCHING & CROSSLINKING
(1)
ch2 ch2
CH, C = 0
V
I
ch2=ch
(NVP)
ch2 ch,
I I
CH, C= O
V
vwwvww^*- CH2 CH
N-VP
CH,
CH- CH,
I I
CH, C = O
V
i
CH
CH,CH,
I I
CH,
CH,
CH-
C= O
PVP Chain =
(2)
CH, -CH,
ch2 c=o
ch2ch.
vwww^.
CH, CH,
I I
CH, C O
V
ch2ch
+H-
N-VP
CH, CH,
CH, C = 0
V
"CH2 C
ch2-ch, ch2
xn-ch
/
CH,C
PVP
ch2 ch2
CH, C = O
V
-CH,
CH
CH,
ch2 ch2
CH, C=Q
V
CH-
Figure 5. y-Radiation Effects in Solution.
163
D. Conduct further characterization of PVP y-radiation
graft coating to aid in understanding of mechanism for
tissue-materials adhesion and its prevention.
1. Measure % hydration of the coated surfaces to find
correlation with coating thickness and with tissue
adhesion properties.
2. Use contact angle hysteresis as another measure of
hydrophilicity for characterizing graft coatings.
6.2 RF Plasma Coatings
A. Prepare coatings under systematic study of power,
vapor pressure and duration of exposure to determine effect
of each variable on extent and nature of graft coating, its
surface character and tissue adhesion properties.
B. Utilize flexibility of plasma polymerization with
use of other vapors (monomers such as HEMA and gases such as
oxygen and acetylene) to prepare coatings; characterize
surface and test tissue adhesion properties to determine
possible use in biomedical applications.
6.3 Further Characterization of Graft Coatings
A. Conduct tissue culture and implantation in cornea
stroma to test biocompatibility of graft coated surfaces.
B. Test mechanical properties of graft coatings:
abrasion resistence and long-term stability. Shows
permanence of the graft.
Cosd
Figure 21. Zisman Plot of Critical Surface Tension (yc) of PMMA.
-j
Table 6. Contact Angle of Water for Various Biomedical Polymers.
Polymer
Average Contact Angle ()
Standard Deviation ()
Urethane
37
7
Silicone Rubber
101
2
Thermoplastic Elastomer
90
2
Teflon
105
3
*Pellthane TM, Upjohn Corporation, Kalamazoo, MI.
**Silastic 500-5 TM, Dow-Corning, Midland, MI.
***Proprietary hydrocarbon block copolymer, C-flex TM, Concept Inc., Clearwater,
FL.
****TM, FEP, E.I. DuPont de Nemours & Co., Wilmington, DE.
119
Figure 34.11. RF Plasma Coating of HEMA on PMMA: 25 Watts
CPower)/500y(Pressure)/15 Minutes(Duration).
Figure 34.12. RF Plasma Coating of HEMA on PMMA: 25 Watts
(Power)/5Q0y(Pressure)/15 Minutes(Duration).
(1) Plasma Induced Polymerization
ch2 ch2
CH, C =0
V
ch2=ch
(N-VP)
Plasma
CH,CH,
I I
CH, C = O
V
CH, CH
CH,CH,
I
CH
, C = 0
^ /
N
CH2= CH
CH, CH,
I I
CH, C = 0
V
CH,
ch2ch2
CH, C =
V
CH
CH,
CH-
(2) Plasma State Polymerization
(Possible Reaction)
ch2 CH2
CH, C=0
V
ch2=ch
Plasma
CH= CH
ch2ch2-n:
c=o
CH, CH,
I I
CH, C = 0
V
CH=CH
CH2 ch2
ch2 C=0
\ /
N
CH2=CH CH2CH *
CH, CH,
Figure 6. Plasma Polymerization Processes
159
tissue damage produced by materials contact can be measured
quantitatively.
5. The tissue adhesion properties were measured for a
range of polymers. Values for the endothelium-polymer
2
adhesive force ranged from a high of 499 mg/cm for a
2
thermoplastic elastomer to 60 mg/cm for a Y-radiation graft
coating of poly(vinyl pyrrolidone) (PVP) on poly(methyl
methacrylate) (PMMA). Uncoated PMMA showed an adhesive
2
force of 405 mg/cm The extent of endothelium damage for
contact with these polymers was highest (85% of cells
destroyed in contact area) for silicone rubber and lowest
(5%) for a Y-radiation graft coating of PVP on PMMA. PMMA
contact produced 60% damage.
6. Gamma-radiation graft coatings of PVP on PMMA
intraocular lens material were prepared and characterized.
These coating were found to be thin (<1 y) and adherent to
the acrylic substrate and significantly altered the surface
of the acrylic, increasing hydrophilicity and surface
energy, and decreasing tissue adhesion and tissue damage.
7. A large reduction in adhesive force (from
2 2
405 mg/cm for PMMA to less 100 mg/cm ) and extent of
endothelium damage (from 60% to less than 45%) was exhibited
by Y-radiation graft coatings of PVP on PMMA which showed
appreciable graft polymerization (molecular weight of
external homopolymer formed of above 600,000) and a
reduction in the contact angle for water of more than 10.
e so
Surface Tension (dynes/cm)
Zisman Plot of Critical Surface Tension (y ) of y-Radiation
Graft Coating: 5% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
Figure 23.
Table 15. Quantitative Cornea Endothelium Damage by SEM for Various Biomedical
Polymers.
Polymer
Cells Destroyed
Average (%)
in Contacted Area
Number of
Tests
Urethane*
60
2
Silicone Rubber**
85
2
Thermoplastic Elastoner***
65
2
Teflon****
40
1
*Pellthane TM, Upjohn Corporation, Kalamazoo, MI.
**Silastic 500-5TM, Dow-Corning, Midland, MI.
***Proprietary hydrocarbon block copolymer, C-flex TM, Concept Inc., Clearwater,
FL.
****TM, FEP, E.I. duPont de Nemours & Co., Wilmington, DE.
CH-
CH-
CH2C
CH.
C = O
C
I
c
VWWWVAAA^
CH.
CH.
C* + *CH.
C = 0
CH-
-CH-
Figure 4. Predominant Reaction Scheme for y-Radiation Graft Coatings of PVP on PMMA.
Mw(x106)
(Molecular Weight by Intrinsic Viscosity)
Figure 16. External Polymer Viscosity Molecular Weight
vs. Monomer Concentration for y-Radiation
Graft Coatings of PVP on PMMA.
58
Table 1. Dog Model for Assessment of Polymer Solutions to
Prevent Adhesions.
Wound
Small
Intestine
Avg.
Score
Rel. Score
Ave. Score
Rel. Score
Drya
3.0
10
2.5
8
Saline3
3.0
10
1.5
5
PVPb
o
CM
7
0.3
1
Dextran
2.0
7
1.3
4
aAverage
of
4
surviving dogs.
^Average
of
3
surviving dogs.
Table 2.
Rat Model-Assessment of PVP Solution to Prevent
Adhesions.
Average Score/Animal
Relative Score
Control3
3.5
5.0
PVP Coating*5 0.7
1.0
aAverage of 4 animals surviving anesthesia and 8 day
maintenance with healed wounds.
Average of 3 surviving animals.
157
energy similar to the y-radiation graft coatings, exhibited
high values in adhesive force testing.
The proposed structure of the PVP y-radiation graft
coating explains the difference in biophysical behavior.
The. graft coating not only presents a "soft" deformable
hydrophilic surface, but also holds water in a "gel"
structure formed by the PVP. The non-adhesive properties of
this structure can be explained in several different ways.
The water held in the graft may be "squeezed-out" with
contact to a tissue surface; the water may then act as a
lubricating boundary layer. Additionally, the smooth and
mechanically-soft surface would act as a cushion to prevent
any mechanical damage.
Alternatively, the diffuse nature of the coating itself
may provide a highly hydrated layer at the surface which
acts much as the viscous PVP solutions used in the
preliminary experiments.
In either case, the hydrated coating acts to prevent
the hydrophobic and/or electrostatic interactions which
might bind contacting endothelium to the polymer surface.
Further investigation into the structure of the PVP
y-radiation graft coatings should offer important insights
into the mechanism of the tissue-materials adhesion
phenomenon.
110
Figure 33.31. 5% N-VP Initial Monomer Concentration and
0.5 Mrad Dose.
Figure 33.32. 10% N-VP Initial Monomer Concentration and
0.5 Mrad Dose.
3500
200
Wavenumber (cm'1)
3000 2500 2000 1800 1600 1400 1200 1000 800 600 400
Wavelength in microns
Figure 7. Infrared Spectrum of N-VP Monomer (As Distilled).
122
showing only that relatively equal amounts of oxygen were
present on the PMMA and the PVP-coated surfaces.
The numbers for the N/C ratio do show a significant
difference between the two types of surfaces in the amount
of nitrogen present. The N/C ratio for the PVP-coated
surfaces is much greater than that for the uncoated PMMA,
for either angle (depth of sampling).
Examination of the ESCA spectra for carbon and nitrogen
shows the real significance of the difference in the N/C
ratios. The nitrogen spectra for the uncoated PMMA (Figure
35) show no real peaks for either angle and can be regarded
as background levels. The nitrogen peaks for both angles of
the two y-radiation graft coatings of PVP on PMMA are
substantial (Figures 36 & 37). The spectra show that
nitrogen is present on these surfaces, assuredly in the
pyrrolidone ring.
The N/C ratio for the two depths of sampling is
effectively the same, so it can be concluded that the PVP
O
coating is uniform down to the sampling depth of 50A.
4.7 Infrared Spectroscopy
The use of attenuated total reflectance (ATR) with
infrared spectroscopy (IR) was selected as the most sensi
tive means to obtain the infared spectra of the y-radiation
graft and RF plasma coatings. Comparison of the spectra of
PMMA and PVP shows the great similarity between the two'; the
best chance at finding a resolvable absorption peak occurs
REFERENCES
1. Jaffe, N.S., Galin, M.A., Hirschman, H., and dayman,
. H.M., Pseudophakos, C.V. Mosby Company, St. Louis
(1978).
2. Shepard, D.S., The Intraocular Lens Manual, Astro
Printers, Santa Maria, California (1977).
3. Bourne, W.M., and Kaufman, H.E., Am. J. Ophthalmol.,
81, 482 (1976).
4. Forsot, S.L., Blackwell, W.L., Jaffe, N.S., and
Kaufman, H.E., Trans. Am. Acad. Ophthalmol.
Otolaryngol, 83, OP-195 (1977).
5. Katz, J., Kaufman, H.E., Goldberg, E.P., and Sheets,
J.W., Trans. Am. Acad. Ophthalmol. Otolaryngol., 83,
OP-204 (1977).
6. Kaufman, H.E., Katz, J., Valenti, J., Sheets, J.W., and
Goldberg, E.P., Science, 198, 525 (1977).
7. Sugar, A., Burnett, J., and Forsot, S.L., Arch.
Ophthalmol., 93, 449 (1978).
8. Irvine, R., Kratz, R.P., and O'Donnell, J.J., Arch.
Ophthalmol., 96, 1023 (1978).
9. Bourne, W.M., Brubaker, R.F., and O'Falon, W.M., Arch.
Ophthalmol., 97, 1473 (1979).
10. Levy, P.L., Roth, A.M., and Gangitano, J.L.,
Ophthalmology, j3j5, 219 (1979) .
11. Soil, D.B., Harrison, S.E., Arturi, F.C., and Clinch,
T., Am. Intra-Ocular Implant Soc. J., jj, 239 (1980).
12. Stegmann, R.R., and Miller, D., Ophth. Surg., 11, 19
(1980).
13. Kirk, S., Burde, R.M., and Waltman, S.R., Invest.
Ophthalmol. Vis. Sci., 16, 1053 (1977).
14. Miller, D., and Stegmann, R.R., Ann. Ophthalmol., 13,
811 (1981).
167
35
placed in test tubes (16 x 125 ram; volume 18 ml) sealed with
aluminum foil and rubber stoppers. No attempt was made to
de-gas the solutions or containers. These test tubes were
irradiated at the dose levels studied (0.1 to 0.5 Mrad)
while held in a specially constructed carousel.
The homopolymer of PVP formed for each set of
conditions was precipitated with 100 ml of acetone in a
small glass container (volume 500 ml) on a blender (Waring,
New Hartford, CN) at high speed. The precipitated polymer
was washed twice with 150 ml of acetone in the blender at
high speed. The polymer was removed from the blender and
loosely sealed in aluminum foil packets. The polymer was
air dried for twelve hours and then dried overnight in a
vacuum oven at 50C and less than 150 torr.
Using the method of the Appendix, solutions for a
viscometry were prepared with methanol as the solvent, and
were filtered twice through Gelman spectroglass filters
(#934-AH, Gelman Instrument Co., Ann Arbor, MI). An
Ubbelholde viscometer, size OB (Fisher Scientific,
Pittsburgh, PA) was used for viscosity measurement, at 30C,
(0.1C) maintained in a water bath. Four dilutions were
used to obtain a In nrel/c vs. concentration plot. The
intercept at zero concentration was found via a best-fit
linear plot with a HP-35c calculator (Hewlett Packard,
Corvallis, OR). Coefficients* for the Mark-Houwink-Sakurada
*These values, for the equation [n] = KMa, were K = 23 x
10J ml/g and a = 0.65.
Wavenumber (cm'1)
4000 3500 3000 2500 2000 1800 1600 1400 1200 1000. 800 600
Wavelength in Microns
Figure 39. Attenuated Total Response-Infrared Spectra of PMMA, y-Radiation Graft
Coating of PVP on PMMA and PVP Cast Film on PMMA.
128
94
these hydrophilic coatings. Still, the value of these
measurements lies not as an accurate measurement of the
surface energy, but as an indication of the (a) increase
relative to the surface of the uncoated PMMA, and (b)
increasing polar interaction with increasing hydrophilicity.
4.5 Scanning Electron Microscopy
All of the following scanning electron micrographs are
of PMMA "chisel-cut" specimens and provide a qualitative
measurement of the coating thickness and the surface mor
phology of the coatings. The appearance of the Y-radiation
graft coatings seems to show little correspondence to the
coating conditions, but some trends do exist (Figure 33).
The majority of the coatings produced at 0.1 and 0.25
Mrad are uniform and show the coverage of the PVP coating
(Figure 33.4-33.27). Their surfaces are very similar in
appearance to uncoated PMMA (Figure 33.1-33.3). At the 0.25
Mrad dose level (Figure 33.16-33.27), holes or gaps do
appear in the coating in some places; this may be due to an
increased thickness compared to the 0.1 Mrad coatings
(Figure 33.4-33.15). All of the coatings seem to show good
adherence to the PMMA substrate and for the 20% and 30% N-VP
coatings, this adherence prevented complete removal, by
scraping, of the coating. Low magnification photographs
(x500) (Figure 33.10, 33.13, 33.22 and 33.25) show the
remnants of the coating remaining in the scraped area; high
magnification photographs (x2000 and x5000) show that the
(above), were used in the construction of a Zisman plot for
each surface tested.
3.9 Scanning Electron Microscopy
Scanning electron microscopy was used to observe the
surface morphology of the different coatings and their
thickness. Coated stubs were scraped with a "chisel" made
from the same PMMA rod stock as the coated substrate. These
specimens were then prepared for microscopy by fixing the
sample to aluminum stubs with colloidal carbon paint
(Structure Probe, Inc., West Chester, PA) to assure mounting
O
and to provide an electrical ground. Gold-palladium (200 A)
was deposited on the surface by a Hummer V sputter coater
(Technics, Alexandria, VA).
Following these preparations, the specimens were
examined with a JEOL JSM-35c SEM (JEOL, Boston, MA). The
scratch made by the PMMA "chisel" was photographed at 500X
and 2000X to show both the appearance of the coating and to
reveal the coating thickness. For these examinations, the
SEM conditions were 15 kV accelerating voltage, condenser at
the 12 o'clock position and the sample held at 0 tilt.
Photographs of the scraped area were taken at 500X, 2000X
and 5000X magnifications to reveal both surface morphology
and the coating thickness.
3E
Epithelium
Bowman's
Membrane
Stroma
Descemet's
Membrane
Endothelium
. Optic nerve
Figure 1. Cross Section of Eye and Cornea: (A) Lense, (B) Anterior Chamber, (C) Posterior
Chamber and (D) Endothelium.
72
dose levels but also for the 0.5 Mrad dose level coatings
(Figures 17-20). The 0.5 Mrad coatings showed a much
greater decrease in contact angle, although the gelation of
the external homopolymer prevented measurement of the
contact angle at 20% and 30% N-VP concentrations.
Comparison of this linear relationship between N-VP
concentration and contact angle for the different dose
levels showed this greater decrease in contact angle at the
5 and 10% N-VP concentrations for the 0.5 Mrad coatings
(Figure 20). The unexpected result that the 0.1 Mrad
coatings show more of a decrease in contact angle than do
the 0.25 Mrad coatings can be explained only on the basis of
the standard deviation of the contact angle measurements.
The standard deviation for this measurement shows that the
values for the 0.1 and 0.25 Mrad coatings do overlap and the
difference is not significant.
Measurement of the contact angle for the RF plasma
coatings shows that these coatings are hydrophilic, like the
Y-radiation graft coatings (Table 5). Only a preliminary
study of RF graft coating was made. Correlations between
reaction conditions and contact angle were not developed.
However, the results of these experiments, along with the
results of biophysical tests, will be discussed later.
Contact angles for a number of common biomedical
polymers were also measured and compared with graft coated
PVP on PMMA (Table 6). The majority of these materials
possess a surface which is more hydrophobic than the
166
f.
g-
n
sp
Cn]
to.
- 1
= intrinsic viscosity
(I.V.)
Determination of I.V.
a. Use linear curve fitting to extrapolate to
zero.
2.
by several techniques, including measurement of the contact
angle for water, electron scattering for chemical analysis
(ESCA), infrared spectroscopy (IR), ultraviolet-visible
spectroscopy (UV-Vis), examination by scanning electron
microscopy (SEM) and measurement of the surface energy.
In addition, a new instrument and technique was
developed for the evaluation of the force of adhesion and
the extent of resultant tissue damage of these coatings and
other plastic materials. These measurements were made for
contact with rabbit cornea endothelium under well controlled
conditions.
Thin graft coatings of PVP on PMMA were found to be
adherent to the acrylic substrate and significantly
increased the hydrophilicity and surface energy as compared
with uncoated PMMA. The altered surface properties of the
PVP graft coated PMMA also resulted in reduction of the
adhesive force and extent of tissue damage, as tested by the
newly devised adhesive force instrument.
Results of this study emphasize the need for develop
ment of hydrophilic polymer graft coatings for clinical IOL
use. ^n vivo animal testing of y-radiation graft coatings
of PVP on PMMA are now in progress. The application of the
hydrophilic polmer coating concepts for other surgical
materials and procedures is also under investigation.
xiv
137
culture, and various methods have been used to measure the
extent of the cell adhesion to material surfaces
(24,36,37,65). None of these techniques are applicable to
the study of the tissue-materials adhesion phenomenon, since
they cannot duplicate the actual clinical conditions of the
phenomenon or measure the significant properties of the
phenomenon in the adhesive force and the resultant cell
damage.
The measurement of the adhesive force for cornea
endothelium contact with PMMA, y-radiation graft coatings of
PVP on PMMA, RF plasma coatings on PMMA and with various
other biomedical polymers have been made. Because of the
expected variation of such measurements with biological
material (cornea endothelium from rabbit eyes), at least ten
measurements were made for each material and coating tested.
The significance of the adhesive force measurement for a
material is in its relative magnitude to the other surfaces
and its correlation to the other properties of that surface,
particularly the other biophysical property measured the
amount of cornea endothelium damage resulting from the
contact.
Measurement of quantitative cornea endothelium damage
by scanning electron microscopy (SEM), also developed for
this study, measures the definitive quantity involved with
the contact adhesion phenomenon. The extent of cell
destruction determines the significance of the adhesion
between the tissue surface and the material since this
53
This crosshair monitored the deflection of the glass
fiber until the adhesive force at the tissue-material
interfaced was exceeded by the flexural stresses in the
glass fiber. When that occured, the crosshair micrometer
was read and recorded, since the interface was broken and
two surfaces had parted.
The "corneal hemisphere was then rotated approximately
60 to bring a fresh area of tissue into position for
measurement. The process was then repeated and continued
until all six positions on the tissue were utilized for
measurements.
The reading recorded during these measurements was
converted into force measurements by a calibration curve
prepared for the glass fiber in use. Such curves were
prepared for each glass fiber; the deflection of each fiber
was measured with known weights hung from the fiber. The
weights were hung on the glass fiber at the same point at
which the gold chain (which suspended the test pieces
experimentally) was located. Then, just as in the procedure
for the measurement of the adhesive force, the micrometer
crosshair of the measuring microscope was used to measure
the deflection for that weight.
These values yielded a linear plot of force versus
deflection. The values for deflection in the adhesive force
measurements were converted into force by use of linear best
fit program and a HP-35c programmable calculator
(Hewlett-Packard, Corvalis, OR). Then, assuming complete
108
Figure 33.27. 30% N-VP Initial Monomer Concentration and
Q.25 Mrad Dose.
Figure 33.28. 5% N-VP Initial Monomer Concentration and
0.5 Mrad Dose.
Cos#
Surface Tension (Dynes/cm)
Figure 26. Zisman Plot of Critical Surface Tension (y ) of y-Radiation
Graft Coating: 30% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
114
Figure 34.1. RF Plasma Coating of N-VP on PMMA: 50 Watts
(.Power)/lOOOy(Pressurel/10 Minutes(Duration).
Figure 34.2. RF Plasma Coating of N-VP on PMMA: 5Q Watts
(Power)/lOOOu(Pressure)/1Q Minutes(Duration).
Figure 34. Scanning Electron Micrographs of Scraped RF
Plasma Coatings on PMMA.
9
Figure 2. Scanning Electron Micrograph of Cornea Endothelium,
fxqnoi
Figure
3. Scanning Electron Micrograph of Cornea Endothelium
Damaged by Contact to Acrylic Intraocular Lens.
(X2000)
HYDROPHILIC POLYMER COATINGS TO PREVENT TISSUE ADHESION
BY
JOHN WESLEY SHEETS, JR.
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1983
Cos?
Figure 27. Zisman Plot of Critical Surface Tension (y ) of 0.25 Mrad
y-Radiation Graft Coatings of Different N-^P Concentrations.
00
cn
55
gold-palladium. The samples were then viewed with a JEOL
JSM-35C SEM.
SEM photomicrographs at 20-3OX were taken of each
specimen and evaluated. The initial assessment of the
extent of damage was done by averaging two independently
made estimates. The accuracy of these assessments was
confirmed by grid counting; this technique employed a grid
and the number of squares within the areas of damage were
compared to the number of squares within the entire area of
contact. This grid counting technique was used on a
selected sample of photomicrographs and percent of tissue
damage obtained by these measurements matched those made by
experimenters' assessments.
78
extremely close to the literature value of 39 dynes/cm (57)
(Figure 21). Measurement of the critical surface tension
for the 0.25 Mrad radiation graft series begins with a
slight decrease in the surface tension for the 5% N-VP
concentration graft coating and then begins to increase to
greater critical surface tension for the 10% and greater
N-VP concentration graft coatings (Figures 22-26). However,
the surface energies obtained for the higher concentrations
are not realistic, as the numbers approach and exceed
100 dynes/cm (Table 7).
The Zisman plots for the 20% and 30% N-VP concentra
tions show a change in slope from the normal, negative slope
to a positive slope (Figures 25-28). The coefficient of
determination for these two coatings' plots is also
considerably less than for the other plots. The polar
interactions that result from the use of polar liquids on
these increasingly hydrophilic surfaces disrupt the
usefulness of the Zisman plot as an accurate measurement of
the critical surface tension of these coatings.
In an attempt to accurately calculate the critical
surface tension for these surfaces, the contact angle data
for the different liquids were treated by the method of
Baszkin and Lyman (60). This method separates the
dispersive and nondispersive (polar) interactions between
the liquid and the. solid and has been performed on surfaces
of high critical surface tension. The polar contribution to
the critical surface tension has been reported as high as
Figure Page
41. Ultraviolet-Visible Transmission Spectra of
y-Radiation Graft Coatings of PVP on PMMA,
0.25 Mrad Dose Level 131
42. Ultraviolet-Visible Transmission Spectra of
y-Radiation Graft Coatings of PVP on PMMA,
0.5 Mrad Dose 132
43. Ultraviolet-Visible Transmission Spectra of
y-Irradiated PMMA 133
44. Ultraviolet-Visible Transmission Spectra of
RF Plasma Coatings on PMMA 135
45. Adhesive Force vs. Monomer Concentration for
PVP y-Radiation Graft Coatings on PMMA 141
46. Representative Scanning Electron Micrographs
of Cornea Endothelium Damage 142
47. Quantitative Cornea Endothelium Damage by SEM
vs. N-VP Monomer Concentration for 0.25 Mrad
Dose y-Radiation Graft Coatings on PMMA .... 147
48. Schematic View of PVP y-Radiation Graft on
PMMA 155
xii
Cos#
1.0
Figure 31. Zisman Plot for Critical Surface Tension (y ) of RF Plasma
Coatings of N-VP on PMMA: 35 Watts(Power)/§00y(Pressure)/
60 Minutes(Duration).
VO
HYDROPHILIC POLYMER COATINGS TO PREVENT TISSUE ADHESION
BY
JOHN WESLEY SHEETS, JR.
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1983
Copyright 1983
by
John Wesley Sheets, Jr.
To my Mother and Father,
who made it all possible.
ACKNOWLEDGEMENTS
The author would like to acknowledge the following
individuals who were invaluable to him and this research:
Dr. Eugene P. Goldberg, special thanks for his patience and
guidance as my mentor; Drs. Moshe Levy and Shimon Reich, for
their paternal and scholarly assistance; Dr. Moshe Yalon,
for his friendship and rare talents; the faculty and staff,
especially Drs. Stanley R. Bates, Christopher D. Batich,
Charles L. Beatty and John J. Hren, for showing how much fun
science can be; Dr. Larry L. Hench and June Wilson, for
their encouragement and advice; and Drs. Mutaz B. Habal,
Randall Olson, Jeffery Katz and Herbert E. Kaufman, for
their collaboration in the preliminary research.
Also, the author wishes to express his sincere
appreciation of his friends: Thomas C. Saitta (counselor),
William E. Longo (confidant), Walter J. McCracken
(compatriot), Cindy L. Flenniken (colleague), Alan D. Walsh
(crony) and Ronald A. Palmer (coach).
The author is also indebted to the National Institutes
of Health-Eye-Institute and Intermedies Intraocular, Inc.
for grants which provided partial financial support for this
study.
iv
TABLE OF CONTENTS
g*2_e.
ACKNOWLEGEMENTS iv
LIST OF TABLES . vii
LIST OF FIGURES ix
ABSTRACT xiii
1. INTRODUCTION 1
2. BACKGROUND 2
2.1 Intraocular Lens Development and Problems ... 2
2.2 Biological Adhesion 11
2.3 Radiation Polymerized Graft Coatings 13
2.4 Plasma Polymerized Graft Coatings 18
3. MATERIALS AND METHODS 22
3.1 Preliminary Studies 22
3.1.1 Prevention of Cornea Endothelium Damage
by Poly(vinyl pyrrolidone) Solutions 22
3.1.2 Peritoneal Adhesions in Abdominal
Surgery 24
3.2 PMMA Substrates 27
3.3 Purification of Monomer (N-VP) 28
3.4 y-Radiation Graft Coatings 29
3.5 Intrinsic Viscosity Molecular Weight of PVP . 34
3.6 RF Plasma Coatings 36
3.7 Contact Angle for Water 38
3.8 Critical Surface Tension ..... 39
3.9 Scanning Electron Microscopy 41
3.1Q ESC (Electron Scattering for Chemical
Analysis) 42
3.11 Infrared Spectroscopy 42
3.12 Ultraviolet-Visible Spectroscopy 43
3.13 PVP-Iodine Interaction 43
3.14 Biophysical Measurements 44
v
Page
3.14.1 Instrument for Biophysical Measure
ments 44
3.14.2 Preparation of Material Sample for
Measurement 44
3.14.3 Preparation of Tissue Samples 47
3.14.4 Adhesive Force Measurement 50
3.14.5 Quantitative Cornea Endothelium
Damage by SEM 54
4. RESULTS AND DISCUSSION . 56
4.1 Preliminary Studies and Overview of Thesis
Research 56
4.1.1 Prevention of Cornea Endothelium Damage
by Poly(vinyl pyrrolidone) Solutions 56
4.1.2 Peritoneal Adhesions in Abdominal
Surgery 57
4.1.3 Other Related Research 60
4.1.4 Overview of Thesis Research 61
4.2 Intrinsic Viscosity Molecular Weight of PVP . 65
4.3 Contact Angle for Water 68
4.4 Critical Surface Tension 77
4.5 Scanning Electron Microscopy 94
4.6 ESCA . 120
4.7 Infrared Spectroscopy 122
4.8 Ultraviolet-Visible Spectroscopy 129
4.9 PVP-Iodine Interaction 134
4.10 Biophysical Measurements 136
4.11 Structure and Properties of y-Radiation
Graft Coatings 153
5. CONCLUSIONS 158
6. FUTURE RESEARCH 162
6.1 y-Radiation Graft Coatings for IOL Use .... 162
6.2 RF Plasma Coatings 163
6.3 Further Characterization of Graft Coatings . 163
APPENDIX 164
REFERENCES 167
BIOGRAPHICAL SKETCH 172
vi
LIST OF TABLES
Table
Page
1. Dog Model for Assessment of Polymer Solutions
, to Prevent Adhesions . 58
2. Rat Model-Assessment of PVP Solution to Prevent
Adhesions 58
3. Viscosity Molecular Weight* of PVP. External
Polymer Formed in Solution During y-Radiation
Graft Coating 66
4. Contact Angle of Water on y-Radiation Graft
Coating of PVP on PMMA 69
5. Contact Angle of Water for RF Plasma Coatings
on PMMA 70
6. Contact Angle of Water for Various Biomedical
Polymers . . 71
7.Critical Surface Tension from Zisman Plots*
for y-Radiation Graft Coatings of PVP on PMMA 85
8. Summary of Zisman Plots for Critical Surface
Tension of RF Plasma Coatings on PMMA 93
9. ESCA Results 121
10. Summary of Adhesive Force Measurements for
Cornea Endothelium Contact with y-Radiation
Graft Coatings of PVP on PMMA . 139
11. Quantitative Cornea Endothelium Damage by SEM
for Contact with y-Radiation Graft Coatings
of PVP on PMMA 145
12. Summary of Adhesive Force Measurements for
Cornea Endothelium Contact with RF Plasma
Coating on PMMA 148
13. Quantitative Cornea Endothelium Damage by SEM
for RF Plasma Coatings on PMMA 149
vii:
Table
Pase
14. Summary of Adhesive Force Measurements for
Cornea Endothelium Contact with Various
Biomedical Polymers 151
15. Quantitative Cornea Endothelium Damage by SEM
for Various Biomedical Polymers 152
16. Summary of Measurements for y-Radiation
Graft Coatings of PVP on PMMA 156
viii
LIST OF FIGURES
Figure Page
1* Cross Section of Eye and Cornea: (A) Lense,
(B) Anterior Chamber, (C) Posterior Chamber and
(D) Endothelium . 7
2. Scanning Electron Micrograph of Cornea Endothel
ium (X900) 9
3. Scanning Electron Micrograph of Cornea
Endothelium Damaged by Contact to Acrylic Intra
ocular Lens (X2000) 9
4. Predominant Reaction Scheme for y-Radiation
Graft Coatings of PVP on PMMA 15
5. y-Radiation Effects in Solution 16
6. Plasma Polymerization Processes 19
7. Infrared Spectrum of N-VP Monomer (As Distilled). 30
8. Infrared Spectrum of N-VP Monomer (Literature,
Reference 53) 31
9. Schematic Drawing of Co-60 Source, Department
of Radiation Biology, University of Florida ... 33
10. Schematic Drawing of KF Plasma Apparatus .... 37
11. Tissue-Polymer Adhesion Measurement Instrument 45
12. Acrylic Sample in Holder 46
13. Plexiglass Tank (Filled with Saline) with
Tissue Sample 48
14. Loading of Acrylic-Endothelium Interface .... 51
15. Deflection of Glass Fiber due to Acrylic-
Endothelium Adhesion with Lowering of Micrometer
Stage 52
ix
Figure
Page
16. External Polymer Viscosity Molecular Weight vs.
Monomer Concentration for y-Radiation Graft
Coatings of PVP on PMMA 67
17. Contact Angle for Water vs. % Monomer for
0.1 Mrad Dose y-Radiation Graft Coatings of PVP
and PMMA 73
18. Contact Angle for Water vs % Monomer for
. 0.25 Mrad Dose y-Radiation Graft Coatings of PVP
and PMMA . 74
19. Contact Angle for Water vs. % Monomer for
0.5 Mrad Dose y-Radiation Graft Coatings of PVP
and PMMA 75
20. Summary of Contact Angle for Water vs. % Monomer
for Various Dose Levels of y-Radiation Graft
Coatings of PVP on PMMA 76
21. Zisman Plot of Critical Surface Tension (y )
of PMMA ..?... 79
22. Zisman Plot of Critical Surface Tension (y )
of y-Irradiated 0.25 Mrad PMMA ? . 80
23. Zisman Plot for Critical Surface Tension (Y )
of y-Radiation Graft Coating: 5% N-VP Initial
Monomer Concentration and 0.25 Mrad Dose .... 81
24. Zisman Plot for Critical Surface Tension (y )
of y-Radiation Graft Coating: 10% N-VP Initial
Monomer Concentration and 0.25 Mrad Dose .... 82
25. Zisman Plot for Critical. Surf ace Tension (.Y )
of y-Radiation Graft Coating: 20% N-VP Initial
Monomer Concentration and 0.25 Mrad Dose .... 83
26. Zisman Plot for Critical Surface Tension (y )
of y-Radiation Graft Coating: 30% N-VP Initial
Monomer Concentration and 0.25 Mrad Dose .... 84
27. Zisman Plots for Critical Surface Tensions (y )
of 0.25 Mrad y-Radiation Graft Coatings of c
Different N-VP Concentrations 86
28. Summary of Zisman Plots for Critical Surface
Tensions (y ) of PVP on PMMA vs. % N-VP for
y-RadiationCGraft Coatings of 0.25 Mrad Dose . 87
x
Figure
Page
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
Zisman Plot for Critical Surface Tension (y )
of RF Plasma Coating of N-VP on PMMA:
50 Watts(Power)/1000y(Pressure)/10 Minutes
(Duration)
Zisman Plot for Critical Surface Tension (y )
of RF Plasma Coating of N-VP on PMMA: c
100 Watts(Power)/20Oy(Pressure)/20 Minutes
(Duration)
Zisman Plot for Critical Surface Tension (y j
of RF Plasma Coating of N-VP on PMMA:
35 Watts(Power)/500y(Pressure)/60 Minutes
(Duration)
Zisman Plot for Critical Surface Tension (y )
of RF Plasma Coating of HEMA on PMMA:
25 Watts(Power)/500y(Pressure)/15 Minutes
(Duration)
89
90
91
92
Scanning Electron Micrographs of Scraped
y-Radiation Graft Coatings of PVP on PMMA .... 95
Scanning Electron Micrographs of Scraped
RF Plasma Coatings on PMMA 114
ESCA Spectra for PMMA (Uncoated) (B.E. = Bonding
Energy; N/C = Nitrogen to Carbon Atomic Ratio) 123
ESCA Spectra for y-Radiation Graft Coating of
PVP on PMMA: 10% N-VP Initial Monomer Concen
tration and Q.25 Mrad Dose (B.E. = Binding Energy;
N/C = Nitrogen to Carbon Atomic Ratio) 124
ESCA Spectra for y-Radiation Graft Coating of
PVP on PMMA: 10% N-VP Initial Monomer Concen
tration and 0.5 Mrad Dose (B.E. = Binding Energy;
N/C = Nitrogen to Carbon Atomic Ratio) 125
Infrared Spectra for PMMA and PVP (Literature,
Reference 62) 127
Attenuated Total Response-Infrared Spectra of
PMMA, y-Radiation Graft Coating of PVP on PMMA
and PVP Cast Film on PMMA 128
Ultraviolet-Visible Transmission Spectra of y-
Radiation Graft Coatings of PVP on PMMA, 0.1 Mrad
Dose level 130
xi
Figure Page
41. Ultraviolet-Visible Transmission Spectra of
y-Radiation Graft Coatings of PVP on PMMA,
0.25 Mrad Dose Level 131
42. Ultraviolet-Visible Transmission Spectra of
y-Radiation Graft Coatings of PVP on PMMA,
0.5 Mrad Dose 132
43. Ultraviolet-Visible Transmission Spectra of
y-Irradiated PMMA 133
44. Ultraviolet-Visible Transmission Spectra of
RF Plasma Coatings on PMMA 135
45. Adhesive Force vs. Monomer Concentration for
PVP y-Radiation Graft Coatings on PMMA 141
46. Representative Scanning Electron Micrographs
of Cornea Endothelium Damage 142
47. Quantitative Cornea Endothelium Damage by SEM
vs. N-VP Monomer Concentration for 0.25 Mrad
Dose y-Radiation Graft Coatings on PMMA .... 147
48. Schematic View of PVP y-Radiation Graft on
PMMA 155
xii
Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
HYDROPHILIC POLYMER COATINGS TO PREVENT TISSUE ADHESION
By
John Wesley Sheets, Jr.
April 1983
Chairman: Eugene P. Goldberg
Major Department: Materials Science and Engineering
Adhesion following contact between plastic surgical
materials and endothelium surfaces has been shown to result
in extensive and critical tissue damage in intraocular lens
(IOL) insertion and abdominal surgery. Coatings of
hydrophilic polymer solutions were shown effective in
reducing such damage by acting as lubricants and barriers to
adhesion. The primary emphasis of this study was on the
preparation, characterization and evaluation of permanent
graft coatings of hydrophilic polymers for the prevention of
IOL adhesion to the cornea endothelium because of the
importance of this adhesion problem in IOL surgery.
Gamma-radiation and radio-frequency (RF) plasma
polymerized graft coatings of the hydrophilic polymer,
poly(vinyl pyrrolidone) (PVP) were prepared on the IOL
material, poly(methyl methacrylate) (PMMA). The chemical
and physical properties of these coatings were characterized
xiii
by several techniques, including measurement of the contact
angle for water, electron scattering for chemical analysis
(ESCA), infrared spectroscopy (IR), ultraviolet-visible
spectroscopy (UV-Vis), examination by scanning electron
microscopy (SEM) and measurement of the surface energy.
In addition, a new instrument and technique was
developed for the evaluation of the force of adhesion and
the extent of resultant tissue damage of these coatings and
other plastic materials. These measurements were made for
contact with rabbit cornea endothelium under well controlled
conditions.
Thin graft coatings of PVP on PMMA were found to be
adherent to the acrylic substrate and significantly
increased the hydrophilicity and surface energy as compared
with uncoated PMMA. The altered surface properties of the
PVP graft coated PMMA also resulted in reduction of the
adhesive force and extent of tissue damage, as tested by the
newly devised adhesive force instrument.
Results of this study emphasize the need for develop
ment of hydrophilic polymer graft coatings for clinical IOL
use. ^n vivo animal testing of y-radiation graft coatings
of PVP on PMMA are now in progress. The application of the
hydrophilic polmer coating concepts for other surgical
materials and procedures is also under investigation.
xiv
1.
INTRODUCTION
The use of synthetic polymers for prosthetic and
cosmetic implants and for surgical devices has increased
rapidly over the past three decades. The wide range of
advantageous chemical and physical properties possessed by
polymers has shown them to be uniquely suitable for many
medical applications. Some of the most demanding
requirements to be met by polymers are inertness and
biocompatibility in contact with tissues of the body.
Research on the biocompatibility of polymers (and other
materials) has been focused on short and long term toxicity,
tissue acceptance and thrombogenicity. However, very short
term, temporary and transient contacts between materials and
tissue have not been carefully investigated to date. This
study has therefore emphasized the effects of momentary
contacts between polymer and tissue surfaces, and tissue
damage resulting from such contacts.
This area of research was initiated through an
investigation of post-operative complications following
intraocular lens (IOL) insertion. An IOL is a poly(methyl
methacrylate) lens used to replace the natural crystalline
lense following cataract surgery.
1
2
Our preliminary studies showed that contact between an
IOL and the corneal endothelium could result in adhesion of
endothelium cells to the acrylic polymer surface.
Manipulation caused tearing of these adherent cells away
from the tissue surface. Loss of endothelium cells poses a
critical problem since these cells are a non-regenerative
monlayer and serve an essential function in maintaining
fluid balance and clarity of the cornea. Prevention of this
tissue damage is therefore essential to the success of
ophthalmic IOL implantation surgery.
The major focus of this research has therefore been on
acrylic IOL-cornea endothelium surface interactions and
prevention of tissue damage using hydrophilic polymer
coatings. However, the findings in ophthalmic surgery have
also been extended to other surgical procedures. For
example, the possibility of similar types of tissue damage
in abdominal surgery which results in peritoneal adhesions
was questioned. Manipulations within the peritoneal cavity
were shown to result in tissue adherence to latex rubber
gloves used in surgery. Normal healing of the damaged
tissue produces adhesions between adjacent, highly
regenerative tissue surfaces, with the potential for severe
post-operative complications. Other areas of tissue
adhesion to polymer surfaces with resulting damage were
found in vascular catherization and endotrachea tube
intubation, resulting in desquamation or "stripping" of the
3
endothelium cells. The resulting tissue damage may lead to
infections following catherization.
The adhesion phenomenon we have discovered appears to
occur extensively in all types of surgery and poses a
significant but yet unappreciated problem for instrument or
device materials in surgery. The results of this study
suggest that modification of the surfaces of biomedical
devices with hydrophilic polymer coatings can reduce or
prevent adhesive tissue damage for IOLs, catheters and
surgical gloves.
The objectives of this research have been to
investigate the adhesion phenomenon, and its prevention, and
especially to develop practical and clinically viable
permanent hydrophilic coatings for safer IOLs. The critical
nature of damage to the corneal endothelium resulting from
adhesion to an IOL and the large number of IOL insertions
currently performed (over 200,000 annually) underscore the
importance of research to modify the IOL surface.
Both temporary and permanent modification with
hydrophilic polymer coatings have been studied. Although
transient soluble polymer coatings were effective in
reducing tissue damage, we chose to emphasize permanent
hydrophilic graft coatings which would be inherently safer
and easier to use clinically. Gamma-radiation graft
coatings of poly(vinyl pyrrolidone) (PVP) on poly(methyl
methacrylate) (PMMA) were studied as well as hydrophilic
polymer coatings produced by radio-frequency (RF) plasma
4
graft polymerization. Both methods produced cross-linked,
covalently bound graft coatings on the PMMA substrate.
Polymerization conditions were investigated in detail for
y-radiation graft coating.
The chemical and physical nature of these coatings were
characterized. Contact angle measurements were used to
assess the relative hydrophilicity of the surface. Scanning
electron microscopy (SEM) was used to determine surface
morphology. ATR infrared spectroscopy and electron
scattering for chemical analysis (ESCA) were employed to
chemically analyze the surface composition.
A new instrument and technique devised by Dr. S. Reich
was built for the quantitative determination, for the first
time, of the polymer-tissue adhesive force. This method
used the rabbit cornea endothelium as a model endothelium
surface for quantitative measurement of the adhesive force
at the polymer-tissue interface under controlled conditions
of contact. This instrument was also used in conjunction
with SEM to assess the cell damage resulting from polymer
contacts. Adhesive force was correlated with cell damage
and with the chemical and physical nature of the coatings to
understand the adhesion phenomenon and to determine the
coating characteristics necessary to minimize tissue damage.
These studies have shown that the adhesion phenomenon
has broad implications in surgery, although the mechanism is
not fully understood. Hydrophilic polymer coatings for
surgical materials should prove useful to reduce adhesion
5
and tissue damage which may result in post-operative
complications. This study has opened up an interesting new
area of medical materials science that deserves further
study.
2. BACKGROUND
2,1 Intraocular Lens Development and Problems
A cataract is a clouding of the natural crystalline
lense of the eye which necessitates surgical removal of the
lense (Figure 1-A). The optical function of the natural
lense must be replaced if visual acuity is to be restored.
Three alternatives exist for this purpose: spectacle
lenses, contact lenses or intraocular lenses (IOL). An IOL
is placed within the eye, in either the anterior or
posterior chamber, and serves as the optical replacement for
the natural crystalline lense.
The history of the modern intraocular lense began in
1949, when the British surgeon Dr. Harold Ridley implanted a
lens of poly(methyl methacrylate) (PMMA) in an eye following
cataract surgery (1,2). His use of the plastic PMMA had
been suggested by his experiences during the Second World
War, when he found pieces of shattered PMMA aircraft
canopies lodged in the eyes of fliers. He noted that these
fragments did not cause any reaction in the eye and would be
suitable, because of their optical properties, as a
replacement for the natural lense.
The acceptance and use of PMMA as the material for
intraocular lenses grew from that first implantation.
6
3E
Epithelium
Bowman's
Membrane
Stroma
Descemet's
Membrane
Endothelium
. Optic nerve
Figure 1. Cross Section of Eye and Cornea: (A) Lense, (B) Anterior Chamber, (C) Posterior
Chamber and (D) Endothelium.
8
Although other materials have been suggested, PMMA is the
material which is currently used for nearly all IOLs.* The
style or shape of lens, its placement in either the anterior
or posterior chamber (Figure 1-B & C), the surgical
technique for insertion and the method of fixation within
the eye have been varied extensively since the first
implantation, but the optical portion of the lens has
remained as originally fabricated from PMMA.
Studies have shown that after IOL insertion that there
was an average loss of almost half of the central corneal
endothelial cells (3,4); these cells form a monolayer on the
posterior surface of the cornea (Figure 1-D). These cells
serve an essential function in maintaining fluid balance
within the cornea. Damage to the endothelium will result in
leakage of fluid into the corneal layers and will cause
swelling and blurred vision. This monolayer is nonregen-
erative in man so that the damage is irreversible.
Our initial studies (5,6) revealed that this cell loss
was due to contact with.the acrylic IOL surface a't the time
of the insertion surgery (Figures 2 & 3). Others have
confirmed that such contact causes endothelium cell loss
(7,8) and efforts to prevent the contact and damage were
undertaken.
The only exceptions are poly(propylene) support loops and a
commercially available IOL made of glass with poly(imide)
supports (Lunell, NYC).
9
Figure 2. Scanning Electron Micrograph of Cornea Endothelium,
fxqnoi
Figure
3. Scanning Electron Micrograph of Cornea Endothelium
Damaged by Contact to Acrylic Intraocular Lens.
(X2000)
10
When the adhesive damage phenomenon was first observed,
we showed that aqueous solutions of the hydrophilic polymer
poly(vinyl pyrrolidone) (PVP) were effective in preventing
endothelium damage. Since the first publication of our work
on adhesive endothelium damage in 1977 (5,6), research in
many laboratories has resulted in several methods for
protecting the endothelium during IOL insertion.
Surgical techniques have been developed to minimize the
chance for damaging contacts between IOL and endothelium
surfaces. The most widely-used technique employs an air
bubble (9), which is formed and maintained in the anterior
chamber during insertion surgery to keep the endothelium
from collapsing against the lens being inserted. This
technique has been fairly successful but as with any
surgical' technique, it is not fool-proof. Another surgical
technique employed a soft contact lens shield to protect the
endothelium (10).
The other methods developed to protect the endothelium
were adaptations of our initial suggestion to use
hydrophilic polymer coatings. Bovine serum albumin,
chondroitin sulfate, human gamma globulin, hyaluronic acid
and whole plasma are among the different natural polymer
solutions studied as protective agents (11-16). Of these,
the sodium salt of hyaluronic acid has found extensive use,
and is marketed under the trademark, Healon (Pharmacia,
Piscataway, NJ) (14-16). More recently, chondroitin sulfate
solutions have been introduced commercially (Cilco,
11
Huntington, WV) as an alternative to the high-priced and
hard to obtain Healon. However, a major disadvantage of
these solutions is that the surgeon is obligated to use an
additional procedure, application of a protective coating in
the operating room.
Dry, cast coatings of bovine submaxillary mucin (17)
and'poly(vinyl alcohol) (PVA) (18,19) have also been used;
these coatings hydrate and dissolve in situ. While these
transient and soluble coatings have proven effective for
reducing the adhesion and damage from corneal
endothelium-IOL contact, the protection is only temporary.
Damage from contact following the implant surgery, such as
contact occurring by IOL detachment, is not prevented by any
of the temporary coatings. The shortcomings of temporary
protective coatings demonstrate the need to develop
permanently bound coating which can be applied prior to
surgery and provide a protective, hydrophilic surface for
the life of the implant.
2.2 Biological Adhesion
Tissue-materials interface studies to date have
emphasized long-term polymer biocompatibility for
nonthrombogenic and tissue-compatible implants or prostheses
(20). Short-term toxicology and thrombogenicity as well as
long-term tissue acceptance have been major points of
concern (21-25). In the testing recommended for the
evaluation of materials intended for use in contact with
12
blood and other tissues, issued by the U. S. Department of
Health and Human Services, extensive and exhaustive
protocols for the testing of toxicity and thrombogenicity
are detailed (26,27), but no testing of cell adhesion to
materials (other than blood cell adhesion) is recognized or
recommended. The adhesion of cells and tissue to materials
has* been relatively ignored.
However, some research has been devoted to the study of
cell adhesion to surfaces and reviews are available (28,29).
This research has been relatively limited dealing only with
cell adhesion pertinent to tissue culture (30), studies of
surgical adhesives and dental restoratives (31-33),
bacterial adhesion to otic implants (34), marine bacteria
and barnacle adhesion (35), and changes in cell surfaces
accompanying tumor cell metastasis (21). Few studies have
attempted to measure the actual strength of cell adhesion to
surfaces (36).
In these studies, tissue cultures were grown on
different substrates and then were subjected to shear
forces. Although a number of methods have been used, the
most extensive studies have used the shearing force
generated by a metal disc immersed in the culture medium and
lowered to a measured distance from the cells and then
rotated at a known velocity (36) The percentage of cells
of the tissue culture detached by this constant shear force
for a set period of time was measured. For a shear force of
2
7.4 dynes/cm applied for twenty seconds, 27.8% of the
cells from a rat fibroblast culture were detached from a
Pyrex glass substrate (37).
The importance of the forces which may be involved in
cell adhesion to a surface has been reviewed (38). Deemed
most important are chemical (electrostatic, covalent and
hydrogen) bonds and van der Waals1 interactions of the
London type (hydrophobic bonds) (39). Of these factors, the
hydrophobic-hydrophilic nature of the surfaces and the
resultant interactions with cells has received considerable
attention but there is no general agreement whether cells
adhere better to hydrophilic or hydrophobic surfaces.
However, in the examples of muscle cell adhesion to hydrogel
graft-coated surfaces (24) and with the strength of surgical
adhesives to liver tissue (32), an increased hydrophilicity,
relative to other surfaces, was found to favor reduced cell
adhesion.
2.3 Radiation Polymerized Graft Coatings
Permanent, covalently bound coatings of PVP on PMMA
were produced in this study by Y-irradiation of the
substrate immersed in aqueous solutions of the PVP monomer,
N-vinyl pyrrolidone (N-VP). Gamma-Radiation produces free
radicals in both the PMMA and the N-VP monomer, creating
graft sites in PMMA and initiating N-VP polymerization.
Free radicals in the bulk PMMA tend to degrade PMMA,
since scission predominates over crosslinking for PMMA.
However, free radicals produced on the surface of the PMMA
serve as initiation sites for graft polymerization of the
N-VP (Figure 4).
Free radicals in the N-VP are created with opening of
the vinyl group double bond and homopolymerization of the
N-VP. The PVP chain will continue to grow until termination
occurs, and may react with the acrylic substrate, producing
the- graft polymer. The Y-radiation may also cause
crosslinking of the PVP, by hydrogen extraction and
branching of the PVP chain (Figure 5).
Graft polymerization by Y-radiation of PVP on PMMA was
first studied by Henglein et al. in 1958 (40). However,'
these graft coatings were made using the monomer in methanol
following long periods (up to 69 hours) of allowing
diffusion of monomer into the PMMA substrate (with swelling
of the PMMA). This process, along with the high radiation
doses (up to 4 megarads (Mrad)), resulted in severe
distortion of the PMMA (41). While such extreme conditions
would not be suitable for preserving the optical properties
of an acrylic IOL, some penetration of the N-VP into PMMA
may be of interest in future work.
Use of aqueous solutions of N-VP (0-20%) was shown to
reduce diffusion into the substrate when PVP was grafted to
poly(dimethylsiloxane), using high energy (3 MEV) and 1.5 to
4.5 Mrad dose, from a Van de Graff generator (42). These
coatings were shown to be hydrophilic in porportion to the
amount of graft. The irradiation conditions (high doses and
very high dose rates) produced thick coatings, several
CH-
CH-
CH2C
CH.
C = O
C
I
c
VWWWVAAA^
CH.
CH.
C* + *CH.
C = 0
CH-
-CH-
Figure 4. Predominant Reaction Scheme for y-Radiation Graft Coatings of PVP on PMMA.
(1) HOMOPOLYMERIZATION
(2) BRANCHING & CROSSLINKING
(1)
ch2 ch2
CH, C = 0
V
I
ch2=ch
(NVP)
ch2 ch,
I I
CH, C= O
V
vwwvww^*- CH2 CH
N-VP
CH,
CH- CH,
I I
CH, C = O
V
i
CH
CH,CH,
I I
CH,
CH,
CH-
C= O
PVP Chain =
(2)
CH, -CH,
ch2 c=o
ch2ch.
vwww^.
CH, CH,
I I
CH, C O
V
ch2ch
+H-
N-VP
CH, CH,
CH, C = 0
V
"CH2 C
ch2-ch, ch2
xn-ch
/
CH,C
PVP
ch2 ch2
CH, C = O
V
-CH,
CH
CH,
ch2 ch2
CH, C=Q
V
CH-
Figure 5. y-Radiation Effects in Solution.
17
mils thick, but low molecular weight graft polymers (less
than 100,000) were found in the coatings.
PVP has also been grafted onto
poly(tetrafluoroethylene) (PTFE) using Y-radiation (43).
Here, the grafting conditions produced a heavy coating (over
70% weight gain for a 50 micron film) and hydrophilic
modification of the surface. Biological testing of the
graft coated surface showed a decrease in the adsorption of
fibrinogen and immunoglobulins from blood serum, but an
increase in albumin absorption.
Extensive research has been conducted by Ratner and
Hoffman on the use of Y-radiation to graft PVP and other
hydrophilic polymers onto silicone rubber (24). Aqueous
solutions of N-VP (20%) were irradiated to 0.25 Mrad using a
20,000 Curie Co-60 source and heavy graft coatings
2
(1.5 mg/cm ) of PVP were produced. These studies have also
shown the effect of low concentrations (0.005 Molar) of
cupric ion to minimize gelation in the external homopolymer
(44). Biological testing of these surfaces showed a
reduction in the adhesion of chick embryo muscle cells. The
desirability of this low cell adhesion and the aforemen
tioned reduction in protein adsorption has prompted studies
with radiation grafted hydrogels for a broad range of
biomedical applications (45).
The strategy for our use of y-radiation graft coatings
was to use low radiation doses (<0.5 Mrad) and low N-VP
monomer concentrations in order to minimize degradation of
the PMMA substrate and crosslinking of the PVP. Swelling
and distortion of the substrate was also minimized by the
use of the low monomer concentrations in aqueous solution.
Diffusion of monomer into the acrylic was limited by
immersion of the substrate in the polymerization medium only
at the time of irradiation. The goal was production of very
thin, uniform adherent PVP coatings on the acrylic
substrate.
2.4 Plasma Polymerized Graft Coatings
In addition to Y-radiation grafting, plasma polymerized
coatings were studied. A plasma is an excited low pressure
gas created by radio-frequency (RF) discharge or a direct
current (DC) electrical glow discharge. The plasma produces
ions, electrons, ion radicals and other excited species
which will readily polymerize. Yasuda has classified the
polymerization reactions of monomers in a plasma into two
categories: plasma-induced polymerization and plasma state
polymerization (Figure 6) (46).
In a plasma-induced polymerization, the vapor of a
monomer is excited to the form in which it would polymerize
by conventional polymerization, maintaining its original
structure. For example, the plasma serves to open the
double bond of the vinyl group in N-VP and promote
conventional radical polymerization.
Plasma state polymerization does not follow
conventional polymerization kinetics and relies on the
(1) Plasma Induced Polymerization
ch2 ch2
CH, C =0
V
ch2=ch
(N-VP)
Plasma
CH,CH,
I I
CH, C = O
V
CH, CH
CH,CH,
I
CH
, C = 0
^ /
N
CH2= CH
CH, CH,
I I
CH, C = 0
V
CH,
ch2ch2
CH, C =
V
CH
CH,
CH-
(2) Plasma State Polymerization
(Possible Reaction)
ch2 CH2
CH, C=0
V
ch2=ch
Plasma
CH= CH
ch2ch2-n:
c=o
CH, CH,
I I
CH, C = 0
V
CH=CH
CH2 ch2
ch2 C=0
\ /
N
CH2=CH CH2CH *
CH, CH,
Figure 6. Plasma Polymerization Processes
highly excited state of the plasma. Here, the plasma
creates reactive species which may be only fragments of the
original monomer. These fragments then react to form a
polymer. The polymer may be formed from molecules which
would not normally react or polymerize. Polymers formed by
these reactions are usually highly cross-linked and are
difficult to characterize because of their complex
structures.
There are a number of advantages to the use of plasma
polymerization in the production of polymer coatings. Most
important is the ability to produce thin and uniform
coatings. Moreover, the coatings can be produced on a wide
range of substrates. By varying the coating conditions,
coatings with a wide range of properties can be produced
from a single starting material.
A large body of literature exists about the use of
plasma polymerization for the production of coatings (see
review by Shen and Bell) (47), and a number of studies are
O
relevant to this study. Thin (less than 1000A) coatings
have been produced on contact lenses by RF plasma
polymerization of a mixture of acetylene, water vapor and
nitrogen (48). The surfaces produced were shown to be
hydrophilic (contact angle for water at 38) and reduced
both protein adsorption and cell adhesion.
In other biomedical applications, RF plasma polymerized
films of poly(ethylene), poly(styrene) and poly(chloro-
trifluoroethylene) have been shown to produce no change from
uncoated silicone rubber in biocompatibility studies done
with intramuscular implants of up to two years (49). Other
studies have shown that RF plasma polymerized films of
hexamethylcyclotrisiloxane on poly(propylene) and silicone
rubber membranes reduce the adhesion of platelets and
leucocytes in in vitro canine testing (50).
These different examples of the use of RF plasma
polymerized films to coat a variety of different materials
for a range of biomedical applications demonstrate the
usefulness of this technique in producing thin and uniform
coatings for biomedical applications. In this study, we
have shown that RF plasma grafting may be an alternative
method for producing hydrophilic polymer coatings on
intraocular lenses.
3. MATERIALS AND METHODS
3.1 Preliminary Studies
Although the emphasis of this research has been on the
development and characterization of permanent hydrophilic
polymer graft coatings for intraocular lenses (IOLs),
preliminary experiments were conducted which first
demonstrated the tissue-materials adhesion phenomenon and
the effectiveness of hydrophilic polymer solutions to
prevent tissue damage. The two most significant experiments
are describe here and elsewhere (5,6,51); the remainder of
Part 3, deals with the subsequent research based on findings
of these initial experiments.
3,1.1 Prevention of Cornea Endothelium Damage by
Poly(vinyl pyrrolidone) Solutions
In these experiments, cornea endothelium from rabbit
and human eye bank eyes were contacted to acrylic
hemispheres to simulate IOL insertion (5). These
hemispheres were machined and highly polished to a radius of
curvature of 7.3 mm; their visually smooth surface and
curvature facilitated total contact between the acrylic and
endothelium surfaces.
22
23
To obtain the tissue samples, adult New Zealand albino
rabbits were sacrificed by an intravenous overdose of sodium
pentobarbital and the eyes were enucleated immediately upon
death. The corneas were removed with a 3 mm rim of sclera
and the lense-iris diaphragm was peeled from the cornea.
The corneas were placed endothelium-side up on a teflon
blo'ck and the central areas were punched out with a 7 mm
trephine. An identical procedure was employed in isolating
7 mm cornea buttons from fresh human eyes. Only paired (two
eyes from the same donor) human eyes were used, so that one
could serve as a control for the other. The corneas were
then placed on the acrylic surfaces, described as follows.
Endothelium damage was determined by either optical
microscopy with nitro-blue tetrazolium (NBT) staining or
scanning electron microscopy (SEM).
The cornea buttons were contacted to acrylic hemi
spheres which had been dipped in either a balanced salt
solution or a 40 weight % solution of poly(vinyl
pyrrolidone) (PVP) (GAF Corporation, New York; PVP K 29-32,
40,000 MW) in a balanced salt solution. The acrylic
hemisphere was held upright following dipping and the cornea
button was placed, endothelium side down, on the acrylic
surface. Contact time was varied from 1 to 60 seconds.
NBT staining (52) was performed by incubating the
corneas, following contact to an acrylic surface, at 37C
for 15 minutes with a drop of NBT stock solution and a drop
of reduced diphosphopyridine (DPNH) (0.3 mg/ml in water) on
the endothelium surface. Endothelium damage was then
estimated by observation with low-power optical microscopy.
Corneas were prepared for SEM by fixation in 2.5% cold
glutaraldehyde for one hour and osmium tetroxide for
90 minutes, both prepared with Millionig buffer. Corneas
were then dehydrated in serial baths of increasing
concentration of ethanol in water solutions. The samples
were then critical point dried using a Bomar SPC-900/EX
critical point apparatus (Bomar, Tacoma, WA) and then coated
with palladium-gold. Samples were then viewed with a Zeiss
Novoscan 30 scanning electron microscope (Carl Zeiss, Inc.,
New York).
3.1.2 Peritoneal Adhesions in Abdominal Surgery
The experimental plan was to perform exploratory
manipulations within the peritoneal cavity on rat and canine
animal models under simulated operating room conditions in
which the surgeon performs a clean, noncontaminating surgery
and yet obtains massive intra-abdominal adhesions, a major
post-operative complication (51). Such adhesions are
defined as the fibrous collagenous connective tissue that
develops after serosal trauma; in severe cases, these
adhesions can result in mechanical constriction of bowel
function, with potentially fatal consequences. Our intent
was to compare the conventional clinical technique with
procedures wherein hydrophilic polymer solutions of PVP and
dextran were applied to tissue surfaces, surgical gloves and
sponges prior to contact with internal organs and tissue.
Female mongrel dogs (12-20 kg) were anesthesized
initially with thiamylal sodium (Surital, Parke-Davis,
Morris Plains, NJ) (20 mg; 0.5 cc/kg of dogs weight) and
maintained on halothane with controlled ventilation
following endotracheal tube insertion. Abdomens were
prepped with a PVP-iodine (Betadine, Purdue Frederick,
Norwalk, CT) solution and draped for a sterile
intra-abdominal procedure. A midline incision was made with
a cold knife and bleeding vessels were electrocoagulated.
Dogs were divided into the following 4 groups of 4 dogs
each, and all were subjected to a complete exteriorized
bowel exploration as follows:
I)Organs manipulated with dry gloves and dry sponges.
II)Manipulation with dry gloves but using saline-wet
sponges and organs wet with sterile saline.
Ill)Gloves and sponges wet with 25 weight % soluble
dextran (Sigma, St. Louis; 200,000-300,000 MW) in
saline solution. Immediately after opening, 50 ml
dextran solution was sprayed on exposed organs
before manipulation. Just before closure 50 ml
dextran solution was again sprayed on organs.
IV)Same as III using 25 weight % PVP (GAF, New York;
40,000 MW) in saline.
A standardized procedure for organ manipulation was
used for each dog involving palpation of all organs,
26
exteriorization of bowel, bimanual palpation, replacement of
bowel, and the abdominal wall was closed in layers.
Experimental animals were housed separately and fed routine
diets. Surviving animals were sacrificed and examined
8 days after initial surgery and adhesions of the wound and
small intestine graded on a numerical scale based upon
degree and extent of adhesion bands.'
Additional experiments using rats were conducted for
futher evaluation of PVP solution. Twenty female
Sprague-Dawley rats (185-200 g) were anesthetized with
pentobarbital and divided into 2 groups. For the control
group of 10, the abdominal wall was opened and all
intra-abdominal organs bimanually examined. Exteriorization
was necessary to simulate clinical conditions for
exploratory surgeries. Abdominal contents were exposed,
gently explored and manipulated using dry gloves and sponges
for 5-10 minutes, the abdominal wall was closed in layers,
and the incision sutured. For the second group, the
procedure was repeated except that immediately after the
abdominal wall incision, 5 ml of 25 weight % PVP solution in
saline was sprayed into the abdominal cavity and all
manipulations were performed using gloves and sponges wet
with the PVP solution. Just before closure, an additional
5 ml of PVP solution was sprayed over the exposed organs.
Rats were maintained in individual cages for observation.
Eight days after surgery surviving rats were sacrificed and
intestines and organs examined for adhesions. The presence
27
of intra-abdominal adhesions was evaluated using a numerical
scale similar to that for the dog experiments.
3.2 PMMA Substrates
Acrylic test piece of several geometries were required
for the various methods of testing studied. Two test pieces
used most were "stubs" from 1/8" diameter rod, lathe cut to
1/8" length (PMMA from Modern Plastics, Orlando, Florida;
140,000 MW) and "slabs" (1/2" x 7/8" x 1/8"), saw cut from
Perspex CQ sheet (Medical grade PMMA from ICI, England;
2,400,000 MW). The stubs were used for characterization of
coatings by contact angle (for water), electron scattering
for chemical analysis (ESCA), examination by SEM, and
biophysical testing of adhesive force and SEM assessment of
damage from contact to cornea endothelium tissue. The slabs
were used in measurement of critical surface energy and
ultraviolet-visible (UV-Vis) and infrared (IR) spectroscopy.
Following lathe cutting to 1/8" lengths, the stubs were
individually end-polished in two steps. Coarse polishing
was done with water-wet 600 grit silica carbide paper
(Carbimet, Buehler Ltd., Evanston, ILL). A final dry
polish, on a napped cloth wheel (Microcloth, Buehler Ltd.)
using 40 y diamond paste (Astro-Met, GCA-Precision
Scientific, Chicago) was performed until visual inspection
revealed no scratches or surface irregularities. The
polishing was followed by an ultrasonic cleaning procedure:
five minutes in a soap water bath and then two five-minute
rinses in distilled water. Examination of the polished
surfaces by SEM was performed on random samples to assure
smooth surfaces.
The slabs were cut from medical grade acrylic sheet and
were rinsed with distilled water before use.
' 3.3 Purification of Monomer (N-VP)
The N-vinyl pyrrolidone (N-VP) monomer was obtained
from Eastman Kodak Chemicals (Rochester, NY) and contained
0.1% NaOH as an inhibitor to prevent polymerization. The
monomer was purified by distillation under vacuum prior to
use to remove this additive and any other impurities.
Vacuum distillation was performed in a batch process; a
1000 ml Erlenmeyer flask with a teflon-coated magnetic
stir-bar was charged with approximately 500 ml of the
as-received N-VP. The flask was heated by an electric
mantle and supported by a magnetic stirrer. The flask was
connected by a Claisen tube with a 6" high sidearm (which
held a thermometer to register distillation temperature) to
a water cooled Leibig condenser (40 cm jacket length). This
condenser was lined by a connection joint to a 500 ml
Erlenmeyer receiving flask in an ice bath and to the vacuum
source. Vacuum was supplied by a mechanical pump (Model
D-150, Precision Scientific, Chicago) via a liquid nitrogen
cold trap. The vacuum attained was approximately
100 micron.
Distillation was conducted until a stable,
constant-boiling temperature was achieved typically at
66-80C depending on the vacuum achieved. The first 100 ml
of distillate collected after this point was discarded,
along with any previous distillate. The next 250 ml of
distillate was collected and preserved for use. The
remaining distillant was discarded and the system was
recharged for continued distillation following the same
procedure.
Purity of the distillate was checked by index of
refraction and infared spectroscopy. The index of
refraction was measured by an Abbe-type refractometer
(accuracy 0.0001; Fisher Scientific, Pittsburgh, PA) and
was found in all cases to be within 0.0001 of the literature
value of n^5 = 1.5120 (53).
Infrared spectra of the distilled N-VP were recorded
with a Perkin-Elmer Model 283B infrared spectrophotometer
under standard operating conditions (Figure 7). These
spectra matched the spectrum for N-vinyl pyrrolidone found
in the literature (Figure 7) (53).
The monomer was stored in sealed flasks at < 0C in a
freezer until use.
3.4 y-Radiation Graft Coatings
Gamma-Radiation graft polymerization coatings were
prepared by irradiation of the PMMA substrate in an aqueous
solution of the N-VP monomer with y-radiation from a 600
3500
200
Wavenumber (cm'1)
3000 2500 2000 1800 1600 1400 1200 1000 800 600 400
Wavelength in microns
Figure 7. Infrared Spectrum of N-VP Monomer (As Distilled).
Wavenumber (cm*1)
4000 3000 2500 2000 150014001300 1200 1100 1000 950 900 850 800 750 700 650
Figure 8. Infrared Spectrum of N-VP Monomer (Literature, Reference 53).
32
Curie Co-60 source (Department of Radiation Biology, Univer
sity of Florida) (Figure 9). Radiation doses used were from
0.1 to 0.5 megarads (Mrad), as measured by Fricke dosimetry,
(54) performed by the Department of Radiation Biology. To
minimize the irradiation times, and also maximize the number
of samples irradiated, the irradiations were usually made at
a distance of two inches from the source. The dose rate at
this distance was 1360 rads per minute.
Solutions of N-VP in water were prepared volumetrically
using freshly distilled water and distilled monomer (pre
pared as previously described). The solutions were prepared
immediately prior to irradiation and were not degassed.
Different irradiation containers were used for the two
kinds of PMMA substrates which were coated. The short rod
"stubs" were held in a 1/8" teflon sheet which fit in the
slots of a Coplin staining jar (Volume 90 ml; Fisher
Scientific, Pittsburgh); the "stubs" were held such that
both ends were exposed to the polymerization media. The
rectangular PMMA "slabs" were placed for coating in
borosilicate glass test tubes (16 x 125 mm, volume ^ 18 ml)
(Fisher Scientific, Pittsburgh). The test tubes were then
placed in a specially constructed carousel which held the
tubes upright at a distance of two inches from the source
during the irradiation. The Coplin jars were free-standing
within the Co-60 source and were placed so that the source
would be two inches from the PMMA "stubs." In all cases,
Figure 9. Schematic Drawing of Co-60 Source, Department of Radiation Biology,
University of Florida.
34
the PMMA substrates were not placed in the irradiation
containers (filled with the polymerization media) until
immediately prior to irradiation.
Following irradiation, the PMMA substrates were removed
from the irradiation containers and holders and immersed
immediately in distilled water. The wash water was changed
several times until all external homopolymer was removed.
The completion of the cleaning procedure was checked by
allowing the samples to air-dry and then inspecting for
debris of external homopolymer. Cleaning by washing and
soaking in distilled water was continued until all
homopolymer had been removed. The cleaned samples were
stored until use in polyethylene beakers filled with
distilled water. The homopolymer and solution remaining
following irradiation were removed from the irradiation
container and saved in capped, sealed amber glass vials.
3.5 Intrinsic Viscosity Molecular Weight of PVP
The molecular weight of the PVP homopolymer formed in a
solution by Y-radiation induced polymerization was
calculated by measurement of intrinsic viscosity and use of
the Mark-Houwink-Sakurada equation. The polymer was
obtained by precipitation of polymer formed under conditions
identical to those used for graft coating of PMMA, but
without the acrylic substrate present.
Solutions of distilled N-VP (5-30%, distilled as
previously described) and freshly distilled water were
35
placed in test tubes (16 x 125 ram; volume 18 ml) sealed with
aluminum foil and rubber stoppers. No attempt was made to
de-gas the solutions or containers. These test tubes were
irradiated at the dose levels studied (0.1 to 0.5 Mrad)
while held in a specially constructed carousel.
The homopolymer of PVP formed for each set of
conditions was precipitated with 100 ml of acetone in a
small glass container (volume 500 ml) on a blender (Waring,
New Hartford, CN) at high speed. The precipitated polymer
was washed twice with 150 ml of acetone in the blender at
high speed. The polymer was removed from the blender and
loosely sealed in aluminum foil packets. The polymer was
air dried for twelve hours and then dried overnight in a
vacuum oven at 50C and less than 150 torr.
Using the method of the Appendix, solutions for a
viscometry were prepared with methanol as the solvent, and
were filtered twice through Gelman spectroglass filters
(#934-AH, Gelman Instrument Co., Ann Arbor, MI). An
Ubbelholde viscometer, size OB (Fisher Scientific,
Pittsburgh, PA) was used for viscosity measurement, at 30C,
(0.1C) maintained in a water bath. Four dilutions were
used to obtain a In nrel/c vs. concentration plot. The
intercept at zero concentration was found via a best-fit
linear plot with a HP-35c calculator (Hewlett Packard,
Corvallis, OR). Coefficients* for the Mark-Houwink-Sakurada
*These values, for the equation [n] = KMa, were K = 23 x
10J ml/g and a = 0.65.
36
equation were taken from the Polymer Handbook (55) The
molecular weight of PVP was determined for all radiation
graft conditions that produced a soluble homopolymer.
3.6 RF Plasma Coatings
Modification of the PMMA substrate was performed by
graft coating with hydrophilic polymers formed by
radio-frequency (RF) plasma polymerization of N-VP and
2-hydroxyethyl methacrylate (HEMA). A vertical, "bell jar"
reaction chamber was constructed for this procedure; the
schematic diagram shows the apparatus and its set-up
(Figure 10). The plasma was created by induction of a
13.56 MHz signal of a 100 watt RF generator (Tegal,
Richmond, CA) through a ten-turn coil of copper tubing. The
output load was adjusted for optimal conditions and the
power level was controlled by a matching network (Tegal) and
a SWR meter (Heath, Benton Harbor, MI).
PMMA samples (both "stubs" and "slabs") to be coated
were supported on a stand 3.5 inches directly below the
plasma generation region (the plasma was generated in the
tube surrounded by the copper tubing coil); the samples were
secured with double-sided adhesive tape. The chamber was
evacuated to 100 micron pressure by a mechanical pump (Model
D-150, Precision Scientific, Chicago) connected with a
liquid nitrogen cold trap. The vacuum was monitored by a
thermocouple vacuum gauge (Series 270, Granville Phillips,
Figure 10. Schematic Drawing of RE Plasma Apparatus.
To Lina
100 Watt RF
Generator
38
Boulder, CO), and maintained until residual water vapor was
eliminated from the system.
Monomer vapor from monomer held in a 20 ml long-necked
Erlenmeyer flask was then bled into the system and adjusted
by a micro-metering valve (SS-22RS4, Whitey Co., Highland
Heights, OH) until the desired pressure was achieved. The
monomers, N-VP (Eastman Kodak Chemicals, Rochester, NY) and
HEMA (Aldrich Chemical Co., Milwaukee, WI) were used as
received. The monomer vapor available was sufficiently
pure, and the monomers, as received, contained inhibitors
which prevented polymerization when the monomer was heated
(by a hot air gun) to achieve the desired vapor pressures.
After the desired pressure was achieved, the RF power
was switched on and adjusted to the selected power level, as
monitored by the SWR meter. Monomer vapor pressure and RF
power were maintained at constant levels over the duration
of the plasma polymerization and deposition of coating.
Following the completion of the coating procedure, the
chamber was opened to the atmosphere, and the samples were
cleaned by rinsing with distilled water. The samples were
stored with the side toward the plasma generation region
facing up; testing and characterization was performed only
on this face of the sample.
3.7 Contact Angle for Water
Contact angle measurements were made to assess the
relative water wettability of the different materials' and
39
coatings' surfaces. These measurements were made with a NRL
contact angle goniometer (Rame'-Hart, Mountain Lakes, NJ) at
ambient temperature and humidity (usually % 23C and 60%
relative humidity).
The Y-radiation and RF plasma graft coated and uncoated
acrylic samples were equilibrated overnight in distilled
water; water droplets adhering to the surface were shaken
off prior to testing. All other materials tested were used
as received.
Using a Gilmont micrometer syringe (Gilmont, Great
Neck, NY), a 0.002 ml drop of freshly distilled water was
deposited on the plastic surface. The contact angle on each
side of the drop was measured to the nearest degree and
recorded. The surface was then dried by touching a small
piece of filter paper to the periphery of the drop; any
remaining water was blown from the surface with compressed
gas (Manostat, NYC). Five drops were used on each sample,
yielding ten measurements of the contact angle.
3.8 Critical Surface Tension
The critical surface tensions of both the y-radiation
graft coatings and the RF plasma coatings were determined by
the construction of a Zisman plot of the cosine of the
contact angle for liquids of different surface tensions
versus the surface tensions of the liquids (56). The
intercept at cosine of the contact angle equal to one (as
determined with a best-fit, linear plot, using a HP-35c
40
calculator (Hewlett Packard, Corvallis, OR)) was the
critical surface tension for that surface.
The liquids used were selected from a wide range of
liquids used for such measurements (57). The liquids used
were chosen on the basis of their being non-solvents for
both PMMA and PVP and their non-spreading behavior
(formation of a finite contact angle) on both polymers'
surfaces. The value of the surface tensions for these
liquids was taken as the value for 20C, although the
measurements were made at ambient conditions. This was
possible since the change in surface tension with
temperature for organic liquids is only ^0.1 dynes/cm/C.
The five liquids used and their surface tensions are water
(72.6 dynes/cm), glycerol (63.4 dynes/cm), formamide
(58.2 dynes/cm), 2.2'-thiodiethanol (54.0 dynes/cm), and
2
methylene iodide (58.2 dynes/cm).
The larger surface area acrylic slabs were used for
these measurements; the sample was stored at ambient
conditions prior to testing. The contact angle measurement
for each liquid was made according to the same procedure as
for the contact angle for water measurements, except two
drops, yielding four angles for measurement, were used for
each liquid. A fresh, clean area of the sample surface was
used for each drop.
Following the measurements of these angles for the five
liquids, the average for each liquid was calculated. These
values, along with the surface tensions of the five liquids
(above), were used in the construction of a Zisman plot for
each surface tested.
3.9 Scanning Electron Microscopy
Scanning electron microscopy was used to observe the
surface morphology of the different coatings and their
thickness. Coated stubs were scraped with a "chisel" made
from the same PMMA rod stock as the coated substrate. These
specimens were then prepared for microscopy by fixing the
sample to aluminum stubs with colloidal carbon paint
(Structure Probe, Inc., West Chester, PA) to assure mounting
O
and to provide an electrical ground. Gold-palladium (200 A)
was deposited on the surface by a Hummer V sputter coater
(Technics, Alexandria, VA).
Following these preparations, the specimens were
examined with a JEOL JSM-35c SEM (JEOL, Boston, MA). The
scratch made by the PMMA "chisel" was photographed at 500X
and 2000X to show both the appearance of the coating and to
reveal the coating thickness. For these examinations, the
SEM conditions were 15 kV accelerating voltage, condenser at
the 12 o'clock position and the sample held at 0 tilt.
Photographs of the scraped area were taken at 500X, 2000X
and 5000X magnifications to reveal both surface morphology
and the coating thickness.
42
3.10 ESCA (Electron Scattering for Chemical Analysis)
Samples for ESCA were sent to Dr. Christopher Batich*
of Dupont Central Research (Wilmington, Delaware) for
analysis. These samples were mounted to copper wire to
eliminate any background interference in the spectra. A
Kratos ES-300 (Kratos, Manchester, England) ESCA
spectrometer was used, with aluminum (Kot) as the x-ray
source. Data for each peak were collected for twenty
minutes.
Analysis of the raw data was performed by Dr. Batich.
Peak heights and areas were determined for C, N, O, Cl, Cu
and Si. The ratio relative to carbon was then calculated
for each element.
3.11 Infrared Spectroscopy
Surface analysis of the coated surfaces was made by
infrared spectroscopy using a Perkin-Elmer Model 283B
spectrometer (Perkin-Elmer, Stamford, CN). The analysis was
made on the sample surfaces by use of an attentuated total
reflectance (ATR) accessory (No. 185-0382, Perkin-Elmer).
Maximum sensitivity was obtained using KRS-5 crystals as the
prism elements at angles of 60 and 45. Samples, in the
form of coated acrylic "slabs," were placed on both sides of
the element to maximize the spectra's strength. The
greatest slit width available was used, along with a
*Present address, Dept, of Materials Science, University of
Florida.
43
reference beam attenuator, to compensate for the loss of
signal due to the ATR accessory.
Coated samples were air-dried and stored at ambient
conditions before spectra were obtained. For comparison of
spectra for these graft coated samples, cast films of PVP
were made and analyzed by ATR-IR. These films were made
from 5 and 10% solution of PVP (MW 40,000) (GAF, NYC) in
methanol with a doctor blade. These films were cast on the
acrylic "slabs" and were dried in a vacuum oven at 50C and
less than 150 torr prior to testing. The spectra were
analyzed as above, and were used to aid in resolving
spectral peaks and analysis of the spectra for the graft
coated samples.
3.12 Ultraviolet-Visible Spectroscopy
The ultraviolet-visible (UV-Vis) spectra of the PMMA
and the coated samples were taken from 900 to 190 nanometers
with a Model 552 Perkins-Elmer spectrophotometer
(Perkin-Elmer, Stanford, CN). In addition, the effect of
y-radiation alone was analyzed spectroscopically at
radiation doses of 0.25, 2.5 and 5.0 Mrad.
3.13 PVP-Iodine Interaction
Samples of PMMA and y-radiation graft coated PMMA were
equilibrated overnight in saturated solutions of iodine in
water. Both UV-Vis and ATR-IR spectra were obtained, as
previously described, to use the complexing of the iodine by
44
PVP as a measure of the amount of PVP present in the
coating.
3.14 Biophysical Measurements
*
3.14.1 Instrument for Biophysical Measurements
The instrument was constructed from a design developed
at the Weizmann Institute, Rehovot, Israel by visiting
Professor Dr. Shimon Reich. The instrument functioned to
allow both the measurement of adhesive force and to provide
quantification of damage from contact of endothelium to a
material's surface.
The instrument consisted of two parts: a mechanism
mounted on a scavenged microscope and a measuring microscope
with a micrometer-controlled crosshair, mounted in its
ocular. Figure 11 shows the mechanism mounted on the
scavenged microscope. A detailed description of the
instrument and its operation follows.
3.14.2 Preparation of the Material Sample for Measurement
Three forms of materials (coated and uncoated acrylic,
and other polymers) were tested, but the holder design
required a sample in the form of a 1/8" x 1/8" rod or
"stub." Acrylic stubs were prepared from rod as discussed
previously and were friction-fit into a conical holder at
the point of the cone (Figure 12). Materials not in
suitable rod form were cut from sheet into 1/8" discs and
Adjusted Support
Stage for
Tripod Weight
Tripod Weight
Glass Fiber
Material Holder
Transparent
Plexiglass
Tank
Cornea
Endothelium
Microscope
Stage
Water Immersion Test Cell
Cornea Endothelium
Polymer Sample
Rack and Pinion
Micrometer Staqe
(for lowering and raising
tissue sample tank)
Lateral View of Section (A)
X-Y Micrometer Stage
(for centering of material
sample over endothelium)
Microscope Base
Tissue-Polymer Adhesion Measurement Instrument.
Figure 11
Ln
47
attached with cyanoacrylate glue (Loctite, Cleveland, OH) to
acrylic stubs for use.
The holder was suspended, cone pointing down, from a
drawn glass fiber by a fine gold chain. Above this holder a
tripod weight of 17 grams was freely held, such that the
flat base of the cone would support the weight when contact
with the tissue sample was made. Therefore, contact would
be made while this weight was loading the interface with a
known pressure.
The prepared tissue sample was held on the stage of the
microscope in a acrylic tank filled with physiological
saline (0.8%) (Figure 13). A micrometer stage allowed fine
adjustment of the vertical position of acrylic tank and the
tissue sample enclosed. An X-Y micrometer stage allowed
precise centering of the tissue beneath the conical stub
holder. By filling the acrylic tank with saline, surface
tension effects in adhesion were eliminated and tissue
freshness was maintained during the measurements.
3.14.3 Preparation of Tissue Samples
Two different forms of tissue samples were used in the
measurements of adhesive force and for damage
quantification. However, the type and source of the tissue
was the same: Cornea endothelium obtained from adult male
New Zealand albino rabbits. The rabbits were sacrificed by
an intravenous overdose of pentobarbital and the eyes were
Figure 13. Plexiglass Tank (filled with Saline) with
Tissue Sample.
49
then enucleated. The eyes were used less than two hours
postmortem.
For tissue damage measurements, the corneas were
removed from the eyes and placed in saline. Then an 8 mm
corneal trephine (Storz, St. Louis, MO) was used to punch
out a disc from the central portion of the cornea. This
cornea "button" was then glued, endothelial side up, to an
aluminum SEM stub with cyanoacrylate glue. A flat, uniform
surface of endothelium was formed that would allow precise
contact with a material surface. This sample was then
positioned and held within the saline-filled acrylic tank
for testing.
In measurement of the adhesive force, a sample holder
was constructed so that multiple (up to six) measurements
could be made from a single cornea. Corneas were removed
from the enucleated rabbit eyes and trimmed to leave a
narrow (1 mm) scleral rim. The corneas were then inverted
and placed on a supporting plastic hemisphere of matching
curvature. The cornea was held at the limbus with a metal
collar, secured with a screw plate. In this manner, a
"corneal hemisphere" was created; by holding the hemisphere
tilted within the acrylic tank, rotation of the hemisphere
on its axis would provide multiple contact points along a
latitudinal line. Contact was made normal to the surface by
the test piece held above.
50
3.14.4 Adhesive Force Measurement
Operation of the instrument to measure adhesive force
was made after the freshly prepared tissue sample was
positioned within the acrylic tank and it was placed on the
micrometer-controlled stage. The material test piece was
friction-fit into its holder and suspended above the tissue
tank.
The rack and pinion was used to lower the test piece to
within a few millimeters of the tissue surface. At this
point, the crosshair of the measuring microscope was zeroed
and aligned with a reference point of the glass fiber. This
was the resting position of the glass fiber, which was
stressed only by the weight of the test piece and holder.
Contact between the test piece and the tissue sample was
then made.
The stage supporting the acrylic tank and the tissue
sample was slowly raised by operation of its micrometer
control. As the tissue sample contacted the test piece
surface, the elevation was continued until the tripod weight
was supported by the test piece holder. This loading was
maintained for thirty seconds.
The micrometer-controlled stage was then lowered
slowly. Adhesion between the two surfaces caused the test
piece to accompany the tissue in its downward travel. This
movement was simultaneously tracked by the micrometer
crosshairs of the measuring microscope (Figure 15).
51
Figure 14. Loading of Acrylic-Endothelium Interface.
52
Figure 15. Deflection of Glass Fiber due to Acrylic-
Endothelium Adhesion with Lowering of
Micrometer Stage.
53
This crosshair monitored the deflection of the glass
fiber until the adhesive force at the tissue-material
interfaced was exceeded by the flexural stresses in the
glass fiber. When that occured, the crosshair micrometer
was read and recorded, since the interface was broken and
two surfaces had parted.
The "corneal hemisphere was then rotated approximately
60 to bring a fresh area of tissue into position for
measurement. The process was then repeated and continued
until all six positions on the tissue were utilized for
measurements.
The reading recorded during these measurements was
converted into force measurements by a calibration curve
prepared for the glass fiber in use. Such curves were
prepared for each glass fiber; the deflection of each fiber
was measured with known weights hung from the fiber. The
weights were hung on the glass fiber at the same point at
which the gold chain (which suspended the test pieces
experimentally) was located. Then, just as in the procedure
for the measurement of the adhesive force, the micrometer
crosshair of the measuring microscope was used to measure
the deflection for that weight.
These values yielded a linear plot of force versus
deflection. The values for deflection in the adhesive force
measurements were converted into force by use of linear best
fit program and a HP-35c programmable calculator
(Hewlett-Packard, Corvalis, OR). Then, assuming complete
54
2
contact over the entire 0.079 cm of the test pieces, the
2
adhesive force in terms of milligrams/cm was calculated.
3.14.5 Quantitative Cornea Endothelium Damage by SEM
The procedure for measuring the tissue damage resulting
from a material contact began by contacting the material
sample with the endothelium just as iri the measurement of
adhesive force. However, only one contact was made on each
tissue sample and following that single touch, the tissue
was removed from the instrument. The measurement made by
the micrometer crosshair was noted, but was used only to
confirm that proper contact had been made.
The tissue sample was then immediately prepared for
scanning electron microscopy, while still glued to the
aluminum stub. The tissue was fixed for 1-3 hours in 2.5%
glutaraldehyde in Millionig buffer, and was then placed in a
buffer rinse for equal period of time. The tissue was then
dehydrated by a series of increasingly concentrated
ethanol-water baths. The samples were held in baths of 50,
70, 80, 90, 95 and 100 percent ethanol solutions for
ten minutes each. Following an additional ten minutes fresh
ethanol, the samples were then critical point dried using a
Tousimis Samdri PVT-3 (Tousimis, Rockville, MD).
The tissue specimens were then reglued to SEM stubs
with colloidal carbon paint, and after sufficient drying,
O
the samples were then sputter-coated with 200 A of
55
gold-palladium. The samples were then viewed with a JEOL
JSM-35C SEM.
SEM photomicrographs at 20-3OX were taken of each
specimen and evaluated. The initial assessment of the
extent of damage was done by averaging two independently
made estimates. The accuracy of these assessments was
confirmed by grid counting; this technique employed a grid
and the number of squares within the areas of damage were
compared to the number of squares within the entire area of
contact. This grid counting technique was used on a
selected sample of photomicrographs and percent of tissue
damage obtained by these measurements matched those made by
experimenters' assessments.
4. RESULTS AND DISCUSSION
4,. 1 Preliminary Studies and Overview of Thesis Research
In the preliminary studies the tissue-materials
adhesion phenomenon was discovered and the use of
hydrophilic polymer solutions to reduce adhesion was
demonstrated (5/ 6, 51). The results of these and related
studies are presented and discussed here since they
established the research goals and plan for this
dissertation research.
4.1.1 Prevention of Cornea Endothelium Damage by Poly(vinyl
pyrollidone) Solutions
The discovery of the tissue-material adhesion
phenomenon dealt with the intraocular lens-cornea
endothelium interface (5). In these studies, rabbit, and
human cornea endothelium was examined by optical microscopy
with nitro-blue tetrazolium staining and scanning electron
microscopy to reveal damaged cells following contact to
poly(methyl methacrylate) (PMMA) surfaces, dipped in either
balanced salt solution or solutions of the hydrophilic
polymer, poly(vinyl pyrrolidone) (PVP).
Examination by both methods showed that 30-35% of the
endothelium cells were damaged by 60 second contact with
56
57
acrylic surfaces dipped in balanced salt solution. Even a
one second contact produced extensive damage (approximately
20-25%). However, when the'acrylic surface was dipped in a
PVP solution (40 weight %), examination revealed almost no
damage to the endothelium, regardless of the time of
contact. Balanced salt solution would not wet the acrylic
surface while the PVP solution produced an adhesion
barrier-lubricating boundary layer.
4.1.2 Peritoneal Adhesions in Abdominal Surgery
These experiments evaluated the ability of solutions of
the hydrophilic polymers PVP and dextran to prevent
peritoneal adhesions resulting from trauma due to surgical
51
manipulations. Tables 1 and 2 show the results for the
canine and rat models, respectively.
A numerical rating scale was used to derive objective
data from subjective observations. Ratings were based upon
examination of the wound and small intestine and scored on
the.following scale:
0 No adhesions
1 Very slight evidence for adhesions
2 Slight to moderate number of adhesions
3 Moderate to extensive adhesions
4 Massive and extensive adhesions.
For rat experiments, the score for each animal
represented a consensus of 3 surgical observers. The dog
experiments were scored by 2 surgical observers. Scores
58
Table 1. Dog Model for Assessment of Polymer Solutions to
Prevent Adhesions.
Wound
Small
Intestine
Avg.
Score
Rel. Score
Ave. Score
Rel. Score
Drya
3.0
10
2.5
8
Saline3
3.0
10
1.5
5
PVPb
o
CM
7
0.3
1
Dextran
2.0
7
1.3
4
aAverage
of
4
surviving dogs.
^Average
of
3
surviving dogs.
Table 2.
Rat Model-Assessment of PVP Solution to Prevent
Adhesions.
Average Score/Animal
Relative Score
Control3
3.5
5.0
PVP Coating*5 0.7
1.0
aAverage of 4 animals surviving anesthesia and 8 day
maintenance with healed wounds.
Average of 3 surviving animals.
59
were averaged for each test group; the lowest average score
was assigned a value of 1.0 and higher scores assigned
appropriate relative values compared to the base low score.
In both the dog and rat experiments, the controls
showed moderate to extensive adhesions to the wound and of
the small intestine. Comparative results with PVP coating
were highly meaningful in both cases in that virtual
elimination of adhesions of the small intestine was
demonstrated. The dextran-treated dogs also showed some
reduction in wound adhesions. However, the saline-wet
control and the dextran-treated group both tended to have
less severe intestinal adhesions as compared with the dog
control group but were still much inferior to the
PVP-treated group.
The peritoneal adhesions were eliminated by the action
of the PVP solution to prevent serosal rubber-glove damage
during surgical manipulations. Rubbing of an intestinal-
surface was shown by SEM to produce considerable serosal
abrasion; virtually no tissue damage was observed following
rubbing by rubber gloves wet with PVP solution. As with the
acrylic-cornea endothelium contact, tissue damage was
eliminated by the interposition of nonadhesive, lubricating
boundary layer in the form of a surface-coating hydrophilic
polymer solution.
These experiments are significant in demonstrating for
the first time that surgical tissue damage and resulting
adhesions may be substantially reduced by application of
60
protective polymer coatings prior to manipulation of tissue
and organs in surgery.
4.1.3 Other Related Research
Unpublished research has shown the occurrence of the
tissue-materials adhesion phenomenon in the use of
endotracheal catheters. In these experiments, inflated
catheter cuffs were contacted with sections of trachea,
which had been removed from ferrets and then laid open
longitudinally. This contact was made with a loading force
which would simulate the clinical inflation of the cuff
within the trachea. These contacts, using cuffs dry and wet
with a 25 weight % solution of PVP in phosphate-buffered
saline, showed that desquamation of tracheal endothelium was
reduced by the use of the PVP solution. Quantitative data
were not obtained because the area of contact was difficult
to assess.
Quantitative measurement of the IOL-endothelium inter
action was also difficult; attempts to measure the adhesive
force at an acrylic-cornea endothelium interface using an
Instron tensile test machine with a one gram full-scale load
cell were unsuccessful. The results of these tests were not
reproducible, because of difficulty in mounting the tissue
and dehydration of the tissue during the time required for
measurement. Loading of the interface was difficult and
extremely variable as was estimation of the actual area of
contact. Similarly, attempts to measure cornea endothelium
61
damage quantitatively by optical microscopy with NBT
staining were unsuccessful; counting of damaged cells at
high magnification was unable to distinguish between
materials of widely different tissue adhesion properties.
The. failure of these experiments demonstrated the need for
an in vitro method for evaluation and accurate measurement
of the tissue-material adhesion phenomenon.
4.1.4 Overview of Thesis Research
As described above, surface modification by solutions
of hydrophilic polymers was shown effective in reducing or
eliminating tissue-material adhesion in several different
areas of surgery. However, a clinical need for preventing
IOL-cornea endothelium adhesion and its damaging conse
quences existed and efforts were focused on this problem.
Although solutions of PVP used clinically showed promise
(5), a permanent modification of the IOL surface eliminates
the need to apply coating by the surgeon in the operating
room. Greater safety would also be assured by permanent
modification if the IOL became detached following
implantation, when damage to the endothelium from contact,
just as produced during the original insertion, might occur.
Cast films of a hydrophilic polymer would eliminate the
application of the coating by the surgeon, but would not be
a permanent means to prevent any post-surgical occurrence of
cornea endothelium-IOL adhesion.
62
PMMA is very difficult to modify by conventional
wet-chemistry; attempts in this laboratory to perform
hydrophilic surface modifications by such means were
unsuccessful. For example, the acrylic surface appeared
unltered by exposure to a boiling aqueous solution of
concentrated sodium hydroxide. Gamma-radiation graft
polymerization was chosen as a method to produce hydrophilic
surfaces by polymerization on the surface of the PMMA. This
investigation emphasized the use of y-radiation to form
graft coatings of PVP on the PMMA surface and so produce a
hydrophilic modification of the acrylic lens surface.
A study of radiation dose and monomer concentration was
undertaken for y-radiation graft polymerization of PVP on
PMMA. Low radiation doses (<0.5 megarad (Mrad)) and low
monomer concentrations were chosen to minimize PMMA
degradation and gelation in the external PVP homopolymer.
Distortion of the PMMA by solvent swelling was also
minimized by using monomer concentrations of < 30 volume %
in aqueous solution and immersion of the PMMA in the
polymerization medium only at the time of irradiation.
Hydrophilic surface modification of PMMA was also
performed by radio-frequency (RF) plasma coatings. These
coatings were produced from the monomers of PVP and
poly(hydroxyethyl methacrylate) (PHEMA) to show an
additional possible method of surface modification of
hydrophilic polymer graft coating.
63
The hydrophilic polymer graft coatings were
characterized to determine the nature of the coating, most
particularly the properties which might influence the
biophysical behavior. The different tests which were
performed are discussed in the following sections.
Measurement of the intrinsic viscosity of the external
polymer formed in solution in y-radiation graft coating was
made to understand the relative amounts of polymer formed
under the different conditions in the study of radiation
dose and monomer concentration. The molecular weight of the
external polymer determined by these measurements would be a
relative measure of the extent of polymerization for the
graft coatings themselves and could be correlated to the
biophysical behavior and other properties of the graft
coating.
The contact angle of water was a surface property
measured to show the water wettability of the coatings and
their hydrophilicity, relative to PMMA. Contact angle was
measured for hydrated samples. Closely related to the
measurement of contact angle for water, the critical surface
tension for the graft coated surfaces was determined. It
was felt that the surface energy of the material might have
a significant influence on the contact adhesion resulting
from contact to an endothelium surface.
Scanning electron microscopy was used to examine the
surfaces and detect and record any changes in the surface
64
morphology by the graft coating. The examination by SEM
also was used to reveal the coating thickness, by scraping
the coating away and viewing the border of the scrape.
Three different spectroscopic techniques were used to
detect the presence of PVP graft coating on the PMMA
surface. Electron scattering for chemical analysis (ESCA)
and infrared spectroscopy (IR) were used for this purpose.
Ultraviolet-visible (UV-Vis) spectroscopy was used as a
possible means to show the coating presence but also to show
if the coating and the coating process had any effect on the
light transmissivity of the PMMA. Difficulties in detecting
the PVP by IR and UV-Vis lead to use of iodine to complex to
the PVP and enhance detection; the results of these
experiments will also be discussed.
In addition to the preparation and characterization of
hydrophilic polymer graft coatings on PMMA, an instrument
and technique for measurement of adhesive force and
endothelium damage were developed out of necessity; as
previously described, attempts to quantify tissue-materials
adhesion had been unsuccessful and such a means was
necessary for evaluation of the biophysical properties of
hydrophilic polymer graft coatings. The results of the
biophysical testing of adhesive force and quantitative
endothelium damage measurement of these materials will be
presented and the correlations to the other characteriza
tions described above will be discussed.
65
The tissue adhesion properties of several common
biomedical polymers were also evaluated for comparison to
the graft coated acrylic system. The contact angle for
water of these materials was also measured for the same
purpose.
4.2 Intrinsic Viscosity Molecular Weight of PVP
The molecular weight of the external polymer formed in
solution during Y-radiation graft coating gave an indication
of the relative molecular weight of the polymer (PVP)
actually grafted to the PMMA substrate; the molecular
weights of PVP formed under Y-radiation graft coating
conditions are presented in Table 3. The linearity of the
molecular weight versus initial N-VP concentration plot,
shown in Figure 16, coincides with previously reported data
(58), as did results for 0.1 and 0.25 Mrad dose levels. The
molecular weights for a given initial monomer concentration
were consistently higher for the higher dose level. The
difference became larger with increasing concentration.
In addition, the insolubility of the precipitated
homopolymer of the 5% and 10% N-VP concentrations at the
0.5 Mrad dose level in methanol, the solvent used for the
viscometry, indicated, along with the gelation of the
external homopolymer at 20% and 30% monomer concentrations
for that radiation dose level, the tendency for increased
crosslinking at the 0.5 Mrad dose level. Gelation of PVP in
solution has previously been shown to be a function of both
Table 3. Viscosity Molecular Weight* of PVP.
External Polymer Formed in Solution During Y-Radiation Graft. Coating.
% N-VP (Initial Concentration in
Polymerization Media)/YRadiation Dose (Mrad)
M
W
5%/0.1
309,470
10%/0.1
693,473
20%/0.1
1,028,743
30%/0.1
1,307,795
5%/0.25
577,992
10%/0.25
714,313
20%/0.25
1,371,339
30%/0.25
2,278,969
The external polymers formed for 0.5 Mrad radiation doses were insoluble and
therefore unfit for intrinsic viscosity measurement.
Calculated by Mark-Houwink-Sakurda equation from intrinsic viscosity in
methanol at 30C.
Mw(x106)
(Molecular Weight by Intrinsic Viscosity)
Figure 16. External Polymer Viscosity Molecular Weight
vs. Monomer Concentration for y-Radiation
Graft Coatings of PVP on PMMA.
68
radiation dose and concentration (59). Experiments using
cupric and ferrous ions as chain transfer agents to reduce
gelation (as reported (44)) were performed but not pursued
because of possible toxicity of the cupric ion and
difficulty in removing the ferrousion.
4.3 Contact Angle for Water
The measurement of the contact angle of water for the
different surfaces was important since it was felt that it
was the hydrophobic nature of the PMMA that made it adherent
to the endothelium of the cornea (Tables 4-6). Measurement
of the advancing angle of water for hydrated samples would
be valuable in correlating the hydrophilic nature of the
coatings with their biophysical behavior in the adhesive
force testing and the endothelium damage produced (and
evaluated by SEM).
The measurements revealed a significant decrease in
contact angle and therefore an increase in hydrophilicity
was achieved using Y-radiation- graft coatings of PVP on the .
PMMA substrate. Uncoated PMMA showed a contact angle of 72
and the y-radiation graft coatings showed a decrease in
angle down to 30 for a coating prepared at 30% N-VP
concentration and 0.1 Mrad dose.
As with the molecular weight for the external
homopolymer formed in solution, at 0.1 and 0.25 Mrad doses
the change in contact angle for water was proportional to
increasing monomer concentration for a given dose level.
This linearity was found not only for the 0.1 and 0.25 Mrad
Table 4. Contact Angle of Water on Y-Radiation Graft Coatings of PVP on PMMA.
Polymer
Average Contact Angle ()
Standard Deviation ()
PMMA*
72
8
5%**/0.1 Mrad***
61
3
10%/0.1
57
2
20%/0.1
40
3
30%/0.1
30
1
0%/0.25
65
2
5%/0.25
59
6
10%/0.25
59
7
20%/0.25
. 44
3
30%/0.25
37
5
5%/0.5
46
3
10%/0.5
38
7
*Uncoated; agrees with literature value of 78 (57).
**Initial monomer concentration in polymerization media.
***Radiation dose in megarad (Mrad).
Table 5. Contact Angle of Water for RF Plasma Coatings on PMMA.
Plasma Conditions
Power*/Pressure**/Time***
(Monomer)****
Average Contact Angle ()
Standard Deviation ( )
PMMA
72
8
50W/1000y/10min.(N-VP)
40
6
100W/200y/20min.(N-VP)
26
9
35W/500y/60min.(N-VP)
60
3
25W/500y/15min.(HEMA)
42
14
*RF power to plasma (Watts).
**Pressure in reactor (Microns).
***Time of exposure to plasma.
****Monomer vapor used.
Table 6. Contact Angle of Water for Various Biomedical Polymers.
Polymer
Average Contact Angle ()
Standard Deviation ()
Urethane
37
7
Silicone Rubber
101
2
Thermoplastic Elastomer
90
2
Teflon
105
3
*Pellthane TM, Upjohn Corporation, Kalamazoo, MI.
**Silastic 500-5 TM, Dow-Corning, Midland, MI.
***Proprietary hydrocarbon block copolymer, C-flex TM, Concept Inc., Clearwater,
FL.
****TM, FEP, E.I. DuPont de Nemours & Co., Wilmington, DE.
72
dose levels but also for the 0.5 Mrad dose level coatings
(Figures 17-20). The 0.5 Mrad coatings showed a much
greater decrease in contact angle, although the gelation of
the external homopolymer prevented measurement of the
contact angle at 20% and 30% N-VP concentrations.
Comparison of this linear relationship between N-VP
concentration and contact angle for the different dose
levels showed this greater decrease in contact angle at the
5 and 10% N-VP concentrations for the 0.5 Mrad coatings
(Figure 20). The unexpected result that the 0.1 Mrad
coatings show more of a decrease in contact angle than do
the 0.25 Mrad coatings can be explained only on the basis of
the standard deviation of the contact angle measurements.
The standard deviation for this measurement shows that the
values for the 0.1 and 0.25 Mrad coatings do overlap and the
difference is not significant.
Measurement of the contact angle for the RF plasma
coatings shows that these coatings are hydrophilic, like the
Y-radiation graft coatings (Table 5). Only a preliminary
study of RF graft coating was made. Correlations between
reaction conditions and contact angle were not developed.
However, the results of these experiments, along with the
results of biophysical tests, will be discussed later.
Contact angles for a number of common biomedical
polymers were also measured and compared with graft coated
PVP on PMMA (Table 6). The majority of these materials
possess a surface which is more hydrophobic than the
Contact Angle ()
S % N*VP (Initial Concentration in Polymerization Media)
a.
Figure 17. Contact Angle for Water vs. % Monomer for
0.1 Mrad Dose y-Radiation Graft Coatings of
PVP on PMMA.
74
E % IM-VP (Initial Concentration in Polymerization Media)
Figure 18. Contact Angle for Water vs. % Monomer for
0.25 Mrad Dose y-Radiation Graft Coatings of
PVP on PMMA.
Uncleanable gel formed at monomer
concentrations of 20% and greater.
Figure 19. Contact Angle for Water vs. % Monomer for
0.5 Mrad Dose y-Radiation Graft Coatings of
PVP on PMMA.
£ %N-VP
(Initial Concentration in Polymerization Media)
Coefficient of Determination
Figure 20. Summary of Contact Angle for Water vs. %
Monomer for Various Dose Levels of Y"Radia
tion Graft Coatings of PVP on PMMA.
77
uncoated PMMA and therefore furnish information about
materials with surfaces much different than the hydrophilic
coatings prepared by Y-radiation graft and RF plasma
polymerizations. These materials are all polymers which are
important or potentially important for biomedical use in
applications in which the contact adhesion phenomena could
play a significant role.
4.4 Critical Surface Tension
Critical surface tension measurements were made to
additionally characterize the 0.25 Mrad Y-radiation graft
coating series and the RF plasma coatings for correlation to
the biophysical measurements and the other surface
characterizations. However, the high surface energies of
these surfaces made the results of these measurements
difficult to interpret.
In order to measure the contact angle over a range of
different surface tension liquids, liquids which showed a
finite contact angle and would not spread (contact angle of
zero) were required. These liquids were all of relatively
high surface tension (greater than 50 dynes/cm) and the
majority of which are polar in nature. The result is polar
interaction with the hydrophilic PVP graft coated surfaces.
The Zisman plot for PMMA, tested at ambient conditions,
showed a critical surface tension of 38 dynes/cm, which is
78
extremely close to the literature value of 39 dynes/cm (57)
(Figure 21). Measurement of the critical surface tension
for the 0.25 Mrad radiation graft series begins with a
slight decrease in the surface tension for the 5% N-VP
concentration graft coating and then begins to increase to
greater critical surface tension for the 10% and greater
N-VP concentration graft coatings (Figures 22-26). However,
the surface energies obtained for the higher concentrations
are not realistic, as the numbers approach and exceed
100 dynes/cm (Table 7).
The Zisman plots for the 20% and 30% N-VP concentra
tions show a change in slope from the normal, negative slope
to a positive slope (Figures 25-28). The coefficient of
determination for these two coatings' plots is also
considerably less than for the other plots. The polar
interactions that result from the use of polar liquids on
these increasingly hydrophilic surfaces disrupt the
usefulness of the Zisman plot as an accurate measurement of
the critical surface tension of these coatings.
In an attempt to accurately calculate the critical
surface tension for these surfaces, the contact angle data
for the different liquids were treated by the method of
Baszkin and Lyman (60). This method separates the
dispersive and nondispersive (polar) interactions between
the liquid and the. solid and has been performed on surfaces
of high critical surface tension. The polar contribution to
the critical surface tension has been reported as high as
Cosd
Figure 21. Zisman Plot of Critical Surface Tension (yc) of PMMA.
-j
Cos#
Figure 22. Zisman Plot of Critical Surface Tension (y ) of y-Irradiated
(.0.25 Mr ad) PMMA. c
e so
Surface Tension (dynes/cm)
Zisman Plot of Critical Surface Tension (y ) of y-Radiation
Graft Coating: 5% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
Figure 23.
Cos#
Surface Tension (Dynes/cm)
Figure 24. Zisman Plot of Critical Surface Tension (y ) of y-Radiation
Graft Coating: 10% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
00
to
Cos#
Figure 25. Zisman Plot of Critical Surface Tension (y ) of y-Radiation
Graft Coating: 20% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
Cos#
Surface Tension (Dynes/cm)
Figure 26. Zisman Plot of Critical Surface Tension (y ) of y-Radiation
Graft Coating: 30% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
*
Table 7. Critical Surface Tension from Zisman Plots for y
Coatings of PVP on PMMA.
-Radiation Graft
% N-VP (Initial Concentration in
Polymerization Media)
Surface Energy
(Dynes/cm)
PMMA
38
0%
43
5
28
10
46
20
68
30
104
*Using water, formanide, glycerol, thiodiethanol and methylene iodide as
solvents. Instantaneous contact angles were measured.
00
U1
Cos?
Figure 27. Zisman Plot of Critical Surface Tension (y ) of 0.25 Mrad
y-Radiation Graft Coatings of Different N-^P Concentrations.
00
cn
110
100
90
80
70
60
50
40
30
20
10
PMMA
30%/0.25 104
1
_L
0
_J 1 _J L_
5 10 20 30
% N-VP (Initial Concentration in Polymerization Media)
¡ummary of Zisman Plots for Critical Surface Tensions (y ) of PVP on PMMA
rs. % N-VP for y-Radiation Graft Coatings of 0.25 Mrad Dose.
00
88
90-95 dynes/cm and the dispersive contribution to be 25-39
dynes/cm for poly(hydroxyethyl methacrylate) surfaces (61).
Treatment of the data in this manner yields comparable
numbers for the polar contribution for the 20% and 30% N-VP
concentration graft coatings (about 120 dynes/cm.)/ and the
dispersive contribution is low, less than 5 dynes/cm.
However, the treatment of the data for the PMMA surface by
this method yields values for the two contributions which
are relatively lower (87 dynes/cm for the polar contribution
and 2 dynes/cm for the dispersive contribution), but are
unreasonably high for a PMMA surface.
It is clear that the polar interactions resulting from
the increased hydrophilicity of the graft coatings greatly
complicate the measurement and determination of the critical
surface tension for those surfaces. Both methods of
treating the data show that there is a relative increase in
the surface tension for the graft coated surfaces over the
critical surface tension for uncoated PMMA. Accurate
determination of the critical surface tensions for these
surfaces will require further investigation.
The complicating effects of these polar interactions
are also present in the critical surface tension
measurements for the RF plasma coatings, although the slopes
and linearity of the Zisman plots for these coatings are
normal and acceptable (Table 8) (Figures 29-32). The
critical surface tensions calculated by these plots (all
near 50 dynes/cm) show the increased surface tension for
Cos Q
1.0
Figure 29. Zisman Plot for Critical Surface Tension (y ) of RF Plasma
Coatings of N-VP on PMMA: 50 Watts(Power)/I000p(Pressure)/
10 Minutes(Duration).
Cos 6
Figure 30. Zisman Plot for Critical Surface Tension (y ) of RF Plasma
Coatings of N-VP on PMMA: 100 Watts (Power)^200y(Pressure)/
20 Minutes(Duration).
Cos#
1.0
Figure 31. Zisman Plot for Critical Surface Tension (y ) of RF Plasma
Coatings of N-VP on PMMA: 35 Watts(Power)/§00y(Pressure)/
60 Minutes(Duration).
VO
Cos d
Figure 32. Zisman Plot for Critical Surface Tension (y ) of RF Plasma Coating
of HEMA on PMMA: 25 Watts (Power)/500y (Pressure)/15 Minutes
(Duration).
Table 8. Summary of Zisman Plots for Critical Surface Tension of RF Plasma
Coatings on PMMA.
Plasma Conditions
Power*/Pressure**/Time***(Monomer)****
Surface Energy
(Dynes/cm)
PMMA
38
50W/1000U/10Min.
(N-VP)
46
100W/ 200y/20Min.
(N-VP)
53
35W/ 500u/60Min.
(N-VP)
50
25W/ 500P/15Min.
(HEMA)
50
*RF power to plasma.
**Pressure in reactor.
***Time of exposure to plasma.
****Monomer vapor used.
94
these hydrophilic coatings. Still, the value of these
measurements lies not as an accurate measurement of the
surface energy, but as an indication of the (a) increase
relative to the surface of the uncoated PMMA, and (b)
increasing polar interaction with increasing hydrophilicity.
4.5 Scanning Electron Microscopy
All of the following scanning electron micrographs are
of PMMA "chisel-cut" specimens and provide a qualitative
measurement of the coating thickness and the surface mor
phology of the coatings. The appearance of the Y-radiation
graft coatings seems to show little correspondence to the
coating conditions, but some trends do exist (Figure 33).
The majority of the coatings produced at 0.1 and 0.25
Mrad are uniform and show the coverage of the PVP coating
(Figure 33.4-33.27). Their surfaces are very similar in
appearance to uncoated PMMA (Figure 33.1-33.3). At the 0.25
Mrad dose level (Figure 33.16-33.27), holes or gaps do
appear in the coating in some places; this may be due to an
increased thickness compared to the 0.1 Mrad coatings
(Figure 33.4-33.15). All of the coatings seem to show good
adherence to the PMMA substrate and for the 20% and 30% N-VP
coatings, this adherence prevented complete removal, by
scraping, of the coating. Low magnification photographs
(x500) (Figure 33.10, 33.13, 33.22 and 33.25) show the
remnants of the coating remaining in the scraped area; high
magnification photographs (x2000 and x5000) show that the
Figure 33.1. Untreated PMMA.
Figure 33.2. Untreated PMMA.
Figure 33. Scanning Electron Micrographs of Scraped
y-Radiation Graft Coatings of PVP on PMMA.
96
Figure 33.4. 5% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.'
97
Figure 33.5. 5% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
98
^ 4
15KU X500 1001 10.0U UFMSE
*,
Figure 33.7. 10% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
15KU X2000 1001 10.0U UFMSE
Figure 33.8. 10% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
99
Figure 33.9. 10% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
Figure 33.10. 2Q% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
100
< s
5 l *
' t
£
0
: .1
# f.
4 >
4 /
- :
f ,
Ii
si
f i
10.0U UFMSE
15KU X2000 2001
Figure 33.11. 20% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
Figure 33.12. 20% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
1Q1
Figure 33.13. 30% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
Figure 33.14. 30% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
102
Figure 33.15. 30% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
Figure 33.16. 5% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
103
Figure 33.17.
5% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
Figure 33.18. 5% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
104
Figure 33.19. 10% N-VP Initial Monomer Concentration and
Q.25 Mrad Dose.
Figure 33.20. 10% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
105
Figure 33.21. 10% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
Figure 33.22. 20% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
106
i J
15IW X2090^i ' 2jg^ 6 0IJ UFUSE
Figure 33.23. 20% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
107
Figure 33.25. 30% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
Af<; 0 ** ry .
V r .
*T?v ' /
Jr '
f y
ff? *
/ /--'V
*
i t f
* /
r
'
(
^ *
* *
' ¥.> *
15KU X2000 3025
* ,
10.0U UFMSE
Figure 33.26. 30% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
108
Figure 33.27. 30% N-VP Initial Monomer Concentration and
Q.25 Mrad Dose.
Figure 33.28. 5% N-VP Initial Monomer Concentration and
0.5 Mrad Dose.
109
33.29. 5% NVP Initial Monomer Concentration and
0.5 Mrad Dose.
Figure 33.30. 5% N-VP Initial Monomer Concentration and
0.5 Mrad Dose.
110
Figure 33.31. 5% N-VP Initial Monomer Concentration and
0.5 Mrad Dose.
Figure 33.32. 10% N-VP Initial Monomer Concentration and
0.5 Mrad Dose.
Ill
Figure 33.33. 10% N-VP Initial Monomer Concentration and
0.5 Mrad Dose.
Figure 33.34. 10% N-VP Initial Monomer Concentration and
0.5 Mrad Dose.
112
Figure 33.35. 10% NVP Initial Monomer Concentration and
0.5 Mrad Dose.
113
PMMA substrate is not visiblesome of the coating still
covers the substrate surface (Figure 33.11, 33.12, 33.14,
33.15, 33.23, 33.24, 33.26 and 33.27).
The strong adherence of the Y-radiation graft coatings
is also evident in the examination of the 0.5 Mrad coatings
(Figure 33.28 and 33.25). For these coatings, at both the
5% and 10% N-VP concentrations, the scraping procedure would
not remove the coating but only "scuffed" it. The gelation,
which occurred in polymerization media at higher N-VP -
concentrations for the 0.5 Mrad dose level,* suggests that
these coatings were highly crosslinked and probably would be
more difficult to scrape away. The "scuffing" behavior
seems likely for a coating which is still relatively thin,
but both highly coherent and adherent.
The coatings prepared by RF plasma polymerization also
appear to be very adherent to the PMMA substrate (Fig
ure 34). The polishing scratches visible on the PMMA
substrate where the coating has been scraped away, as on the
5% N-VP concentration and 0.1 Mrad radiation graft coating
(Figure 33.4-33.6), are not apparent on the scraped areas of
*As well as the insolubility in methanol of the homopolymer
precipitated from the polymerization media for these
coatings.
114
Figure 34.1. RF Plasma Coating of N-VP on PMMA: 50 Watts
(.Power)/lOOOy(Pressurel/10 Minutes(Duration).
Figure 34.2. RF Plasma Coating of N-VP on PMMA: 5Q Watts
(Power)/lOOOu(Pressure)/1Q Minutes(Duration).
Figure 34. Scanning Electron Micrographs of Scraped RF
Plasma Coatings on PMMA.
115
Figure 34.3. RF Plasma Coating of N-VP on PMMA: 50 Watts
(Power}/1000]i CPressure)/10 Minutes(Duration).
Figure 34.4. RF Plasma Coating of N-VP on PMMA: 10Q Watts
(Power)/2Q0]i CPressure)/20 Minutes (Duration) .
116
Figure 34.5. RF Plasma Coating of N-VP on PMMA: 1QQ Watts
(Power)/200y(Pressure)/20 Minutes(Duration).
Figure 34.6. RF Plasma Coating of N-VP on PMMA: 10Q Watts
(Power)/2QQj (Pressure)/2Q Minutes (Duration) .
117
Figure 34.7. RF Plasma Coating of N-VP on PMMA: 35 Watts
(Power)/5Q0y(Pressure)/60 Minutes(Duration).
Figure 34.8. RF Plasma Coating of N-VP on PMMA: 35 Watts
(Power)/500y(Pressure)/60 Minutes(Duration).
118
Figure 34.9. RF Plasma Coating of N-VP on PMMA: 35 Watts
(Power)/500y(Pressure)/60 Minutes (Duration).
Figure 34.10. RF Plasma Coating of N-VP on PMMA: 25 Watts
(Power)/500y(Pressure)/15 Minutes(Duration).
119
Figure 34.11. RF Plasma Coating of HEMA on PMMA: 25 Watts
CPower)/500y(Pressure)/15 Minutes(Duration).
Figure 34.12. RF Plasma Coating of HEMA on PMMA: 25 Watts
(Power)/5Q0y(Pressure)/15 Minutes(Duration).
120
the RF plasma coatings. The RF plasma coatings showed
different mechanical properties and did not seem to be
totally scraped away also because the scrape was not clearly
defined. The lack of a clearly defined edge made judgement
of the coating thickness difficult, but seem to be on the
same order as the y-radiation graft coating thicknesses of a
micton or less.
4.6 ESCA
Surface analysis of the y-radiation graft coatings by
ESCA shows the presence of the PVP on their surfaces.
Nitrogen, present in the pyrrolidone ring of PVP, can be
detected at significant levels on these coated surfaces,
relative to a lack of nitrogen on uncoated PMMA.
Table 9 shows the results for the ESCA analysis of an
uncoated PMMA sample and two y-radiation graft coatings of
PVP on PMMA, made under the following conditions: 10% N-VP
concentration with 0.25 Mrad y-radiation dose and 10% N-VP
with 0.5 Mrad y-radiation dose. Two angles of incident
radiation were used in the ESCA to sample the surfaces: a
O
0 angle sampled at a mean depth of 50A and a 70 angle
O
sampled at a mean depth of 25A. The data are expressed in
terms of atomic ratio relative to carbon.
The data show trace amounts of silicon and chlorine,
which were contaminants of unknown origin. The copper, also
detected only at trace levels, comes from the copper wire
used as a sample holder. The 0/C ratio, showing the amount
of oxygen present, will be considered in this discussion as
121
Table 9. ESCA Results
Molar Ratio to Carbon PMMA and y-Radiation Graft Coatings
PVP on PMMA
Sample Angle
PMMA 00
70
10% N-VP Concentration/ 0
0.25 Mrad Dose 70
10% N-VP Concentration/ 0
0.5 Mrad Dose 70
O/C
N/C
Si/C
Cl/C
Cu/C
.20
.13
.02
.014
.009
.023
.020
.016
.030
.25
.24
.049
.045
.007
.014
.005.
.011
.037
.003
.41
.36
.053
.078
.074
.095
.023
.012
.020
.003
122
showing only that relatively equal amounts of oxygen were
present on the PMMA and the PVP-coated surfaces.
The numbers for the N/C ratio do show a significant
difference between the two types of surfaces in the amount
of nitrogen present. The N/C ratio for the PVP-coated
surfaces is much greater than that for the uncoated PMMA,
for either angle (depth of sampling).
Examination of the ESCA spectra for carbon and nitrogen
shows the real significance of the difference in the N/C
ratios. The nitrogen spectra for the uncoated PMMA (Figure
35) show no real peaks for either angle and can be regarded
as background levels. The nitrogen peaks for both angles of
the two y-radiation graft coatings of PVP on PMMA are
substantial (Figures 36 & 37). The spectra show that
nitrogen is present on these surfaces, assuredly in the
pyrrolidone ring.
The N/C ratio for the two depths of sampling is
effectively the same, so it can be concluded that the PVP
O
coating is uniform down to the sampling depth of 50A.
4.7 Infrared Spectroscopy
The use of attenuated total reflectance (ATR) with
infrared spectroscopy (IR) was selected as the most sensi
tive means to obtain the infared spectra of the y-radiation
graft and RF plasma coatings. Comparison of the spectra of
PMMA and PVP shows the great similarity between the two'; the
best chance at finding a resolvable absorption peak occurs
INTENSITY
NITROGEN
SPECTRA
(0 ANGLE)
N/C = 0.020 N/C = 0.017
Figure 35. ESCA Spectra for PMMA (Uncoated) (B.E. = Binding Energy; N/C = Nitrogen to
Carbon Atomic Ratio).
123
B.E.
INTENSITY
NITROGEN
SPECTRA
(70 ANGLE)
B.E.
N/C = 0.049 N/C = 0.045
ESCA Spectra for y-Radiation Graft Coating of PVP on PMMA: 10% N-VP
Initial Monomer Concentration and 0.25 Mrad Dose (B.E. = Binding Energy
N/C = Nitrogen to Carbon Atomic Ratio).
Figure 36.
INTENSITY
NITROGEN
SPECTRA
(0 ANGLE)
B.E.
B. E.
N/C = 0.053 N/C = 0.078
Figure 37. ESCA Spectra for y-Radiation Graft Coating of PVP on PMMA: 10% N-VP Initial
Monomer Concentration and 0.5 Mrad Dose (B.E. = Binding Energy; N/C = Nitrogen
to Carbon Atomic Ratio).
126
at the 5.8 micron and 6.0 micron peaks for PMMA and PVP,
respectively (Figure 38) (62). However, all efforts to
detect the PVP coating by a peak at that or any
characteristic wavelength were unsuccessful (Figure 39).
The thickness of a coating sufficient for detection on
an absorbing substrate has been determined by Harrick (63).
(sin^O -
where 6.^ is the thickness necessary for detection, is the
wavelength of the peak to be detected in the reflection
element, 0 is the incident angle for internal reflection and
n2^ is the ratio of the refractive index of the coating to
the refractive index of the internal reflection element.
For maximum sensitivity, the number of reflections
incident to sample surface was maximized. Use of internal
reflection elements at 60 and 45 provided 25 and 15
reflections, respectively. Using the above equation and the
conditions for detection of the 6.0 micron peak with KRS-5
(n=2.37) internal reflection elements at 60 and 45, the
coating thickness required for detection was calculated.
This calculation yielded thicknesses of 1.2 micron for the
60 and 2.8 microns for the 45 element. If the thickness
of the coatings as observed by SEM are accurate, their
thicknesses of less than a micron are indetectable by
Wavenumber (cm'1) .
4000 3000 2500 2000 1500 1300 1100 1000 900 800 700 650 625
Figure 38.- Infrared Spectra for PMMA and PVP (Literature, Reference 62).
Wavenumber (cm'1)
4000 3500 3000 2500 2000 1800 1600 1400 1200 1000. 800 600
Wavelength in Microns
Figure 39. Attenuated Total Response-Infrared Spectra of PMMA, y-Radiation Graft
Coating of PVP on PMMA and PVP Cast Film on PMMA.
128
129
infrared spectroscopy, particularly in light of the high
absorption of the PMMA substrate.
In an attempt to corroborate this conclusion, films of
PVP cast from chloroform solution onto PMMA were made over a
range of thicknesses, and ATR-IR spectra were taken
(Figure 39). The thinest detectable film was made from a 5%
PVP' solution with a 1 mil doctor blade; its spectrum showed
a strongly absorbing peak at 6.0 micron, indicating the
detection of both PVP and the PMMA substrate. SEM
examination of this surface, abraded to remove only the PVP
film, showed that the film was on the order of a few microns
in thickness, considerably thicker than the y-radiation
graft and RF plasma coatings which had been examined under
the same conditions. The coating thickness must therefore
be below the limit of detection of ATR-IR, that is less than
one micron.
4.8 Ultraviolet-Visible Spectroscopy
The ultraviolet-visible (UV-Vis) spectrum of PMMA shows
a sharp loss in transmission at 300 nanometers. The UV-Vis
spectra of the radiation graft coated samples show no marked
changes in the spectrum, although the severity of the drop
in transmission is lessened (Figures 40-42). This change in
spectrum is independent of the N-VP concentration in the
polymerization media and is related only to the level of
Y-radiation dose. The effects of y-radiation are more
pronounced at higher doses as can be seen in Figure 43 and
% Transmission
Wavelength (ntn)
Figure 40. Ultraviolet-Visible Transmission Spectra of YRadiation Graft Coatings of
PVP on PMMA, 0.1 Mrad Dose Level.
130
% Transmission
Wavelength (nm)
Figure 41. Ultraviolet-Visible Transmission Spectra of y-Radiation Graft Coatings of
PVP on PMMA, 0.25 Mrad Dose Level.
131
% Transmission
Wavelength (nm)
Figure 42. Ultraviolet-Visible Transmission Spectra of y-Radiation Graft Coatings of
PVP on PMMA, 0.5 Mrad Dose Level.
132
% Transmission
Wavelength (nm)
Figure 43. Ultraviolet-Visible Transmission Spectra of y-Irradiated PMMA.
133
134
in the literature (41). This spectral change is attributed
to radiolysis of the PMMA. The effects of the y-radiation
at the dose levels used in the production of the coatings in
this study are relatively insignificant, but higher
radiation doses, as for sterilization, may have detrimental
effects in some uses of the PMMA.
The UV-Vis spectra of the RF plasma coatings show that
these coatings also have no effect on the transmission of
the material over this spectral region (Figure 44).
4.9 PVP-Iodine Interaction
PVP forms a stable chemical complex with iodine, the
exact nature of which remains to be completely understood.
Our attempts to detect and measure the complexing of iodine
by the Y-radiation graft coatings and the RF plasma coatings
were unsuccessful by both UV-Vis and ATR-IR spectroscopy.
The spectra for the coatings were unaltered by soaking the
samples in a saturated solution of iodine in water
overnight.
Iodine has been shown to alter both spectra for
solutions of PVP. However, since the PVP alone was
undetected by both methods of spectroscopy, it was unlikely
that the PVP-Iodine complex would be detected. The spectra
obtained confirmed this, as the spectra were unchanged.
Other research has shown that the addition of iodine
does alter the spectra for PVP in solution (64). The IR
spectral changes are slight and the presence of the highly
Wavelength |nm)
Figure 44. Ultraviolet-Visible Transmission Spectra of RF Plasma Coatings on PMMA.
135
136
absorbing PMMA substrate has been shown in this research to
mask the detection of the PVP alone. Iodine broadens the
peak at 6.0 microns, which unfortunately coincides with a
PMMA peak, so the effect of the iodine is lost by this
interference.
Changes in the ultraviolet-visible spectrum are also
not apparent for the samples of either y-radiation graft or
RF plasma coatings of PVP on PMMA. The UV-Vis spectrum of
the iodine alone shows absorption peaks at 450 and 350
nanometers,* but these peaks are not shown on the spectra
obtained for the coated samples equilibrated with the
saturated iodine solutions. The explanation for the lack of
these peaks lies either in the amount of PVP present in the
coating or in the amount of iodine absorbed. Since other
measurements have shown the coatings to be less than a
micron thick, the lack of sufficient coating for complexing
a detectable amount of iodine seems that most likely reason.
4.10 Biophysical Measurements
The development of the instrument and technique capable
of producing quantitative measurements of the contact
adhesion phenomenon was an important step toward
understanding its nature and mechanism. Other research has
examined the adhesive nature of materials to cells in
*Peaks at shorter wavelengths could not be seen due to the
absorptivity of PMMA below 300 nm.
137
culture, and various methods have been used to measure the
extent of the cell adhesion to material surfaces
(24,36,37,65). None of these techniques are applicable to
the study of the tissue-materials adhesion phenomenon, since
they cannot duplicate the actual clinical conditions of the
phenomenon or measure the significant properties of the
phenomenon in the adhesive force and the resultant cell
damage.
The measurement of the adhesive force for cornea
endothelium contact with PMMA, y-radiation graft coatings of
PVP on PMMA, RF plasma coatings on PMMA and with various
other biomedical polymers have been made. Because of the
expected variation of such measurements with biological
material (cornea endothelium from rabbit eyes), at least ten
measurements were made for each material and coating tested.
The significance of the adhesive force measurement for a
material is in its relative magnitude to the other surfaces
and its correlation to the other properties of that surface,
particularly the other biophysical property measured the
amount of cornea endothelium damage resulting from the
contact.
Measurement of quantitative cornea endothelium damage
by scanning electron microscopy (SEM), also developed for
this study, measures the definitive quantity involved with
the contact adhesion phenomenon. The extent of cell
destruction determines the significance of the adhesion
between the tissue surface and the material since this
138
destruction leads to the other complications from the
contact.
Only one such measurement could be made from each
cornear placing a limitation on the number of measurements
that could be practically made. Even with the limited
number of tests for a given surface, the standard deviations
were reasonable, considering the type of test and were, for
the most part, approximately 15%.
For all of the biophysical testing, the baselines for
comparison are the data for PMMA. The greatest number of
tests were made on the PMMA, because during the development
and refinement of the adhesive force measurement technique,
each cornea was tested with an experimental coating and then
three control measurements were made with PMMA. After the
uniformity in the results for the PMMA was assured by
testing in this manner over a period, only random sampling
was done to provide a control measurement. Comparison to
PMMA is valuable in that the PMMA is the material currently
being used. For an alternative material to be chosen, its
biophysical behavior must be more favorable than that of the
PMMA.
In the adhesive force testing, PMMA shows an average
2
adhesive force of 405 mg/cm The values for the
Y-radiation graft coatings are shown in Table 10 and all
show a reduction in the adhesive force below that value.
The sample which was given a radiation dose of 0.25 Mrad
while in water had an average value of 490 mg/cm but this
Table 10. Summary of Adhesive Force Measurements for Cornea Endothelium Contact
with Y-Radiation Graft Coatings of PVP on PMMA.
Polymer Surface
(% N-VP Monomer/
Y-Radiation Dose (Mrad))
Average Adhesive
Force (mg/crn )
Standard _
Deviation (mg/cin )
Number of
tests
PMMA
405
162
93
5%/0.1
362
80
21
10%/0.1
179
48
21
20%/0.1
174
118
29
30%/0.1
108
34
29
0%/0.25
490
212
21
5%/0.25
156
40
20
10%/0.25
70
83
24
20%/0.25
105
29
22
30%/0.25
111
19
25
5%/0.5
209
29
26
10%/0.5
146
27
28
139
140
was easily within the standard deviation of the measurements
and shows no significant difference. However, as a plot of
N-VP concentration versus the adhesive force for each of the
three radiation doses shows, the adhesive force drops
significantly with increasing monomer concentration; the
decline in adhesive force is greatest for the 0.25 Mrad
coatings, as the average adhesive force decreases to about
one-fourth of the adhesive force for PMMA (Figure 45).
The decrease in adhesive force with the change in
coating conditions parallels the change in contact angle for
water for the same change in coating conditions. The change
in contact angle is more linear, but the decrease in
adhesive force with increasing hydrophilicity is a
correlation which the initial hypothesis of the usefulness
in developing a hydrophilic coating to prevent the contact
adhesion phenomenon proposed.
Quantitative cornea endothelium damage measurement by
SEM for the contact with these same Y-radiation graft
coatings of PVP on the PMMA confirm the ability of the
coatings to reduce the severity of the contact adhesion.
Figure 46 shows representative SEM photos of damaged
endothelium. As seen in Table 11, PMMA produced an average
destruction of 60% of the cells within the area of contact;
this number does not immediately decrease with the coatings
produced at only 5% N-VP concentration and 0.25 Mrad, but as
monomer concentrations increase past this level, the extent
of endothelium damage shows a dramatic decrease. Although
141
490 r
o.
Figure 45. Adhesive Force vs. Monomer Concentration for
PVP y-Radiation Graft Coatings on PMMA.
142
Figure 46.2. 10% Cornea Endothelium Damage after Contact
with 30% N-VP/Q.25 Mrad Graft Coating.
Figure 46. Representative Scanning Electron Micrographs
of Cornea Endothelium Damage.
Figure 46.1. 0% Cornea Endothelium Damage after Contact
with 30% N-VP/Q.25 Mrad Graft Coating.
143
Figure 46.3. 3Q% Cornea Endothelium Damage after Contact
with 2Q% N-VP/0.25 Mrad Graft Coating.
50% Cornea Endothelium Damage after Contact
with 1Q% N-VP/Q.25 Mrad Graft Coating.
144
Figure 46.5. 7Q% Cornea Endothelium Damage after Contact
with Untreated PMMA.
Table 11. Quantitative Cornea Endothelium Damage by SEM For Contact with
Y-Radiation Graft Coatings of PVP on PMMA. .
N-VP Monomer Concentration/
Y-Radiation Dose (Mrad)
Number of
Tests
Cells Destroyed in Contacted Area
Average(%) Standard Deviation(%)
PMMA
5
60
17
10%/0.1
4
45
13
30%/0.1
4
13
13
0%/0.25
3
65
31
5%/0.25
3
63
21
10%/0.25
4
30
20
20%/0.25
3
17
12
30%/0.25
3
5
5
10%/0.5
3
14
6
145
146
the data is only complete for the 0.25 Mrad series.
Figure 47 shows that the decline follows the increasing
monomer concentration, just as the adhesive force declined
and the hydrophilicity increased.
Table 11 also shows that a radiation dose of 0.25 Mrad
produces no significant change in the contact adhesion
behavior of PMMA, as assessed by endothelium damage. Taken
with the other information available for the material
irradiated in water, the radiation has no appreciable effect
on either the physical or biophysical behavior of the PMMA
at such a low radiation dose level.
The correlations between contact angle, the adhesive
force and extent of endothelium damage are not as clear for
the RF plasma coatings as for the Y-radiation graft
coatings. All of the surfaces produced by RF plasma coating
show a reduction in the contact angle for water (Table 5).
However, there is no accompanying decrease in the adhesive
force to the cornea endothelium, as shown with the
Eradiation graft coatings, although the quantitative
measurement of cornea endothelium damage does show a
decrease (Tables 12 and 13). The correlation between
contact angle for water and the extent of cell destruction
is not present: the RF plasma coating of the lowest contact
angle for water (26), shows no change in the percent of
cell destroyed from PMMA. The RF plasma coating of PVP
which showed the greatest reduction in the cell destruction
Figure 47. Quantitative Cornea Endothelium Damage by SEM
vs. N-VP Monomer Concentration for 0.25 Mrad
Dose y-Radiation Graft Coatings on PMMA.
Table 12. Summary of Adhesive Force Measurements for Cornea Endothelium Contact
with RF Plasma Coating on PMMA.
Plasma Conditions
Power */Pre s sure */Time * *
(Monomer****)
Average Adhesive
Force (mg/crn )
Standard ^
Deviation (mg/cin )
Number of
Tests
50W/1000y/10min.
(N-VP)
221
50
21
100W/200y/20min.
(N-VP)
513
245
9
35W/500y/60min.
(N-VP)
277
147
11
25W/500y/15min.
(HEMA)
221
87
10
*RF power to plasma (Watts).
**Pressure in reactor (Microns).
***Time of exposure to plasma.
****Monomer vapor used.
148
Table 13. Quantitative Cornea Endothelium Damage by SEM for RF Plasma Coatings
on PMMA.
Plasma Conditions % Cells Destroyed in Contacted Area
Power*/Pressure**
/Time***(Monomer****)
Average (%)
Number of Tests
50W/1000y/10min.
(N-VP)
28
4
100W/200y/20min.
(N-VP)
61
3
35W/500V/60min.
(N-VP)
25
2
25W/500 y/15min.
(HEMA)
25
2
*RF power to plasma (Watts).
**Pressure in reactor (Microns).
***Time of exposure to plasma.
****ilonomer vapor used.
(down to 25%) had only a small reduction in contact angle
for water, to 60.
Because there was no systematic variation of parameters
involved in the RF plasma coating process, it is impossible
to determine, at this point, the reasons for the anonamlies
to the previous correlations. The reduction in the
endothelium damage without a similar reduction in the
adhesive force does not seem logical, and will require
further research to provide an explanation. Still, the RF
plasma coatings, produced from both N-VP and HEMA monomers,
are effective in the reduction of endothelium damage;
whether or the correlation exist between surface properties
and the biophysical behavior of these RF plasma coatings
will require a more exhaustive study.
The testing of the various biomedical polymers by these
biophysical tests is shown in Tables 14 and 15. The
adhesive force for urethane is down within the range of the
values found for the Y-radiation graft coatings, and its
contact angle is quite low (37). However, the damage to
the cornea endothelium produced by contact with urethane is
60% which matches that for PMMA. The rest of the materials
tested are hydrophobic, with contact angles as high as 105,
for teflon. The adhesive force values are also high, on the
same level as the PMMA, as are the extent of damage to the
endothelium, measured by SEM. The only exception to this is
the 40% damage assessment for teflon, which is the result of
Table 14. Summary of Adhesive Force Measurements for Cornea Endothelium Contact
with Various Biomedical Polymers.
Polymer
Average Adhesive
Force (mg/cin )
Standard ^
Deviation (mg/cm j
Number of
Tests
Urethane*
187
36
12
Silicone Rubber**
320
45
16
Thermoplastic
Elastomer***
499
211
22
Teflon****
268
93
10
*Pellthane TM, Upjohn Corporation, Kalamazoo, MI.
**Silastic 500-5TM, Dow-Corning, Midland, MI.
***Proprietary hydrocarbon block copolymer, C-flex TM, Concept Inc., Clearwater,
FL.
****TM, FEP, E.I. duPont de Nemours & Co., Wilmington, DE.
151
Table 15. Quantitative Cornea Endothelium Damage by SEM for Various Biomedical
Polymers.
Polymer
Cells Destroyed
Average (%)
in Contacted Area
Number of
Tests
Urethane*
60
2
Silicone Rubber**
85
2
Thermoplastic Elastoner***
65
2
Teflon****
40
1
*Pellthane TM, Upjohn Corporation, Kalamazoo, MI.
**Silastic 500-5TM, Dow-Corning, Midland, MI.
***Proprietary hydrocarbon block copolymer, C-flex TM, Concept Inc., Clearwater,
FL.
****TM, FEP, E.I. duPont de Nemours & Co., Wilmington, DE.
153
only one measurement, and is difficult to regard as
reliable.
In general, these materials still show the correlation
between contact angle, for water, and the biophysical
behavior, although the correlation is not nearly as good as
for the system of y-radiation graft coatings. Part of this
may be due to the difference in the mechanical properties of
these polymers and the sample form tested. The behavior in
the biophysical tests, which depended greatly on a sure
contact between the material and the endothelium surfaces,
could be affected by the difference in these materials'
ability to do so, relative to the PMMA and coated PMMA
surfaces..
4.11 Structure and Properties of Y-Radiation
Graft Coatings
Evaluation of the characterizations performed on the
Y-radiation graft coatings of PVP on PMMA suggests a
structure and an explanation for the non-adhesive properties
of the graft coating. However, further research is needed
to definitively establish the mechanism for the
tissue-materials adhesion phenomenon and its prevention.
Measurement of the external homopolymer molecular
weight formed during the Y-radiation graft coating reaction
indicates that the graft coating is composed of long (high
molecular weight) chains. In addition, the gelation of the
homopolymer at high radiation doses and N-VP monomer
concentrations suggests some crosslinking of the grafted
chains, even at lesser dose and monomer concentration
levels.
The Y-radiation graft coatings appear thin and adherent
to the acrylic substrate in SEM examination. In addition,
the majority of the surfaces are also smooth and uniform,
although holes and surface roughness are visible for some
coatings.
A structure for the PVP graft, based on its appearance
and physical and chemical properties is proposed in
Figure 48: the graft coating is composed of long molecules
forming a crosslinked network extending from the acrylic
surface. The density of the network decreases as the number
of chains decreases with the length of the chain and the
distance from the acrylic substrate. Although the %
hydration of the coatings has yet to be measured, the
hydrophilic-polymer coating probably contains a large amount
of water, particularly at the surface.
The proposed structure offers possible explanation of
the non-adhesive properties of the Y-radiation graft
coatings. Prevention of adhesion solely on the basis of a
hydrophilic surface was found to be an insufficient
explanation; other factors are involved. For example,
glass, generally considerd a hydrophilic surface, and
urethane (contact angle for water measured here of 37),
both produced extensive endothelium damage from adhesive
contact. Even the RF plasma coatings of PVP, which showed
increases in hydrophilicity and surface energy similar to
PMMA
SUBSTRATE
CROSS-LINKED
PVP GRAFT
COATING
SMOOTH HYDRATED PVP
'DIFFUSE' BOUNDARY
LAYER AT SURFACE
Figure 48. Schematic View of PVP YRadiation Graft on PMMA.
155
Table 16. Summary of Measurements for -Radiation Graft Coatings of PVP on PMMA.
Average
Average Adhesive
Average Cells
Polymer
M *
w
Contact Angle
() Force (mg/cni )
Destroyed (%)
PMMA
72
405
60
5%**/0.1 Mrad***
309/470
61
362
10%/0.1
693/473
57
179
45
20%/0.1
1,028.743
40
174
30%/0.1
1,307,795
30
108
13
0%/0.25
_
65
490
65
5%/0.25
577,992
59
156
63
10%/0.25
714,313
59
70
30
20%/0.25
1,371/339
44
105
17
30%/0.25
2,278,969
37
111
5
5%/0.5
46
209
_ _
10%/0.5
38
146
14
Molecular weight
of external
homopolymer by
intrinsic viscosity.
**Initial monomer concentration in polymerization media.
***Radiation dose in megarad (Mrad).
157
energy similar to the y-radiation graft coatings, exhibited
high values in adhesive force testing.
The proposed structure of the PVP y-radiation graft
coating explains the difference in biophysical behavior.
The. graft coating not only presents a "soft" deformable
hydrophilic surface, but also holds water in a "gel"
structure formed by the PVP. The non-adhesive properties of
this structure can be explained in several different ways.
The water held in the graft may be "squeezed-out" with
contact to a tissue surface; the water may then act as a
lubricating boundary layer. Additionally, the smooth and
mechanically-soft surface would act as a cushion to prevent
any mechanical damage.
Alternatively, the diffuse nature of the coating itself
may provide a highly hydrated layer at the surface which
acts much as the viscous PVP solutions used in the
preliminary experiments.
In either case, the hydrated coating acts to prevent
the hydrophobic and/or electrostatic interactions which
might bind contacting endothelium to the polymer surface.
Further investigation into the structure of the PVP
y-radiation graft coatings should offer important insights
into the mechanism of the tissue-materials adhesion
phenomenon.
5. CONCLUSIONS
1. It was discovered that momentary contact between an
endothelium tissue surface and a polymer surface results in
damage to the tissue by adhesion of the endothelium cells to
the contacting surface. The phenomenon was shown to be a
clinically significant problem in two currently used
surgical procedures: intraocular lens (IOL) insertion and
abdominal surgery.
2. The reduction and prevention of the damaging
adhesion phenomenon was shown through the use of hydrophilic
polymer coatings. Coatings in the form of viscous aqueous
solutions, cast films, or covalently bound grafts of
hydrophilic polymers were effective in acting as lubricants
and barriers to tissue-material adhesion.
3. A model system for the adhesion phenomenon was
established using rabbit cornea endothelium for the
evaluation of tissue-polymer adhesion and the efficacy of
hydrophilic polymer coatings in reducing this adhesion.
4. A new adhesive force instrument and measurement
technique were developed to quantify tissue adhesion
properties of materials under well controlled conditions.
Using scanning electron microscopy (SEM), the extent of
158
159
tissue damage produced by materials contact can be measured
quantitatively.
5. The tissue adhesion properties were measured for a
range of polymers. Values for the endothelium-polymer
2
adhesive force ranged from a high of 499 mg/cm for a
2
thermoplastic elastomer to 60 mg/cm for a Y-radiation graft
coating of poly(vinyl pyrrolidone) (PVP) on poly(methyl
methacrylate) (PMMA). Uncoated PMMA showed an adhesive
2
force of 405 mg/cm The extent of endothelium damage for
contact with these polymers was highest (85% of cells
destroyed in contact area) for silicone rubber and lowest
(5%) for a Y-radiation graft coating of PVP on PMMA. PMMA
contact produced 60% damage.
6. Gamma-radiation graft coatings of PVP on PMMA
intraocular lens material were prepared and characterized.
These coating were found to be thin (<1 y) and adherent to
the acrylic substrate and significantly altered the surface
of the acrylic, increasing hydrophilicity and surface
energy, and decreasing tissue adhesion and tissue damage.
7. A large reduction in adhesive force (from
2 2
405 mg/cm for PMMA to less 100 mg/cm ) and extent of
endothelium damage (from 60% to less than 45%) was exhibited
by Y-radiation graft coatings of PVP on PMMA which showed
appreciable graft polymerization (molecular weight of
external homopolymer formed of above 600,000) and a
reduction in the contact angle for water of more than 10.
160
8. Radiation dose and monomer concentration were
studied for the y-radiation graft coatings of PVP on PMMA.
These variables were shown to have a direct correlation on
the extent of endothelium damage. Increased radiation dose
or initial monomer concentration in the preparation of the
coating resulted in a reduction of endothelium damage.
9. Radio-frequency (RF) plasma coatings of PVP and
poly(hydroxyethyl methacrylate) (PHEMA) on PMMA were also
prepared and characterized. These coatings increased both
hydrophilicity and surface energy similar to the Y-radiation
graft coatings and also reduced both the adhesive force and
tissue damage in testing with contact to rabbit cornea
endothelium.
10. Testing of hydrophobic biomedical polymers,
silicone rubber, teflon and a thermoplastic elastomer, along
with the testing of the hydrophilic polymer coated acrylic
surfaces, suggested a general correlation between
hydrophilicity and biophysical behavior. Cornea endothelium
contact to the more hydrophobic surfaces showed greater
adhesiveness and resulted in greater amounts of tissue
damage.
11. Plasma and y-radiation graft coatings of PVP, with
approximately the same contact angles of water showed
different levels of adhesive force and tissue damage. The
nature of the y-radiation graft coatings, as seen in Fig
ure 48, may explain the difference.
161
The Yradiatin graft coatings may form a diffuse,
boundary layer which not only modified the surface by its
presence as a coating, but also by acting as a thin hydrated
gel, which when compressed will release water to act as a
lubricant and prevent adhesion. RF plasma coatings,
particularly those formed at high energy levels are very
heavily crosslinked and may not be able to act as a
lubricating layer, with a high water content as well as the
Y-radiation graft coatings.
12. The reduction of adhesive force and prevention of
tissue damage by the permanent graft coatings of hydrophilic
polymers prepared and characterized here show that safer
IOLs are possible. These thin and adherent coatings seem to
provide a barrier to adhesion with tissue surfaces, without
interfering with IOL function.
13. Preliminary studies conducted on peritoneal
adhesions from abdominal surgery and desquamation from
endotrachea tube intubation have shown additional areas
where the use of hydrophilic polymer coatings is
advantageous in preventing tissue damage. The permanent
hydrophilic polymer graft coatings developed here may be
suitable for use in these applications.
6. FUTURE RESEARCH
* Specific tasks for future research are listed.
6.1 y-Radiation Graft Coatings for IOL Use
A. Prepare y-radiation graft coatings of PVP on PMMA
at higher irradiation doses rates, with alternative solvent
systems (such as methanol or methanol/water) and with
complete de-gassing (remove oxygen) of polymerization
reaction vessel to determine effects on molecular weight and
structure of graft, coating thickness and surface character
(hydrophilicity and surface energy). Determine if these
graft polymerization conditions improve tissue adhesion
behavior.
B. Change monomer to prepare graft polymer surfaces of
different iongenicities (such as poly(hydroxyethyl
methacrylate) and poly(acrylamide) and poly(acrylic acid))
in order to establish effect on tissue adhesion properties.
C. In conjunction with the use of alternative solvents
and monomers, prepare coatings after limited diffusion of
monomer into the acrylic substrate to show effect of
penetration of graft coating on overall coating thickness,
hydrophilicity and surface energy and tissue adhesion
properties.
162
163
D. Conduct further characterization of PVP y-radiation
graft coating to aid in understanding of mechanism for
tissue-materials adhesion and its prevention.
1. Measure % hydration of the coated surfaces to find
correlation with coating thickness and with tissue
adhesion properties.
2. Use contact angle hysteresis as another measure of
hydrophilicity for characterizing graft coatings.
6.2 RF Plasma Coatings
A. Prepare coatings under systematic study of power,
vapor pressure and duration of exposure to determine effect
of each variable on extent and nature of graft coating, its
surface character and tissue adhesion properties.
B. Utilize flexibility of plasma polymerization with
use of other vapors (monomers such as HEMA and gases such as
oxygen and acetylene) to prepare coatings; characterize
surface and test tissue adhesion properties to determine
possible use in biomedical applications.
6.3 Further Characterization of Graft Coatings
A. Conduct tissue culture and implantation in cornea
stroma to test biocompatibility of graft coated surfaces.
B. Test mechanical properties of graft coatings:
abrasion resistence and long-term stability. Shows
permanence of the graft.
APPENDIX
INTRINSIC VISCOSITY MEASUREMENT
I. Sample Preparation
A. Concentration
C=0.5-1.0g/100ml
B. Enough sample to prepare, a concentration in above
range is weighted out exactly. The sample is then
dissolved in an appropriate solvent in a
volumetric flask. Because at least 8 ml is needed
for the measurement/ the volume of solution should
be greater than 10 ml (20-25 ml is recommended).
C. The solution should be filtered with a solvent-
resistant filter prior to measurement.
II. Thermostat for Constant Temperature Bath
A. The temperature should be set at the appropriate
temperature. Fluctuation of temperature should be
within 0.01C.
III. Viscometer
A.
It
should be kept clean and free from dust.
B.
Cleaning Procedure
1.
Chromerge cleaning solution can be used for
cleaning
2.
Rinse
3. -Dry with vacuum aspirator. Do Not Heat in
drying oven.
IV. Viscosity Measurement
A. Viscometer must be set exactly perpendicular.
B. Solvent Viscosity:
1. 10 ml of the chosen solvent is added to the
day viscometer
164
165
2. Solvent flow time should be measured after
equilibration of solvent temperature (at least
15 minutes)
3. Solvent flow time should be over 100 second.
If times are less than 100 seconds, then a
viscometer of a thinner capillary should be
used.
4. The flow time fluctuation of three serial
measurements should be within 0.1 seconds.
(If not, check for dust in the capillary and
temperature fluctuation in the constant
temperature bath).
C. Sample Viscosity
1. First step
a. 10 ml sample solution is added to the dry
viscometer.
b. Sample flow time should be measured after
temperature equilibration.
c. Three serial flow times should be
measured. Fluctuation must be within
0.1 seconds (if not see IV-B(4).
2. Second step
a. 3 ml of solvent is added to viscometer to
dilute sample solution. It should be
mixed completely with sample solution.
b. Prior to measurement, capillary should be
washed three times with the mixed
solution.
c. Repeat IV-C(l)b and IV-C(l)c.
3. Third step
a.Repeat IV-C(2).
4. Fourth step
a. Repeat IV-C(2).
D. Calculations
1. Definitions
a. tQ = solvent flow time
b. t? = sample flow time
c. n£el = relative viscosity
d. nsp = specific viscosity
t.
nrel =
o
e.
166
f.
g-
n
sp
Cn]
to.
- 1
= intrinsic viscosity
(I.V.)
Determination of I.V.
a. Use linear curve fitting to extrapolate to
zero.
2.
REFERENCES
1. Jaffe, N.S., Galin, M.A., Hirschman, H., and dayman,
. H.M., Pseudophakos, C.V. Mosby Company, St. Louis
(1978).
2. Shepard, D.S., The Intraocular Lens Manual, Astro
Printers, Santa Maria, California (1977).
3. Bourne, W.M., and Kaufman, H.E., Am. J. Ophthalmol.,
81, 482 (1976).
4. Forsot, S.L., Blackwell, W.L., Jaffe, N.S., and
Kaufman, H.E., Trans. Am. Acad. Ophthalmol.
Otolaryngol, 83, OP-195 (1977).
5. Katz, J., Kaufman, H.E., Goldberg, E.P., and Sheets,
J.W., Trans. Am. Acad. Ophthalmol. Otolaryngol., 83,
OP-204 (1977).
6. Kaufman, H.E., Katz, J., Valenti, J., Sheets, J.W., and
Goldberg, E.P., Science, 198, 525 (1977).
7. Sugar, A., Burnett, J., and Forsot, S.L., Arch.
Ophthalmol., 93, 449 (1978).
8. Irvine, R., Kratz, R.P., and O'Donnell, J.J., Arch.
Ophthalmol., 96, 1023 (1978).
9. Bourne, W.M., Brubaker, R.F., and O'Falon, W.M., Arch.
Ophthalmol., 97, 1473 (1979).
10. Levy, P.L., Roth, A.M., and Gangitano, J.L.,
Ophthalmology, j3j5, 219 (1979) .
11. Soil, D.B., Harrison, S.E., Arturi, F.C., and Clinch,
T., Am. Intra-Ocular Implant Soc. J., jj, 239 (1980).
12. Stegmann, R.R., and Miller, D., Ophth. Surg., 11, 19
(1980).
13. Kirk, S., Burde, R.M., and Waltman, S.R., Invest.
Ophthalmol. Vis. Sci., 16, 1053 (1977).
14. Miller, D., and Stegmann, R.R., Ann. Ophthalmol., 13,
811 (1981).
167
168
15.
Pape, L.G., and Balazs, E.P., Ophthalmology, 87,
(1980).
699
16.
Pape, L.G., Am. Intra-Ocular Implant Soc. J., 6,
(1980) .
342
17.
Peyman, G., and Zweig, K., Am. J. Ophthalmol., 87
(1979).
, 561
18.'
Knight, P.M., and Link, W.J., Am. Intra-Ocular Implant
Soc. J., 5, 123 (1979).
19.
Olson, R., Kolodner, H., Morgan, K.S., Escapini,
Sevel, D., and Kaufman, H.E., Arch. Ophthalmol.,
H.,
98,
1840 (1980).
20. Bruck, S.D., Trans. Am. Soc. Artif. Intern. Organs, 18,
1 (1972).
21. Weiss, L., and Blumenson, L.E., J. Cell Physiol., 70,
23 (1967).
22... George, J.N., Blood, 40, 862-874 (1972).
23. Kim, S.W., Lee, R.G., Oster, H., Coleman, D., Andrade,
J.N., Lentz, D.J., and Olsen, D., Trans. Am. Soc.
Artif. Intern. Organs, 20_, 449 (1974).
24. Ratner, B.D., Horbett, T., Hoffman, A.S., and Hauschkg,
S., J. Biomed. Mater. Res., S_, 407 (1975).
25. Brash, J.L., Brophy, J.M., and Fuerstein, I.A., J.
Biomed. Mater. Res., JU), 429 (1976).
26. National Heart, Lung and Blood Institute Working Group,
Guidelines for BloodMaterials Interactions, U.S.
Dept, of Health and Human Services, Washington, D.C.,
1980.
27. National Heart, Lung and Blood Institute Working Group,
Guidelines for Physiochemical Characterization of
Biomaterials, U.S. Dept, of Health and Human Services,
Washington, D.C., 1980.
28. Manly, R.S., ed., Adhesion in Biological Systems,
Academic Press, New York, 1970.
29. Berkeley, R.C.W., Lynch, J.M., Melling, J., Rutter,
P.R., and Vincent, B., eds., Microbial Adhesion to
Surfaces, Ellis Horwood Ltd., London 1980.
30. Kruse, P.F., and Patterson, M.K., eds., Tissue Culture
Methods and Applications, Academic Press, New York
1973.
169
31. James. R.A., and Kelin, E.E., J. Biomed. Mater. Res.,
5, 373 (1974).
32. Wang, P.Y., and Forrester, D.H., Trans. Am. Soc. Artif.
Organs, _20i, -504 (1974).
33. Grode, G.A., Falb, R.D., Cooper, C.W. and Lynn, J., in
Adhesion in Biological Systems, Manly, R.S., ed.,
Academic Press, New York, 1970, pg. 153.
34. Karlan, M.C., Skoble, B., Grizzard, M., Cassis, N.J.,
Singleton, G.T., Buscemi, P., and Goldberg, E.P.,
Otolaryngol. Head Neck Surg., ^8, 783 (1980).
35. Corpe, W.A., in Adhesion in Biological Systems,
Manly, R.S., ed., Academic Press, New York, 1970, .
pg. 74.
36. Taylor, A.C., in Adhesion in Biological Systems,
Manly, R.S., ed., Academic Press, New York, 1970,
pg. 51.
37. Weiss, L., Exp. Cell Res., Suppl. 8, 153 (1961).
38. Pethica, B.A., Exp. Cell Res., Suppl. 8, 123 (1961).
39. Weiss, L., in Adhesion in Biological Systems,
Manly, R.S., ed., Academic Press, New York, 1970. pg.
1.
40. Henglein, A., Schnabel, W., and Heine, K., Angnew.
Chem., 15, 461 (1958) .
41. Chpiro, A., Radiation Chemistry of Polymeric Systems,
John Wiley and Sons, Inc., New York, 1962, p. 667.
42. Yasuda, H.K., and Refojo, J., Polym. Sci.: Part A, 2,
5093 (1964).
43. Boffa, G.A., Lucien, N., Faure, A., Boffa, M.C.,
Jozefonvicz, J., Szubarga, A., Maudon, P., and Larrieu,
M.J., J. Biomed. Mater. Res., 11, 317 (1977).
44. Ratner, B.C., and Hoffman, A.S., J. Appl. Polym. Sci.,
18, 3183 (1974).
45. Ratner, B.D., and Hoffman, A.S., in Hydrogels for
Medical and Related Applications, Andrade, J.D., ed.,
American Chemical Society, Washington, D.C., 1976,
pg. 1.
46. Yasuda, H.K., in Plasma Polymerization, Shen, M. and
Bell, A.T., eds., American Chemical Society,
Washington, D.C., 1979, pg. 37.
170
47. Shen, M., and Bell, A.T., eds., Plasma Polymerization/
American Chemical Society, Washington, D.C., 1979.
48. Yasuda, H.K., Bumgarner, M.O., Marsh, H.C., Yamanashi,
B.S., Devito, D.P., Wolbarsht, M.L., Reed, J.W.,
Bessler, M., Landers, M.B., Hercules, D.M., and Carver,
J., J. Biomed. Mater. Res., _9, 629 (1975).
49.. Nichols, M.F., Hahn, A.W., Easley, J.R., and Mayhan,
K.G., J. Biomed. Mater. Res., 13^, 299 (1979).
50.. Chawla, A.S., Biomaterials, 2, 83 (1981).
51. Goldberg, E.P., Sheets, J.W., and Habal, M.B., Arch.
Surg., 115, 776 (1980).
52. Gabe, M., Histological Techniques, Masson and
SpringerVerlag, New York, 1976, pg. 652.
53. GAF Corporation, V-Pyrol, Author, New York.
54. Wilson, J.E., Radiation Chemistry of Monomers, Polymers
and Plastics, Marcel Dekker, Inc., New York, 1974,
pg. 111.
55. Brandrup, J., and Immergut, E.H., eds., Polymer
Handbook, John Wiley and Sons, Inc., New York, 1975,
pg. IV-1.
56. Zisman, W.A., in Contact Angle, Wettability and
Adhesion, Fowkes, F.M., ed., American Chemical Society,
Washington, D.C., 1964, pg. 1.
57. Jarvis, N.L., Fox, R.B., and Zisman, W.A., in Contact
Angle, Wettability and Adhesion, Fowkes, F.M., ed.,
American Chemical Society, Washington, D.C., 1964,
pg. 323.
58. Ballantine, D.S., Further Studies of the Effect of
Gamma Radiation on Vinyl Polymer Systems, Brookhaven
National Laboratory, Upton, New York, 1954.
59. Ballantine, D.S., and Manowitz, Status Report on the
Gamma Ray Initiated Polymerization of
N-Vinylpyrrolidone, Brookhaven National Laboratory,
Upton, New York, 1954.
60. Bazkin, A., and Lyman, D.J., J. Biomed. Mater. Res.,
14, 393 (1980) .
61. Andrade, J.D., King. R.N., Gregonis, D.E., and Coleman,
D.L., J. Polym. Sci. Polym. Symp., 66, 313 (1979).
171
62. Pouchert, C.J., Aldrich Library of Infared Spectra,
Aldrich Chemical Company, Milwaukee, Wisconsin, 1975.
63. Harrick, N.J., Internal Reflection Spectroscopy, John
Wiley and Sons, Inc., New York, 1967, pg. 284.
64. Kaneniwa, N., and Ikekawa, A., Chem. Pharm. Bull., 22,
2990 (1974).
65. Tanzawa, H., Nagaoka, S., Suzuki, J., Kobayashi, S.,
Masubuchi, Y., and Kikuchi, T., in Biomedical Polymers,
Goldberg, E.P. and Nakajima, A., eds., Academic Press,
New York, pg. 189.
BIOGRAPHICAL SKETCH
\ John Wesley Sheets, Jr., was born on September 17,
1953, in Jacksonville, Florida. He was educated in
Jacksonville and graduated in 1971 from S.W. Wolfson High
School.
Following graduation, the author attended Vanderbilt
University in Nashville, Tennessee, for two years. He then
spent the summer of 1973 at Jacksonville University,
Jacksonville, Florida, before entering the University of
Florida in September of 1973. He graduated with the degree
of Bachelor of Science in zoology in June of 1975. After
one year of post-baccalaurate studies in chemistry, he
entered graduate studies in the Department of Materials
Science and Engineering, University of Florida, in the
Spring of 1976.
While pursuing M.S. and Ph.D. studies at the University
of Florida, the author has served as a graduate research
associate. He is a member of the Society of the Sigma Xi,
Tau Beta Pi, Alpha Sigma Mu and the Society for Plastics
Engineers.
172
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
s J (. \
Eugene P. Goldberg, Chairman
Professor of Materials Science
and Engineering
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
( ^.^'i.afry
Professor/of Materials Science
and Engineering
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
George B. Butler
Professor of Chemistry
This dissertation was submitted to the Graduate Faculty of
the College of Engineering and to the Graduate Council, and
was accepted as partial fulfillment of the requirements for
the degree of Doctor of Philosophy.
April, 1983
Dean, College of Engineering
Dean for Graduate Studies and
Research
Internet Distribution Consent Agreement
In reference to the following dissertation:
AUTHOR: Sheets, John
TITLE: Hydrophilic polymer coatings to prevent tissue adhesion / (record
number: 473800)
PUBLICATION DATE: 1983
<^Ak\
, as copyright holder for the aforementioned dissertation,
hereby grant specific and limited archive and distribution rights to the Board of Trustees of the University of
Florida and its agents. I authorize the University of Florida to digitize and distribute the dissertation described
above for nonprofit, educational purposes via the Internet or successive technologies.
This is a non-exclusive grant of permissions for specific off-line and on-line uses for an indefinite term. Off-line
uses shall be limited to those specifically allowed by "Fair Use" as prescribed by the terms of United States
copyright legislation (cf, Title 17, U.S. Code) as well as to the maintenance and preservation of a digital archive
copy. Digitization allows the University of Florida to generate image- and text-based versions as appropriate
and to provide and enhance access using search software.
This/^rant of permissions prohibits use of the digitized versions for commercial use or profit.
Signature-of Copyright Holder
Printed draped Name of Copyright Holder/Licensee
Personal information blurred
iff 'll- ot
Date of Signature
Please print, sign and return to:
Cathleen Martyniak
UF Dissertation Project
Preservation Department
University of Florida Libraries
P.O. Box 117007
Gainesville, FL 32611-7007
3
170
47. Shen, M., and Bell, A.T., eds., Plasma Polymerization/
American Chemical Society, Washington, D.C., 1979.
48. Yasuda, H.K., Bumgarner, M.O., Marsh, H.C., Yamanashi,
B.S., Devito, D.P., Wolbarsht, M.L., Reed, J.W.,
Bessler, M., Landers, M.B., Hercules, D.M., and Carver,
J., J. Biomed. Mater. Res., _9, 629 (1975).
49.. Nichols, M.F., Hahn, A.W., Easley, J.R., and Mayhan,
K.G., J. Biomed. Mater. Res., 13^, 299 (1979).
50.. Chawla, A.S., Biomaterials, 2, 83 (1981).
51. Goldberg, E.P., Sheets, J.W., and Habal, M.B., Arch.
Surg., 115, 776 (1980).
52. Gabe, M., Histological Techniques, Masson and
SpringerVerlag, New York, 1976, pg. 652.
53. GAF Corporation, V-Pyrol, Author, New York.
54. Wilson, J.E., Radiation Chemistry of Monomers, Polymers
and Plastics, Marcel Dekker, Inc., New York, 1974,
pg. 111.
55. Brandrup, J., and Immergut, E.H., eds., Polymer
Handbook, John Wiley and Sons, Inc., New York, 1975,
pg. IV-1.
56. Zisman, W.A., in Contact Angle, Wettability and
Adhesion, Fowkes, F.M., ed., American Chemical Society,
Washington, D.C., 1964, pg. 1.
57. Jarvis, N.L., Fox, R.B., and Zisman, W.A., in Contact
Angle, Wettability and Adhesion, Fowkes, F.M., ed.,
American Chemical Society, Washington, D.C., 1964,
pg. 323.
58. Ballantine, D.S., Further Studies of the Effect of
Gamma Radiation on Vinyl Polymer Systems, Brookhaven
National Laboratory, Upton, New York, 1954.
59. Ballantine, D.S., and Manowitz, Status Report on the
Gamma Ray Initiated Polymerization of
N-Vinylpyrrolidone, Brookhaven National Laboratory,
Upton, New York, 1954.
60. Bazkin, A., and Lyman, D.J., J. Biomed. Mater. Res.,
14, 393 (1980) .
61. Andrade, J.D., King. R.N., Gregonis, D.E., and Coleman,
D.L., J. Polym. Sci. Polym. Symp., 66, 313 (1979).
27
of intra-abdominal adhesions was evaluated using a numerical
scale similar to that for the dog experiments.
3.2 PMMA Substrates
Acrylic test piece of several geometries were required
for the various methods of testing studied. Two test pieces
used most were "stubs" from 1/8" diameter rod, lathe cut to
1/8" length (PMMA from Modern Plastics, Orlando, Florida;
140,000 MW) and "slabs" (1/2" x 7/8" x 1/8"), saw cut from
Perspex CQ sheet (Medical grade PMMA from ICI, England;
2,400,000 MW). The stubs were used for characterization of
coatings by contact angle (for water), electron scattering
for chemical analysis (ESCA), examination by SEM, and
biophysical testing of adhesive force and SEM assessment of
damage from contact to cornea endothelium tissue. The slabs
were used in measurement of critical surface energy and
ultraviolet-visible (UV-Vis) and infrared (IR) spectroscopy.
Following lathe cutting to 1/8" lengths, the stubs were
individually end-polished in two steps. Coarse polishing
was done with water-wet 600 grit silica carbide paper
(Carbimet, Buehler Ltd., Evanston, ILL). A final dry
polish, on a napped cloth wheel (Microcloth, Buehler Ltd.)
using 40 y diamond paste (Astro-Met, GCA-Precision
Scientific, Chicago) was performed until visual inspection
revealed no scratches or surface irregularities. The
polishing was followed by an ultrasonic cleaning procedure:
five minutes in a soap water bath and then two five-minute
51
Figure 14. Loading of Acrylic-Endothelium Interface.
Copyright 1983
by
John Wesley Sheets, Jr.
6. FUTURE RESEARCH
* Specific tasks for future research are listed.
6.1 y-Radiation Graft Coatings for IOL Use
A. Prepare y-radiation graft coatings of PVP on PMMA
at higher irradiation doses rates, with alternative solvent
systems (such as methanol or methanol/water) and with
complete de-gassing (remove oxygen) of polymerization
reaction vessel to determine effects on molecular weight and
structure of graft, coating thickness and surface character
(hydrophilicity and surface energy). Determine if these
graft polymerization conditions improve tissue adhesion
behavior.
B. Change monomer to prepare graft polymer surfaces of
different iongenicities (such as poly(hydroxyethyl
methacrylate) and poly(acrylamide) and poly(acrylic acid))
in order to establish effect on tissue adhesion properties.
C. In conjunction with the use of alternative solvents
and monomers, prepare coatings after limited diffusion of
monomer into the acrylic substrate to show effect of
penetration of graft coating on overall coating thickness,
hydrophilicity and surface energy and tissue adhesion
properties.
162
57
acrylic surfaces dipped in balanced salt solution. Even a
one second contact produced extensive damage (approximately
20-25%). However, when the'acrylic surface was dipped in a
PVP solution (40 weight %), examination revealed almost no
damage to the endothelium, regardless of the time of
contact. Balanced salt solution would not wet the acrylic
surface while the PVP solution produced an adhesion
barrier-lubricating boundary layer.
4.1.2 Peritoneal Adhesions in Abdominal Surgery
These experiments evaluated the ability of solutions of
the hydrophilic polymers PVP and dextran to prevent
peritoneal adhesions resulting from trauma due to surgical
51
manipulations. Tables 1 and 2 show the results for the
canine and rat models, respectively.
A numerical rating scale was used to derive objective
data from subjective observations. Ratings were based upon
examination of the wound and small intestine and scored on
the.following scale:
0 No adhesions
1 Very slight evidence for adhesions
2 Slight to moderate number of adhesions
3 Moderate to extensive adhesions
4 Massive and extensive adhesions.
For rat experiments, the score for each animal
represented a consensus of 3 surgical observers. The dog
experiments were scored by 2 surgical observers. Scores
59
were averaged for each test group; the lowest average score
was assigned a value of 1.0 and higher scores assigned
appropriate relative values compared to the base low score.
In both the dog and rat experiments, the controls
showed moderate to extensive adhesions to the wound and of
the small intestine. Comparative results with PVP coating
were highly meaningful in both cases in that virtual
elimination of adhesions of the small intestine was
demonstrated. The dextran-treated dogs also showed some
reduction in wound adhesions. However, the saline-wet
control and the dextran-treated group both tended to have
less severe intestinal adhesions as compared with the dog
control group but were still much inferior to the
PVP-treated group.
The peritoneal adhesions were eliminated by the action
of the PVP solution to prevent serosal rubber-glove damage
during surgical manipulations. Rubbing of an intestinal-
surface was shown by SEM to produce considerable serosal
abrasion; virtually no tissue damage was observed following
rubbing by rubber gloves wet with PVP solution. As with the
acrylic-cornea endothelium contact, tissue damage was
eliminated by the interposition of nonadhesive, lubricating
boundary layer in the form of a surface-coating hydrophilic
polymer solution.
These experiments are significant in demonstrating for
the first time that surgical tissue damage and resulting
adhesions may be substantially reduced by application of
*
Table 7. Critical Surface Tension from Zisman Plots for y
Coatings of PVP on PMMA.
-Radiation Graft
% N-VP (Initial Concentration in
Polymerization Media)
Surface Energy
(Dynes/cm)
PMMA
38
0%
43
5
28
10
46
20
68
30
104
*Using water, formanide, glycerol, thiodiethanol and methylene iodide as
solvents. Instantaneous contact angles were measured.
00
U1
10
When the adhesive damage phenomenon was first observed,
we showed that aqueous solutions of the hydrophilic polymer
poly(vinyl pyrrolidone) (PVP) were effective in preventing
endothelium damage. Since the first publication of our work
on adhesive endothelium damage in 1977 (5,6), research in
many laboratories has resulted in several methods for
protecting the endothelium during IOL insertion.
Surgical techniques have been developed to minimize the
chance for damaging contacts between IOL and endothelium
surfaces. The most widely-used technique employs an air
bubble (9), which is formed and maintained in the anterior
chamber during insertion surgery to keep the endothelium
from collapsing against the lens being inserted. This
technique has been fairly successful but as with any
surgical' technique, it is not fool-proof. Another surgical
technique employed a soft contact lens shield to protect the
endothelium (10).
The other methods developed to protect the endothelium
were adaptations of our initial suggestion to use
hydrophilic polymer coatings. Bovine serum albumin,
chondroitin sulfate, human gamma globulin, hyaluronic acid
and whole plasma are among the different natural polymer
solutions studied as protective agents (11-16). Of these,
the sodium salt of hyaluronic acid has found extensive use,
and is marketed under the trademark, Healon (Pharmacia,
Piscataway, NJ) (14-16). More recently, chondroitin sulfate
solutions have been introduced commercially (Cilco,
Table 12. Summary of Adhesive Force Measurements for Cornea Endothelium Contact
with RF Plasma Coating on PMMA.
Plasma Conditions
Power */Pre s sure */Time * *
(Monomer****)
Average Adhesive
Force (mg/crn )
Standard ^
Deviation (mg/cin )
Number of
Tests
50W/1000y/10min.
(N-VP)
221
50
21
100W/200y/20min.
(N-VP)
513
245
9
35W/500y/60min.
(N-VP)
277
147
11
25W/500y/15min.
(HEMA)
221
87
10
*RF power to plasma (Watts).
**Pressure in reactor (Microns).
***Time of exposure to plasma.
****Monomer vapor used.
148
Figure
Page
16. External Polymer Viscosity Molecular Weight vs.
Monomer Concentration for y-Radiation Graft
Coatings of PVP on PMMA 67
17. Contact Angle for Water vs. % Monomer for
0.1 Mrad Dose y-Radiation Graft Coatings of PVP
and PMMA 73
18. Contact Angle for Water vs % Monomer for
. 0.25 Mrad Dose y-Radiation Graft Coatings of PVP
and PMMA . 74
19. Contact Angle for Water vs. % Monomer for
0.5 Mrad Dose y-Radiation Graft Coatings of PVP
and PMMA 75
20. Summary of Contact Angle for Water vs. % Monomer
for Various Dose Levels of y-Radiation Graft
Coatings of PVP on PMMA 76
21. Zisman Plot of Critical Surface Tension (y )
of PMMA ..?... 79
22. Zisman Plot of Critical Surface Tension (y )
of y-Irradiated 0.25 Mrad PMMA ? . 80
23. Zisman Plot for Critical Surface Tension (Y )
of y-Radiation Graft Coating: 5% N-VP Initial
Monomer Concentration and 0.25 Mrad Dose .... 81
24. Zisman Plot for Critical Surface Tension (y )
of y-Radiation Graft Coating: 10% N-VP Initial
Monomer Concentration and 0.25 Mrad Dose .... 82
25. Zisman Plot for Critical. Surf ace Tension (.Y )
of y-Radiation Graft Coating: 20% N-VP Initial
Monomer Concentration and 0.25 Mrad Dose .... 83
26. Zisman Plot for Critical Surface Tension (y )
of y-Radiation Graft Coating: 30% N-VP Initial
Monomer Concentration and 0.25 Mrad Dose .... 84
27. Zisman Plots for Critical Surface Tensions (y )
of 0.25 Mrad y-Radiation Graft Coatings of c
Different N-VP Concentrations 86
28. Summary of Zisman Plots for Critical Surface
Tensions (y ) of PVP on PMMA vs. % N-VP for
y-RadiationCGraft Coatings of 0.25 Mrad Dose . 87
x
109
33.29. 5% NVP Initial Monomer Concentration and
0.5 Mrad Dose.
Figure 33.30. 5% N-VP Initial Monomer Concentration and
0.5 Mrad Dose.
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
George B. Butler
Professor of Chemistry
This dissertation was submitted to the Graduate Faculty of
the College of Engineering and to the Graduate Council, and
was accepted as partial fulfillment of the requirements for
the degree of Doctor of Philosophy.
April, 1983
Dean, College of Engineering
Dean for Graduate Studies and
Research
Table 4. Contact Angle of Water on Y-Radiation Graft Coatings of PVP on PMMA.
Polymer
Average Contact Angle ()
Standard Deviation ()
PMMA*
72
8
5%**/0.1 Mrad***
61
3
10%/0.1
57
2
20%/0.1
40
3
30%/0.1
30
1
0%/0.25
65
2
5%/0.25
59
6
10%/0.25
59
7
20%/0.25
. 44
3
30%/0.25
37
5
5%/0.5
46
3
10%/0.5
38
7
*Uncoated; agrees with literature value of 78 (57).
**Initial monomer concentration in polymerization media.
***Radiation dose in megarad (Mrad).
3. MATERIALS AND METHODS
3.1 Preliminary Studies
Although the emphasis of this research has been on the
development and characterization of permanent hydrophilic
polymer graft coatings for intraocular lenses (IOLs),
preliminary experiments were conducted which first
demonstrated the tissue-materials adhesion phenomenon and
the effectiveness of hydrophilic polymer solutions to
prevent tissue damage. The two most significant experiments
are describe here and elsewhere (5,6,51); the remainder of
Part 3, deals with the subsequent research based on findings
of these initial experiments.
3,1.1 Prevention of Cornea Endothelium Damage by
Poly(vinyl pyrrolidone) Solutions
In these experiments, cornea endothelium from rabbit
and human eye bank eyes were contacted to acrylic
hemispheres to simulate IOL insertion (5). These
hemispheres were machined and highly polished to a radius of
curvature of 7.3 mm; their visually smooth surface and
curvature facilitated total contact between the acrylic and
endothelium surfaces.
22
49
then enucleated. The eyes were used less than two hours
postmortem.
For tissue damage measurements, the corneas were
removed from the eyes and placed in saline. Then an 8 mm
corneal trephine (Storz, St. Louis, MO) was used to punch
out a disc from the central portion of the cornea. This
cornea "button" was then glued, endothelial side up, to an
aluminum SEM stub with cyanoacrylate glue. A flat, uniform
surface of endothelium was formed that would allow precise
contact with a material surface. This sample was then
positioned and held within the saline-filled acrylic tank
for testing.
In measurement of the adhesive force, a sample holder
was constructed so that multiple (up to six) measurements
could be made from a single cornea. Corneas were removed
from the enucleated rabbit eyes and trimmed to leave a
narrow (1 mm) scleral rim. The corneas were then inverted
and placed on a supporting plastic hemisphere of matching
curvature. The cornea was held at the limbus with a metal
collar, secured with a screw plate. In this manner, a
"corneal hemisphere" was created; by holding the hemisphere
tilted within the acrylic tank, rotation of the hemisphere
on its axis would provide multiple contact points along a
latitudinal line. Contact was made normal to the surface by
the test piece held above.
Ill
Figure 33.33. 10% N-VP Initial Monomer Concentration and
0.5 Mrad Dose.
Figure 33.34. 10% N-VP Initial Monomer Concentration and
0.5 Mrad Dose.
Table 8. Summary of Zisman Plots for Critical Surface Tension of RF Plasma
Coatings on PMMA.
Plasma Conditions
Power*/Pressure**/Time***(Monomer)****
Surface Energy
(Dynes/cm)
PMMA
38
50W/1000U/10Min.
(N-VP)
46
100W/ 200y/20Min.
(N-VP)
53
35W/ 500u/60Min.
(N-VP)
50
25W/ 500P/15Min.
(HEMA)
50
*RF power to plasma.
**Pressure in reactor.
***Time of exposure to plasma.
****Monomer vapor used.
uncoated silicone rubber in biocompatibility studies done
with intramuscular implants of up to two years (49). Other
studies have shown that RF plasma polymerized films of
hexamethylcyclotrisiloxane on poly(propylene) and silicone
rubber membranes reduce the adhesion of platelets and
leucocytes in in vitro canine testing (50).
These different examples of the use of RF plasma
polymerized films to coat a variety of different materials
for a range of biomedical applications demonstrate the
usefulness of this technique in producing thin and uniform
coatings for biomedical applications. In this study, we
have shown that RF plasma grafting may be an alternative
method for producing hydrophilic polymer coatings on
intraocular lenses.
Wavelength |nm)
Figure 44. Ultraviolet-Visible Transmission Spectra of RF Plasma Coatings on PMMA.
135
Table 3. Viscosity Molecular Weight* of PVP.
External Polymer Formed in Solution During Y-Radiation Graft. Coating.
% N-VP (Initial Concentration in
Polymerization Media)/YRadiation Dose (Mrad)
M
W
5%/0.1
309,470
10%/0.1
693,473
20%/0.1
1,028,743
30%/0.1
1,307,795
5%/0.25
577,992
10%/0.25
714,313
20%/0.25
1,371,339
30%/0.25
2,278,969
The external polymers formed for 0.5 Mrad radiation doses were insoluble and
therefore unfit for intrinsic viscosity measurement.
Calculated by Mark-Houwink-Sakurda equation from intrinsic viscosity in
methanol at 30C.
141
490 r
o.
Figure 45. Adhesive Force vs. Monomer Concentration for
PVP y-Radiation Graft Coatings on PMMA.
117
Figure 34.7. RF Plasma Coating of N-VP on PMMA: 35 Watts
(Power)/5Q0y(Pressure)/60 Minutes(Duration).
Figure 34.8. RF Plasma Coating of N-VP on PMMA: 35 Watts
(Power)/500y(Pressure)/60 Minutes(Duration).
5
and tissue damage which may result in post-operative
complications. This study has opened up an interesting new
area of medical materials science that deserves further
study.
PMMA
SUBSTRATE
CROSS-LINKED
PVP GRAFT
COATING
SMOOTH HYDRATED PVP
'DIFFUSE' BOUNDARY
LAYER AT SURFACE
Figure 48. Schematic View of PVP YRadiation Graft on PMMA.
155
INTENSITY
NITROGEN
SPECTRA
(0 ANGLE)
B.E.
B. E.
N/C = 0.053 N/C = 0.078
Figure 37. ESCA Spectra for y-Radiation Graft Coating of PVP on PMMA: 10% N-VP Initial
Monomer Concentration and 0.5 Mrad Dose (B.E. = Binding Energy; N/C = Nitrogen
to Carbon Atomic Ratio).
INTENSITY
NITROGEN
SPECTRA
(0 ANGLE)
N/C = 0.020 N/C = 0.017
Figure 35. ESCA Spectra for PMMA (Uncoated) (B.E. = Binding Energy; N/C = Nitrogen to
Carbon Atomic Ratio).
123
88
90-95 dynes/cm and the dispersive contribution to be 25-39
dynes/cm for poly(hydroxyethyl methacrylate) surfaces (61).
Treatment of the data in this manner yields comparable
numbers for the polar contribution for the 20% and 30% N-VP
concentration graft coatings (about 120 dynes/cm.)/ and the
dispersive contribution is low, less than 5 dynes/cm.
However, the treatment of the data for the PMMA surface by
this method yields values for the two contributions which
are relatively lower (87 dynes/cm for the polar contribution
and 2 dynes/cm for the dispersive contribution), but are
unreasonably high for a PMMA surface.
It is clear that the polar interactions resulting from
the increased hydrophilicity of the graft coatings greatly
complicate the measurement and determination of the critical
surface tension for those surfaces. Both methods of
treating the data show that there is a relative increase in
the surface tension for the graft coated surfaces over the
critical surface tension for uncoated PMMA. Accurate
determination of the critical surface tensions for these
surfaces will require further investigation.
The complicating effects of these polar interactions
are also present in the critical surface tension
measurements for the RF plasma coatings, although the slopes
and linearity of the Zisman plots for these coatings are
normal and acceptable (Table 8) (Figures 29-32). The
critical surface tensions calculated by these plots (all
near 50 dynes/cm) show the increased surface tension for
cells from a rat fibroblast culture were detached from a
Pyrex glass substrate (37).
The importance of the forces which may be involved in
cell adhesion to a surface has been reviewed (38). Deemed
most important are chemical (electrostatic, covalent and
hydrogen) bonds and van der Waals1 interactions of the
London type (hydrophobic bonds) (39). Of these factors, the
hydrophobic-hydrophilic nature of the surfaces and the
resultant interactions with cells has received considerable
attention but there is no general agreement whether cells
adhere better to hydrophilic or hydrophobic surfaces.
However, in the examples of muscle cell adhesion to hydrogel
graft-coated surfaces (24) and with the strength of surgical
adhesives to liver tissue (32), an increased hydrophilicity,
relative to other surfaces, was found to favor reduced cell
adhesion.
2.3 Radiation Polymerized Graft Coatings
Permanent, covalently bound coatings of PVP on PMMA
were produced in this study by Y-irradiation of the
substrate immersed in aqueous solutions of the PVP monomer,
N-vinyl pyrrolidone (N-VP). Gamma-Radiation produces free
radicals in both the PMMA and the N-VP monomer, creating
graft sites in PMMA and initiating N-VP polymerization.
Free radicals in the bulk PMMA tend to degrade PMMA,
since scission predominates over crosslinking for PMMA.
However, free radicals produced on the surface of the PMMA
126
at the 5.8 micron and 6.0 micron peaks for PMMA and PVP,
respectively (Figure 38) (62). However, all efforts to
detect the PVP coating by a peak at that or any
characteristic wavelength were unsuccessful (Figure 39).
The thickness of a coating sufficient for detection on
an absorbing substrate has been determined by Harrick (63).
(sin^O -
where 6.^ is the thickness necessary for detection, is the
wavelength of the peak to be detected in the reflection
element, 0 is the incident angle for internal reflection and
n2^ is the ratio of the refractive index of the coating to
the refractive index of the internal reflection element.
For maximum sensitivity, the number of reflections
incident to sample surface was maximized. Use of internal
reflection elements at 60 and 45 provided 25 and 15
reflections, respectively. Using the above equation and the
conditions for detection of the 6.0 micron peak with KRS-5
(n=2.37) internal reflection elements at 60 and 45, the
coating thickness required for detection was calculated.
This calculation yielded thicknesses of 1.2 micron for the
60 and 2.8 microns for the 45 element. If the thickness
of the coatings as observed by SEM are accurate, their
thicknesses of less than a micron are indetectable by
4. RESULTS AND DISCUSSION
4,. 1 Preliminary Studies and Overview of Thesis Research
In the preliminary studies the tissue-materials
adhesion phenomenon was discovered and the use of
hydrophilic polymer solutions to reduce adhesion was
demonstrated (5/ 6, 51). The results of these and related
studies are presented and discussed here since they
established the research goals and plan for this
dissertation research.
4.1.1 Prevention of Cornea Endothelium Damage by Poly(vinyl
pyrollidone) Solutions
The discovery of the tissue-material adhesion
phenomenon dealt with the intraocular lens-cornea
endothelium interface (5). In these studies, rabbit, and
human cornea endothelium was examined by optical microscopy
with nitro-blue tetrazolium staining and scanning electron
microscopy to reveal damaged cells following contact to
poly(methyl methacrylate) (PMMA) surfaces, dipped in either
balanced salt solution or solutions of the hydrophilic
polymer, poly(vinyl pyrrolidone) (PVP).
Examination by both methods showed that 30-35% of the
endothelium cells were damaged by 60 second contact with
56
153
only one measurement, and is difficult to regard as
reliable.
In general, these materials still show the correlation
between contact angle, for water, and the biophysical
behavior, although the correlation is not nearly as good as
for the system of y-radiation graft coatings. Part of this
may be due to the difference in the mechanical properties of
these polymers and the sample form tested. The behavior in
the biophysical tests, which depended greatly on a sure
contact between the material and the endothelium surfaces,
could be affected by the difference in these materials'
ability to do so, relative to the PMMA and coated PMMA
surfaces..
4.11 Structure and Properties of Y-Radiation
Graft Coatings
Evaluation of the characterizations performed on the
Y-radiation graft coatings of PVP on PMMA suggests a
structure and an explanation for the non-adhesive properties
of the graft coating. However, further research is needed
to definitively establish the mechanism for the
tissue-materials adhesion phenomenon and its prevention.
Measurement of the external homopolymer molecular
weight formed during the Y-radiation graft coating reaction
indicates that the graft coating is composed of long (high
molecular weight) chains. In addition, the gelation of the
homopolymer at high radiation doses and N-VP monomer
concentrations suggests some crosslinking of the grafted
Distillation was conducted until a stable,
constant-boiling temperature was achieved typically at
66-80C depending on the vacuum achieved. The first 100 ml
of distillate collected after this point was discarded,
along with any previous distillate. The next 250 ml of
distillate was collected and preserved for use. The
remaining distillant was discarded and the system was
recharged for continued distillation following the same
procedure.
Purity of the distillate was checked by index of
refraction and infared spectroscopy. The index of
refraction was measured by an Abbe-type refractometer
(accuracy 0.0001; Fisher Scientific, Pittsburgh, PA) and
was found in all cases to be within 0.0001 of the literature
value of n^5 = 1.5120 (53).
Infrared spectra of the distilled N-VP were recorded
with a Perkin-Elmer Model 283B infrared spectrophotometer
under standard operating conditions (Figure 7). These
spectra matched the spectrum for N-vinyl pyrrolidone found
in the literature (Figure 7) (53).
The monomer was stored in sealed flasks at < 0C in a
freezer until use.
3.4 y-Radiation Graft Coatings
Gamma-Radiation graft polymerization coatings were
prepared by irradiation of the PMMA substrate in an aqueous
solution of the N-VP monomer with y-radiation from a 600
xml record header identifier oai:www.uflib.ufl.edu.ufdc:UF0008602400001datestamp 2009-02-16setSpec [UFDC_OAI_SET]metadata oai_dc:dc xmlns:oai_dc http:www.openarchives.orgOAI2.0oai_dc xmlns:dc http:purl.orgdcelements1.1 xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.openarchives.orgOAI2.0oai_dc.xsd dc:title Hydrophilic polymer coatings to prevent tissue adhesiondc:creator Sheets, John Wesleydc:publisher John Wesley Sheetsdc:date 1983dc:type Bookdc:identifier http://www.uflib.ufl.edu/ufdc/?b=UF00086024&v=0000111665750 (oclc)000473800 (alephbibnum)dc:source University of Floridadc:language English
serve as initiation sites for graft polymerization of the
N-VP (Figure 4).
Free radicals in the N-VP are created with opening of
the vinyl group double bond and homopolymerization of the
N-VP. The PVP chain will continue to grow until termination
occurs, and may react with the acrylic substrate, producing
the- graft polymer. The Y-radiation may also cause
crosslinking of the PVP, by hydrogen extraction and
branching of the PVP chain (Figure 5).
Graft polymerization by Y-radiation of PVP on PMMA was
first studied by Henglein et al. in 1958 (40). However,'
these graft coatings were made using the monomer in methanol
following long periods (up to 69 hours) of allowing
diffusion of monomer into the PMMA substrate (with swelling
of the PMMA). This process, along with the high radiation
doses (up to 4 megarads (Mrad)), resulted in severe
distortion of the PMMA (41). While such extreme conditions
would not be suitable for preserving the optical properties
of an acrylic IOL, some penetration of the N-VP into PMMA
may be of interest in future work.
Use of aqueous solutions of N-VP (0-20%) was shown to
reduce diffusion into the substrate when PVP was grafted to
poly(dimethylsiloxane), using high energy (3 MEV) and 1.5 to
4.5 Mrad dose, from a Van de Graff generator (42). These
coatings were shown to be hydrophilic in porportion to the
amount of graft. The irradiation conditions (high doses and
very high dose rates) produced thick coatings, several
112
Figure 33.35. 10% NVP Initial Monomer Concentration and
0.5 Mrad Dose.
62
PMMA is very difficult to modify by conventional
wet-chemistry; attempts in this laboratory to perform
hydrophilic surface modifications by such means were
unsuccessful. For example, the acrylic surface appeared
unltered by exposure to a boiling aqueous solution of
concentrated sodium hydroxide. Gamma-radiation graft
polymerization was chosen as a method to produce hydrophilic
surfaces by polymerization on the surface of the PMMA. This
investigation emphasized the use of y-radiation to form
graft coatings of PVP on the PMMA surface and so produce a
hydrophilic modification of the acrylic lens surface.
A study of radiation dose and monomer concentration was
undertaken for y-radiation graft polymerization of PVP on
PMMA. Low radiation doses (<0.5 megarad (Mrad)) and low
monomer concentrations were chosen to minimize PMMA
degradation and gelation in the external PVP homopolymer.
Distortion of the PMMA by solvent swelling was also
minimized by using monomer concentrations of < 30 volume %
in aqueous solution and immersion of the PMMA in the
polymerization medium only at the time of irradiation.
Hydrophilic surface modification of PMMA was also
performed by radio-frequency (RF) plasma coatings. These
coatings were produced from the monomers of PVP and
poly(hydroxyethyl methacrylate) (PHEMA) to show an
additional possible method of surface modification of
hydrophilic polymer graft coating.
4
graft polymerization. Both methods produced cross-linked,
covalently bound graft coatings on the PMMA substrate.
Polymerization conditions were investigated in detail for
y-radiation graft coating.
The chemical and physical nature of these coatings were
characterized. Contact angle measurements were used to
assess the relative hydrophilicity of the surface. Scanning
electron microscopy (SEM) was used to determine surface
morphology. ATR infrared spectroscopy and electron
scattering for chemical analysis (ESCA) were employed to
chemically analyze the surface composition.
A new instrument and technique devised by Dr. S. Reich
was built for the quantitative determination, for the first
time, of the polymer-tissue adhesive force. This method
used the rabbit cornea endothelium as a model endothelium
surface for quantitative measurement of the adhesive force
at the polymer-tissue interface under controlled conditions
of contact. This instrument was also used in conjunction
with SEM to assess the cell damage resulting from polymer
contacts. Adhesive force was correlated with cell damage
and with the chemical and physical nature of the coatings to
understand the adhesion phenomenon and to determine the
coating characteristics necessary to minimize tissue damage.
These studies have shown that the adhesion phenomenon
has broad implications in surgery, although the mechanism is
not fully understood. Hydrophilic polymer coatings for
surgical materials should prove useful to reduce adhesion
118
Figure 34.9. RF Plasma Coating of N-VP on PMMA: 35 Watts
(Power)/500y(Pressure)/60 Minutes (Duration).
Figure 34.10. RF Plasma Coating of N-VP on PMMA: 25 Watts
(Power)/500y(Pressure)/15 Minutes(Duration).
11
Huntington, WV) as an alternative to the high-priced and
hard to obtain Healon. However, a major disadvantage of
these solutions is that the surgeon is obligated to use an
additional procedure, application of a protective coating in
the operating room.
Dry, cast coatings of bovine submaxillary mucin (17)
and'poly(vinyl alcohol) (PVA) (18,19) have also been used;
these coatings hydrate and dissolve in situ. While these
transient and soluble coatings have proven effective for
reducing the adhesion and damage from corneal
endothelium-IOL contact, the protection is only temporary.
Damage from contact following the implant surgery, such as
contact occurring by IOL detachment, is not prevented by any
of the temporary coatings. The shortcomings of temporary
protective coatings demonstrate the need to develop
permanently bound coating which can be applied prior to
surgery and provide a protective, hydrophilic surface for
the life of the implant.
2.2 Biological Adhesion
Tissue-materials interface studies to date have
emphasized long-term polymer biocompatibility for
nonthrombogenic and tissue-compatible implants or prostheses
(20). Short-term toxicology and thrombogenicity as well as
long-term tissue acceptance have been major points of
concern (21-25). In the testing recommended for the
evaluation of materials intended for use in contact with
64
morphology by the graft coating. The examination by SEM
also was used to reveal the coating thickness, by scraping
the coating away and viewing the border of the scrape.
Three different spectroscopic techniques were used to
detect the presence of PVP graft coating on the PMMA
surface. Electron scattering for chemical analysis (ESCA)
and infrared spectroscopy (IR) were used for this purpose.
Ultraviolet-visible (UV-Vis) spectroscopy was used as a
possible means to show the coating presence but also to show
if the coating and the coating process had any effect on the
light transmissivity of the PMMA. Difficulties in detecting
the PVP by IR and UV-Vis lead to use of iodine to complex to
the PVP and enhance detection; the results of these
experiments will also be discussed.
In addition to the preparation and characterization of
hydrophilic polymer graft coatings on PMMA, an instrument
and technique for measurement of adhesive force and
endothelium damage were developed out of necessity; as
previously described, attempts to quantify tissue-materials
adhesion had been unsuccessful and such a means was
necessary for evaluation of the biophysical properties of
hydrophilic polymer graft coatings. The results of the
biophysical testing of adhesive force and quantitative
endothelium damage measurement of these materials will be
presented and the correlations to the other characteriza
tions described above will be discussed.
17
mils thick, but low molecular weight graft polymers (less
than 100,000) were found in the coatings.
PVP has also been grafted onto
poly(tetrafluoroethylene) (PTFE) using Y-radiation (43).
Here, the grafting conditions produced a heavy coating (over
70% weight gain for a 50 micron film) and hydrophilic
modification of the surface. Biological testing of the
graft coated surface showed a decrease in the adsorption of
fibrinogen and immunoglobulins from blood serum, but an
increase in albumin absorption.
Extensive research has been conducted by Ratner and
Hoffman on the use of Y-radiation to graft PVP and other
hydrophilic polymers onto silicone rubber (24). Aqueous
solutions of N-VP (20%) were irradiated to 0.25 Mrad using a
20,000 Curie Co-60 source and heavy graft coatings
2
(1.5 mg/cm ) of PVP were produced. These studies have also
shown the effect of low concentrations (0.005 Molar) of
cupric ion to minimize gelation in the external homopolymer
(44). Biological testing of these surfaces showed a
reduction in the adhesion of chick embryo muscle cells. The
desirability of this low cell adhesion and the aforemen
tioned reduction in protein adsorption has prompted studies
with radiation grafted hydrogels for a broad range of
biomedical applications (45).
The strategy for our use of y-radiation graft coatings
was to use low radiation doses (<0.5 Mrad) and low N-VP
monomer concentrations in order to minimize degradation of
the endothelium surface. Endothelium damage was then
estimated by observation with low-power optical microscopy.
Corneas were prepared for SEM by fixation in 2.5% cold
glutaraldehyde for one hour and osmium tetroxide for
90 minutes, both prepared with Millionig buffer. Corneas
were then dehydrated in serial baths of increasing
concentration of ethanol in water solutions. The samples
were then critical point dried using a Bomar SPC-900/EX
critical point apparatus (Bomar, Tacoma, WA) and then coated
with palladium-gold. Samples were then viewed with a Zeiss
Novoscan 30 scanning electron microscope (Carl Zeiss, Inc.,
New York).
3.1.2 Peritoneal Adhesions in Abdominal Surgery
The experimental plan was to perform exploratory
manipulations within the peritoneal cavity on rat and canine
animal models under simulated operating room conditions in
which the surgeon performs a clean, noncontaminating surgery
and yet obtains massive intra-abdominal adhesions, a major
post-operative complication (51). Such adhesions are
defined as the fibrous collagenous connective tissue that
develops after serosal trauma; in severe cases, these
adhesions can result in mechanical constriction of bowel
function, with potentially fatal consequences. Our intent
was to compare the conventional clinical technique with
procedures wherein hydrophilic polymer solutions of PVP and
168
15.
Pape, L.G., and Balazs, E.P., Ophthalmology, 87,
(1980).
699
16.
Pape, L.G., Am. Intra-Ocular Implant Soc. J., 6,
(1980) .
342
17.
Peyman, G., and Zweig, K., Am. J. Ophthalmol., 87
(1979).
, 561
18.'
Knight, P.M., and Link, W.J., Am. Intra-Ocular Implant
Soc. J., 5, 123 (1979).
19.
Olson, R., Kolodner, H., Morgan, K.S., Escapini,
Sevel, D., and Kaufman, H.E., Arch. Ophthalmol.,
H.,
98,
1840 (1980).
20. Bruck, S.D., Trans. Am. Soc. Artif. Intern. Organs, 18,
1 (1972).
21. Weiss, L., and Blumenson, L.E., J. Cell Physiol., 70,
23 (1967).
22... George, J.N., Blood, 40, 862-874 (1972).
23. Kim, S.W., Lee, R.G., Oster, H., Coleman, D., Andrade,
J.N., Lentz, D.J., and Olsen, D., Trans. Am. Soc.
Artif. Intern. Organs, 20_, 449 (1974).
24. Ratner, B.D., Horbett, T., Hoffman, A.S., and Hauschkg,
S., J. Biomed. Mater. Res., S_, 407 (1975).
25. Brash, J.L., Brophy, J.M., and Fuerstein, I.A., J.
Biomed. Mater. Res., JU), 429 (1976).
26. National Heart, Lung and Blood Institute Working Group,
Guidelines for BloodMaterials Interactions, U.S.
Dept, of Health and Human Services, Washington, D.C.,
1980.
27. National Heart, Lung and Blood Institute Working Group,
Guidelines for Physiochemical Characterization of
Biomaterials, U.S. Dept, of Health and Human Services,
Washington, D.C., 1980.
28. Manly, R.S., ed., Adhesion in Biological Systems,
Academic Press, New York, 1970.
29. Berkeley, R.C.W., Lynch, J.M., Melling, J., Rutter,
P.R., and Vincent, B., eds., Microbial Adhesion to
Surfaces, Ellis Horwood Ltd., London 1980.
30. Kruse, P.F., and Patterson, M.K., eds., Tissue Culture
Methods and Applications, Academic Press, New York
1973.
44
PVP as a measure of the amount of PVP present in the
coating.
3.14 Biophysical Measurements
*
3.14.1 Instrument for Biophysical Measurements
The instrument was constructed from a design developed
at the Weizmann Institute, Rehovot, Israel by visiting
Professor Dr. Shimon Reich. The instrument functioned to
allow both the measurement of adhesive force and to provide
quantification of damage from contact of endothelium to a
material's surface.
The instrument consisted of two parts: a mechanism
mounted on a scavenged microscope and a measuring microscope
with a micrometer-controlled crosshair, mounted in its
ocular. Figure 11 shows the mechanism mounted on the
scavenged microscope. A detailed description of the
instrument and its operation follows.
3.14.2 Preparation of the Material Sample for Measurement
Three forms of materials (coated and uncoated acrylic,
and other polymers) were tested, but the holder design
required a sample in the form of a 1/8" x 1/8" rod or
"stub." Acrylic stubs were prepared from rod as discussed
previously and were friction-fit into a conical holder at
the point of the cone (Figure 12). Materials not in
suitable rod form were cut from sheet into 1/8" discs and
B.E.
INTENSITY
NITROGEN
SPECTRA
(70 ANGLE)
B.E.
N/C = 0.049 N/C = 0.045
ESCA Spectra for y-Radiation Graft Coating of PVP on PMMA: 10% N-VP
Initial Monomer Concentration and 0.25 Mrad Dose (B.E. = Binding Energy
N/C = Nitrogen to Carbon Atomic Ratio).
Figure 36.
106
i J
15IW X2090^i ' 2jg^ 6 0IJ UFUSE
Figure 33.23. 20% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
43
reference beam attenuator, to compensate for the loss of
signal due to the ATR accessory.
Coated samples were air-dried and stored at ambient
conditions before spectra were obtained. For comparison of
spectra for these graft coated samples, cast films of PVP
were made and analyzed by ATR-IR. These films were made
from 5 and 10% solution of PVP (MW 40,000) (GAF, NYC) in
methanol with a doctor blade. These films were cast on the
acrylic "slabs" and were dried in a vacuum oven at 50C and
less than 150 torr prior to testing. The spectra were
analyzed as above, and were used to aid in resolving
spectral peaks and analysis of the spectra for the graft
coated samples.
3.12 Ultraviolet-Visible Spectroscopy
The ultraviolet-visible (UV-Vis) spectra of the PMMA
and the coated samples were taken from 900 to 190 nanometers
with a Model 552 Perkins-Elmer spectrophotometer
(Perkin-Elmer, Stanford, CN). In addition, the effect of
y-radiation alone was analyzed spectroscopically at
radiation doses of 0.25, 2.5 and 5.0 Mrad.
3.13 PVP-Iodine Interaction
Samples of PMMA and y-radiation graft coated PMMA were
equilibrated overnight in saturated solutions of iodine in
water. Both UV-Vis and ATR-IR spectra were obtained, as
previously described, to use the complexing of the iodine by
Cos Q
1.0
Figure 29. Zisman Plot for Critical Surface Tension (y ) of RF Plasma
Coatings of N-VP on PMMA: 50 Watts(Power)/I000p(Pressure)/
10 Minutes(Duration).
Table 11. Quantitative Cornea Endothelium Damage by SEM For Contact with
Y-Radiation Graft Coatings of PVP on PMMA. .
N-VP Monomer Concentration/
Y-Radiation Dose (Mrad)
Number of
Tests
Cells Destroyed in Contacted Area
Average(%) Standard Deviation(%)
PMMA
5
60
17
10%/0.1
4
45
13
30%/0.1
4
13
13
0%/0.25
3
65
31
5%/0.25
3
63
21
10%/0.25
4
30
20
20%/0.25
3
17
12
30%/0.25
3
5
5
10%/0.5
3
14
6
145
Table 5. Contact Angle of Water for RF Plasma Coatings on PMMA.
Plasma Conditions
Power*/Pressure**/Time***
(Monomer)****
Average Contact Angle ()
Standard Deviation ( )
PMMA
72
8
50W/1000y/10min.(N-VP)
40
6
100W/200y/20min.(N-VP)
26
9
35W/500y/60min.(N-VP)
60
3
25W/500y/15min.(HEMA)
42
14
*RF power to plasma (Watts).
**Pressure in reactor (Microns).
***Time of exposure to plasma.
****Monomer vapor used.
Wavenumber (cm*1)
4000 3000 2500 2000 150014001300 1200 1100 1000 950 900 850 800 750 700 650
Figure 8. Infrared Spectrum of N-VP Monomer (Literature, Reference 53).
Table 13. Quantitative Cornea Endothelium Damage by SEM for RF Plasma Coatings
on PMMA.
Plasma Conditions % Cells Destroyed in Contacted Area
Power*/Pressure**
/Time***(Monomer****)
Average (%)
Number of Tests
50W/1000y/10min.
(N-VP)
28
4
100W/200y/20min.
(N-VP)
61
3
35W/500V/60min.
(N-VP)
25
2
25W/500 y/15min.
(HEMA)
25
2
*RF power to plasma (Watts).
**Pressure in reactor (Microns).
***Time of exposure to plasma.
****ilonomer vapor used.
160
8. Radiation dose and monomer concentration were
studied for the y-radiation graft coatings of PVP on PMMA.
These variables were shown to have a direct correlation on
the extent of endothelium damage. Increased radiation dose
or initial monomer concentration in the preparation of the
coating resulted in a reduction of endothelium damage.
9. Radio-frequency (RF) plasma coatings of PVP and
poly(hydroxyethyl methacrylate) (PHEMA) on PMMA were also
prepared and characterized. These coatings increased both
hydrophilicity and surface energy similar to the Y-radiation
graft coatings and also reduced both the adhesive force and
tissue damage in testing with contact to rabbit cornea
endothelium.
10. Testing of hydrophobic biomedical polymers,
silicone rubber, teflon and a thermoplastic elastomer, along
with the testing of the hydrophilic polymer coated acrylic
surfaces, suggested a general correlation between
hydrophilicity and biophysical behavior. Cornea endothelium
contact to the more hydrophobic surfaces showed greater
adhesiveness and resulted in greater amounts of tissue
damage.
11. Plasma and y-radiation graft coatings of PVP, with
approximately the same contact angles of water showed
different levels of adhesive force and tissue damage. The
nature of the y-radiation graft coatings, as seen in Fig
ure 48, may explain the difference.
116
Figure 34.5. RF Plasma Coating of N-VP on PMMA: 1QQ Watts
(Power)/200y(Pressure)/20 Minutes(Duration).
Figure 34.6. RF Plasma Coating of N-VP on PMMA: 10Q Watts
(Power)/2QQj (Pressure)/2Q Minutes (Duration) .
highly excited state of the plasma. Here, the plasma
creates reactive species which may be only fragments of the
original monomer. These fragments then react to form a
polymer. The polymer may be formed from molecules which
would not normally react or polymerize. Polymers formed by
these reactions are usually highly cross-linked and are
difficult to characterize because of their complex
structures.
There are a number of advantages to the use of plasma
polymerization in the production of polymer coatings. Most
important is the ability to produce thin and uniform
coatings. Moreover, the coatings can be produced on a wide
range of substrates. By varying the coating conditions,
coatings with a wide range of properties can be produced
from a single starting material.
A large body of literature exists about the use of
plasma polymerization for the production of coatings (see
review by Shen and Bell) (47), and a number of studies are
O
relevant to this study. Thin (less than 1000A) coatings
have been produced on contact lenses by RF plasma
polymerization of a mixture of acetylene, water vapor and
nitrogen (48). The surfaces produced were shown to be
hydrophilic (contact angle for water at 38) and reduced
both protein adsorption and cell adhesion.
In other biomedical applications, RF plasma polymerized
films of poly(ethylene), poly(styrene) and poly(chloro-
trifluoroethylene) have been shown to produce no change from
3
endothelium cells. The resulting tissue damage may lead to
infections following catherization.
The adhesion phenomenon we have discovered appears to
occur extensively in all types of surgery and poses a
significant but yet unappreciated problem for instrument or
device materials in surgery. The results of this study
suggest that modification of the surfaces of biomedical
devices with hydrophilic polymer coatings can reduce or
prevent adhesive tissue damage for IOLs, catheters and
surgical gloves.
The objectives of this research have been to
investigate the adhesion phenomenon, and its prevention, and
especially to develop practical and clinically viable
permanent hydrophilic coatings for safer IOLs. The critical
nature of damage to the corneal endothelium resulting from
adhesion to an IOL and the large number of IOL insertions
currently performed (over 200,000 annually) underscore the
importance of research to modify the IOL surface.
Both temporary and permanent modification with
hydrophilic polymer coatings have been studied. Although
transient soluble polymer coatings were effective in
reducing tissue damage, we chose to emphasize permanent
hydrophilic graft coatings which would be inherently safer
and easier to use clinically. Gamma-radiation graft
coatings of poly(vinyl pyrrolidone) (PVP) on poly(methyl
methacrylate) (PMMA) were studied as well as hydrophilic
polymer coatings produced by radio-frequency (RF) plasma
TABLE OF CONTENTS
g*2_e.
ACKNOWLEGEMENTS iv
LIST OF TABLES . vii
LIST OF FIGURES ix
ABSTRACT xiii
1. INTRODUCTION 1
2. BACKGROUND 2
2.1 Intraocular Lens Development and Problems ... 2
2.2 Biological Adhesion 11
2.3 Radiation Polymerized Graft Coatings 13
2.4 Plasma Polymerized Graft Coatings 18
3. MATERIALS AND METHODS 22
3.1 Preliminary Studies 22
3.1.1 Prevention of Cornea Endothelium Damage
by Poly(vinyl pyrrolidone) Solutions 22
3.1.2 Peritoneal Adhesions in Abdominal
Surgery 24
3.2 PMMA Substrates 27
3.3 Purification of Monomer (N-VP) 28
3.4 y-Radiation Graft Coatings 29
3.5 Intrinsic Viscosity Molecular Weight of PVP . 34
3.6 RF Plasma Coatings 36
3.7 Contact Angle for Water 38
3.8 Critical Surface Tension ..... 39
3.9 Scanning Electron Microscopy 41
3.1Q ESC (Electron Scattering for Chemical
Analysis) 42
3.11 Infrared Spectroscopy 42
3.12 Ultraviolet-Visible Spectroscopy 43
3.13 PVP-Iodine Interaction 43
3.14 Biophysical Measurements 44
v
the PMMA substrate and crosslinking of the PVP. Swelling
and distortion of the substrate was also minimized by the
use of the low monomer concentrations in aqueous solution.
Diffusion of monomer into the acrylic was limited by
immersion of the substrate in the polymerization medium only
at the time of irradiation. The goal was production of very
thin, uniform adherent PVP coatings on the acrylic
substrate.
2.4 Plasma Polymerized Graft Coatings
In addition to Y-radiation grafting, plasma polymerized
coatings were studied. A plasma is an excited low pressure
gas created by radio-frequency (RF) discharge or a direct
current (DC) electrical glow discharge. The plasma produces
ions, electrons, ion radicals and other excited species
which will readily polymerize. Yasuda has classified the
polymerization reactions of monomers in a plasma into two
categories: plasma-induced polymerization and plasma state
polymerization (Figure 6) (46).
In a plasma-induced polymerization, the vapor of a
monomer is excited to the form in which it would polymerize
by conventional polymerization, maintaining its original
structure. For example, the plasma serves to open the
double bond of the vinyl group in N-VP and promote
conventional radical polymerization.
Plasma state polymerization does not follow
conventional polymerization kinetics and relies on the
97
Figure 33.5. 5% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
100
< s
5 l *
' t
£
0
: .1
# f.
4 >
4 /
- :
f ,
Ii
si
f i
10.0U UFMSE
15KU X2000 2001
Figure 33.11. 20% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
Figure 33.12. 20% N-VP Initial Monomer Concentration and
0.1 Mrad Dose.
54
2
contact over the entire 0.079 cm of the test pieces, the
2
adhesive force in terms of milligrams/cm was calculated.
3.14.5 Quantitative Cornea Endothelium Damage by SEM
The procedure for measuring the tissue damage resulting
from a material contact began by contacting the material
sample with the endothelium just as iri the measurement of
adhesive force. However, only one contact was made on each
tissue sample and following that single touch, the tissue
was removed from the instrument. The measurement made by
the micrometer crosshair was noted, but was used only to
confirm that proper contact had been made.
The tissue sample was then immediately prepared for
scanning electron microscopy, while still glued to the
aluminum stub. The tissue was fixed for 1-3 hours in 2.5%
glutaraldehyde in Millionig buffer, and was then placed in a
buffer rinse for equal period of time. The tissue was then
dehydrated by a series of increasingly concentrated
ethanol-water baths. The samples were held in baths of 50,
70, 80, 90, 95 and 100 percent ethanol solutions for
ten minutes each. Following an additional ten minutes fresh
ethanol, the samples were then critical point dried using a
Tousimis Samdri PVT-3 (Tousimis, Rockville, MD).
The tissue specimens were then reglued to SEM stubs
with colloidal carbon paint, and after sufficient drying,
O
the samples were then sputter-coated with 200 A of
26
exteriorization of bowel, bimanual palpation, replacement of
bowel, and the abdominal wall was closed in layers.
Experimental animals were housed separately and fed routine
diets. Surviving animals were sacrificed and examined
8 days after initial surgery and adhesions of the wound and
small intestine graded on a numerical scale based upon
degree and extent of adhesion bands.'
Additional experiments using rats were conducted for
futher evaluation of PVP solution. Twenty female
Sprague-Dawley rats (185-200 g) were anesthetized with
pentobarbital and divided into 2 groups. For the control
group of 10, the abdominal wall was opened and all
intra-abdominal organs bimanually examined. Exteriorization
was necessary to simulate clinical conditions for
exploratory surgeries. Abdominal contents were exposed,
gently explored and manipulated using dry gloves and sponges
for 5-10 minutes, the abdominal wall was closed in layers,
and the incision sutured. For the second group, the
procedure was repeated except that immediately after the
abdominal wall incision, 5 ml of 25 weight % PVP solution in
saline was sprayed into the abdominal cavity and all
manipulations were performed using gloves and sponges wet
with the PVP solution. Just before closure, an additional
5 ml of PVP solution was sprayed over the exposed organs.
Rats were maintained in individual cages for observation.
Eight days after surgery surviving rats were sacrificed and
intestines and organs examined for adhesions. The presence
Figure
Page
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
Zisman Plot for Critical Surface Tension (y )
of RF Plasma Coating of N-VP on PMMA:
50 Watts(Power)/1000y(Pressure)/10 Minutes
(Duration)
Zisman Plot for Critical Surface Tension (y )
of RF Plasma Coating of N-VP on PMMA: c
100 Watts(Power)/20Oy(Pressure)/20 Minutes
(Duration)
Zisman Plot for Critical Surface Tension (y j
of RF Plasma Coating of N-VP on PMMA:
35 Watts(Power)/500y(Pressure)/60 Minutes
(Duration)
Zisman Plot for Critical Surface Tension (y )
of RF Plasma Coating of HEMA on PMMA:
25 Watts(Power)/500y(Pressure)/15 Minutes
(Duration)
89
90
91
92
Scanning Electron Micrographs of Scraped
y-Radiation Graft Coatings of PVP on PMMA .... 95
Scanning Electron Micrographs of Scraped
RF Plasma Coatings on PMMA 114
ESCA Spectra for PMMA (Uncoated) (B.E. = Bonding
Energy; N/C = Nitrogen to Carbon Atomic Ratio) 123
ESCA Spectra for y-Radiation Graft Coating of
PVP on PMMA: 10% N-VP Initial Monomer Concen
tration and Q.25 Mrad Dose (B.E. = Binding Energy;
N/C = Nitrogen to Carbon Atomic Ratio) 124
ESCA Spectra for y-Radiation Graft Coating of
PVP on PMMA: 10% N-VP Initial Monomer Concen
tration and 0.5 Mrad Dose (B.E. = Binding Energy;
N/C = Nitrogen to Carbon Atomic Ratio) 125
Infrared Spectra for PMMA and PVP (Literature,
Reference 62) 127
Attenuated Total Response-Infrared Spectra of
PMMA, y-Radiation Graft Coating of PVP on PMMA
and PVP Cast Film on PMMA 128
Ultraviolet-Visible Transmission Spectra of y-
Radiation Graft Coatings of PVP on PMMA, 0.1 Mrad
Dose level 130
xi
Adjusted Support
Stage for
Tripod Weight
Tripod Weight
Glass Fiber
Material Holder
Transparent
Plexiglass
Tank
Cornea
Endothelium
Microscope
Stage
Water Immersion Test Cell
Cornea Endothelium
Polymer Sample
Rack and Pinion
Micrometer Staqe
(for lowering and raising
tissue sample tank)
Lateral View of Section (A)
X-Y Micrometer Stage
(for centering of material
sample over endothelium)
Microscope Base
Tissue-Polymer Adhesion Measurement Instrument.
Figure 11
Ln
Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
HYDROPHILIC POLYMER COATINGS TO PREVENT TISSUE ADHESION
By
John Wesley Sheets, Jr.
April 1983
Chairman: Eugene P. Goldberg
Major Department: Materials Science and Engineering
Adhesion following contact between plastic surgical
materials and endothelium surfaces has been shown to result
in extensive and critical tissue damage in intraocular lens
(IOL) insertion and abdominal surgery. Coatings of
hydrophilic polymer solutions were shown effective in
reducing such damage by acting as lubricants and barriers to
adhesion. The primary emphasis of this study was on the
preparation, characterization and evaluation of permanent
graft coatings of hydrophilic polymers for the prevention of
IOL adhesion to the cornea endothelium because of the
importance of this adhesion problem in IOL surgery.
Gamma-radiation and radio-frequency (RF) plasma
polymerized graft coatings of the hydrophilic polymer,
poly(vinyl pyrrolidone) (PVP) were prepared on the IOL
material, poly(methyl methacrylate) (PMMA). The chemical
and physical properties of these coatings were characterized
xiii
39
coatings' surfaces. These measurements were made with a NRL
contact angle goniometer (Rame'-Hart, Mountain Lakes, NJ) at
ambient temperature and humidity (usually % 23C and 60%
relative humidity).
The Y-radiation and RF plasma graft coated and uncoated
acrylic samples were equilibrated overnight in distilled
water; water droplets adhering to the surface were shaken
off prior to testing. All other materials tested were used
as received.
Using a Gilmont micrometer syringe (Gilmont, Great
Neck, NY), a 0.002 ml drop of freshly distilled water was
deposited on the plastic surface. The contact angle on each
side of the drop was measured to the nearest degree and
recorded. The surface was then dried by touching a small
piece of filter paper to the periphery of the drop; any
remaining water was blown from the surface with compressed
gas (Manostat, NYC). Five drops were used on each sample,
yielding ten measurements of the contact angle.
3.8 Critical Surface Tension
The critical surface tensions of both the y-radiation
graft coatings and the RF plasma coatings were determined by
the construction of a Zisman plot of the cosine of the
contact angle for liquids of different surface tensions
versus the surface tensions of the liquids (56). The
intercept at cosine of the contact angle equal to one (as
determined with a best-fit, linear plot, using a HP-35c
120
the RF plasma coatings. The RF plasma coatings showed
different mechanical properties and did not seem to be
totally scraped away also because the scrape was not clearly
defined. The lack of a clearly defined edge made judgement
of the coating thickness difficult, but seem to be on the
same order as the y-radiation graft coating thicknesses of a
micton or less.
4.6 ESCA
Surface analysis of the y-radiation graft coatings by
ESCA shows the presence of the PVP on their surfaces.
Nitrogen, present in the pyrrolidone ring of PVP, can be
detected at significant levels on these coated surfaces,
relative to a lack of nitrogen on uncoated PMMA.
Table 9 shows the results for the ESCA analysis of an
uncoated PMMA sample and two y-radiation graft coatings of
PVP on PMMA, made under the following conditions: 10% N-VP
concentration with 0.25 Mrad y-radiation dose and 10% N-VP
with 0.5 Mrad y-radiation dose. Two angles of incident
radiation were used in the ESCA to sample the surfaces: a
O
0 angle sampled at a mean depth of 50A and a 70 angle
O
sampled at a mean depth of 25A. The data are expressed in
terms of atomic ratio relative to carbon.
The data show trace amounts of silicon and chlorine,
which were contaminants of unknown origin. The copper, also
detected only at trace levels, comes from the copper wire
used as a sample holder. The 0/C ratio, showing the amount
of oxygen present, will be considered in this discussion as
5. CONCLUSIONS
1. It was discovered that momentary contact between an
endothelium tissue surface and a polymer surface results in
damage to the tissue by adhesion of the endothelium cells to
the contacting surface. The phenomenon was shown to be a
clinically significant problem in two currently used
surgical procedures: intraocular lens (IOL) insertion and
abdominal surgery.
2. The reduction and prevention of the damaging
adhesion phenomenon was shown through the use of hydrophilic
polymer coatings. Coatings in the form of viscous aqueous
solutions, cast films, or covalently bound grafts of
hydrophilic polymers were effective in acting as lubricants
and barriers to tissue-material adhesion.
3. A model system for the adhesion phenomenon was
established using rabbit cornea endothelium for the
evaluation of tissue-polymer adhesion and the efficacy of
hydrophilic polymer coatings in reducing this adhesion.
4. A new adhesive force instrument and measurement
technique were developed to quantify tissue adhesion
properties of materials under well controlled conditions.
Using scanning electron microscopy (SEM), the extent of
158
165
2. Solvent flow time should be measured after
equilibration of solvent temperature (at least
15 minutes)
3. Solvent flow time should be over 100 second.
If times are less than 100 seconds, then a
viscometer of a thinner capillary should be
used.
4. The flow time fluctuation of three serial
measurements should be within 0.1 seconds.
(If not, check for dust in the capillary and
temperature fluctuation in the constant
temperature bath).
C. Sample Viscosity
1. First step
a. 10 ml sample solution is added to the dry
viscometer.
b. Sample flow time should be measured after
temperature equilibration.
c. Three serial flow times should be
measured. Fluctuation must be within
0.1 seconds (if not see IV-B(4).
2. Second step
a. 3 ml of solvent is added to viscometer to
dilute sample solution. It should be
mixed completely with sample solution.
b. Prior to measurement, capillary should be
washed three times with the mixed
solution.
c. Repeat IV-C(l)b and IV-C(l)c.
3. Third step
a.Repeat IV-C(2).
4. Fourth step
a. Repeat IV-C(2).
D. Calculations
1. Definitions
a. tQ = solvent flow time
b. t? = sample flow time
c. n£el = relative viscosity
d. nsp = specific viscosity
t.
nrel =
o
e.
32
Curie Co-60 source (Department of Radiation Biology, Univer
sity of Florida) (Figure 9). Radiation doses used were from
0.1 to 0.5 megarads (Mrad), as measured by Fricke dosimetry,
(54) performed by the Department of Radiation Biology. To
minimize the irradiation times, and also maximize the number
of samples irradiated, the irradiations were usually made at
a distance of two inches from the source. The dose rate at
this distance was 1360 rads per minute.
Solutions of N-VP in water were prepared volumetrically
using freshly distilled water and distilled monomer (pre
pared as previously described). The solutions were prepared
immediately prior to irradiation and were not degassed.
Different irradiation containers were used for the two
kinds of PMMA substrates which were coated. The short rod
"stubs" were held in a 1/8" teflon sheet which fit in the
slots of a Coplin staining jar (Volume 90 ml; Fisher
Scientific, Pittsburgh); the "stubs" were held such that
both ends were exposed to the polymerization media. The
rectangular PMMA "slabs" were placed for coating in
borosilicate glass test tubes (16 x 125 mm, volume ^ 18 ml)
(Fisher Scientific, Pittsburgh). The test tubes were then
placed in a specially constructed carousel which held the
tubes upright at a distance of two inches from the source
during the irradiation. The Coplin jars were free-standing
within the Co-60 source and were placed so that the source
would be two inches from the PMMA "stubs." In all cases,
113
PMMA substrate is not visiblesome of the coating still
covers the substrate surface (Figure 33.11, 33.12, 33.14,
33.15, 33.23, 33.24, 33.26 and 33.27).
The strong adherence of the Y-radiation graft coatings
is also evident in the examination of the 0.5 Mrad coatings
(Figure 33.28 and 33.25). For these coatings, at both the
5% and 10% N-VP concentrations, the scraping procedure would
not remove the coating but only "scuffed" it. The gelation,
which occurred in polymerization media at higher N-VP -
concentrations for the 0.5 Mrad dose level,* suggests that
these coatings were highly crosslinked and probably would be
more difficult to scrape away. The "scuffing" behavior
seems likely for a coating which is still relatively thin,
but both highly coherent and adherent.
The coatings prepared by RF plasma polymerization also
appear to be very adherent to the PMMA substrate (Fig
ure 34). The polishing scratches visible on the PMMA
substrate where the coating has been scraped away, as on the
5% N-VP concentration and 0.1 Mrad radiation graft coating
(Figure 33.4-33.6), are not apparent on the scraped areas of
*As well as the insolubility in methanol of the homopolymer
precipitated from the polymerization media for these
coatings.
136
absorbing PMMA substrate has been shown in this research to
mask the detection of the PVP alone. Iodine broadens the
peak at 6.0 microns, which unfortunately coincides with a
PMMA peak, so the effect of the iodine is lost by this
interference.
Changes in the ultraviolet-visible spectrum are also
not apparent for the samples of either y-radiation graft or
RF plasma coatings of PVP on PMMA. The UV-Vis spectrum of
the iodine alone shows absorption peaks at 450 and 350
nanometers,* but these peaks are not shown on the spectra
obtained for the coated samples equilibrated with the
saturated iodine solutions. The explanation for the lack of
these peaks lies either in the amount of PVP present in the
coating or in the amount of iodine absorbed. Since other
measurements have shown the coatings to be less than a
micron thick, the lack of sufficient coating for complexing
a detectable amount of iodine seems that most likely reason.
4.10 Biophysical Measurements
The development of the instrument and technique capable
of producing quantitative measurements of the contact
adhesion phenomenon was an important step toward
understanding its nature and mechanism. Other research has
examined the adhesive nature of materials to cells in
*Peaks at shorter wavelengths could not be seen due to the
absorptivity of PMMA below 300 nm.
40
calculator (Hewlett Packard, Corvallis, OR)) was the
critical surface tension for that surface.
The liquids used were selected from a wide range of
liquids used for such measurements (57). The liquids used
were chosen on the basis of their being non-solvents for
both PMMA and PVP and their non-spreading behavior
(formation of a finite contact angle) on both polymers'
surfaces. The value of the surface tensions for these
liquids was taken as the value for 20C, although the
measurements were made at ambient conditions. This was
possible since the change in surface tension with
temperature for organic liquids is only ^0.1 dynes/cm/C.
The five liquids used and their surface tensions are water
(72.6 dynes/cm), glycerol (63.4 dynes/cm), formamide
(58.2 dynes/cm), 2.2'-thiodiethanol (54.0 dynes/cm), and
2
methylene iodide (58.2 dynes/cm).
The larger surface area acrylic slabs were used for
these measurements; the sample was stored at ambient
conditions prior to testing. The contact angle measurement
for each liquid was made according to the same procedure as
for the contact angle for water measurements, except two
drops, yielding four angles for measurement, were used for
each liquid. A fresh, clean area of the sample surface was
used for each drop.
Following the measurements of these angles for the five
liquids, the average for each liquid was calculated. These
values, along with the surface tensions of the five liquids
38
Boulder, CO), and maintained until residual water vapor was
eliminated from the system.
Monomer vapor from monomer held in a 20 ml long-necked
Erlenmeyer flask was then bled into the system and adjusted
by a micro-metering valve (SS-22RS4, Whitey Co., Highland
Heights, OH) until the desired pressure was achieved. The
monomers, N-VP (Eastman Kodak Chemicals, Rochester, NY) and
HEMA (Aldrich Chemical Co., Milwaukee, WI) were used as
received. The monomer vapor available was sufficiently
pure, and the monomers, as received, contained inhibitors
which prevented polymerization when the monomer was heated
(by a hot air gun) to achieve the desired vapor pressures.
After the desired pressure was achieved, the RF power
was switched on and adjusted to the selected power level, as
monitored by the SWR meter. Monomer vapor pressure and RF
power were maintained at constant levels over the duration
of the plasma polymerization and deposition of coating.
Following the completion of the coating procedure, the
chamber was opened to the atmosphere, and the samples were
cleaned by rinsing with distilled water. The samples were
stored with the side toward the plasma generation region
facing up; testing and characterization was performed only
on this face of the sample.
3.7 Contact Angle for Water
Contact angle measurements were made to assess the
relative water wettability of the different materials' and
Contact Angle ()
S % N*VP (Initial Concentration in Polymerization Media)
a.
Figure 17. Contact Angle for Water vs. % Monomer for
0.1 Mrad Dose y-Radiation Graft Coatings of
PVP on PMMA.
74
E % IM-VP (Initial Concentration in Polymerization Media)
Figure 18. Contact Angle for Water vs. % Monomer for
0.25 Mrad Dose y-Radiation Graft Coatings of
PVP on PMMA.
146
the data is only complete for the 0.25 Mrad series.
Figure 47 shows that the decline follows the increasing
monomer concentration, just as the adhesive force declined
and the hydrophilicity increased.
Table 11 also shows that a radiation dose of 0.25 Mrad
produces no significant change in the contact adhesion
behavior of PMMA, as assessed by endothelium damage. Taken
with the other information available for the material
irradiated in water, the radiation has no appreciable effect
on either the physical or biophysical behavior of the PMMA
at such a low radiation dose level.
The correlations between contact angle, the adhesive
force and extent of endothelium damage are not as clear for
the RF plasma coatings as for the Y-radiation graft
coatings. All of the surfaces produced by RF plasma coating
show a reduction in the contact angle for water (Table 5).
However, there is no accompanying decrease in the adhesive
force to the cornea endothelium, as shown with the
Eradiation graft coatings, although the quantitative
measurement of cornea endothelium damage does show a
decrease (Tables 12 and 13). The correlation between
contact angle for water and the extent of cell destruction
is not present: the RF plasma coating of the lowest contact
angle for water (26), shows no change in the percent of
cell destroyed from PMMA. The RF plasma coating of PVP
which showed the greatest reduction in the cell destruction
138
destruction leads to the other complications from the
contact.
Only one such measurement could be made from each
cornear placing a limitation on the number of measurements
that could be practically made. Even with the limited
number of tests for a given surface, the standard deviations
were reasonable, considering the type of test and were, for
the most part, approximately 15%.
For all of the biophysical testing, the baselines for
comparison are the data for PMMA. The greatest number of
tests were made on the PMMA, because during the development
and refinement of the adhesive force measurement technique,
each cornea was tested with an experimental coating and then
three control measurements were made with PMMA. After the
uniformity in the results for the PMMA was assured by
testing in this manner over a period, only random sampling
was done to provide a control measurement. Comparison to
PMMA is valuable in that the PMMA is the material currently
being used. For an alternative material to be chosen, its
biophysical behavior must be more favorable than that of the
PMMA.
In the adhesive force testing, PMMA shows an average
2
adhesive force of 405 mg/cm The values for the
Y-radiation graft coatings are shown in Table 10 and all
show a reduction in the adhesive force below that value.
The sample which was given a radiation dose of 0.25 Mrad
while in water had an average value of 490 mg/cm but this
52
Figure 15. Deflection of Glass Fiber due to Acrylic-
Endothelium Adhesion with Lowering of
Micrometer Stage.
68
radiation dose and concentration (59). Experiments using
cupric and ferrous ions as chain transfer agents to reduce
gelation (as reported (44)) were performed but not pursued
because of possible toxicity of the cupric ion and
difficulty in removing the ferrousion.
4.3 Contact Angle for Water
The measurement of the contact angle of water for the
different surfaces was important since it was felt that it
was the hydrophobic nature of the PMMA that made it adherent
to the endothelium of the cornea (Tables 4-6). Measurement
of the advancing angle of water for hydrated samples would
be valuable in correlating the hydrophilic nature of the
coatings with their biophysical behavior in the adhesive
force testing and the endothelium damage produced (and
evaluated by SEM).
The measurements revealed a significant decrease in
contact angle and therefore an increase in hydrophilicity
was achieved using Y-radiation- graft coatings of PVP on the .
PMMA substrate. Uncoated PMMA showed a contact angle of 72
and the y-radiation graft coatings showed a decrease in
angle down to 30 for a coating prepared at 30% N-VP
concentration and 0.1 Mrad dose.
As with the molecular weight for the external
homopolymer formed in solution, at 0.1 and 0.25 Mrad doses
the change in contact angle for water was proportional to
increasing monomer concentration for a given dose level.
This linearity was found not only for the 0.1 and 0.25 Mrad
Page
3.14.1 Instrument for Biophysical Measure
ments 44
3.14.2 Preparation of Material Sample for
Measurement 44
3.14.3 Preparation of Tissue Samples 47
3.14.4 Adhesive Force Measurement 50
3.14.5 Quantitative Cornea Endothelium
Damage by SEM 54
4. RESULTS AND DISCUSSION . 56
4.1 Preliminary Studies and Overview of Thesis
Research 56
4.1.1 Prevention of Cornea Endothelium Damage
by Poly(vinyl pyrrolidone) Solutions 56
4.1.2 Peritoneal Adhesions in Abdominal
Surgery 57
4.1.3 Other Related Research 60
4.1.4 Overview of Thesis Research 61
4.2 Intrinsic Viscosity Molecular Weight of PVP . 65
4.3 Contact Angle for Water 68
4.4 Critical Surface Tension 77
4.5 Scanning Electron Microscopy 94
4.6 ESCA . 120
4.7 Infrared Spectroscopy 122
4.8 Ultraviolet-Visible Spectroscopy 129
4.9 PVP-Iodine Interaction 134
4.10 Biophysical Measurements 136
4.11 Structure and Properties of y-Radiation
Graft Coatings 153
5. CONCLUSIONS 158
6. FUTURE RESEARCH 162
6.1 y-Radiation Graft Coatings for IOL Use .... 162
6.2 RF Plasma Coatings 163
6.3 Further Characterization of Graft Coatings . 163
APPENDIX 164
REFERENCES 167
BIOGRAPHICAL SKETCH 172
vi
Table 16. Summary of Measurements for -Radiation Graft Coatings of PVP on PMMA.
Average
Average Adhesive
Average Cells
Polymer
M *
w
Contact Angle
() Force (mg/cni )
Destroyed (%)
PMMA
72
405
60
5%**/0.1 Mrad***
309/470
61
362
10%/0.1
693/473
57
179
45
20%/0.1
1,028.743
40
174
30%/0.1
1,307,795
30
108
13
0%/0.25
_
65
490
65
5%/0.25
577,992
59
156
63
10%/0.25
714,313
59
70
30
20%/0.25
1,371/339
44
105
17
30%/0.25
2,278,969
37
111
5
5%/0.5
46
209
_ _
10%/0.5
38
146
14
Molecular weight
of external
homopolymer by
intrinsic viscosity.
**Initial monomer concentration in polymerization media.
***Radiation dose in megarad (Mrad).
121
Table 9. ESCA Results
Molar Ratio to Carbon PMMA and y-Radiation Graft Coatings
PVP on PMMA
Sample Angle
PMMA 00
70
10% N-VP Concentration/ 0
0.25 Mrad Dose 70
10% N-VP Concentration/ 0
0.5 Mrad Dose 70
O/C
N/C
Si/C
Cl/C
Cu/C
.20
.13
.02
.014
.009
.023
.020
.016
.030
.25
.24
.049
.045
.007
.014
.005.
.011
.037
.003
.41
.36
.053
.078
.074
.095
.023
.012
.020
.003
144
Figure 46.5. 7Q% Cornea Endothelium Damage after Contact
with Untreated PMMA.
50
3.14.4 Adhesive Force Measurement
Operation of the instrument to measure adhesive force
was made after the freshly prepared tissue sample was
positioned within the acrylic tank and it was placed on the
micrometer-controlled stage. The material test piece was
friction-fit into its holder and suspended above the tissue
tank.
The rack and pinion was used to lower the test piece to
within a few millimeters of the tissue surface. At this
point, the crosshair of the measuring microscope was zeroed
and aligned with a reference point of the glass fiber. This
was the resting position of the glass fiber, which was
stressed only by the weight of the test piece and holder.
Contact between the test piece and the tissue sample was
then made.
The stage supporting the acrylic tank and the tissue
sample was slowly raised by operation of its micrometer
control. As the tissue sample contacted the test piece
surface, the elevation was continued until the tripod weight
was supported by the test piece holder. This loading was
maintained for thirty seconds.
The micrometer-controlled stage was then lowered
slowly. Adhesion between the two surfaces caused the test
piece to accompany the tissue in its downward travel. This
movement was simultaneously tracked by the micrometer
crosshairs of the measuring microscope (Figure 15).
34
the PMMA substrates were not placed in the irradiation
containers (filled with the polymerization media) until
immediately prior to irradiation.
Following irradiation, the PMMA substrates were removed
from the irradiation containers and holders and immersed
immediately in distilled water. The wash water was changed
several times until all external homopolymer was removed.
The completion of the cleaning procedure was checked by
allowing the samples to air-dry and then inspecting for
debris of external homopolymer. Cleaning by washing and
soaking in distilled water was continued until all
homopolymer had been removed. The cleaned samples were
stored until use in polyethylene beakers filled with
distilled water. The homopolymer and solution remaining
following irradiation were removed from the irradiation
container and saved in capped, sealed amber glass vials.
3.5 Intrinsic Viscosity Molecular Weight of PVP
The molecular weight of the PVP homopolymer formed in a
solution by Y-radiation induced polymerization was
calculated by measurement of intrinsic viscosity and use of
the Mark-Houwink-Sakurada equation. The polymer was
obtained by precipitation of polymer formed under conditions
identical to those used for graft coating of PMMA, but
without the acrylic substrate present.
Solutions of distilled N-VP (5-30%, distilled as
previously described) and freshly distilled water were
LIST OF TABLES
Table
Page
1. Dog Model for Assessment of Polymer Solutions
, to Prevent Adhesions . 58
2. Rat Model-Assessment of PVP Solution to Prevent
Adhesions 58
3. Viscosity Molecular Weight* of PVP. External
Polymer Formed in Solution During y-Radiation
Graft Coating 66
4. Contact Angle of Water on y-Radiation Graft
Coating of PVP on PMMA 69
5. Contact Angle of Water for RF Plasma Coatings
on PMMA 70
6. Contact Angle of Water for Various Biomedical
Polymers . . 71
7.Critical Surface Tension from Zisman Plots*
for y-Radiation Graft Coatings of PVP on PMMA 85
8. Summary of Zisman Plots for Critical Surface
Tension of RF Plasma Coatings on PMMA 93
9. ESCA Results 121
10. Summary of Adhesive Force Measurements for
Cornea Endothelium Contact with y-Radiation
Graft Coatings of PVP on PMMA . 139
11. Quantitative Cornea Endothelium Damage by SEM
for Contact with y-Radiation Graft Coatings
of PVP on PMMA 145
12. Summary of Adhesive Force Measurements for
Cornea Endothelium Contact with RF Plasma
Coating on PMMA 148
13. Quantitative Cornea Endothelium Damage by SEM
for RF Plasma Coatings on PMMA 149
vii:
£ %N-VP
(Initial Concentration in Polymerization Media)
Coefficient of Determination
Figure 20. Summary of Contact Angle for Water vs. %
Monomer for Various Dose Levels of Y"Radia
tion Graft Coatings of PVP on PMMA.
47
attached with cyanoacrylate glue (Loctite, Cleveland, OH) to
acrylic stubs for use.
The holder was suspended, cone pointing down, from a
drawn glass fiber by a fine gold chain. Above this holder a
tripod weight of 17 grams was freely held, such that the
flat base of the cone would support the weight when contact
with the tissue sample was made. Therefore, contact would
be made while this weight was loading the interface with a
known pressure.
The prepared tissue sample was held on the stage of the
microscope in a acrylic tank filled with physiological
saline (0.8%) (Figure 13). A micrometer stage allowed fine
adjustment of the vertical position of acrylic tank and the
tissue sample enclosed. An X-Y micrometer stage allowed
precise centering of the tissue beneath the conical stub
holder. By filling the acrylic tank with saline, surface
tension effects in adhesion were eliminated and tissue
freshness was maintained during the measurements.
3.14.3 Preparation of Tissue Samples
Two different forms of tissue samples were used in the
measurements of adhesive force and for damage
quantification. However, the type and source of the tissue
was the same: Cornea endothelium obtained from adult male
New Zealand albino rabbits. The rabbits were sacrificed by
an intravenous overdose of pentobarbital and the eyes were
Cos 6
Figure 30. Zisman Plot for Critical Surface Tension (y ) of RF Plasma
Coatings of N-VP on PMMA: 100 Watts (Power)^200y(Pressure)/
20 Minutes(Duration).
115
Figure 34.3. RF Plasma Coating of N-VP on PMMA: 50 Watts
(Power}/1000]i CPressure)/10 Minutes(Duration).
Figure 34.4. RF Plasma Coating of N-VP on PMMA: 10Q Watts
(Power)/2Q0]i CPressure)/20 Minutes (Duration) .
Cos#
Figure 22. Zisman Plot of Critical Surface Tension (y ) of y-Irradiated
(.0.25 Mr ad) PMMA. c
2. BACKGROUND
2,1 Intraocular Lens Development and Problems
A cataract is a clouding of the natural crystalline
lense of the eye which necessitates surgical removal of the
lense (Figure 1-A). The optical function of the natural
lense must be replaced if visual acuity is to be restored.
Three alternatives exist for this purpose: spectacle
lenses, contact lenses or intraocular lenses (IOL). An IOL
is placed within the eye, in either the anterior or
posterior chamber, and serves as the optical replacement for
the natural crystalline lense.
The history of the modern intraocular lense began in
1949, when the British surgeon Dr. Harold Ridley implanted a
lens of poly(methyl methacrylate) (PMMA) in an eye following
cataract surgery (1,2). His use of the plastic PMMA had
been suggested by his experiences during the Second World
War, when he found pieces of shattered PMMA aircraft
canopies lodged in the eyes of fliers. He noted that these
fragments did not cause any reaction in the eye and would be
suitable, because of their optical properties, as a
replacement for the natural lense.
The acceptance and use of PMMA as the material for
intraocular lenses grew from that first implantation.
6
1.
INTRODUCTION
The use of synthetic polymers for prosthetic and
cosmetic implants and for surgical devices has increased
rapidly over the past three decades. The wide range of
advantageous chemical and physical properties possessed by
polymers has shown them to be uniquely suitable for many
medical applications. Some of the most demanding
requirements to be met by polymers are inertness and
biocompatibility in contact with tissues of the body.
Research on the biocompatibility of polymers (and other
materials) has been focused on short and long term toxicity,
tissue acceptance and thrombogenicity. However, very short
term, temporary and transient contacts between materials and
tissue have not been carefully investigated to date. This
study has therefore emphasized the effects of momentary
contacts between polymer and tissue surfaces, and tissue
damage resulting from such contacts.
This area of research was initiated through an
investigation of post-operative complications following
intraocular lens (IOL) insertion. An IOL is a poly(methyl
methacrylate) lens used to replace the natural crystalline
lense following cataract surgery.
1
60
protective polymer coatings prior to manipulation of tissue
and organs in surgery.
4.1.3 Other Related Research
Unpublished research has shown the occurrence of the
tissue-materials adhesion phenomenon in the use of
endotracheal catheters. In these experiments, inflated
catheter cuffs were contacted with sections of trachea,
which had been removed from ferrets and then laid open
longitudinally. This contact was made with a loading force
which would simulate the clinical inflation of the cuff
within the trachea. These contacts, using cuffs dry and wet
with a 25 weight % solution of PVP in phosphate-buffered
saline, showed that desquamation of tracheal endothelium was
reduced by the use of the PVP solution. Quantitative data
were not obtained because the area of contact was difficult
to assess.
Quantitative measurement of the IOL-endothelium inter
action was also difficult; attempts to measure the adhesive
force at an acrylic-cornea endothelium interface using an
Instron tensile test machine with a one gram full-scale load
cell were unsuccessful. The results of these tests were not
reproducible, because of difficulty in mounting the tissue
and dehydration of the tissue during the time required for
measurement. Loading of the interface was difficult and
extremely variable as was estimation of the actual area of
contact. Similarly, attempts to measure cornea endothelium
36
equation were taken from the Polymer Handbook (55) The
molecular weight of PVP was determined for all radiation
graft conditions that produced a soluble homopolymer.
3.6 RF Plasma Coatings
Modification of the PMMA substrate was performed by
graft coating with hydrophilic polymers formed by
radio-frequency (RF) plasma polymerization of N-VP and
2-hydroxyethyl methacrylate (HEMA). A vertical, "bell jar"
reaction chamber was constructed for this procedure; the
schematic diagram shows the apparatus and its set-up
(Figure 10). The plasma was created by induction of a
13.56 MHz signal of a 100 watt RF generator (Tegal,
Richmond, CA) through a ten-turn coil of copper tubing. The
output load was adjusted for optimal conditions and the
power level was controlled by a matching network (Tegal) and
a SWR meter (Heath, Benton Harbor, MI).
PMMA samples (both "stubs" and "slabs") to be coated
were supported on a stand 3.5 inches directly below the
plasma generation region (the plasma was generated in the
tube surrounded by the copper tubing coil); the samples were
secured with double-sided adhesive tape. The chamber was
evacuated to 100 micron pressure by a mechanical pump (Model
D-150, Precision Scientific, Chicago) connected with a
liquid nitrogen cold trap. The vacuum was monitored by a
thermocouple vacuum gauge (Series 270, Granville Phillips,
104
Figure 33.19. 10% N-VP Initial Monomer Concentration and
Q.25 Mrad Dose.
Figure 33.20. 10% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
2
Our preliminary studies showed that contact between an
IOL and the corneal endothelium could result in adhesion of
endothelium cells to the acrylic polymer surface.
Manipulation caused tearing of these adherent cells away
from the tissue surface. Loss of endothelium cells poses a
critical problem since these cells are a non-regenerative
monlayer and serve an essential function in maintaining
fluid balance and clarity of the cornea. Prevention of this
tissue damage is therefore essential to the success of
ophthalmic IOL implantation surgery.
The major focus of this research has therefore been on
acrylic IOL-cornea endothelium surface interactions and
prevention of tissue damage using hydrophilic polymer
coatings. However, the findings in ophthalmic surgery have
also been extended to other surgical procedures. For
example, the possibility of similar types of tissue damage
in abdominal surgery which results in peritoneal adhesions
was questioned. Manipulations within the peritoneal cavity
were shown to result in tissue adherence to latex rubber
gloves used in surgery. Normal healing of the damaged
tissue produces adhesions between adjacent, highly
regenerative tissue surfaces, with the potential for severe
post-operative complications. Other areas of tissue
adhesion to polymer surfaces with resulting damage were
found in vascular catherization and endotrachea tube
intubation, resulting in desquamation or "stripping" of the
Cos#
Figure 25. Zisman Plot of Critical Surface Tension (y ) of y-Radiation
Graft Coating: 20% N-VP Initial Monomer Concentration and
0.25 Mrad Dose.
Table 14. Summary of Adhesive Force Measurements for Cornea Endothelium Contact
with Various Biomedical Polymers.
Polymer
Average Adhesive
Force (mg/cin )
Standard ^
Deviation (mg/cm j
Number of
Tests
Urethane*
187
36
12
Silicone Rubber**
320
45
16
Thermoplastic
Elastomer***
499
211
22
Teflon****
268
93
10
*Pellthane TM, Upjohn Corporation, Kalamazoo, MI.
**Silastic 500-5TM, Dow-Corning, Midland, MI.
***Proprietary hydrocarbon block copolymer, C-flex TM, Concept Inc., Clearwater,
FL.
****TM, FEP, E.I. duPont de Nemours & Co., Wilmington, DE.
151
169
31. James. R.A., and Kelin, E.E., J. Biomed. Mater. Res.,
5, 373 (1974).
32. Wang, P.Y., and Forrester, D.H., Trans. Am. Soc. Artif.
Organs, _20i, -504 (1974).
33. Grode, G.A., Falb, R.D., Cooper, C.W. and Lynn, J., in
Adhesion in Biological Systems, Manly, R.S., ed.,
Academic Press, New York, 1970, pg. 153.
34. Karlan, M.C., Skoble, B., Grizzard, M., Cassis, N.J.,
Singleton, G.T., Buscemi, P., and Goldberg, E.P.,
Otolaryngol. Head Neck Surg., ^8, 783 (1980).
35. Corpe, W.A., in Adhesion in Biological Systems,
Manly, R.S., ed., Academic Press, New York, 1970, .
pg. 74.
36. Taylor, A.C., in Adhesion in Biological Systems,
Manly, R.S., ed., Academic Press, New York, 1970,
pg. 51.
37. Weiss, L., Exp. Cell Res., Suppl. 8, 153 (1961).
38. Pethica, B.A., Exp. Cell Res., Suppl. 8, 123 (1961).
39. Weiss, L., in Adhesion in Biological Systems,
Manly, R.S., ed., Academic Press, New York, 1970. pg.
1.
40. Henglein, A., Schnabel, W., and Heine, K., Angnew.
Chem., 15, 461 (1958) .
41. Chpiro, A., Radiation Chemistry of Polymeric Systems,
John Wiley and Sons, Inc., New York, 1962, p. 667.
42. Yasuda, H.K., and Refojo, J., Polym. Sci.: Part A, 2,
5093 (1964).
43. Boffa, G.A., Lucien, N., Faure, A., Boffa, M.C.,
Jozefonvicz, J., Szubarga, A., Maudon, P., and Larrieu,
M.J., J. Biomed. Mater. Res., 11, 317 (1977).
44. Ratner, B.C., and Hoffman, A.S., J. Appl. Polym. Sci.,
18, 3183 (1974).
45. Ratner, B.D., and Hoffman, A.S., in Hydrogels for
Medical and Related Applications, Andrade, J.D., ed.,
American Chemical Society, Washington, D.C., 1976,
pg. 1.
46. Yasuda, H.K., in Plasma Polymerization, Shen, M. and
Bell, A.T., eds., American Chemical Society,
Washington, D.C., 1979, pg. 37.
chains, even at lesser dose and monomer concentration
levels.
The Y-radiation graft coatings appear thin and adherent
to the acrylic substrate in SEM examination. In addition,
the majority of the surfaces are also smooth and uniform,
although holes and surface roughness are visible for some
coatings.
A structure for the PVP graft, based on its appearance
and physical and chemical properties is proposed in
Figure 48: the graft coating is composed of long molecules
forming a crosslinked network extending from the acrylic
surface. The density of the network decreases as the number
of chains decreases with the length of the chain and the
distance from the acrylic substrate. Although the %
hydration of the coatings has yet to be measured, the
hydrophilic-polymer coating probably contains a large amount
of water, particularly at the surface.
The proposed structure offers possible explanation of
the non-adhesive properties of the Y-radiation graft
coatings. Prevention of adhesion solely on the basis of a
hydrophilic surface was found to be an insufficient
explanation; other factors are involved. For example,
glass, generally considerd a hydrophilic surface, and
urethane (contact angle for water measured here of 37),
both produced extensive endothelium damage from adhesive
contact. Even the RF plasma coatings of PVP, which showed
increases in hydrophilicity and surface energy similar to
Cos d
Figure 32. Zisman Plot for Critical Surface Tension (y ) of RF Plasma Coating
of HEMA on PMMA: 25 Watts (Power)/500y (Pressure)/15 Minutes
(Duration).
|
