Download The use of bioerodible polymers and 5-fluorouracil in

Investigative Ophthalmology & Visual Science, Vol. 29, No. 11, November 1988
Copyright © Association for Research in Vision and Ophthalmology
The Use of Bioerodible Polymers and 5-Fluorouracil in
Glaucoma Filtration Surgery
David A. Lee,* Karri W. Leong,f William C. Panek,* Calvin T. Eng,* and Ben J. Glasgow^
A study was performed to examine the effect of a localized and sustained delivery of 5-fluorouracil
(5-FU) on the success of glaucoma filtration surgery in 18 rabbits in a prospective, randomized,
double-masked and placebo-controlled fashion. A bioerodible polyanhydride composed of bis (p-carboxyphenoxy) propane and sebacic acid was used as the drug carrier. The polymer and 5-FU (20% by
weight) were compressed into 3 mm diameter discs, 1 mm thick. The polymer with the 5-FU was
randomized to one eye and the fellow eye received the blank polymer. The results showed that
intraocular pressures (IOP) were lower in the experimental eyes during the 5th through 17th postoperative days, but eventually both experimental and control eyes returned to preoperative levels. Filtration blebs lasted longer in experimental eyes when compared to control eyes. Implant disappearance occurred after IOP elevations and bleb failure. Eventually, the filtration surgery failed in both the
experimental and control rabbit eyes. Invest Ophthalmol Vis Sci 29:1692-1697,1988
The antimetabolite 5-fluorouracil (5-FU), a fluorinated pyrimidine which competitively inhibits thymidylate synthetase, resulting in an inhibition of deoxyribonucleic acid (DNA) synthesis, has been found
to improve the success of glaucoma filtration surgery
in the eyes of nonhuman primates 1 and in the eyes of
glaucoma patients with poor surgical prognoses.2-3
The presumed mechanism for the increase in success
rate is the inhibition of fibroblast proliferation and
subsequent scarring at the site of surgery by 5-FU.
Subconjunctival administration of 5-FU postoperatively may improve the success rate of glaucoma
filtration surgery, but also has the disadvantages of
frequent administration, patient discomfort, and ocular surface problems. 4 These disadvantages may be
eliminated by different drug delivery systems which
provide a controlled and localized release of drug
over an extended period of time.5"7 Drug delivery
systems using a collagen implant 8 or bioerodible
polymers 9 have been investigated. This paper
presents the results of a series of experiments using
5-FU impregnated bioerodible polymer discs in a
rabbit model of glaucoma filtration surgery.
Materials and Methods
Chemicals were obtained as follows: 5-fluorouracil
from Sigma Chemical Company (St. Louis, MO);
ketamine from Parke-Davis (Morris Plains, NJ); xylazine from Haver-Lockhart (Shawnee, KA); pentobarbital from Abbott Laboratories (North Chicago,
IL); paraformaldehyde, glutaraldehyde and sodium
cacodylate from Tousimis (Rockville, MD). All other
chemicals were reagent grade. All animal subjects
were treated in accordance to the ARVO Resolution
on the Use of Animals in Research.
The copolyanhydride of bis (p-carboxyphenoxy)
propane (PCPP) and sebacic acid (SA), in a ratio of
50:50 by weight, was used as the carrier matrix. The
polymers were synthesized by adapting the method
described by Conix.10 The implants were fabricated
by compressing the polymer (100 Kpsi and room
temperature) combined with solid 5-FU (20% by
weight) into discs of the dimension 3 mm in diameter
and 1 mm thick. Each implant therefore contained
1.5 ± 0.1 mg of 5-FU. The control implants were
made in an identical manner in the absence of 5-FU.
All implants were stored desiccated in a bottle.
A prospective, randomized, double-masked and
placebo-controlled study was performed by comparing a bioerodible polymer impregnated with 20%
5-FU to the same bioerodible polymer without any
drug. The intraocular pressures of 18 normal Dutch
rabbits, each weighing between 1.5 and 2.5 kg, were
From the *Jules Stein Eye Institute, UCLA School of Medicine
Los Angeles, California, the fDepartment of Biomedical Engineering, Johns Hopkins University, Baltimore, Maryland, and the JDepartment of Pathology, UCLA School of Medicine, Los Angeles,
California.
Supported in part by the American Philosophical Society, the
Elsie B. Ballantyne Fund, Research to Prevent Blindness, and the
Lucille Simon Glaucoma Research Fund.
Submitted for publication: January 21, 1988; accepted June 13,
1988.
Reprint requests: David A. Lee, MD, Jules Stein Eye Institute,
UCLA School of Medicine, Room 2-118 B, Los Angeles, CA
90024-1771.
1692
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933368/ on 05/04/2017
No. 11
BIOERODIBLE POLYMERS AND 5-FLUOROURACIL / Lee er ol
measured by pneumotonometry (Alcon Applanation
Pneumatonograph, Digilab Inc., Cambridge, MA).
All of the rabbits underwent preoperative eye examinations with a Kowa hand-held slit-lamp biomicroscope. Prior to each intraocular pressure measurement, one drop of 0.5% proparacaine hydrochloride
was topically applied to each eye.
General anesthesia was given using ketamine 50
mg/kg I.M. and xylazine 15 mg/kg I.M. A posterior
lip sclerectomy was performed on both eyes of each
animal by the same surgeon. A lid speculum was inserted and the superotemporal conjunctiva was incised near the fornix with Wescott scissors. The conjunctiva was carefully dissected anteriorly to the
limbus. Excessive Tenon's tissue overlying the sclera
was excised. A limbal groove was made with a 57
Beaver blade and extended anteriorly into the corneal
stroma. Before the anterior chamber was entered, a
paracentesis was made through peripheral clear cornea away from the filtering site. Then the anterior
chamber was entered through thefilteringsite and a 1
X 3 mm block of scleral tissue and trabecular meshwork was excised with Vannas scissors. The edges of
the sclerectomy were cauterized to control hemostasis. Then a peripheral iridectomy was performed. The
disc-shaped polymer was placed over the sclera adjacent to the sclerostomy site in the subconjunctival
space (Fig. 1). Both eyes of each rabbit had the polymer inserted, the experimental eye had the polymer
containing 5-FU and the control eye had the polymer
without any 5-FU. The polymer with 20% 5-FU was
randomly assigned to one eye and the fellow eye received the control polymer without the drug. The
conjunctival incision was closed over the polymer
with a running 10-0 nylon suture (Fig. 2). The anterior chamber was reformed with sterile saline solution through the paracentesis site and the bleb was
checked for leaks. At the conclusion of the surgical
procedure, topical erythromycin ointment was applied to both eyes of each animal in order to minimize the risk of postoperative infection.
Postoperative follow-up included ocular examinations with a Kowa hand-held slit-lamp biomicroscope and intraocular pressure measurements by
pneumotonometry the first day after surgery and
every other day thereafter for 3 weeks. Observations
were recorded on individualized charts for each animal, with special attention given to the appearance of
the bleb, conjunctiva, implant, cornea and anterior
chamber.
The animals were sacrificed with an overdose of
sodium pentobarbital 75 mg/kg I.V. Two animals
were sacrificed 2 weeks after surgery and 16 animals
were sacrificed 5 weeks after surgery. The two animals which were sacrificed 2 weeks after surgery were
1690
Fig. 1. Disc-shaped polymer before insertion into glaucoma filtration surgical site.
enucleated and the eyes were processed for light microscopy.
A similar but separate series of ten Dutch rabbits
were studied to compare the effects of glaucoma filtration surgery with the disc polymer without drugs
to glaucoma filtration surgery without any polymer.
This series often animals was followed in an identical
manner.
The eyes of two study animals (four eyes) were
enucleated after sacrifice and fixed in 10% neutral
buffered formalin for 48 hr prior to sectioning. The
eyes were sectioned near the sclerostomy site. The cut
specimens were dehydrated, infiltrated, and embedded in paraffin. Eight micron step sections of the bleb
(wound) were cut with a rotary microtome. Sections
were prepared with hematoxylin and eosin and Mas-
Fig. 2. Disc-shaped polymer under conjunctiva at the end of
glaucoma filtration surgery.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933368/ on 05/04/2017
1694
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / November 1988
Vol. 29
o Polymer
• No Polymer
Mean t S.E.
I
5
7
9
II
3
Time Postoperative (days)
5
7
9
II
13
15
17
19
Time Postoperative (days)
13
Fig. 3. Graph of postoperative intraocular pressures over time.
Statistically significant lower pressures occurred in experimental
eyes between postoperative days 5 through 17.
Fig. 4. A graph of postoperative intraocular pressures following
glaucoma filtration surgery comparing control polymer with no
drug. There was no statistically significant difference over time
after surgery.
son's trichrome stain and studied by light microscopy.
As the animals were sacrificed, the implants were
retrieved and examined for remaining drug and extent of matrix degradation. The dry weight loss of the
implant was recorded after rinsing with distilled
water and drying under vacuum overnight. The
amount of 5-FU remaining in the matrix was determined by HPLC as described previously.9
The in vitro drug release kinetics were determined
by placing the drug-loaded discs in vials containing
10 ml of 0.1 M, phosphate buffer, pH 7.4, at 37°C.
The periodically changed buffer solutions were then
subjected to HPLC analysis as described previously.9
Statistical analysis was done to compare experimental eyes with the fellow control eyes, using the
paired t-test, paired Wilcoxon test, and McNemar's
test, as appropriate for different variables." The fol-
Fig. 5. Photomicrograph of a histologic section from the sclerostomy site of a rabbit eye with a polymer containing 5-FU. Cornea (C), ciliary
body (CB), and open sclerostomy (arrow) are shown (Masson's trichome, original magnification X48).
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933368/ on 05/04/2017
No. 11
OIOERODIDLE POLYMERS AND 5-FLUOROURACIL / Lee er ol
1695
Fig. 6. Photomicrograph of anterior segment of a rabbit eye treated with the control polymer shows cornea (C), ciliary body (CB) and
sclerostomy site (arrows). Sclerostomy site was replaced byfibroblastsand collagen (Masson's trichrome, original magnification X48).
lowing variables were analyzed: intraocular pressure,
time to bleb failure, time to implant disappearance,
conjunctival injection, corneal haze and pigmentation, and hyphema. A probability of less than 0.05
was considered to be statistically significant.
Results
Of the 18 rabbits in this study, postoperative hyphemas occurred in three of the 5-FU eyes and two of
the control eyes. The polymer was extruded in only
one eye, which had contained the control polymer.
There were no intraoperative complications of conjunctival button holes or vitreous loss. There were no
postoperative complications of wound leaks, corneal
haze or pigmentation or endophthalmitis.
The postoperative pressures are shown in Figure 3.
From day 5 to day 17 after surgery the intraocular
pressures in the experimental 5-FU eyes were significantly lower than in the control eyes. However, eventually the intraocular pressures returned to their preoperative levels in both eyes. The blebs in the 5-FU
eyes lasted significantly longer than in the control
eyes. There was no significant difference in the rate of
implant disappearance between experimental and
control eyes, nor was there a temporal relationship
between implant disappearance and the return of the
intraocular pressure to preoperative levels. The polymer lasted during the entire five week follow-up period. Eventually, filtration surgery failed in both experimental and control eyes.
There was no difference in the postoperative intraocular pressure course in the eyes which received a
control polymer disc and fellow eyes which received
no disc (Fig. 4). The intraoperative and postoperative
complication rates between the experimental and
control eyes were identical.
Four representative eyes of the 18 rabbits which
had the polymer with 5-FU in one eye and the blank
polymer in the fellow eye were harvested 14 days after
surgery and studied by light microscopy. Sclerotomy
sites were identified. The 5-FU treated eyes had open
sclerostomies (Fig. 5), while the control eyes had sclerostomy sites occluded by granulation tissue (Fig. 6).
Due to difficulty in separating the implant completely from the fibrotic tissue, we could only esti-
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933368/ on 05/04/2017
INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / November 1988
1696
100
90 80
70
60
50
40
30
20
10
0
()
1
2
_ ^ » -
'
6
8
4
1
Time, Day
Fig. 7. A graph of in vitro drug release kinetics of 5-FU from
bioerodible polymers. Most of the drug was released by 1 week.
mate the extent of matrix degradation. The weight
loss of the carrier was observed to be 12% ± 8%. The
determination of the 5-FU remaining in the matrix
was plagued by the inability of the implant to completely dissolve in a good solvent such as chloroform;
the implants were then extracted with acetonitrile,
which is a good solvent for 5-FU, but a nonsolvent for
the polymer. The HPLC analysis showed that only
traces of 5-FU were left behind in some samples. This
observation is consistent with the in vitro release
study which showed that most of the drug was released by one week (Fig. 7).
Discussion
Preliminary studies have shown that 5-FU may be
efficacious in improving the success rate of glaucoma
filtration surgery in patients who are at increased risk
of failure.1"3 When 5-FU is administered by frequent
subconjunctival injection, however, complications
may occur.4
Different drug delivery vehicles have been tried to
deliver 5-FU in such a manner as to avoid frequent
subconjunctival injections and to decrease the complication rate.8-9 Collagen implants and bioerodible
polymers have been used to deliver 5-FU over an
extended period of time to a localized area at the site
of glaucoma filtration surgery. Cylindrically shaped
bioerodible polymers which were inserted into the
sclerostomy at the time of glaucoma filtration surgery
had an increased number of complications.9 However, those polymers which were impregnated with
5-FU temporarily prolonged the function of glaucoma filtration surgery as evidenced by lower intraocular pressure, filtration blebs of longer duration and
higher outflow facility.9
Vol. 29
In this study, there were significantly fewer postoperative complications than in the previous study,
probably because the polymer shape was less traumatic to the ocular structures. The shape of the polymer probably did not have a mechanical effect on the
success rate because there was no difference in intraocular pressure levels between eyes with control polymer versus eyes without polymer following glaucoma
filtration surgery.
Also in this study, the intraocular pressures were
lower in the experimental eyes for 12 days, which was
longer than in the previous study in which 10% 5-FU
bioerodible polymers were used.9 This improvement
in the duration of lowered intraocular pressure in the
experimental eyes is probably due to the higher concentration of 5-FU in the polymer and longer duration of release. However, it is noted that eventually
both experimental and control eyes in this study returned to preoperative intraocular pressure levels. No
tonographic measurements were performed in this
study; however, based on the results of the previous
study,9 the lower intraocular pressure in experimental
eyes was probably due to an increase in outflow facility rather than a decrease in aqueous humor production. This is supported by the histopathological finding at 2 weeks after surgery of an open sclerostomy in
the experimental eyes. No abnormal morphological
changes were seen in the ciliary body epithelium and
trabecular meshwork of 5-FU treated eyes when
compared to control eyes by light microscopy.
In summary, this study further confirms the hypothesis that a localized and controlled release of an
antifibroblastic agent to the sclerostomy site is beneficial to glaucoma filtration surgery. Further refinements in this new drug delivery system are needed. It
is speculated that the most significant factors are the
release characteristics and the type of antifibroblastic
agents. Work is in progress to optimize these two
aspects. From the in vitro release studies, it is strongly
suspected that the onset of bleb failure coincides with
the exhaustion of the drug. To prolong the duration
of drug release, a more hydrophobic polymeric carrier may be required to slow down the diffusion of
hydrophilic 5-FU. However, the use of a hydrophobic
carrier conflicts with the desire to have an implant
disappearing soon after the exhaustion of the drug
supply. The use of other antifibroblastic agents may
alleviate the delivery problem if the drug is less hydrophilic. The potential of using this polymeric controlled delivery technology in enhancing the success
of glaucoma filtration surgery looks promising.
Key words: 5-fluorouracil, experimental glaucoma surgery,
bioerodible polymers, rabbits,filteringbleb, wound healing
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933368/ on 05/04/2017
No. 11
BIOERODIBLE POLYMERS AND 5-FLUOROURACIL / Lee er ol
Acknowledgments
The authors would like to thank Ms. Jeanne M. Cooper
for her secretarial assistance in typing the manuscript, Mr.
Onelio Clark for his help in biostatistical analysis, and Dr.
Ana Maria Zaragoza for technical assistance in processing
specimens for light microscopy.
References
1. Gressel MG, Parrish RK, and Folberg R: 5-fluorouracil and
glaucoma filtering surgery: I. An animal model. Ophthalmology 91:378, 1984.
2. Heuer DK, Parrish RK, Gressel MG, Hodapp E, Palmberg PF,
and Anderson DR: 5-fluorouracil and glaucoma filtering surgery: II. A pilot study. Ophthalmology 91:384, 1984.
3. Heuer DK, Parrish RK, Gressel MG, Hodapp E, Desjardins
DC, Skuta GL, Palmberg PF, Nevarez JA, and Rockwood EJ:
5-fluorouracil and glaucomafilteringsurgery: III. Intermediate
follow-up of a pilot study. Ophthalmology 93:1537, 1986.
4. Lee DA, Hersh P, Kersten D, and Melamed S: Complications
of subjunctival 5-fluorouracil following glaucoma filtering
surgery. Ophthalmic Surg 18:187, 1987.
1697
5. Leong KW, Kost J, and Mathiowitz E: Polyanhydrides for
controlled release of bioactive agents. Biomaterials 7:364,
1986.
6. Leong KW, Brott BC, and Langer R: Bioerodible polyanhydrides as drug-carrier matrices: I. Characterization, degradation and release characteristics. J Biomed Mater Res 19:941,
1985.
7. Leong KW, D'Amore P, and Marietta M: Bioerodible polyanhydrides as drug-carrier matrices: II. Biocompatibility and
chemical reactivity. J Biomed Mater Res 20:51, 1986.
8. Kay JS, Litin BS, Jones MA, Fryczkowski AW, Chvapil M,
and Herschler J: Delivery of antifibroblast agents as adjuncts to
filtration surgery: Part II. Delivery of 5-fluorouracil and bleomycin in a collagen implant: Pilot study in the rabbit. Ophthalmic Surg 17:796, 1986.
9. Lee DA, Flores RA, Anderson PJ, Leong KW, Teekhasaenee
C, de Kater AW, and Hertzmark E: Glaucoma filtration surgery in rabbits using bioerodible polymers and 5-fluorouracil.
Ophthalmology 94:1523, 1987.
10. Conix A: Poly [1,3 bis (p-carboxyphenoxy) propane anhydride]. Macro Synth 2:95, 1966.
11. Snedecor GW and Cochran WG: Statistical Methods, 6th ed.
Ames, Iowa State University Press, 1974.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933368/ on 05/04/2017