`
`Dynamics of ocular surface pH
`
`WILLIAM H . COLES AND PATRICIA A . JAROS
`
`From the Department of Ophthalmology, Emory University School of Medicine, Atlanta, Georgia, USA
`
`SUMMARY . We studied ocular surface pH in 161 subjects. The mean pH for 133 normal volunteers
`was 7 ·11, SO 1·5. We found that older women had a more alkaline pH than other subjects, that the
`pH shifted from acid to alkaline during the day, that one hour of eyelid closure caused an acid shift
`in pH, and that pH recovered to baseline values within 40 minutes after acid drop instillation. We
`explored the mechanism of pH regulation, and we believe that pH changes could affect contact lens
`toleration, drug effectiveness, and clinical signs in disease processes.
`
`and rinsed with distilled water after use on each
`subject.
`The ocular surface pH was tested in 133 health
`volunteers. One hundred and seventeen were selected
`from Grady Memorial Hospital and 16 were subjects
`not seeking eye care. Patients with inflammation due
`to allergy, conjunctivitis, lid abnormalities, or pro(cid:173)
`longed eyelid closure were excluded from this group.
`The subjects were divided into eight groups by age
`and sex. Effects on pH of age, sex, order or readings,
`first eye measured, and time of day were studied .
`Six patients without ocular disease , who had been
`admitted to an orthopaedic ward , were tested eight
`times throughout one day. The measurements were
`taken every hour in both eyes from 0900 to 1700.
`Effects on pH of time of day were studied.
`to
`Fourteen healthy volunteers were
`tested
`determine the effects of eyelid closure on ocular
`
`Using the direct contact microelectrode we have
`studied ocular surface pH in the inferior lateral fornix .
`Shifts in pH could have therapeutic and diagnostic
`significance . Both antibiotic effectiveness' ~ and cell
`viability ' are affected by pH .
`pH changes may also help in understanding mech(cid:173)
`anisms of disease processes, since rosacea, 4 allergy ,
`and bacterial infection ' alter pH. Even toleration of
`soft contact lenses might be related to pH, because
`lens hydration decreases in acid pH ."
`We present pH findings on 322 eyes (161 subjects)
`and explore mechanisms for the maintenance of, as
`well as deviations from, the steady state pH.
`
`Materials and methods
`
`Subjects' age, sex, race, topical and systemic medica(cid:173)
`tion, and time of day were recorded. Patients taking
`substances known to be excreted in tears (metho(cid:173)
`trextate, aspirin , ampicillin, bacampicillin 7 ") were
`excluded from the study . Informed consent was signed
`by all subjects.
`Both eyes were tested . The eye to be measured first
`was selected at random by coin toss . A digital readout
`(Corning 125) pH meter was standardised with pH 7
`and pH 10 solutions and adjusted to read samples at
`32°C"' before each measurement. A microelectrode
`pH probe (Microelectrodes, Inc. 410) was placed in
`the lateral inferior fornix (Fig. I), and after a 5-10-
`second stabilisation period for the digital readout the
`pH was measured . Three consecutive readings were
`taken in each eye . The probe was cleaned with alcohol
`
`Correspondence to William H . Coles. MD. Emory Universit y Clinic.
`25 Prescott Street. Room J417. Atlanta. Georgia JOJOH . USA .
`
`Fig. I Placement of pH microelectrode in lateral inj(•rior
`fornix.
`549
`
`APOTEX 1009, pg. 1
`
`
`
`550
`
`William H. Coles and Patricia A. Jaros
`
`surface pH. The eye to be patched was selecte? at
`random by coin toss .. A cotton ball placed ~n the hd m
`combination with an eye patch assured hd closure.
`After one hour the patch was removed and the pH
`was immediately measured, followed by measure-
`ments in the unpatched eye.
`.
`.
`Eight subjects were measured before drop ms~llla
`tion (phenylephrine 10%, pH 4·80, and Mydnac~l
`(tropicamide) I 'Yo, pH 6·00), and subsequ.ent~y u~ttl
`two measurements were found at premst1llat1on
`values. Effects on ocular surface pH after instillation
`of dilating drops with regard to pH recovery were
`studied.
`
`7.50
`
`7.40
`
`7.30
`
`I
`Q. 7.20
`
`7.101
`7.00
`
`6.90
`
`8i·
`
`L_9_:~00--~1-1~:0-0~~1~:0~0~~3~:0~0~~5~:~0~0-
`TIME
`Fig. 2 Changes of pH lo alkaline as measured from 09 00 lo
`1700h. Sixsubjec/s.
`
`opposing surface, air or lid conjunctiva, also va.ries
`conditions on the ocular surface. Although we m1ght
`define the ocular surface as tear film layers and epi(cid:173)
`thelium, the many mechanisms that contribute. to
`alterations in surface pH may have little correlatiOn
`with exact structural definition.
`Previous studies on pH have been limited by an
`anatomical approach to measurements. Those studies
`looked at variation in tear film pH by collecting
`samples at the meniscus, 11 12 a technique ~nown. to
`induce an alkaline error. 10 13 The only stud1es wh1ch
`have used the direct contact microelectrode have not
`investigated the variability of the readings at a specific
`point on the surface. Our data show that the pH on
`the conjunctival surface varies in ways that probably
`
`7.60
`
`7.10
`
`6.60
`~ 6.10
`
`5.60
`
`5.10
`
`4.60
`
`Results
`
`For each of the normal subjects three separate pH
`readings were taken. As the three readings were not
`significantly different from one another, the mean of
`the three readings was obtained. Table 1 presents the
`means, standard deviation, and sample sizes for each
`of the eight groups. The mean pH values for the .two
`age groups differed for males and females ( a.nalys1s of
`variance F=5·52, p<0·03). The pH mcreased
`significantly with age in women.
`The pH shifted from acidic to alkaline values when
`six individuals were tested throughout one day
`(p<0·05) (Fig. 2). There was a significant acidic shift
`in pH following one hour of lid closure. The patched
`eyes had a mean pH 7 · 20 before patching and pH 7 ·06
`after one hour of eyelid closure (SE=0·16). The
`control eyes had a mean pH 7 ·19 before and pH 7 ·19
`after one hour (SE=0·16).
`Ocular surface pH recovered to predrop instillation
`levels within 40 minutes in six subjects (Fig. 3).
`
`Discussion
`
`As regards pH on the eye, the ocular surface is more
`than the interface between eye and air or eye and
`conjunctiva. At any point on the conjunctiva or
`cornea the pH, measured by the direct contact micro(cid:173)
`electrode, is probably affected by all the layers of tear
`film and epithelial cell metabolism. In addition the
`
`Table I Ocular surface pH measuremenls on subjec/s
`<40 and ?;40
`
`Race
`
`Sex
`
`B
`w
`
`B
`w
`
`M
`
`M
`
`F
`
`F
`
`A!:e
`
`<40
`"'40
`<40
`"'40
`<40
`"'40
`<40
`"'40
`
`Mean SD
`
`7·070·11
`7·15 0·16
`7·140·13
`7·12 0·23
`7·03 ().()<)
`7·13 0·19
`7·06 0·07
`7·2XO·I4
`
`II
`
`17
`14
`In
`5
`24
`27
`IX
`12
`
`0
`
`30
`
`40
`
`20
`10
`TIME (minutes)
`Fig. 3 pH values before and afler adminislralion of drops.
`A pH reading was laken as baseline. One drop of
`lropicamide (pH 6·0) and one drop of phenylephrine (pH
`4·8) were inslilled in !he lower j(1rnix. 0 represen/s lime of
`drop inslillalion. pH was measured unlit recovery /0 baseline
`values. Six subjec/s.
`
`APOTEX 1009, pg. 2
`
`
`
`Dynamics of ocular surface pH
`
`551
`
`contribute to contact lens intolerance and could
`influence both drug effectiveness and inflammatory
`responses.
`We are not sure that ocular precomeal tear film
`and the pH means at various locations on the surface
`might not vary from the area we studied. Differences
`have been measured between upper and lower
`fornices. 14 We chose to measure ocular surface pH in
`the inferior lateral fornix because it is easily accessible
`to the tip of the probe, and the lateral fornix provides
`consistent, coverage of the probe with lid conjunctiva
`where variation in temperature and atmospheric
`exposure are minimal. Our aim was to decrease
`stimulation and tearing; measurements in the fornix
`avoid the sensitive corneal surface. We found that no
`change occurs among three consecutive pH readings
`and that there is no difference between eyes. This
`indicates that no stimulation occurs that affects the
`readings with the technique as described.
`pH as measured on the ocular surface is not simply
`a reflexive regulation of the tears but an important
`physical measurement influenced by the interaction
`of many factors. To categorise the complexities, our
`data suggest a steady state (with predictable fluctua(cid:173)
`tions around a mean) which is maintained by (1) the
`influences of tear buffer capacity; (2) cellular products
`and secretions on the ocular surface; and (3) factors
`such as age and sex, which affect the mean pH.
`A major stabiliser to the pH on the surface of the
`eye is the bicarbonate buffer system in the aqueous
`portion of tears. 1s Our data and other data 14 1s
`strongly support this concept. The slow rise in pH
`during the day can best be attributed to the diffusion
`of C02 to the atmosphere. With loss of C02 the
`concentration of H + decreases and an alkaline shift in
`pH results ..
`
`Equilibrium: C02+H20~H2COJ~H++HC03-
`
`Eycopcn: ~ C02+ H20 +-H2C03 +- ~ H++HCOJ(cid:173)
`(pH t)
`
`After an hour of eyelid closure, C02 is retained in the
`aqueous tears.
`Eycsclos~d: t C02+ H20-+ H2C03--+ t H++HCOJ(cid:173)
`(pH ~)
`
`A rapid drop in pH occurs as shown by our results.
`The changes in pH associated with lid closure and
`atmospheric exposure are almost surely related to
`C02 concentration primarily involving this buffer
`system. 1s
`The bicarbonate buffer system is most important in
`the maintenance of a consistent pH. But there are
`protein buffers and cellular secretions that contribute
`to changes and fluctuations in pH.
`There are two tear proteins t~at exert a buffer
`
`capacity in the aqueous portion of tears. Lysozyme is
`a basic protein and specific tear albumin is acidic. A
`decrease in lysozyme, which is known in keratocon(cid:173)
`junctivitis sicca and smog eye irritation, may decrease
`tear buffering capacity. 16
`Meibomian glands secrete lipids as a stable compo(cid:173)
`sition of cholesterol and cholesteryl esters on to the
`surface of the tear film. A low-grade infection, how(cid:173)
`ever, changes these Meibomian lipids. 171M Bacteria
`generate enzymes that lyse long-chain components of
`these lipid secretions, forming free fatty acids that are
`irritating to the tear film and corneal surface, creating
`an unstable tear film 17 and probably altering pH on
`the ocular surface.
`Mucin, which is alkaline and secreted by goblet
`cells, is dissolved within the overlying aqueous layer
`and interacts with the outer lipid layer of the tear
`film. 17 Changes in the quality and quantity of mucin,
`as seen with dry eye or acid bums, could influence pH
`on the ocular surface.
`Sex and age affect the mean pH. Our studies show
`that females over 40 have an alkaline shift in the
`mean. We also find a steady increase in the mean
`surface pH with age in all subjects.
`Tear fluid changes that occur with aging include a
`higher incidence of dry eye. This is especially true of
`elderly females. Perhaps mechanisms involved in dry
`eye, such as an increase in mucin or other non-specific
`aging changes, could participate in pH alterations on
`the ocular surface.
`Ocular surface pH means are also affected by time
`of day. We have shown on the ocular surface, as
`others have in tears, 12 that pH becomes more alkaline
`as the day progresses. The bicarbonate buffer system
`probably has the most important influence on this
`alkaline shift in pH. But, since the tear constituents
`of aqueous, mucin, and oils are secreted, other factors
`may also be important. The bodily secretions of
`saliva 19 and urine 211 as well as blood 21 22 show daily
`changes in pH. Known influences include hormonal
`in some cases, metabolic in others. One may speculate
`that the secretions comprising tears could be similarly
`affected.
`If the ocular surface pH is made acid, the return to
`the steady-state levels is probably by reflex tearing.
`Our data showed that lowering the pH with 2 drops
`(Mydriacyl, pH 4·80, and phenylephrine, pH 6·00)
`results in a rapid return within 20-40 minutes to
`initial pH, probably secondary to the tearing induced
`by irritation from the acid substances. The dilution
`effect of the reflex tears is probably most important to
`the return of the pH to its level before the drops were
`given.
`Deviations from the steady state pH can also occur
`from disease processes. pH is acid in rosacea patients,
`as measured by direct contact microelectrode. 4 An
`
`APOTEX 1009, pg. 3
`
`
`
`552
`
`William H. Coles and Patricia A. Jaros
`
`allergic reaction in the eye is seen to cause an alkaline
`shift in pH 5 when studied with a pH sensitive dye,
`bromthymol blue. Further confirmation is needed in
`allergy and other conditions which might alter ocular
`surface pH, such as dry eye and bacterial infection,
`by direct-contact microelectrode measurement.
`The hydration of soft contact lenses can be
`decreased by acid changes in pH. Permalens water
`content decreases 8% when the pH goes from 7·4 to
`6. 6 This has been thought to be a reason for contact
`lens intolerance and possibly cause of the tight lens
`syndrome. 6 Previous reports4 have shown acid
`changes in rosacea on the conjunctival surfaces that
`could be in ranges that would make rosacea patients
`poor contact lens candidates.
`Antibiotic effectiveness in vivo changes with pH. 1
`The aminoglycosides cannot penetrate tissues at pH
`above 7 ·4. 1 Gentamicin is the most commonly used
`aminoglycoside in ophthalmology, and it has been
`shown that conditions of reduced pH may decrease
`the antibacterial effect of this drug. 1 The effectiveness
`of gentamicin as ·well as other ophthalmic amino(cid:173)
`glycosides, such as neomycin and tobramycin, needs
`to be investigated when the ocular surface is altered,
`since an alkaline pH, as shown by our studies, is
`common in older females.
`Cell viability in vitro is best at pH 7·4-7·8; pH
`lower than 6·8 or higher than 7 ·6 slows growth. 3 Cells
`involved in ocular healing are also affected in vitro by
`pH changes. For example, the optimum growth of
`fibroblasts occurs when the pH is in the range of7 ·4 to
`7·8.23
`We conclude that ocular surface pH varies signifi(cid:173)
`cantly in both healthy and disease states. Understand(cid:173)
`ing of the surface pH could have therapeutic and
`diagnostic significance and might help in the under(cid:173)
`standing of complexities of disease processes and
`intolerance of eyes to contact lenses.
`
`We acknowledge the help of the Department of Biometry, Emory
`University, in the statistical analysis.
`
`This study was supported in part by National Institutes of Health;
`National Eye Institute Grant No. EY03923 (Dr Coles); and in part
`by a Departmental Grant from Research to Prevent Blindness.
`
`References
`
`I Vandaux P. Peripheral inactivation of gentamicin. J Antimicrob
`Chemotherl98I;8(supplA): 17-25.
`2 Barza M. Principles of tissue penetration of antibiotics. J Anti·
`microb Chemother 1981; 8 (suppl C): 7-28.
`3 Fischer A. Growth of fibroblasts and hydrogen ion concentration
`of the medium.J Exp Med 1921; 34:447-54.
`4 Jaros PA, Coles WH. Ocular surface pH in rosacea. Contact
`Lens Assoc J 1983; 9: 333-6.
`5 Norn MS. Hydrogen ion concentration of tear fluid. Acta Oph·
`thalmol (Kbh) 1968; 46: 189-200.
`6 McCarey BE, Wilson LA. pH, osmolarity and temperature
`effects on the water content of hydrogel contact lenses. Contact
`lntraocul Lens Med Jl982; 8: 158-67.
`7 Doroshow JH, Gershon YL, Gaasterland DE, Hubbard SP,
`Young RC, Myers EE. Ocular irritation from high-dose metho(cid:173)
`trexate therapy: Pharmokinetics of drug in tear film. Cancer
`1981; 48: 2158-62.
`8 ValenticJP, Leopold IH, Dea FJ. Excretion of salicylic acid into
`tears following oral administration of aspirin. Ophthalmology
`1980; 87: 815-20.
`9 Simon C, Malercyzyk V, Klaus M. Absorption of bacampicillin
`and ampicillin and penetration into body fluids (skin blister fluid,
`saliva, tears) in healthy volunteers. ScandJ Infect Dis 1978; suppl
`14:228-32.
`10 Abelson MB, UdeiiiJ, Weston JH. Normal human tear pH by
`direct measurement. Arch Ophthalmo/1981; 99: 301.
`II Hill RM. Laboratory studies. lnt Ophthalmol Clin 1981; 21:
`223-36.
`12 Carney LG, Hill RM. Human tear pH. Arch Ophtha/mo/1976;
`94:821-4.
`13 Norn MS. Human tear pH. Arch Ophtha/mo/1977; 95: 170.
`14 Fischer FH, Wiederholt M. Human precorneal tear film pH
`measured by microelectrodes. Albrecht von Graefes Arch Klin
`Ophtha/mo/1982; 218: 168-70.
`15 Hill RM, Carney LG. Human tear responses to alkali. Invest
`Ophthalmol Visual Sci 1980; 19: 207-10.
`16 Prctsch RC, Pearlman ME. Human tear lysozyme variables.
`Arch Ophtha/mo/1973; 90:94-6.
`17 Lemp MA. Prccorneal fluid and blinking. lnt Ophthalmol Clin
`1981; 21: 55-64.
`18 Brown SI, Dervichian OG. The oils of the Meibomian glands.
`Arch Ophtha/mo/1969; 82: 537-40.
`19 Dawes C. Circadian rhythms in the flow rate and composition of
`unstimulated and stimulated human submandibular saliva. J
`Physiol (Land) 1975; 244:535-48.
`20 Elliot JS, Sharp RF, Lewis L. Urinary pH. J Ural 1959; 81:
`339-43.
`21 Shock NW, Hastings AB. Studies of the acid-base balance ofthe
`blood. J Bioi Chern 1934; 104:585-601.
`22 Hastings AB, Eisele CW. Diurnal variations in the acid-base
`balance of the blood. ProcSoc Exp Bioi Med 1940;43: 308-12.
`23 Hosford GN, Hicks AM. Hydrogen ion concentration of tears.
`Arch Ophtha/mo/1935; 13: 14-25.
`
`APOTEX 1009, pg. 4
`
`