JOURNAL OF OCULAR PHARMACOLOGY AND THERAPEUTICS
Volume 22, Number 6, 2006
© Mary Ann Liebert, Inc.

Efficiency of Instillation Methods for
Prostaglandin Medications
RICHARD FISCELLA,1,2 JACOB T. WILENSKY,l TINA H CHIANG,3 and JOHN G. WALT3

ABSTRACT

Purpose: The aim of this study was to determine the most efficient methods for instillation
of prostaglandin analogs.
Methods: Drops were dispensed at room temperature from 2.5-mL bottles of bimatoprost,
travoprost, and latanoprost. Two determinations of drop count were each made from bottles
held vertically, at a 45-degree angle, and horizontally. The total volumes of medication dispensed from each bottle were measured.
Results: The mean number of drops dispensed was 111.0, 105.1, and 76.1 drops for bimatoprost bottles; 81.4, 101.1, and 85.3 drops for travoprost bottles; and 94.3, 88.4, and 67.1 drops
for latanoprost bottles, held vertically, at 45 degrees, and horizontally, respectively. The mean
volume of medication dispensed per 2.5-mL bottle was 3.17 mL for bimatoprost, 2.54 mL for
travoprost, and 3.02 mL for latanoprost. The most efficient instillation methods provided 56
days of bilateral therapy per 2.5-mL bottle for bimatoprost, 51 days for travoprost, and 47 days
for latanoprost, with corresponding yearly medication costs of $408 for bimatoprost, $449 for
travoprost, and $475 for latanoprost. Yearly savings of $109 to $192 could be achieved by using the most efficient instillation methods, representing 5.6 months of medication saved for
patients using bimatoprost, 3.0 months for patients using travoprost, and 4.9 months for patients using latanoprost.
Conclusions: Health care providers are urged to instruct glaucoma patients in the most efficient method of instillation. For bimatoprost and latanoprost, vertical instillation is recommended, with 45 degrees nearly as efficient, and for travoprost, instillation at 45 degrees is
recommended.

INTRODUCTION
irreversible ocuG
lar disease that affects the optic nerve, causing loss of visual field and contrast sensitivity.
LAUCOMA IS A PROGRESSIVE,

Therapeutic intervention with intraocular pressure (IOP)-lowering medication slows the progression of the disease and helps glaucoma pa-

tients to retain visual function. l -4 Topically administered prostaglandin analogs, comprising bimatoprost, travoprost, and latanoprost, are the
most effective and widely prescribed treatments
for glaucoma and ocular hypertension.5--7 They
have the added advantage of once-daily administration, which helps to improve patients' adherence to their therapeutic regimen.8-10

lDepartments of Ophthalmology and 2Pharmacy Practice, University of Chicago, Chicago, IL.
Inc., Irvine, CA.
Research for this work was funded by Allergan, Inc. (Irvine, CA).

3 Allergan

477

 478

FlSCELLA ET AL.

The cost-consciousness of managed care
METHODS
providers and of individuals who pay for most
or all of their own prescriptions has prompted
Procedures mimicking normal instillation were
studies analyzing costs and cost-effectiveness of employed to determine the number of drops disthe prostaglandin analog class of lOP-lowering pensed from 2.5-mL bottles of the once-daily
medications.1 1- 13 One area that has not yet been lOP-lowering glaucoma medications, bimatoinvestigated is whether patients instill once-daily prost (Lumigan®; Allergan Inc., Irvine, CA),
lOP-lowering drugs in the most efficient manner travoprost (Travatan®; Alcon Laboratories, Inc.,
possible. Instillation of a smaller-size drop would Ft. Worth, TX), and latanoprost (Xalatan®; Pfizer
be desirable because the volume of individual Inc., New York, NY). Drops were dispensed
drops dispensed by most bottles of ophthalmic slowly and evenly at room temperature into a
medications exceeds the maximum volume that graduated cylinder. Drops were counted from
the palpebral fissure can accommodate, which is eight 2.5-mL bottles of each medication at each
approximately 30 p,L.14 Normal tear volume is es- instillation angle (vertical, 45 degrees, and horitimated to be 7-10 p,L,15,16 giving a net volume zontal), and the total volume of medication disavailable for instillation of 20-23 p,L. Any med- pensed from each bottle was recorded. Mean
ication in excess of 20-23 p,L probably overflows drop volume was calculated by dividing the
immediately after administration. Because nor- mean measured volume of medication per bottle
mal tear volume is restored within 2-3 min by re- by the mean number of drops per bottle. In an
flex blinking, tearing, and drainage through the additional analysis, drops were counted from the
nasal-lacrimal duct, additional medication is bottles held at 45 degrees, only 2 drops at a time,
probably lost as well.14 By waiting 5 min in be- with the bottles placed back on the table between
tween administration of eye drop solutions, the administration of 2 drops, to simulate daily doseffect of diluting the first medication is mini- ing to both eyes.
The yearly cost of lOP-lowering medication
mized.
Studies have shown that the bioavailability and was calculated for each instillation method. Onceefficacy of drops as small as 15 p,L are equivalent daily bilateral treatment requires 730 drops per
to those of larger drops14 Therefore, smaller year. The number of bottles that would be redrops would be preferable to minimize systemic quired per year for once-daily bilateral treatment
exposure and spilled or wasted medication. Ob- was first calculated by dividing 730 drops by the
viously, a smaller drop size would mean that mean number of drops found per 2.5-mL bottle.
more doses could be dispensed from each bottle Then, the calculated number of bottles per year
of medication, providing cost savings to patients (which was not rounded to an integer value) was
multiplied by the published average wholesale
and managed care providers.
Drop sizes of ophthalmic solutions depend on price (AWP; Price Alert 2005, Medi-Span, Indiproperties of the solution itself, such as viscosity anapolis, IN). The A WPs for 2.5-mL bottles were
and surface tension, which probably vary little $62.10 for bima toprost, $62.19 for travoprost, and
among medications with similar formulations. $61.29 for latanoprost.
The dimensions and design of the dropper tip
Statistical analyses employed unpaired Stugreatly influence drop size, as does the angle at dent t tests.
which the bottle is held when the drops are dispensed. 14 The one variable affecting drop size
that patients are able to control is the angle at
which the bottle is held when the drops are inRESULTS
stilled. In this study, we measured the number of
drops dispensed from 2.5-mL bottles of the onceThe mean number of drops dispensed from
daily lOP-lowering drugs, bimatoprost, travo- 2.5-mL bottles of prostaglandin analogs is preprost, and latanoprost, when the bottles were sented in Figure 1. For bimatoprost, the greatest
held vertically, at 45 degrees, and horizontally. mean number of drops (111.0) was dispensed
The results show that patients can achieve sig- from bottles held vertically, with nearly as many
nificant savings in the cost of their lOP-lowering drops (105.1) from bottles held at 45 degrees,
medication by choosing the most efficient angle and the least (76.1) from bottles held horizonfor instillation.
tally. A similar pattern was seen for latanoprost,

 479

INSTILLATION METHODS FOR PROSTAGLANDIN MEDS
Angle 01 instillation
• Horizontal

045 degrees
II Vertical

P< .0001

140
Ql

E

I

120

P

= .0006

I

0

II

P

= .018

.0
...J

E

1.0

N

100
80

(j;
0.
UJ

0.
0

-0
c

60
40

Cll

Ql

:2

20
0
Travoprost

Bimatoprost

Latanoprost

FIG. 1. Numbers of drops dispensed at room temperature from 2.5-mL bottles of prostaglandin analogs held at
three different angles. Each bar represents the mean of 8
determinations, with standard deviations shown as error
bars. P-values for comparisons between angles of instillation that yielded the greatest numbers of drops are
shown in the figure. Other statistically significant comparisons were: bimatoprost, vertical versus 45 degrees
(P = 0.022), vertical versus horizontal, and 45 degrees versus horizontal (P < 0.0001); travoprost, 45 degrees versus
vertical (P < 0.0001), 45 degrees versus horizontal (P =
0.0033); latanoprost, vertical versus horizontal and 45 degrees versus horizontal (P < 0.0001).

with 94.3, 88.4, and 67.1 drops dispensed from
bottles held vertically, at 45 degrees, and horizontally, respectively. For travoprost, the greatest number of drops was dispensed from bottles
held at 45 degrees (101.1), with fewer drops dispensed from bottles held at horizontally (85.3)
and vertically (81.4). The number of drops dispensed at the most efficient instillation angle
would provide 56 days of once-daily bilateral
therapy from a 2.5-mL bottle of bimatoprost, 51
TABLE

Bimatoprost
Travoprost
Latanoprost

days for travoprost, and 47 days for latanoprost
(Table 1) .
In the method where only 2 drops were dispensed from the bottles with table placement between administrations, the number of drops and
recorded volume dispensed from the bimatoprost
and travoprost bottles were similar to the previous
results. However, for latanoprost bottles, fewer
drops (83 vs. 88 drops) and less volume was captured into the graduated cylinder (2.6 vs. 3.0 mL).
Visual observation of the tip of the latanoprost bottle during the procedure demonstrated liquid running down the outside the length of the tip after
replacement of the bottle on the table.
Measurement of the total volume of medication that could be dispensed from 2.5-mL bottles
showed that, on average, bottles of bimatoprost
contained 3.17 mL, bottles of travoprost contained 2.54 mL, and bottles of latanoprost contained 3.02 mL (Fig. 2). The angle of instillation
had no effect on the total volume of medication
per bottle that could be dispensed. When drops
were dispensed at the angles yielding the greatest number of drops, the calculated average drop
volume was 28.7 ILL for bimatoprost, 25.4 ILL for
travoprost, and 32.9 ILL for latanoprost (Table 1).
The yearly costs of treatment with prostaglandin
analogs are presented in Figure 3. For bimatoprost,
the yearly costs were $408, $431, and $596 for medication instilled vertically, at 45 degrees, and horizontally, respectively. For latanoprost, yearly costs
were $475, $506, and $667 for vertical, 45-degree,
and horizontal instillation, respectively. For travoprost, yearly costs were $449, $532, and $558 for 45degree, horizontal, and vertical instillation, respectively. Vertical instead of horizontal instillation
would save $187 per year for patients using bimatoprost, and $192 for patients using latanoprost
(Table 1). Patients using travoprost would save
$108 per year by instilling at 45 degrees instead of
vertically.

1.

DAYS OF THERAPY PER BOTILE, DROP VOLUME, AND YEARLY COST
WHEN THE MOST EFFICIENT INSTILLATION METHOD WAS USED

Maximum days
of therapy per
bottle

Drop
volume

Lowest
yearly cost

Most efficient
instillation
angle

Yearly
savings·

56
51
47

28.7 ILL
25.4 ILL
32.9 ILL

$408
$449
$475

vertical
45 degrees
vertical

$187
$109
$192

"Savings that would be achieved by using the most efficient instillation angle, as compared with the least efficient
angle.

 FISCELLA ET AL.

480

c:

o

""§
u
'6

_~ ~
0
0.0

~....J

-gE

3.5
3.0

2.5
2.0

~11)1

~ ~ .5
~ ~ 1.0
>
c:
~

:i!

0.5
0.0
Sima!opros!

Travoprost

la!anopros!

FIG. 2. Mean total volumes dispensed from 2.S-mL bottles of prostaglandin analogs. Each bar represents the
mean of 24 determinations at all three instillation angles,
with standard deviations shown as error bars.

DISCUSSION

methods. Trying to account for any variations
noted in actual patient administration, latanoprost may not last as long as bimatoprost or
travoprost. This discrepancy may reflect the loss
of latanoprost down the side of the dropper tip
after each administration. This does not appear
to be a problem with the bottle design of bimatoprost or travoprost.
Many aspects of eye drop formation and delivery may affect the size of the drop that falls
from the bottle. 14 These include, changing from a
vertical position to a horizontal position when administering the drop, the design of the outer surface of the dropper tip and the surface tension of
the solution. As an example, a bottle with an annular recess at the tip may dispense a decreased
drop size when tilted from 90 to 45 degrees.1 4
However, tilting may also result in partially
formed drop resulting in a variable drop size, depending upon the surface tension of the solution.
A dropper tip with no annular recess and a lower
surface tension may actually demonstrate an increase in drop size when the bottle is tilted from
90 to 45 degrees. Each bottle of the prostaglandin
analogs studied was a different design. The bimatoprost bottle appeared to have a small
rounded tip with no annular recess. The latanoprost bottle appears to have the largest inner and outer diameter tapered tip of all the
prostaglandin analogs. Although modification of
the latanoprost bottle years before allowed for

Data presented in this study show that the angle at which prostaglandin analogs are dispensed
greatly affects the numbers of drops that can be
obtained from a 2.5-mL bottle. Vertical instillation was most efficient and instillation at 45 degrees was nearly as efficient for bimatoprost and
latanoprost, yielding significantly more drops per
bottle (31%-46%) than horizontal instillation of
these 2 medications (P < 0.0001). By contrast, instillation at 45 degrees was most efficient for
travoprost, yielding significantly more drops per
bottle (19%-24%) than horizontal or vertical instillation (P :s 0.0033). The mean number of drops
Angle of instillation
dispensed by the most efficient instillation angle
• Horizontal
was significantly greater for bimatoprost than
045 degrees
travoprost and latanoprost (P = 0.0006 and
III Vertical
P < 0.0001, respectively), and was significantly
$700
greater for travoprost than latanoprost (P =
$600
0.018). The mean numbers of drops dispensed at
45 degrees from 2.5-mL bottles of bimatoprost, CQl
E $500
travoprost, and latanoprost in this study, 105.1, ~
101.1, and 88.4 drops, respectively, correlated ~ $400
0
well with previously published findings, 103, 98, U5
0
$300
and 92 drops,ll and 111.3, 102.9, and 97.6 drops, u
£
respectively.12.17 The calculated volumes of drops Qlco $200
dispensed at the most efficient angles of instilla- >$100
tion, 28.7 ~L for bimatoprost and 32.9 ~L for latanoprost instilled vertically, and 25.4 ~L for
$0
Simatoprost
Travoprost
travoprost instilled at 45 degrees, exceed the vollatanopros!
ume of the palpebral fissure available for instilFIG. 3. Yearly costs for bilateral administration of
lation, suggesting that fully efficacious volumes prostaglandin analogs instilled at various angles. See
of medication are delivered by these instillation Methods for calculations.

 481

INSTILLATION METHODS FOR PROSTAGLANDIN MEDS

better control of drop administration, the drop itself appears to be larger with the newer bottle design. 12 Both bimatoprost and latanoprost produced more drops in the vertical, 45-degree, and
horizontal positions, respectively, possibly reflecting the bottle tip design. The travoprost bottle has a conical design that appears to have a
slight increase in internal diameter as it approaches the end of the tip. Travoprost produced
the most drops in the 4S-degree position. Although the tip does not appear to have an annular recess, perhaps the slight increase in the inner
diameter at the end of the tip may affect the drop
size as the bottle tilting proceeds from vertical to
45-degree positions. Other considerations previously mentioned, such as surface tension, may
also be variables that may affect the drop size in
various positions. 14
The total volume of once-daily lOP-lowering
medication dispensed from 2.5-mL bottles averaged 3.17 mL for bimatoprost (a 0.67-mL overfill),
a volume significantly greater than that dispensed from bottles of travoprost, 2.54 mL (negligible overfill; P < 0.0001), and latanoprost, 3.02
mL (a 0.62-mL overfill; P = 0.0001). Previous
studies reported medication volumes of 3.06,
2.92, and 2.98 mLY and 3.3, 3.0, and 3.05 mLY
for 2.5-mL bottles of bimatoprost, travoprost, and
latanoprost, respectively.
Of the 3 prostaglandin analogs, bimatoprost
provided the greatest number of days of bilateral
therapy per 2.5-mL bottle, 56 days, and the lowest cost per year, $408, when dispensed vertically.
Vertical instillation of bimatoprost and latanoprost would save $187 and $192 per year,
compared with horizontal instillation. Instillation
of travoprost at a 45-degree angle would save
$108 per year relative to instillation from a vertical angle. These savings are substantial: They represent the cost of 3.0 bottles and 5.6 months of bilateral therapy with bimatoprost, 1.8 bottles and
3.0 months of therapy with travoprost, and 3.1
bottles and 4.9 months of therapy with latanoprost.

CONCLUSIONS
Health care providers are urged to instruct
glaucoma patients in the most efficient method of
instillation for their prostaglandin analogs. For
bimatoprost and latanoprost, the most efficient
method is instillation with the bottle held verti-

cally, with 45 degrees nearly as efficient. For
travoprost, the most efficient method is instillation at 45 degrees.

ACKNOWLEDGMENT
The authors gratefully acknowledge John
Keener, PhD., for assistance with the development of the manuscript.

REFERENCES
1. The Advanced Glaucoma Intervention Study (AGIS).
7. The relationship between control of intraocular
pressure and visual field deterioration. Am. f. Ophthalmol. 130:429-440, 2000.
2. Fiscella, RG. Pharmacological considerations in the
treatment of glaucoma. Manag. Care ll(Supp1. 1):
16-20, 2002.
3. Fiscella, RG. Glaucoma medications: A drug-therapy
review. Manag. Care ll(Suppl. 11):25-31,2002.
4. Kass, M.A., Heuer, O.K., Higginbotham, E.J., et a!. The
Ocular Hypertension Treatment Study: A randomized trial determines that topical ocular hypotensive
medication delays or prevents the onset of primary
open-angle glaucoma. Arch. Ophthalmol. 120:701-713,
2002.
5. Hedman, K., and Aim, A. A pooled-data analysis of
three randomized, double-masked, 6-month clinical
studies comparing the intraocular pressure reducing
effect of latanoprost and timolol. fur. J. Ophthalmol.
10:95-104, 2000.
6. Fellman, RL., Sullivan, E.K., Ratliff, M., et a!., and the
Travoprost Study Group. Comparison of travoprost
0.0015% and 0.004% with timolol 0.5% in patients
with elevated intraocular pressure: A 6-month,
masked, multicenter trial. Ophthalmology 109:9981008,2002.
7. Cohen, J.S., Gross, RL., Cheetham, J.K., et a1. Twoyear, double-masked comparison of bimatoprost with
timolol in patients with glaucoma or ocular hypertension. Surv. Ophthalmol. 49(Suppl. 1):545-S52, 2004.
8. Tsai, J.C, McClure, CA., Ramos, S.E., et a!. Compliance barriers in glaucoma: A systematic classification.
J. Glaucoma 12:393-398, 2003.
9. Jampel, H.D., Parekh, P., Johnson, E., et a!. Preferences
for eye drop characteristics among glaucoma specialists: A willingness-to-pay analysis. f. Glaucoma
14:151-156,2005.
10. Olthoff, CM., Schouten, J.S., van de Borne, B.W., et
al. Noncompliance with ocular hypotensive treatment
in patients with glaucoma or ocular hypertension an
evidence-based review. Ophthalmology 112:953-961,
2005.

 FISCELLA ET AL.

482
11. Mick, A.B., Gonzalez,S., Dunbar, M.T., et a1. A cost
analysis of the prostaglandin analogs. Optometry
73:614-619, 2002.
12. Fiscella, RG., Green, A., Patuszynski, D.H., et aJ.
Medical therapy cost considerations for glaucoma.
Am. J. Ophthnlmol. 136:18-25, 2003.
13. Walt, J.G., and Lee, J.T. A cost-effectiveness comparison of bimatoprost versus latanoprost in patients
with glaucoma or ocular hypertension. Surv. Ophthalmol. (49 Suppl. 1):536--544, 2004.
14. Van Santvliet, L., and Ludwig, A. Determinants of eye
drop size. Surv. Ophtha/mol. 49:197-213,2004.
15. Eter, N., and Gobbels, M. A new technique for tear
film fluorophotometry. Br. J. Ophthalmol. 86:616-619,
2002.
16. Yokoi, N., Bron, A.J., Tiffany, J.M., et al. Relationship
between tear volume and tear meniscus curvature.
Arch. Ophthalmol. 122:1265-1269,2004.

17. Fiscella, RG., Geller, J.L., Gryz, L.L., et a1. Cost considerations of medical therapy for glaucoma. Am. J.
Ophthalmol. 128:426-433, 1999.

Received: July 10, 2006
Accepted: August 22, 2006

Reprint Requests: Richard Fiscella
University of C11icago
833 s. Wood Street
Room 164
Chicago, IL 60612

E-mail: fisc@uic.edu

 Copyright of Journal of Ocular Pharmacology & Therapeutics is the property of Mary Ann
Liebert, Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv
without the copyright holder's express written permission. However, users may print, download, or
email articles for individual use.