Journ31l of in.nder.r.3du.3- Re-_-search
.. Oluine 9, issue 3 - Sprinn. 200ii
Ophthalmic Drug Delivery through Contact Lenses
Current ophthalmic drug delivery systems are insufficient, specifically eye drops, which allow approximately 95%
of the drug contained in the drops to be lost due to absorption through the conjunctiva or through the tear
drainage. The use of poly-2-hydroxyethyl methacrylate hydrogels with drug-laden microemulsions has
been proposed as a method to deliver drugs to the eye. The contact lenses restrict the drug from being lost to
tear drainage by releasing the drug into two tear layers on either side of the contact lens, where it ultimately
diffuses into the eye. Microemulsions require a unique composition of oil, surfactant, and solution to reach a
stable state. This paper focuses on creating stable microemulsions capable of high drug loading, dispersing
these particles throughout a polymer gel, and measuring the release of the drug.
Currently, approximately 90% of all ophthalmic drugs are delivered using eye drops. Although convenient and
easy to use, eye drops are very inefficient, losing about 95% of the applied drug to absorption by the conjunctiva
and through tear drainage. This significant drug dosage is released into the blood stream, where it can
have detrimental side effects elsewhere in the body1-3. For example, timolol is a nonselective beta blocker that
is capable of treating glaucoma by preventing the production of aqueous humor, thus lowering the pressure inside
of the eye. Although useful for this purpose, timolol has several severe side-effects throughout the body, such
as cardiac arrhythmias, bronchiospasms, depression, and heart failure. To target the dosage to the eye, it has
been proposed to use particle-laden contact lenses for ophthalmic drug delivery. Ophthalmic drug formulations
such as timolol can be encapsulated within microemulsion nanoparticles, then entrapped with a soft poly-
2-hydroxyethyl methacrylate (p-HEMA) hydrogel contact lens. Mircoemulsions are clear, thermodynamically
stable liquid mixtures of oil, water, and surfactant. An oil-in-water microemulsion uses the surfactant to decrease
the interfacial tension between droplets of oil and the continuous water phase. These microemulsions allow
highly concentrated drug to diffuse out of the lens allowing for long-term ophthalmic drug delivery. The drug
diffuses from the lens into the tear layers surrounding the lens, ultimately passing into the eye. This paper
focuses on defining the stable compositions of microemulsions used in loading hydrogels and several methods
of trapping microemulsions laden with timolol within p-HEMA gels4.
Due to the numerous harmful side-effects of timolol, researchers have previously tried soaking contact lens in
drug solution to target delivery to the eye. When the lens is placed on the eye, the drug diffuses into the pre-
lens and post-lens tear layers and eventually into the eye. The benefits of the lens is the long residence time
within the eye and in the limited interaction of these tear layers with the outside tear fluid, thus greatly reducing
the undesirable dosage into the bloodstream5,6. A few clinical studies have shown the soaked lens to give
desired therapeutic results7-11. Researchers have also used microemulsions in ophthalmic drug delivery,
although the primary purpose of these studies was to increase the dosage of hydrophobic drugs by entrapping
them in the oil phase. These proved to be ultimately ineffective in decreasing the transfer of drug into
the bloodstream since the microemulsions are washed away by tears. By entrapping the microemulsions in the
lens it will prevent them from enter tear circulation.
MATERIALS AND DESIGN
Hydroxyethyl methacrylate monomer and ethylene glycol dimethacrylate (EGDMA) were purchased from
Aldrich Chemicals (St Louis, MO); ethylbutyrate and benzoyl peroxide (BP) (97%) were purchased from
Aldrich Chemicals (Milwaukee, WI); Timolol Maleate, Pluronic F127, Dulbecco's phosphate buffered saline
(PBS), sodium caprylate, and sodium hydroxide pellets (99.998%) were purchased from Sigma Chemicals (St
Louis, MO); Darocur TPO was kindly provided by Ciba (Tarrytown, NY).
Methodology and Procedures
Decrease Ethyl Butyrate Solubility. To create a stable microemulsion and minimize the amount of drug outside
of the microemulsions, the solubility of the ethyl butyrate in solution must be very low preventing the oil
from solubilizing. Sodium chloride and NaOH were used to increase the ion concentration and pH of the
water phase, which will decrease the solubility of the ethyl butyrate. The solubility limit was determined by
small incremental additions of ethyl butyrate until the solution became opaque. Once the solubility of ethyl
butyrate in 60% wt HEMA and 40% wt water was below 1%, the solution could then be used to create
Loading timolol into the oil phase. To entrap the timolol within the oil phase, the timolol maleate was
fist converted to timolol base. Approximately 80 mg of timolol maleate was mixed in a test tube with 6 ml of
water and 3 g of NaOH. At this low pH, the timolol maleate will convert to a very hydrophobic timolol base and
phase separate in the bottom of the tube. Five milliliters of the aqueous solution was then carefully removed
by pipette ensuring not to remove any of the timolol base, and 400 pl of hyl butyrate was added to extract
the timolol base. The resulting upper phase of ethyl butyrate and timolol base (referred to as TM/EB) was used
in creating the microemulsions.
Synthesis of microemulsions. Every microemulsion was made using the same aqueous solution as
described above, containing 25 g of 60% wt HEMA and 40% wt water, 0.25 g NaCI, and 3 g of 2N NaOH along
with the addition of 0.0425 g of sodium caprylate. This mixture was kept constant throughout the experiments
and well be referred to as the single "pseudo-component" HEMA/water solution.
Since microemulsions will only form in solutions containing specific ratios of HEMA/water solution :
Pluronic Surfacant : TM/EB, a ternary phase diagram was constructed to determine the feasible range
of compositions. At each test point chosen on the diagram, the corresponding amount of Pluronic F127
surfactant and TM/EB was added to give the desired composition. The solution was thoroughly mixed and allowed
to stabilize overnight. Those systems that remained opaque were marked as non-microemulsion. The systems
that became clear were tested further using polarized sheets. Due to the shape of microemulsions, they rotate
light very poorly. Each sample was place between perpendicular polarized sheets. If the sample rotated the
light, allowing it to pass through the sheets, it was ruled also to be a non-microemulsion system. The samples
were then further tested by vigorously mixing the system using a vortex mixer. Any trace amount of the TM/
EB which may have phase separated and was not visible through the first eye examination was forced back into
the solution. This caused unstable or minimally phase-separated samples to become opaque. Any samples
which remained clear were ruled to be microemulsions. Each point on the diagram was labeled to indicate if it was
a successful microemulsion using an "O" for a success, and an "X" for a failure. Visible trends could then be used
to create phase boundaries.
An example of one stable composition contains the 25 g HEMA/water solution along with 3.6 g of Pluronic F127,
0.61 g of TM/EB. Fifty-eight other compositions where tested in order to accurately model the phase diagram.
Entrapment of microemulsions in HEMA gels. The microemulsion particles were entrapped within p-HEMA
gels using UV-polymerization with Darocur TPO initiator. One milliliter of the microemulsion solution was added
to 1.35 mL of HEMA, 5 ml of ethylene glycol dimethacrylate (EGDMA). Dissolved oxygen must be removed
from solution before using free radical polymerization, since oxygen is capable of acting as an inhibitor. To
remove the dissolved oxygen in the solution, the sample was degassed by bubbling nitrogen for 15 minutes.
Three milligrams of Darocur TPO was added o the mixture and allowed to stir for 10 minutes outside of direct
light exposure. Next, the solution was then poured into a mold made of two 3.5" x 3.5" glass plates clapped over
a 200 pm plastic spacer. The mold was set on top of a UVB-light illuminator for 35 minutes.
Polymerization of EGDMA microemulsion. It is possible that the microemulsions could have destabilized
during the polymerization process, and it is very difficult to prove they are intact within the gel. To ensure this
does not happen, the microemulsions themselves were first polymerized in solution. Approximately, 1 ml of
EGDMA was degassed for 15 minutes by bubbling nitrogen. The EGDMA was then loaded with timolol by
substituting it for ethyl butyrate in the same process. The timolol-loaded EGDMA will be referred to as TM/
EGDMA. Eight hundred mL of the TM/EGDMA was combined with 0.0124 g of Darocur TPO and stirred for
ten minutes. A separate 10 g solution of water, NaCI, and 2N NaOH was prepared in the ratio 100 : 1.8 :
10, respectively. Next, 1.310 g of Pluronic F127 was mixed into the solution until fully dissolved. A 0.0239
mL aliquot of TM/EGDMA with initiator was added to this solution and allowed to stabilize overnight covered from
any light exposure. The solution was then placed in a Petri dish on a UVB-light illuminator for 35 minutes.
The solution formed some large aggregates; and these large particles were not used in further experiments.
The ethyl butyrate microemulsion solution was replaced by 1 mL of the polymerized EGDMA microemulsion
solution and entrapped within a HEMA gel using the same technique above.
Measuring timolol release from polymerized EGDMA microparticle laden hydrogels. After polymerization,
the gels were removed from the glass mold and cut in to 1.5 cm x 1.5 cm squares. The gels were dried overnight
to allow any water remaining in the gels to evaporate. The next day, the exact gel dimensions and gel weight
were recorded. Each gel was then placed in vial with 3 mL phosphate buffer solution, which better mimics a
tear environment than DI water. Periodically, a 1.5 mL aliquot of the solution was temporarily removed and
analyzed using UV-visible spectroscopy. Timolol present in solution will yield an absorbance peak at 295
nm proportional to the fraction of drug released from the gel. These periodic measurements were repeated
until changes in timolol concentration became insignificant.
RESULTS AND DISCUSSION
Solubility of Ethyl Butyrate
Table 1 displays the effect of increasing NaCI concentration on the solubility of ethyl butyrate in the HEMA/
water solution. Increasing the NaCI concentration to one percent weight decreased the solubility from over
ten percent to 3.3% wt.
The effect of NaCI on solubility of Ethyl Butyrate in HEMA Solution.
Weight % of NaCI in 60/40 HEMA/Water Weight % of Soluble Ethyl Butyrate in Solution @ 25
Solution deg. C
Table 2 shows a similar effect on ethyl butyrate solubility by adding 2N NaOH to the HEMA/water solution.
The solubility can be decreased to 1.4 percent by adding 4 mL of NaOH to 25 g of HEMA/water. To more
easily achieve the desired 1.0% ethyl butyrate solubility, the effects of both ion concentration and pH were
combined by adding 0.25g of NaCI and 3ml of 2N NaOH to meet the desired specification. This ratio was then
used throughout the rest of the experiments.
Table 2. The effect of NaOH on the solubility of Ethyl Butyrate in HEMA Solution.
The effect of NaOH on the solubility of Ethyl Butyrate in HEMA Solution
Milliliters of 2N NaOH in 60/40 HEMA/
Water Solution (25g Basis)
Weight % of Soluble Ethyl Butyrate in Solution @ 25
Microemulsion pseudo phase diagram
Determination of feasible microemulsion compositions. A total of 59 compositions ranging from 82-97%
HEMA/water solution were used to determine the feasible range of microemulsion compositions. Figure 1 shows
a reasonable estimation of the microemulsion phase boundary at a room temperature of 22.8 degrees Celsius.
34C4mporwml Pig 022j " C - .....
Figure 1. o -o plo of s e a 2. dr, t
/ / \ *.
., ï¿½ . .q
[18o * -- ------- -- 4
Figure 1. Pseudo three component plot of microemulsion system at 22.8 degrees C.
Temperature dependence of microemulsion stability. To increase the drug loading capacity of the contact
lens and minimize the amount of surfactant required, the microemulsions must have a large oil fraction. As
the temperature of the solution is lowered, it becomes possible to stabilize microemulsions with a larger oil
fraction than before (Figures 1-4). By decreasing the temperature of the microemulsion solution to 5
degrees Celsius, it becomes possible to stabilize microemulsions with an oil fraction of 8.4% wt and an oil-
to-surfactant ratio near 1:1. This greatly increases the drug capacity of the contact lenses.
3,Certapo PIll al dedg C -2,D 0
Figure 2. Pseudo three component plot of microemulsion system at 20.0 degrees C.
/ \ i?
/ \ 11
/ * .5\
/ * "si
Figure 3. Pseudo three component plot of microemulsion system at 15.0 degrees C.
F3.cmPm PIM P6Anlo oi dog C
/ / /
S/ ï¿½ "
Following polymerization of the EGDMA, a sample of the microemulsion was prepared and analyzed using a
SEM microscope (Images 1-3). The resulting images show the microemulsions were successfully polymerized
and resulted a uniform dispersion of microparticles approximately 2 pm in diameter. Microemulsions are
typically much smaller in size, therefore these particles are likely aggregates of several microemulsions.
Image 1. SEM image of polymerized EGDMA microemulsion.
Image 2. SEM image of polymerized EGDMA microemulsion.
Image 3. SEM image ot polymerized EGDMA microemulsion.
Drug Release from gels
Drug release from Ethyl Butyrate microemulsion gels. These experiments were performed by Chi Chung-Li
to determine the equilibrium release time for the microemulsion-laden gel. These equilibrium experiments were
done with a gel composition of 25 g HEMA/water solution : 3 g Pluronic F127 : 0.5 g TM/EB. These gels lost
17.5% of the entrapped drug during a 5 hour initial soak in 200 mL of DI water to remove excess HEMA
monomer. The results of drug release experiments (Figure 5) show that about 8% of the entrapped drug diffuses
out in a period of about 10 days. The release profiles are encouraging because the gels continue to release the
drug for a period of about 25 days with water replacement at rates comparable to the conditions in the eye, and
the equilibrium time is 10 days, which implies that the gels will release drug for 10 days even under perfect
sink conditions. However, it must be noted that these release experiments were performed in DI water, which
may not be a good mimic of the tear environment, particularly for ionizable drugs such as timolol.
Figure 5. Timolol released in DI water from ethyl butyrate microemulsion laden pHEMA gels (n = 4).
Drug release from polymerized EGDMA microparticle laden gels. These experiments were performed
to determine the release rate of timolol from the polymerized EGDMA microparticles when entrapped within a
HEMA gel. The resulting data (Figure 6) indicated that the microparticles released a significant amount of timolol
for only the first hour making it unsuitable for contact lens applications. Although this system did not
provide prolonged release of timolol, this method when used with other types of cross-linking within
the microemulsions may provide sufficient resistance to be useful.
0 1 2 3 4
5 6 7 8
Figure 6. Timolol released in PBS from polymerized EGDMA microparticle laden pHEMA gels (n = 2).
The feasible region of compositions for which the HEMA/water solution : Pluronic F127 : TM/EB system will
form stable microemulsion was defined for a series of temperatures. The capability of increasing the oil fraction
by lowering temperature allows a very large amount of timolol to be dissolved within the oil phase. The
stable microemulsions were added to HEMA and polymerized into transparent hydrogels. Although this
system released for 25 days, it was performed in DI water which may not be a good mimic of the eyes' environment.
EGDMA microemulsions were successfully polymerized as evident through the SEM. The resulting particles
were slightly larger than expected and are likely to be aggregates of the original microemulsions. The timolol
release from this system into PBS was rapid and too short for contact lens application, but this method may be
useful in slightly different conditions. By loading these microemulsions with larger drugs or using a more
closely packing cross linker, these particles may provide sufficient resistance resulting in long term drug release.
I would like to thank Dr. Anuj Chauhan for giving me this research opportunity and for the time he has devoted
to helping me. I would also like to thank Chi-Chung Li for teaching me the laboratory techniques used to
study them. Chi Chung-Li also performed the drug release experiments from the ethyl butyrate microemulsion gels.
1. Lang, J. C. "Ocular drug-delivery conventional ocular formulations." Advanced Drug Delivery Reviews 1995. 16
2. Bourlais, C.; Acar, L.; Zia, H.; Sado P.A.; Needham, T.; Leverge, R. "Ophthalmic drug delivery systems. "Progress
in retinal and eye research, 1998. 17 (1), 33-58
3. Segal, M. P. "Pumps and timed release." FDA Consumer magazine. 1991.
4. Li, Chi-Chung; Abrahamson, Michael; Kapoor, Y.; Chauhan, A. "Timolol transport from microemulsions trapped
in HEMA gels." J. of Colloid and Interface Science, 2007, in press.
5. Creech, J. L.; Chauhan, A.; Radke, C. J. Dispersive mixing in the posterior tear film under a soft contact lens. Ind
Eng Chem Res, 2001. 40 (14), 3015-3026
6. Mc Namara, N. A.; Poise, K.A.; Brand, R.D.; Graham, A.D.; Chan, J.S.; Mc Kenney, C.D. "Tear mixing under a
soft contact lens: Effects of lens diameter." Am. J. of Ophth, 1999. 127 (6), 659-665.
7. Hillman, J. S. "Management of acute glaucoma with Pilocarpine-soaked hydrophilic lens,." Brit. J. Ophthal. 1974.
8. Ramer, R.; Gasset, A. "Ocular penetration of pilocarpine." Annals of Ophthalmology. 1974.6: 1325-1327.
9. Montague, R.; Watkins, R. "Pilocarpine dispensation for the soft hydrophilic contact lens." Brit. J. Ophthal. 1975.
10. Hillman, J.; Masters, J.; Broad, A. "Pilocarpine delivery by hydrophilic lens in the management of acute
glaucoma." Transactions of the ophthalmological societies of the United Kingdom. 1975: 79-84.
11. Giambattista, B.; Virno, M.; Pecori-Giraldi, Pellegrino, N.; Motolese, E. "Possibility of isoproterenol therapy with
soft contact lenses: Ocular hypotension without systemic effects." Ann. Ophthalmol. 1976. 8: 819-829.
Back to the Journal of Undergraduate Research
College of Liberal Arts and Sciences I University Scholars Program I University of Florida I
ï¿½ University of Florida, Gainesville, FL 32611; (352) 846-2032.
f |UNIVERSITY of
The" I'vfnifurda s m:or Ihr G41kr NafNe