JOURNAL OF OCULAR PHARMACOLOGY AND THERAPEUTICSVolume 18, Number 4, 2002© Mary Ann Liebert, Inc.

Differences in Bioavailability of Fornixes inDifferent Head Positions

SHENG-YAO HSU and CHANG-PING LIN

Department of Ophthalmology, Kaohsiung Medical University, Kaohsiung, Taiwan

ABSTRACT

Clinically, in infectious corneal or scleral ulcers, we have found some upper le-sions to show a poorer response to eye-drops than lower lesions. This clinical observa-tion stimulated our interest to investigate the differences of drug bioavailability in up-per and lower fornixes in three different head positions. Seventeen people, 34 eyes, wereenrolled in this study. There were three head positions for 0.1% fluorescein eye-dropsapplication, including sitting, supine, and supine-with-chin-up. Schirmer’s test paperwas placed in the fornix to absorb the fluorescein, and the bioavailability was analyzedby fluorescence spectrophotometry. Fluorescein bioavailability of upper-and-lowerfornixes were 1.30 3 1025% and 7.33 3 1025%, 3.93 3 1025% and 9.57 3 1025%,and 23.19 3 1025% and 5.09 3 1025% in sitting, supine, and supine-with-chin-up po-sitions, respectively. Bioavailability of the lower fornix was significantly higher thanthat of the upper in the sitting position, and the bioavailability of the upper fornix wassignificantly higher than for the lower fornix in the supine-with-chin-up position. Thebioavailability of the upper fornix in the supine-with-chin-up position was significantlyhigher than that in the sitting and supine positions, respectively. The total fluoresceinbioavailability of both fornixes in the supine-with-chin-up position was significantlyhigher than that in the sitting position. We postulate that different head positions caninfluence drug bioavailability in the upper and lower fornixes. Ocular surface lesionsin different sites may require different head positions during eye-drop application to ob-tain the best therapeutic results.

INTRODUCTION

Traditionally, when applying eye-drops, we teach patients to tilt their heads backwards, look up-ward, gently grasp the lower outer eyelid below the lashes, pull the eyelid away from the globe, andplace one drop of eye-drop into the lower fornix. Patients must continue to hold the eyelid in this po-sition for a few seconds to allow the solution to gravitate to the deepest position of the lower fornix,then close the eye or press the punctum for several minutes (1). Clinically, in infectious corneal orscleral ulcers, we have found that some upper lesions show a poorer response to eye-drops when com-pared to lower lesions. This clinical observation compelled us to investigate the differences of drugbioavailability in upper and lower fornixes in different head positions.

371

MATERIALS AND METHODS

The 17 volunteers, 34 eyes, enrolled in the study were all young people without histories of oc-ular surgery or trauma, dry eye syndrome, and lid or nasolacrimal duct abnormalities. The mean agewas 23.6 6 0.5 years old with a range of 23 to 24. Fluorescein 0.1% eye-drops, 50 ml per drop, wereapplied in both eyes in three different head positions, including sitting, supine, and supine-with-chin-up. In supine position, the upper and lower fornixes were kept almost in the same level. In supine-with-chin-up position, a pillow was put under the shoulder of volunteer to keep the upper fornix lowerthan the lower. Each position was tested on different days with three days apart. First, we applied onedrop of topical anesthesia, 0.5% proparacaine hydrochloride, into the lower fornix and waited for 5minutes, then instilled 0.1% fluorescein into the lower fornix in the sitting and supine positions, andinto the upper fornix in the supine-with-chin-up position. The volunteers closed their eyes and waitedfor 5 minutes again. At the end of that time, we put a 5 mm 3 12.5 mm Schirmer’s test paper intoeach fornix and left it in place for one minute to absorb the fluorescein. When the Schirmer’s test pa-per soaked with fluorescein was removed, it was put into a test tube with 3 ml of distilled water. Weshook the test tubes on a rotator machine for 60 minutes to extract the fluorescein from the Schirmer’stest paper. The solution was then poured into a quartz tube and the level of fluorescein solution wasmeasured by fluorescence spectrophotometry. Since the readings on a spectrophotometry possess nopersonal bias, the observer was not masked. Before the quartz tube was reused, it was cleaned 6 timeswith 70% alcohol and another 6 times with distilled water.

372

FIGURE 1. The Standard Curve of Fluorescence Spectrophotometer Value (OD: optical density)—Bioavailability.

A primitive procedure was performed to determine the correlation between fluorescence spec-trophotometer levels and the concentration of fluorescein solution. Three milliliters of 10% fluores-cein solution were poured into a quartz tube, and fluorescein levels by fluorescence spectrophotom-etry were measured. Then 0.3 ml of 10% fluorescein solution was mixed with 2.7 ml of distilled waterto become a 1% solution. These 10 times dilutions were repeated to 10215%, and the fluorescein spec-trophotometer level was determined. All of the procedures were repeated 8 times. The standard curveof the correlation between mean fluorescence spectrophotometer levels and fluorescein solution con-centrations was drawn by GRAPHER software (GRAPHER Version 1.26, 2-D Graphing System,Golden Software, Inc, Colorado, USA). The correlation between fluorescence spectrophotometer val-ues and the concentration of fluorescein is shown as a curve (Fig.1).

We analyzed the differences of fluorescein bioavailability in upper and lower fornixes in thethree head positions from the same eye by dependent sample t-test. P values less than 0.05 were con-sidered significant. We also analyzed the differences of fluorescein dosages among the three kinds ofhead positions in the upper, lower, and total fornix groups by ANOVA and Tukey HSD post hoc test.P values less than 0.05 were considered significant.

RESULTS

Table 1 shows the mean bioavailability of upper or lower fornixes in the three head positionsfor eye-drop application. The bioavailabilitiy of upper and lower fornixes was 1.30 3 1025% and7.33 3 1025% respectively in the sitting position. Bioavailabilitiy of the upper and lower fornixeswas 3.93 3 1025% and 9.57 3 1025% respectively in the supine position, and 23.19 3 1025% and5.09 3 1025% respectively in the supine-with-chin-up position. In the sitting position, bioavailabil-ity of the lower fornix was significantly higher than that of the upper fornix (P 5 0.005). There wasno difference between both fornixes in the supine position. In the supine-with-chin-up position, themean bioavailability of the upper fornix was significantly higher than that of lower fornix (P 5 0.003).

When comparing the bioavailability in the upper fornix among different positions, the fluores-cein concentration was significantly higher in the supine-with-chin-up position than in the sitting (P 5

0.002) and supine positions (P 5 0.007), but there was no difference between the sitting and supinepositions. Comparing fluorescein concentrations in the lower fornix, there were no differences among

373

the three positions. In comparing total concentrations for upper and lower fornixes, bioavailabilitywas significantly higher in the supine-with-chin-up position than in the sitting position (P 5 0.046),but there were no differences between the sitting and supine positions and between the supine andsupine-with-chin-up positions (Table 2).

In the supine-with-chin-up position, the upper fornix becomes lower, so we also compared thebioavailability of the upper fornix in the supine-with-chin-up with that of the lower fornix in sitting.Bioavailability of the upper fornix was significantly higher in the supine-with-chin-up than that of thelower fornix in sitting, but there were no significant differences when compared with that of the lowerfornix in the supine position. The mean bioavailability of the upper fornix in the supine-with-chin-upposition was significantly higher than that of other fornixes and positions, except for that of the lowerfornix in the supine position (Table 3).

DISCUSSION

Drug absorption depends on the molecular properties of the drug, the viscosity of its vehicle,and the functional status of the tissue forming the barrier to penetration (2). Drug concentration andthe contact time at the site of action also play major roles (3–6). When a single drop of medicationof 50 ml is applied, the nasolacrimal duct rapidly drains the excess, or some may be blinked out ofthe eye onto the lid (3). If the eyelids are not squeezed after dosing, a total volume of perhaps 30 mlcan be held for a brief time. Tear turnover rate when eye-drops are applied is approximately 16% perminute in the undisturbed eye, and only 40% of the medication are retained 5 minutes after the dropis instilled (4). The tear film concentration can be prolonged by manually blocking the nasolacrimalducts or by tilting the head back to reduce drainage (7). McGraw and Rollin, suggested that sincegravity plays an important role in tear dynamics, the patient can be instructed to direct the head andeyes toward the feet for 3 minutes following topical drug application to increase the bioavailabilityof cycloplegics (8).

Traditionally, we teach patients to look upward and place eye-drops into the lower fornix, butthis does not distribute the drug over the total ocular surface equally, as has been proved in this study.Fluorescein bioavailability of the upper or lower fornixes is significantly different in the sitting andsupine-with-chin-up positions. This implies that different head positions can influence the drug’s dis-

374

tribution over the ocular surface, and is obviously related to gravity and the tear excretion system.The mean bioavailability of the upper fornix in the supine-with-chin-up position is significantly higherthan that in the other fornix and positions, except for that of the lower fornix in the supine posi-tion.Tilting the head back increases the drug’s concentration in the upper fornix, but concentrationsof drug in the lower fornix is no different in any of the three positions. With regard to lacrimal ex-cretory system, tear excretion from the lower punctum is greater than from the upper (9). When eye-drops are applied to the lower fornix, they are excreted into the lacrimal drainage system quickly; andif they are applied to the upper fornix in the supine-with-chin-up position, eye-drop excretion de-creases and the coating is maintained on the ocular surface for a greater length of time. The upperfornix is deeper than the lower. Theoretically, it provides a larger capacity as a drug reservoir andmay also be a factor in the higher bioavailability of the upper fornix.

Although tear flow of the healthy young individual and anesthetic application before test maydiffer from the real clinical situation, our model of 0.1% fluorescein application is convenient forquantitative analysis of the ocular surface drug distribution. The anionic diagnostic agent sodium flu-orescein is a hydrophilic agent and a good example for this study’s design. Once the fluorescein isinstilled, it mixes rapidly with the tears. The amount of fluorescein penetrating the intact epitheliumis therefore small. Approximately 99% of the fluorescein exits by way of the tears, and only a verysmall amount penetrates the corneal epithelial barrier and enters the stroma (3). Fluorescence spec-trophotometry is very sensitive to fluorescein levels and is easy to perform for the study.

We postulate that different head positions can influence drug bioavailability in upper and lowerfornixes. We hypothesize that ocular surface lesions in different sites may require the adjusting ofhead positions during eye-drop application to obtain the best therapeutic results. The total drugbioavailability of both upper and lower fornixes in the supine-with-chin-up position is about 3 timessignificantly higher than that in sitting; and 2 times higher than that in supine, although without sta-tistically significant difference. The supine-with-chin-up position is the best one to achieve the high-est bioavailability.

375

REFERENCES

1. Bartlett, J.D. Ophthalmic drug delivery. In Clinical Ocular Pharmacology, Chapter 3. Bartlett,J.D., Jaanus, S.D., ed., Butterworth-Heinemann, Woburn, 2001, pp. 41–61.

2. Burstein N.L., and Anderson J.A. Review: corneal penetration and ocular bioavailability of drugs.J. Ocular Pharmacol. 1: 309–326, 1985.

3. Bartlett, J.D. Ophthalmic drug delivery. In Clinical Ocular Pharmacology, Chapter 2. Bartlett,J.D., Jaanus, S.D., ed., Butterworth-Heinemann, Woburn, 2001, pp. 19–40.

4. Mindel, J.S. Pharmacokinetics. In Duane’s Foundations of Clinical Ophthalmology, Vol. 3, Chap-ter 23, Tasman, W., Jaeger E.A., ed., Lippincott-Raven Publishers, Pennsylvania, 1995, pp. 1–17.

5. American Academy of Ophthalmology. Basic and Clinical Science Course. Section 2: Funda-mentals and principles of ophthalmology. San Franciso, USA, 1999-2000, pp. 384–386.

6. Ueno, N., Refojo, M.F., and Abelson, M.B. Pharmacokinetics. In Principles and Practice of Oph-thalmology, Albert, D.M., Jakobiec, F.K., Robinson, N.L. ed., W.B. Saunders Company, Penn-sylvania, 1994, pp. 916–929.

7. Fraunfelder F.T. Extraocular fluid dynamics: how best to apply topical ocular medications. Trans.Am. Ophthalmol. Soc. 74: 457–487, 1976.

8. McGraw, B.E., and Rollins C.L. Method for increasing bioavailability of cycloplegics. J. Am.Optom. Physiol. Opt. 55: 795–800, 1978.

9. Kanski, J.J. Clinical ophthalmology, Kanski, J.J., ed., Butterworth-Heinemann, Woburn, 1999,pp. 44–45.

Received: October 27, 2001Accepted for Publication: March 1, 2002

Reprint Requests: Chang-Ping Lin, M.D., Ph.D.Department of OphthalmologyKaohsiung Medical University100 Shih-Chuan 1st Rd.Kaohsiung, TaiwanE-mail: lincp@cc.kmu.edu.tw

376