WO2000035365A1

WO2000035365A1 - Measurement of tear film break-up-time

Info

Publication number: WO2000035365A1
Application number: PCT/US1999/018391
Authority: WO
Inventors: Victor M. Finnemore; Donald R. Korb
Original assignee: Ocular Research Of Boston, Inc.
Priority date: 1998-08-12
Filing date: 1999-08-12
Publication date: 2000-06-22
Also published as: EP1104263A1; EP1104263A4; AU5482099A

Abstract

The invention is a test strip (30), and method for determining break-up-time of the tear film over the cornea (32). The method comprises providing a planar ophthalmic test strip having a length greater than its width, and having a tip (31) on at least one of its end. The tip adsorbs liquids, has a width that is less than the width of the remainder of the strip, and has a surface area not exceeding 30 square mm. The tip is wetted with liquid and vigorously shaken to remove excess liquid. The tip is then placed in contact with the cornea surface and break-up-time is observed. The test strip is designed to deliver a limited dose of from 0.5 ml to 1.0 νl of liquid to the surface of the cornea.

Description

MEASUREMENT OF TEAR FILM BREAK-UP-TIME

Introduction

The invention relates to a process and test device for measuring the break-up-time

of a tear film over the corneal surface. More specifically, the invention relates to a

process and device for measuring break-up-time that is more accurate and reliable than

processes and devices previously used.

Description of the Prior Art

The break-up-time (BUT) of a tear film is defined as the time interval following a

blink to the occurrence of gaps or breaks in the tear film found on the corneal surface.

BUT is therefore a measure of tear film stability.

There are two methods currently used to measure BUT. The method of choice

involves addition of sodium fluorescein (" fluorescein") to the tear film. The fluorescein

colors the tear film yellow-green and causes the film to fluoresce on exposure to blue

light. Observation of the tear film is made with a slit lamp using moderate magnification

and a blue filter to provide transmission from approximately 330 to 400 nm. A relatively

wide light beam is used to scan the entire corneal surface. The elapsed time for the

development of dark areas in the yellow-green colored fluorescent tear film following a

blink is the BUT. A BUT of about 10 seconds is considered normal; a BUT of 5 to 10

seconds abnormal; and a BUT of less than 5 seconds indicative of dry eye syndrome;

Lemp MA, Hamill Jr., "Factors affecting tear film break-up-time in normal eyes", Arch

Ophthalmo (1973) 89:103-105; Rengstorff, "The precorneal tear film: breakup time and location in normal subjects", Am JOptom Physiol Opt 1974; 51-765; Norn, "Diagnosis of

dry eye"; Lemp et al., 77*e Dry Eye. A comprehensive Guide, Berlin: Springer - Verlag,

1992:152, 153, 178; and Tomlinson, "Complications of Contact Lens Wear", St. Louis:

Mosby, 1992:169, 205, each incorporated herein by reference.

The BUT test using fluorescein has been the standard method for evaluating tear

film stability since about 1969. The method was initially termed BUT but currently is

often referred to as fluorescein-break-up-time (FBUT) to distinguish it from another

method used to measure BUT.

The second method used to measure BUT is a non-invasive test (NIBUT)

introduced in 1985. The NIBUT test uses the principal of reflection from the tear layer

covering the corneal surface. A grid pattern is imaged onto the tear layer over the cornea

and observed at low magnification. The quality of the reflected image is dependent upon

the tear film's role and function in providing a smooth optical surface by masking the

biological irregularities on the surface of the cornea. The elapsed time following a blink

to the first change (disruption) in the image is the BUT. Because of the complexity

involved in performing this test, the NIBUT method is used to a far lesser extent than the

fluorescein method.

There are several methods used to apply fluorescein to the eye. A common

method of applying fluorescein for the FBUT test uses a fluorescein sodium ophthalmic

paper strip which is a paper strip impregnated with about 1 mg of dry sodium fluorescein

and provided in sterilized individual units (fluorescein strip). For use, the dry fluorescein

strip is wetted with sterile saline prior to application to the eye to solvate the fluorescein. The portion of the fluorescein strip which is designed to be wetted, and which then acts as

the source of fluorescein, is usually approximately 5 mm wide by 15 mm long. A specific

recommended method for use of a fluorescein strip is given by Lowther, "Examination of

patients and predicting tear film-related problems with hydrogel lens wear", Dryness,

Tears, and Contact Lens Wear, Boston: Butterworth-Heinemann, 1997:39, 41,

incorporated herein by reference.

Fluorescein may also be added to the eye by application of a liquid drop. A digital

micro pipette is used to apply 1 or 2 μl of fluorescein liquid directly to the eye. A

procedure for adding liquid fluorescein to the eye is given by Nelson, "Diagnosis of

keratoconjunctivitis sicca", Int Ophthalmol Clin 1994;34:37-56, incorporated herein by

reference.

Whether fluorescein is added to the eye by use of a fluorescein strip or as a liquid

drop, current procedures often fail to control either volume or concentration of the

fluorescein and this results in a lack of reproducibility. The most accurate and repeatable

results are obtained if 1 μl of liquid fluorescein is added as a liquid drop; Marquardt et al.,

"Modification of tear film break-up time test for increased reliability presented in Holly

FJ, ed. The Precorneal Tear Film in Health, Disease, and Contact Lens Wear. Lubbock,

TX: Dry Eye Institute, 1986:57-63. The authors compared the fluorescein impregnated

strip method to the addition of 1 and 2 μl of liquid fluorescein and found a significant

increase in reliability with the 1 μl technique. However, applying a drop of fluorescein to

the eye in a volume of 1 μl is difficult and current practice recommends the use of 2 μl as

it is a volume more easily controllable. However, even with a dose of 2 μl, the clinician or research scientist must be particularly skilled in the use of the technique and the

clinical subject must be cooperative. Thus, the application of micro volumes of

fluorescein is technically difficult and therefore not a practicable clinical technique. The

use of a fluorescein strip is also not reliable. The strip is melted with saline and based

upon the procedure used, uncontrolled amounts of fluid are added to the eye. In addition,

some researchers concluded that contact of a paper strip with the eye disrupted the tear

film thereby shortening BUT time.

BUT is an important tool to the clinician and researcher and the FBUT test is the

primary method for measuring tear film stability. However, though the FBUT test is

currently the test of choice for tear film diagnosis, the test is considered inaccurate and

not reproducible by the practitioner. In this respect, a survey in 1998 of 68 of the world's

leading practitioners and researchers in tear film found that the FBUT test was used by

only 19% of the respondents, Korb, "A Survey of 68 Renowned Practitioners for

Preferred Diagnostic Tests for Dry Eye", Presented at British Contact Lens Association

22^nd Annual Clinical Conference, Brighton, England, May 28, 1998. Transactions in

Press. Thus, at present there is no clinically reliable or acceptable method of applying

micro amounts of fluorescein to the tear film for the FBUT test. Consequently, an

improved procedure for administration of the FBUT test is needed.

Summary of the Invention

The subject invention provides a new method for performing the FBUT. The new

method is reliable and reproducible, does not require digital pipette or other costly

equipment, and does not use a strip that disrupts the tear film as a consequence of the method of use. The subject method recognizes that the accuracy and reproducibility of the

FBUT test is dose dependent and that a volume of 0.5 to 1 μl of fluorescein at a

concentration of about 2% provides an accurate and repeatable method for quantifying

tear film BUT.

The invention of the subject application is based upon modification of the existing

fluorescein impregnated strip whereby the dose is controlled and the fluorescein

impregnated strip can be applied to the eye without disruption of the tear film and with

minimal or preferably no sensation to the patient. The invention is based upon redesign of

the tip of the standard fluorescein strip and the application of only this tip to the surface

of the eye. The redesign involves an alternative tip shape with a substantial reduction in

surface area. In this respect, it has been found that a surface area in the range of 5 - 30

square mm is desirable and a surface area of 8 - 15 square mm provides optimal results.

By reduction in the surface area of the tip, a controlled and acceptable volume of

fluorescein is added to the eye and disruption to the tear film by contact with the strip is

minimized.

Description of the Drawings

Figure 1 of the drawings represents a standard fluorescein strip;

Figure 2 of the drawings represents a modified fluorescein strip in accordance

with the invention; and

Figure 3 of the drawings represents application of the strip of Figure 2 to the

surface of the eye. Description of the Preferred Embodiments

The standard fluorescein strip is shown in Figure 1 of the drawings where the strip

10 has tip 11 saturated with fluorescein. The tip 11 of the strip is approximately 5 mm

wide and 15 mm long resulting in a surface area of about 75 square mm. When wetted

with saline, it has been found that the standard strip delivers between 3 and 20 μl of

liquid to the eye dependent upon how the strip is moistened, thereafter shaken, and how

much of the surface area of the tip is touched to the ocular surface. In standard practice,

it is not practical to moisten only a portion of the fluorescein strip with a limited volume

of sterile saline to limit the amount of fluid delivered to the eye since a precision micro

volume delivery system would be needed and inherent poor control in both the wetting of

the fluorescein strip and its application would prevent accurate and reproducible results.

In redesigning the fluorescein strip, various sized fluorescein strips were used to

determine the potential of delivering 0.5 to 1.0 μl of liquid fluorescein by size, surface

area, and shape alterations of the standard fluorescein strip shown in Figure 1. It was

found that the longer the strip, the more variable the volume of fluorescein retained on the

strip after the strip had been moistened with saline. It was further found that the ideal

length of the tip of the strip was between 3 and 10 mm and more preferably, between 4

and 8 mm. The ideal width for the strip was found to be from 0.5 to 3.5 mm and more

preferably, between 1.5 and 2.5 mm. Within the bounds set for these dimensions, it was

further found that the total surface area of the tip should vary between 5 and 30 square

mm and more preferably, within a range of 8 to 15 square mm. If the surface area is in

excess of 40 square mm, more than the desired amount of fluorescein was delivered to the eye and the clinical results are variable. If the surface area is less than 7 square mm, the

volume of fluorescein delivered to the eye is frequently inadequate to provide the

fluorescence required for the FBUT test.

A fluorescein strip in accordance with the invention is illustrated in Figure 2 of

the drawings. The strip 20 has tip 21. The particular strip shown in Figure 2 was prepared

from a fluorescein impregnated strip manufactured by Chauvin Pharmaceuticals Ltd. of

Essex, England by reducing the approximate 15 mm length to approximately 5.0 mm the

5 mm width to 2.0 mm. In this way, the surface area of the tip was reduced to 10 square

mm. Thus, over 85% of the surface area of the standard fluorescein strip was eliminated.

By reduction of the tip as shown in Figure 2, the tip is readily visually differentiated from

the remainder of the tip facilitating use of the strip by the clinician by contact of the

corneal surface with the tip only.

A further goal of the invention is to achieve adequate flexibility and softness so

that when the strip is applied to the eye, it will minimize or eliminate ocular sensation and

thus prevent reflex tearing. Even minimal sensation can result in 1 or more μl of tears

which destabilizes the tear film.

The fluorescein strip of the invention is used in a manner similar to the

fluorescein strip of the prior art. A volume of 0.5 to 1.0 μl of liquid fluorescein can be

consistently delivered to the tear film by moistening a fluorescein strip having a tip

designed as described above with one drop of sterile saline, gently shaking in a 1 - 2 inch

vertical plane, and then applying only the tip of the strip to the eye. The method of

application is shown in Figure 3 of the drawings where the strip 30 has its tip 31 applied directly to the cornea 32. The ability to deliver a micro volume of fluorescein utilizing the

method of the small surface area fluorescein strip is the combined result of the following

factors:

1. Excess fluid falls off of the fluorescein strips' small surface area after it is

moistened, since the smaller the surface area, the less fluid will be

retained.

2. The application of the small surface area of the fluorescein strip to the eye

guarantees consistent delivery of the fluorescein, even by inexperienced

personnel after brief training.

3. Because the small area of the fluorescein strip and the rigidity

characteristics induces almost no sensation when applied to the eye, the

ocular tearing response is minimal.

4. For standardization, it is desirable that the fluorescein strip be held

vertically and gently shaken once in the vertical plane over a distance of 1

- 2 inches.

The volume of fluorescein delivered to the eye using the modified fluorescein

strip of the invention was compared by several in- vitro and in- vivo methods to that

delivered by the laboratory ultra micro digital pipette method, and found to be essentially

identical to fluid volumes of 0.5 to 1 μl. The characteristics of the fluorescence after

instillation on the eye was also equal for both techniques. The fluorescein clearance test,

as modified by Pflugfelder et al., Epstein-Barr virus infection and immunologic

dysfunction in patients with aqueous tear deficiency, Ophthalmology 1990, 97:313-23, provided confirmation. The most convincing demonstration of consistent volumes and

concentration delivered by the modified fluorescein strip, however, was the essentially

identical time periods during which the eyes fluoresced adequately for FBUT evaluation

with both the modified strip and liquid volume techniques. Adequate fluorescence for

FBUT observation usually started immediately after instillation and lasted for 2 - 5

minutes for both techniques. The length of time for tear film fluorescence is a function of

the volume and concentration of the fluorescein. Since the longitudinal fluorescence

times were the same for both the reduced size fluorescein strips and the liquid volumes of

0.5 to 1.0 μl, the volumes and concentrations of the fluorescein delivered to the eye by

both methods were necessarily the same. Thus, the modified fluorescein strip technique

duplicates the ultra micro digital pipettes' ability to deliver micro volumes of fluorescein

providing a clinically applicable method of evaluating BUT and tear film stability in a

reproducible manner.

The following protocol was used for clinical evaluation of the fluorescein strips of

the subject invention.

The purpose of the tests was to compare the accuracy and reproducibility of a

fluorescein strip having its tip reduced to 2 mm wide by 5 mm long (providing a surface

area of 10 square mm) to the standard size strip having a tip 5 mm wide by 15 mm long

(providing a surface area of 75 square mm) in a randomized contralateral study. An

additional study population participated in a randomized contralateral crossover study

design. A further purpose was to compare the ability of each method in classifying the

status of the tear film as dry, borderline (marginal or questionable), or adequate. Fluorets brand of fluorescein sodium ophthalmic strips as manufactured by

Chauvin Pharmaceuticals Ltd., of Essex England, were used for the tests. These

fluorescein strips were impregnated with approximately 1 mg of fluorescein sodium USP.

The portion of the fluorescein strip which is designed to deliver the fluorescein to the eye

after moistening with sterile saline measured approximately 5 mm wide by 15 mm long,

providing a surface area of approximately 75 square mm. The fluorescein strip is

provided by the manufacturer in sterile individual paper packages. The fluorescein strip

is designed to be used on only one eye, and then discarded. A second fluorescein strip is

used for the second eye. For this study, the fluorescein strip was used as provided by the

manufacturer for one eye. A modified fluorescein strip, reduced in size from 5 mm wide

by 15 mm long to 2 mm wide and 5 mm long, reducing the total surface area from

approximately 75 square mm to 10 square mm, was used for the second eye. The

reduction in size was made with a surgical scissors using aseptic technique.

The method for measuring FBUT was the method recommended by Nelson in

Diagnosis of keratoconjunctvitis sicca, Int Ophthalmol Clin 1994, 34:37-56, incorporated

herein by reference, for the study of the conventional (standard) fluorescein strip since it

has also been chosen as the method of choice for FBUT measurement in other

contemporary studies. In addition to the measurements conducted after waiting 1 minute

as recommended by Nelson, three additional measurements were made immediately after

instillation of the fluorescein. The time required to initiate the first measurement after the

instillation of fluorescein varies between 10 and 20 seconds, which is the time being

required to place the patient in the appropriate position in the instrument. The primary study design was a randomized contralateral eye design. The

conventional fluorescein strip was evaluated on one eye, and the modified fluorescein

strip on the contralateral (second) eye. The choice of the conventional or modified

fluorescein strip for right or left eye was randomized. Six trials were completed for one

eye prior to commencing evaluation of the contralateral eye.

The first series of three trials were performed immediately after the instillation of

the fluorescein, requiring 10 - 20 seconds for patient positioning. The trials usually

commenced as described 10 to 20 seconds after the instillation of the fluorescein. The

three trials were performed consecutively. Between each trial, the patient was first asked

to close the eyes and then keep the eyes open as recommended by Nelson. If the BUT

reached 20 seconds in duration, the measurement was terminated since the goal was to

evaluate the accuracy and reproducibility of the two methods for the range of values

required to classify the FBUT values as indicating a dry eye (0 - 4 seconds), a borderline

(marginal or questionable) tear film (5 - 9 seconds), or an adequate tear film (over 10

seconds). The usual time for starting the second series of 3 measurements was 60 - 90

seconds after the first instillation. (No further fluorescein was instilled for the second

series of three measurements.)

After completion of the first eye, the evaluation of the second eye was made

utilizing the same protocol, varying only the type of fluorescein strip. The choice of

order of testing for the two different types of fluorescein strips was selected by

randomization. Randomized contralateral crossover studies were conducted with 5 subjects to

evaluate whether the results were repeatable if the conventional and new type of strips

were evaluated not only on the eye chosen by the randomization process, but also on the

other eye (the selection reversed). These randomized contralateral crossover studies were

conducted after a rest period of 1 ^X to 3 hours.

A total of 19 subjects were evaluated. A summary of the results is presented in

Tables I and II below. Table I summarizes the results for the 5 subjects whose FBUT

values with the standard fluorescein strip method were very variable and obviously not

reproducible. Table II summarizes the results for the 14 subjects who demonstrated

reasonably reproducible FBUT values with the standard fluorescein strip method.

TABLE I

Comparison of FBUT Values With The Standard And Modified Fluorescein Strips For Subjects With Non-reproducible FBUT Values With The Standard Fluorescein Strip.

Table I compares the FBUT values found with the standard fluorescein strip to

those found with the new fluorescein strip for the 5 subjects whose FBUT values were

not reproducible with the standard strip. The measurements are for the first series of

trials, conducted immediately after the instillation of the fluorescein as described in the

protocol section. The results of the second series of trials conducted one minute after

fluorescein instillation, were equivalent to the first series of trials, and therefore are not

included. The data for each of the three trials listed individually in Table I provides a

perspective of the large variability between the three trials. The measurements in three

consecutive trials were considered reproducible if the maximum difference in a series of

three consecutive measurements (trials) did not differ by more than 3 seconds or by more

than a factor of 25%. All 5 subjects did not meet these criteria when the FBUT values

from the standard fluorescein strip method were used, but did meet these criteria when

the FBUT values found with the modified fluorescein strip were used. An example was

subject no. 16, whose FBUT findings when the standard fluorescein strip was used were

2, 14 and 5 seconds. When the new 2 by 5 mm fluorescein strip was used the findings

were very reproducible and were 12, 10 and 11 seconds. The absolute values of the

fluorescein break-up-times were much longer with the new fluorescein strip method.

Although averaging data with large deviations is a questionable practice, it was utilized to

offer a comparison of the averages for the three trials by the standard and new fluorescein

strip methods. The comparison of the averages for the three trials shows a marked and

obvious increase in the average values when performed with the new method; for

example from 2.66 to 9.66 seconds (subject no. 15), from 7.0 to 11.0 seconds (subject no.

16), and from 10.66 to over 20 seconds (subject no. 18).

The results with the 5 subjects whose FBUT values with the standard fluorescein

strips were not reproducible establish that for a series of three trials conducted with the

modified strip, immediately after instillation of the fluorescein, the results were accurate

and reproducible. The results of the second series of three trials conducted one minute

after fluorescein instillation were equivalent. The results with the standard method were not accurate or reproducible in either series of measurements conducted immediately after

the instillation of fluorescein, or in the second series conducted one minute later. Thus,

the new method may be used immediately after instillation of the fluorescein or after 1 or

2 minutes. The elimination of the usual waiting period of one minute prior to conducting

the FBUT is a very important clinical advantage. In addition, significantly greater FBUT

values were found with the modified method indicating that the condition of the tear film

was not as severe as diagnosed by the standard method. These results confirmed the

model and hypothesis that if a greater volume of fluorescein is instilled, the greater the

disruption to the tear film and the greater the artificial reduction of the FBUT values.

TABLE II

Comparison of FBUT Values With The Standard And Modified Fluorescein Strips For Subjects With Reproducible FBUT Values With The Standard Fluorescein Strip.

Table II compares FBUT values found with the standard fluorescein strip to those

found with the modified fluorescein strip for 14 subjects whose FBUT values were

reproducible with the standard fluorescein strip. Since the measurements met the criteria

for reproducibility, only the average of the three trials is presented.

These results with the 14 subjects whose FBUT values with the standard

fluorescein strips were reproducible establishes that the new method provided greater

FBUT values, indicating that the condition was not as severe as diagnosed by the

standard method. Four of the six subjects with dry findings with the standard fluorescein

strip would have been misdiagnosed by at least one classification, as would have one of the four in the marginal classification. Thus, the standard method would have

inaccurately diagnosed 5 of the 10 subjects whose findings were reproducible but

artificially reduced. The results for the second series of measurements conducted

immediately after the initial instillation of fluorescein were equivalent to the first series,

and are therefore not included. Thus, the modified method may be used immediately

after instillation of the fluorescein or after 1 or 2 minutes, providing a very important

clinical advantage and eliminating waiting time. Although the findings with the standard

fluorescein strip method were reproducible, very low FBUT values would be expected to

be consistent and reproducible even if larger volumes of fluorescein were used since this

is the result of very poor tear film stability which results in the rapid tear film break-up.

However, as observed in the data in Table II, the rapid tear film break-up which is

associated with the dry eye state is then further compromised by the excess volume of

fluorescein added to the tear film with the standard method, artificially reducing the

FBUT values. The results found with the standard fluorescein strip should be viewed as

artificially reduced from the more realistic values found with the modified method. Thus,

the FBUT values for the standard method and the modified method were 2.44 versus 6.43

seconds for the dry eye category, and 6.0 versus 9.33 seconds for the marginal tear film

category.

The modified method of the invention yielded accurate and reproducible results

for all four series of three individual trials when the eyes were reversed in the second half

of the randomized contralateral crossover design study, further indicating the

reproducibility of the new method. Table III below presents the FBUT data for the four trials, trials 1, 2 and 3 being conducted immediately after the instillation of the

fluorescein, and trials 4, 5 and 6 conducted one minute later. The same six trials were

then repeated, but with the eyes reversed, after a rest period of 1 Vi to 3 hours.

TABLE III

Comparison of FBUT Values With The Standard And Modified Fluorescein Strips For Five Subjects Participating in the Randomized Contralateral Crossover Study

In view of the above, it can be seen that the subject invention provides a

fluorescein strip and process which permits accurate and reproducible use of the FBUT test for clinical purposes.

Claims

1. An ophthalmic test strip for contact with the cornea, said strip being planar with

a length greater than its width and having a tip on at least one of its ends, said tip being

adsorbent of liquids, having a width that is less than the width of the remainder of the

strip, and having a surface area not exceeding 30 square mm.

2. The test strip of claim 1 where the surface area of the tip varies between 5 and

30 square mm.

3. The test strip of claim 1 where the surface area of the tip varies between 8 and

15 square mm.

4. The test strip of claim 1 where the strip is made of paper.

5. The test strip of claim 1 where at least the tip of the strip is impregnated with

dried fluorescein.

6. The test strip of claim 1 where the width of the tip varies between 0.5 to 3.5

mm and the length of the tip varies between 3 and 10 mm.

7. The test strip of claim 1 where the width of the tip varies between 1.5 and 2.5

mm and the length of the tip varies between 4 and 8 mm.

8. The test strip of claim 1 where the surface area of the tip, following saturation

with saline and shaking, transfers from 0.5 to 1.0 μl of liquid to the surface of the cornea.

9. An ophthalmic test strip for contact with the cornea, said strip being planar with

a length greater than its width and having a tip on one end, said tip being impregnated

with dried fluorescein and having a width that is less than the width of the remainder of the strip which width varies between 0.5 to 3.5 mm, said tip having a surface area not

exceeding 30 square mm.

10. The test strip of claim 9 where the surface area of the tip varies between 5 and

30 square mm.

11. The test strip of claim 9 where the surface area of the tip varies between 8 and

15 square mm.

12. The test strip of claim 9 where the length of the tip varies between 3 and 10

mm.

13. The test strip of claim 9 where the width of the tip varies between 1.5 and 2.5

mm and the length of the tip varies between 4 and 8 mm.

14. The test strip of claim 9 where the surface area of the tip, following saturation

15. A method for determining break-up-time of the tear film over the cornea, said

method comprising the steps of providing a planar ophthalmic test strip having a length

greater than its width and having a tip on at least one of its ends, said tip being adsorbent

of liquids, having a width that is less than the width of the remainder of the strip, and

having a surface area not exceeding 30 square mm, wetting said tip with liquid,

vigorously shaking excess liquid from said tip, contacting the only tip of the strip with the

cornea surface and observing break-up-time.

16. The method of claim 15 where the tip is saturated with dried fluorescein

which is then wetted with saline.

17. The method of claim 15 where the surface area of the tip, following saturation

with saline and shaking, is sized to transfer from 0.5 to 1.0 μl of liquid to the surface of

the cornea.

18. The method of claim 17 where the surface area of the tip varies between 5 and

30 square mm.

19. The method of claim 17 where the surface area of the tip varies between 8 and

15 square mm.

20. The method of claim 15 where the width of the tip varies between 0.5 to 3.5

mm and the length of the tip varies between 3 and 10 mm.