Tear dynamics and dry eye

https://doi.org/10.1016/S1350-9462(98)00004-4Get rights and content

Abstract

Tears undergo four processes: production by the lacrimal gland, distribution by blinking, evaporation from the ocular surface and drainage through the nasolacrimal duct. Abnormalities in any of these steps can cause dry eye. There are two kinds of tear production, basic and reflex, which can be distinguished from each other by the Schirmer test with nasal stimulation. Reflex tearing is important because it supplies such essential components as EGF and vitamin A, whose deficiency may cause squamous metaplasia. There is no reflex tearing in Sjogren's syndrome because of destruction of the lacrimal gland. In cases of diminished or absent reflex tearing, topical autologous serum is the treatment of choice. Even when there is adequate tear production, insufficient distribution, such as occurs with the decreased blinking associated with the use of video display terminals (VDT), may cause dry eye. Any process or activity that suppresses blinking interferes with tear distribution. Tear evaporation increases under certain conditions and in some diseases. When the exposed ocular surface area is increased, such as in VDT work, tear evaporation increases. Meibomian gland dysfunction (MGD) also causes increased tear evaporation by altering the quality of the oily layer in tears. Tear evaporation can be suppressed by using a warm compresser or a humidifier, narrowing the palpebral fissure, or wearing protective eyeglasses. The tear clearance rate is measured by fluorescein dye dilution in the conjunctiva. When the tear clearance is low, inflammatory cytokines or preservatives accumulate in the conjunctival sac, resulting in ocular surface diseases. Frequent use of artificial tears without preservative is the key treatment. A differential diagnosis of the abnormalities of tear dynamics can give us a proper understanding of the pathogenesis of dry eye. With this knowledge, we can formulate an efficient therapeutic approach.

Introduction

Dry eye is caused by qualitative and/or quantitative abnormalities in the tear film layer. Previously, it was thought that a simple lack of tear production causes dry eye, but the etiology is not so clear (Lemp et al., 1970; Lemp, 1995). Tears are produced by the lacrimal gland, and distributed over the ocular surface by blinking; some of the tears evaporate and the rest drain through the lacrimal punctum. Abnormalities in any of these processes can thus cause dry eye. Whereas, dry eye is usually multifactional, a primary problem can be identified in certain cases.

In this review, I would like to summarize the current understanding of dry eye in terms of four aspects of tear dynamics: 1) production by the lacrimal gland; 2) distribution by blinking; 3) evaporation from the ocular surface; and 4) drainage through the nasolacrimal duct. New concepts have been proposed in each of these areas over the last ten years which should enhance both the understanding and the treatment of dry eye (Tseng and Tsubota, 1997).

Tear production by the lacrimal gland

It has been suggested that the lacrimal gland steadily produces tears at a certain level without any stimulation (Jordan and Baum, 1980). One possibility is that this basic tearing comes from the accessory lacrimal gland, whereas reflex tearing derives from the main lacrimal gland. Since the accessory lacrimal gland's innervation has not been clearly determined, it has been thought to function independent of nervous stimulation.

This concept was brought into question by the observation that tear production diminishes during sleep and under general or local anesthesia. All tear production is thus apparently initiated by some external or internal stimuli.

Although true basic tearing may not exist, the term is still used, and can refer to that tearing caused by such subtle and constant stimuli as blinking and the temperature change due to tear evaporation (Fujishima et al., 1996a). The ocular surface is thus always covered by a tear film layer. Basic tearing can be measured by the Schirmer test with topical anesthesia (Grayson, 1979; Lamberts et al., 1979). Since topical anesthesia cannot anesthetize the skin at the lid margin, there may be some residual stimulation (Clinch et al., 1983), although for all clinical intents and purposes, this method does measure basic tearing. In practice we instill 10 μl of 0.4% oxybuprocaine into the conjunctival sac and wait for five min (Xu et al., 1995). The Schirmer test strip is then placed in the lid margin for five more min. A wet portion of the strip less than 5 mm is considered abnormally low.

The number of dry eye patients with low basic tearing seems to be increasing (Hikichi et al., 1995). Most of the patients still have good reflex tear production, as is reflected by generally normal lacrimal gland biopsies (Xu et al., 1996a). The pathogenesis of this type of dry eye has not yet been determined, but one possibility is a neurosensory abnormality involving neuropeptide production and/or receptor sensation (Table 1).

Reflex tearing is produced by strong physical or emotional stimulation of the lacrimal gland. The tears thus produced contain essential components, such as vitamin A and EGF, for the proliferation and differentiation of the corneal and conjunctival epithelium (Ohashi et al., 1989; Ubels et al., 1986; van Setten et al., 1989; Wilson, 1991). Even if basic tearing is decreased, accelerating desiccation of the ocular surface, if reflex tears are present, can provide the ocular surface epithelium with substances necessary for proper epithelial wound healing (Toda et al., 1996).

The Schirmer test without topical anesthesia, in which the test strip stimulates the cornea, conjunctiva and lid margin, has generally been used to measure reflex tearing. However, a result of 0 mm does not necessarily mean that the patient is incapable of producing reflex tears. Schirmer described the measurement of reflex tearing by stimulating the nasal mucosa with a camel's hair brush after anesthetising the ocular surface with 4% cocaine. To check maximal reflex tearing, we have modified the Schirmer II test by using a cotton swab to stimulate the nasal mucosa without any anesthetic (Tsubota, 1991; Tsubota et al., 1996a; Xu et al., 1995). Although the Schirmer II test is rarely used because either reflex tearing is assumed to be intact (Grayson, 1979) or the regular Schirmer test is considered more accurate (Tabbara, 1983), we have seen certain dry eye patients who are incapable of reflex tearing (Tsubota et al., 1996b), and for whom our technique is an important test.

Briefly, we screen these patients with the regular Schirmer test for five min without topical anesthesia. A cotton swab (8 cm long, 3.5 mm wide at the top) is then inserted into the patient's nasal cavity, slightly upward and parallel to the cavity's lateral wall (Fig. 1) (Tsubota et al., 1996a). Next, the Schirmer paper is placed in the conjunctival sac for five min while the cotton swab is kept in place. The wet portion of the strip is measured as in the regular Schirmer test. A value of less than 10 mm, indicates decreased reflex tearing.

According to the new classification of dry eye, patients with a tear deficiency can be divided into two categories, those with Sjogren's syndrome (SS) and those without (Lemp, 1995; Farris et al., 1991). The ocular surface abnormalities are much more severe in the former (Pflugfelder et al., 1990a; Tsubota, 1994a; Tsubota et al., 1994b), as evidenced by squamous metaplasia and greater rose bengal and fluorescein staining (Fig. 2a,b; Table 2) (Nelson et al., 1982; Tseng, 1985). Since both groups lack basic tears (Tsubota et al., 1996a), the simple desiccation theory cannot explain the differences. That difference is due to the ability of non-SS dry eye patients to produce reflex tears (Tsubota et al., 1996a).

SS is an autoimmune disorder characterized by infiltration of lymphocytes into the lacrimal and salivary glands (Fox et al., 1986; Tabbara, 1983). Although the etiology is not determined yet, an association with Epstein-Barr virus infection is suspected (Fox et al., 1986; Pflugfelder et al., 1990b, Pflugfelder et al., 1990c; Saito et al., 1989; Tsubota et al., 1995a). Furthermore, there may be a single antigen responsible for both lacrimal and salivary gland disease because T cell receptor (TCR) restriction was found in both glands in the same SS patients by PCR-single strand conformation polyphorism (SSCP), although simple RT-PCR could not detect the restricted epitopes of a common antigen (Fox et al., 1986; Matsumoto et al., 1996; Mizushima et al., 1995; Pflugfelder et al., 1990b; Saito et al., 1989; Tsubota et al., 1995a). Recent animal studies also supported the importance of autoantigens as the pathogenesis of SS (Haneji et al., 1997). It is believed that the infiltrating lymphocytes, which are present in and cause the destruction of the lacrimal glands only in SS dry eye patients (Tsubota et al., 1996a), suppress reflex tears, depriving the ocular surface epithelium in these patients of critical nutrients and hydration (Table 1).

The evidence suggests that tears are important for supplying vitamin A and growth factors such as EGF, HGF and TGF-β that are essential for maintaining the integrity of the ocular surface epithelium. Vitamin A is necessary for the differentiation and maintenance of the mucosal epithelium. EGF and HGF help in the proliferation of epithelial transient amplifying and stem cells (Li and Tseng, 1996). When patients lose both reflex and basic tears, no essential factors can be supplied to the ocular surface. In other words, severe ocular surface alterations in SS are not due to simple desiccation, rather, they are due to the lack of substances essential for epithelial growth.

Superior limbic keratoconjunctivitis (SLK) is a disease of unknown etiology which is characterized by chronic inflammation of the superior bulbar conjunctiva near the limbus (Ostler, 1983; Udell et al., 1986). It is diagnosed by fine punctate rose bengal staining of the superior cornea and conjunctiva. The symptoms are severe and often difficult to control. Various treatments have been attempted, including the application of silver nitrate solution, resection of the superior bulbar conjunctival lesion, pressure patching followed by soft contact lens use, thermocauterization of the superior bulbar conjunctiva and use of vitamin A drops (Ostler, 1983). However, these techniques are not always effective, and in some cases no conventional treatment is successful.

We have observed that the upper part of the tear meniscus in SLK patients is lower than normal. Since SLK is often observed in dry eye (Cher, 1969), we believe that the condition may be caused, at least in part, by tear deficiency. We feel that localized tear deficiency in the upper conjunctiva may result in abnormal friction between the upper lid and the superior corneal limbus (abnormal blink), causing irritation and interfering with the supply of essential chemicals such as growth factors and vitamin A to the upper conjunctiva (lack of tear components). This hypothesis explained the fact that SLK is strongly associated with thyroid abnormalities where no tears can accumulate between the lid and superior limbus. We employed punctal occlusion as a treatment to test our hypothesis in SLK patients who were refractory to other approaches.

We used cautery and sutures to permanently occlude the lacrimal puncta of 11 SLK patients (22 eyes), in whom topical treatment was ineffective. All cases responded positively to the occlusion (Fig. 3a,b). Rose bengal and fluorescein staining decreased dramatically (2.7±1.6 to 1.1±1.8 and 1.4±1.2 to 0.4±0.8, respectively) (Yang et al., 1997). Impression cytology revealed improvement of squamous metaplasia in the superior conjunctiva as well as increased goblet cells in 9 of 13 eyes (69%) examined. Subjective symptoms improved in all 22 eyes (100%). Therefore, we believe selective punctal occlusion to be a highly effective treatment for SLK patients refractory to traditional methods.

Some forms of squamous metaplasia of the ocular surface epithelium may be treatable by replenishing missing essential tear components (Tsubota, 1998a, Tsubota, 1998b). Since prevention of desiccation or use of commercially available artificial tears cannot provide these substances, we must consider an alternate method.

One possibility is to use autologous serum, which contains vitamin A, EGF, and TGF-β. Such an approach was reportedly beneficial in treating SS (Fox et al., 1984). A total of 40 ml of blood is obtained by venipuncture and centrifuged for five min at 1500 r.p.m. The serum is carefully separated in a sterile manner and put into a bottle with colour coating. Since vitamin A is easily degraded by light, patients are instructed to keep the bottle in a dark, cool place. We use either the serum itself or 20% diluted autologous serum to minimize the side effects and increase the yield from a single venipuncture. Patients are told to use the serum drops six to ten times a day in addition to their other regimens. This treatment has caused decreased rose bengal and fluorescein staining (Fig. 4a,b), and improvement of squamous metaplasia (Fig. 5a,b).

Using the above approach, we were able to reconstruct the ocular surface in patients with severe dry eye such as is associated with ocular pemphigoid or Stevens-Johnson syndrome (Tsubota et al., 1996c; Tsubota et al., 1996d). This technique may be combined with stem cell transplantation from the corneal epithelium and substrate transplantation by amniotic membrane (Fig. 6a,b) (Kim and Tseng, 1995; Tsai and Tseng, 1994; Tseng and Tsai, 1991). Even dry eye patients with results of 0 mm on the regular Schirmer test may be candidates for ocular surgery including excimer laser ablation (Toda et al., 1996) as long as they have reflex tearing to allow proper wound healing. Although absolute dry eye might be considered a contraindication to ocular surgery, the use of drops derived from autologous serum may also enable these patients to undergo surgery.

Since even in SS a very small amount of tear production is maintained, lacrimal punctal occlusion may help by maximizing the time that essential tear components are in contact with the ocular surface epithelium. Since SS patients do not have any reflex tears, epiphora should not be a problem. The upper and lower punctum can be occluded with a punctal plug (Hamano et al., 1983). Rose bengal and fluorescein staining has been shown to improve dramatically within 24 h of occlusion of the upper and lower puncta (Fig. 7a,b), with corresponding cellular level improvements. In very advanced cases, punctal occlusion can be combined with autologous serum drops.

Section snippets

Control of blinking

Tears produced by the major and minor lacrimal glands are spread over the ocular surface and drained through the lacrimal punctum by blinking. Blinking proceeds from temporal to nasal, pushing the tears into the punctum. Blinking abnormalities can cause improper tear distribution and hence ocular surface problems (Doane, 1980).

Human blinking is often compared to that in other animals, and averages 15–20 times per min under relaxed conditions (Blount, 1928; Karson, 1988; Tsubota et al., 1996b;

What determines tear evaporation?

It has been shown that approximately 10% of the total tear volume evaporates while 90% drains through the lacrimal punctum (Mishima et al., 1966). In patients with a quantitatively (e.g., decreased production) or qualitatively (e.g., meibomian gland dysfunction) deficient tear film layer, evaporation becomes relatively more important. We have developed a simple method to quantitatively evaluate tear evaporation (Tsubota and Yamada, 1992). Mathers et al. also developed a reliable method for

Principles of tear drainage

Secreted by the lacrimal gland and spread over the ocular surface by blinking, tears either evaporate or migrate to the inner canthus and ultimately to the nasal cavity through the nasolacrimal duct. Tear dynamics are thus determined by four factors: secretion, blinking, evaporation, and drainage; we have already extensively described the first three. For drainage to be effective, gravity, siphonage, capillary attraction, and muscular activity must work effectively to conduct tears down the

Other aspects of tear dynamics

In addition to the four major elements of tear dynamics discussed above, a minor factor in the tear turnover is absorption by the cornea and conjunctiva. Anterior fluorometry studies (MacDonald and Maurice, 1991) have shown that water and electrolytes can pass through the conjunctival epithelium, and to a lesser extent, through the corneal epithelium, which has rigid cell-cell adhesions. The cornea and conjunctiva contribute various inflammatory components to tear, including Il-1, TNF-α and

Future directions

Now that the significant components of tear dynamics have been identified and their characteristics broadly described, interest is focusing on the cellular and molecular levels. Causes of decreased tear production by the lacrimal gland, are attracting attention. Is the problem due to abnormal nerve control or acinar function itself? If it is due to malfunction of the acinar cells, does this result from a problem at the receptor, transcriptional, or translation level? An additional question is

Acknowledgements

This manuscript is supported by a grant from the Oral Health Science Center of Tokyo Dental College, Chiba, Japan.

References (89)

  • S. Sahlin et al.

    Evaluation of the lacrimal drainage function by the drop test

    Am J Ophthalmol.

    (1996)
  • S.C. Tseng

    Staging of conjunctival squamous metaplasia by impression cytology

    Ophthalmology

    (1985)
  • S.C. Tseng et al.

    Important concepts for treating ocular surface and tear disorders

    Am J Ophthalmol.

    (1997)
  • K. Tsubota

    The effect of wearing spectacles on the humidity of the eye

    Am J Ophthalmol.

    (1989)
  • K. Tsubota

    The importance of the Schirmer test with nasal stimulation [letter]

    Am J Ophthalmol.

    (1991)
  • K. Tsubota et al.

    Surgical reconstruction of the ocular surface in advanced ocular cicatricial pemphigoid and Stevens-Johnson syndrome

    Am J Ophthalmol.

    (1996)
  • K. Tsubota et al.

    Treatment of severe dry eye

    Lancet.

    (1996)
  • K. Tsubota et al.

    Poor illumination, VDTs and desiccated eyes

    Lancet

    (1996)
  • I.J. Udell et al.

    Treatment of superior limbic keratoconjunctivitis by thermocauterization of the superior bulbar conjunctiva

    Ophthalmology

    (1986)
  • S. Wilson

    Lacrimal gland epidermal growth factor production and the ocular surface

    Am J Ophthalmol.

    (1991)
  • H. Yang et al.

    Punctal occlusion for the treatment of superior limbic keratoconjunctivitis

    Am J Ophthalmol.

    (1997)
  • W. Blount

    Studies of the movements of the eye lids of animals: blinking

    Quarterly J Exp Physiology.

    (1928)
  • R. Bowman et al.

    Chronic glepharitis and dry eye

    International Ophthalmology Clinics

    (1987)
  • I. Cher

    Clinical features of superior limbic keratoconjunctivitis in Australia

    Arch Ophthalmol.

    (1969)
  • T. Clinch et al.

    Schirmer's Test. A closer look

    Arch Ophthalmol.

    (1983)
  • J. Dougherty et al.

    The role of tetracycline in chronic blepharitis

    Investigative Ophthalmology Vis Sci.

    (1991)
  • R. Farris et al.

    Sjogren's syndrome and keratoconjunctivitis sicca

    Cornea

    (1991)
  • Foster, C. (1983) Corneal manifestations of neurologic diseases. In The Cornea (eds. G. Smolin and R. Thoft), pp....
  • R. Fox et al.

    Beneficial effect of artificial tears made with autologous serum in patients with keratoconjunctivitis sicca

    Arthritis Rheum.

    (1984)
  • R. Fox et al.

    Sjogren's syndrome: proposed criteria for classification

    Arthritis Rheum.

    (1986)
  • H. Fujishima et al.

    Corneal temperature in patients with dry eye evaluated by infrared radiation thermometry

    Br J Ophthalmology.

    (1996)
  • H. Fujishima et al.

    Allergic conjunctivitis and dry eye

    Br J Ophthalmol.

    (1996)
  • J. Gilbard et al.

    Decreased tear osmorality and absence of the inferior marginal tear strip after sleep

    Cornea

    (1992)
  • Grayson, M. (1979) The dry eye. St. Louis: The CV Mosby...
  • H. Hamano et al.

    A new method for measuring tears

    CLAO J.

    (1983)
  • N. Haneji et al.

    Identification of a-fodrin as a candidate autoantigen in primary Sjogren's syndrome

    Science

    (1997)
  • T. Hikichi et al.

    Prevalence of dry eye in Japanese eye centers

    Graefe's Arch Clin Exp Ophthalmol.

    (1995)
  • Karson, C. (1988) Physiology of normal and abnormal blinking. In Advances in Neurology (pp. 25–37). New York: Raven...
  • J. Kim et al.

    Transplantation of preserved human amniotic membrane for surface reconstruction in severely damaged rabbit corneas

    Cornea

    (1995)
  • D. Lamberts et al.

    Schirmer test after topical anesthesia and the tear meniscus height in normal eyes

    Arch Ophthalmol.

    (1979)
  • M.A. Lemp et al.

    Corneal dessication despite normal tear volume

    Ann Ophthamol.

    (1970)
  • M.A. Lemp

    Report of the National Eye Institute/Industry workshop on Clinical Trials in Dry Eyes

    CLAO J.

    (1995)
  • D. Li et al.

    Differential regulation of cytokine and receptor transcript expression in human corneal and limbal fibroblasts by epidermal growth factor, trasnforming growth factor-a, platelet-derived growth factor B, and interleukin-1B

    Invest Ophthalmol Vis Sci.

    (1996)
  • W.D. Mathers et al.

    Meibomian gland dysfunction in chronic blepharitis

    Cornea

    (1991)
  • Cited by (0)

    View full text