Meibomian gland dysfunction (MGD) is a common disorder representing a condition of various etiologies.1 Obstructive meibomian gland disease is considered to be the most common subtype,2 typically caused by altered meibomian secretion and obstruction of the terminal ducts or plugging of the orifices.2,3 Normally, the meibomian gland secretion forms reservoirs along the upper and lower lid margin from which the tear film lipid layer is formed, which acts to stabilize and prevent evaporation of the tear film.2 In obstructive MGD, the secretion delivered to the reservoir may be reduced to a point where the lipid layer is insufficient to prevent evaporation effectively, thus causing dry eye.4
There is considerable variation in the temperature at which the meibomian secretion begins to melt in scientific literature because of the significant variation in chemical composition of the meibum between individuals and even between glands of the same eye.1 However, higher lipid melting points have been found in MGD compared with those in normal eyes.5–7 Therefore, eyelid-warming therapies have been recommended as a treatment for MGD to clear the obstructed glands.7–13 Here, the heat transferred melts the pathologically altered lipids that have become inspissated and stagnant and relieve the dry eye symptoms associated with MGD.8,14 Warming can be achieved by a variety of means, such as warm moist compresses,10,15 warm moist air,7,11 warm compression devices,8 light-emitting diode and chemical reaction–based eye masks, and heat and pulsatile pressure combination therapy.9,13,16
Since 2004, an eyelid-warming device called the MGDRx EyeBag (The EyeBag Company, Halifax, UK) has been commercially available to treat MGD, composed of one silk and one cotton surface, filled with flax seed (Fig. 1). However, there seems to be no scientific evidence to support the efficacy of this device. Therefore, the aim of this study was to determine the temperature of the external and internal upper and lower eyelids after a warm compress with the MGDRx EyeBag. In addition, the short-term effects of warm compresses with the MGDRx EyeBag on the tear film lipid layer thickness (TFLLT) and tear film stability were investigated.
Subjects recruited consisted of staff, students, and optometry clinic patients from Aston University (Birmingham, UK). The study required subjects to be at least 18 years of age, asymptomatic, non–contact lens wearers with no active eye disease, no ocular medications, no systemic medications known to affect the eyes, and no history of eye surgery in the last 3 months. Subjects were enrolled after written informed consent and underwent slit lamp biomicroscope examination to ensure that there was no active eye disease present, including MGD, based on meibomian gland function (quality or expressibility score, <1).17 The primary outcome measure of the study was measurement of ocular surface temperature. To detect a treatment effect of 1°C change in temperature with 80% power at the 5% level of statistical significance (α = 0.05), 22 subjects were required as each subject acted as his or her own control (one test eye, contralateral control eye) (SD of normal values, ±1.1°C).18 The study was approved by the institutional review board and conformed to the tenets of the Declaration of Helsinki.
An eyebag was heated for 40 seconds in an 800-W microwave oven at full power as recommended in the manufacturer’s instructions. The region of the heated eyebag intended to be in contact with the eyes was measured to be 46.2 ± 0.6°C (115.2 ± 1.1°F) immediately after heating using a thermal camera (ThermoTracer 7102MX, NEC, Japan) on five eyebags, repeated on three occasions at least 24 hours apart. After 5 minutes, the eyebag cooled to a mean surface temperature of 38.5 ± 0.7°C (101.3 ± 1.3°F; Fig. 2). The eyebag was applied with the silk surface in contact with the eyelids of one eye selected at random for 5 minutes immediately after heating. At the same time, a nonheated eyebag (mean surface temperature, 18.1 ± 1.0°C; 64.6 ± 1.8°F) was applied (silk side) to the contralateral eye as a control. Care was taken to ensure that the eyebags did not touch. Timing was kept using a digital stop clock. A second masked researcher conducted the study measurements. Measurements were taken before (baseline), immediately after, and 10 minutes after the removal of the eyebags, and the order of which eye was measured first was randomized. The test room was the same for all subjects and had a temperature of 20 to 22°C (68.0 to 71.6°F), 10 to 20% humidity, and no air circulation/wind.
The surface temperature of the central internal and external upper and lower eyelids (external upper eyelid [EUL]; external lower eyelid [ELL]; internal upper eyelid [IUL]; internal lower eyelid [ILL]) of both eyes was measured using the thermal camera (as above) mounted on a slit lamp biomicroscope,18–20 The external upper and lower eyelids were measured by asking the subject to look down and up; the internal upper and lower eyelids were measured by everting the upper and lower eyelids with a sterile cotton bud. At each location, a thermal image was captured immediately to provide a static temperature reading averaged over a 10 by 10–mm area shown on the camera display. Central, nasal, and temporal markers on the thermal camera display were used to ensure that the same central region of the eyelid surfaces was measured to maintain consistency. Temperature measurements of the external eyelid surfaces were made at baseline, within approximately 10 seconds after eyebag removal (immediately after), and 10 minutes after treatment. The inner eyelid surfaces were measured within approximately 10 seconds of eyelid eversion using sterile cotton buds.
The thickness of the tear film lipid layer and tear film stability were measured in both eyes noninvasively using the TearScope Plus (Keeler Ltd, Windsor, UK) mounted on a slit lamp biomicroscope. An interference image of the lipid layer was produced over the cornea as subjects looked at a distant target (spotlight 6 m away) while blinking naturally. Based on the interference pattern observed, the thickness range of the lipid layer was deduced using a grading scale developed by Guillon, where grade 1, open meshwork, lipid layer 13 to 50 nm thick; 2, closed meshwork, 30 to 50 nm; 3, wave, 50 to 70 nm; 4, amorphous, 80 to 90 nm; 5, colored fringes, 90 to 180 nm; 6, globular, greater than 200 nm.21 Open meshwork, wave, and amorphous patterns represent a normal tear film lipid layer.21 A grid pattern insert was used to assess tear film stability by measuring noninvasive tear film breakup time (NITBUT). Subjects were instructed to blink normally and then keep their eyes open for as long as possible. The NITBUT is defined as the period between the last complete blink and the appearance of a break or distortion in the fine grid pattern.21 The period was measured using a digital stop clock. This was repeated three times, and the values were averaged to give a mean NITBUT value. Measurements were made at baseline, immediately after, and 10 minutes after treatment.
Statistical analysis was performed using SPSS for Microsoft Windows. Differences between control and test eyes for eyelid temperature and NITBUT over time were evaluated by repeated-measures analysis of variance, and, where significant, post hoc analysis was performed using t tests with a Bonferroni correction applied (significance level set at p < 0.05). Differences between control and test eyes for lipid layer thickness were evaluated by the Freidman test, and post hoc analysis, where significant, was performed using Wilcoxon signed rank tests with a Bonferroni correction applied (significance level set at p < 0.05).
A total of 22 healthy subjects (50% men) were assessed and had a mean age of 22.0 ± 2.7 years (range, 18 to 27 years) aged matched by sex.
At baseline, there was no statistically significant difference between control and test eyelid temperature at all locations (EUL, 35.0 ± 1.2°C vs. 35.1 ± 1.1°C, p = 0.514; ELL, 35.0 ± 1.2°C vs. 35.0 ± 1.1°C, p = 0.920; IUL, 35.6 ± 1.1°C vs. 35.6 ± 1.1°C, p = 0.275; ILL, 35.6 ± 1.0°C vs. 35.6 ± 1.0°C, p = 1.000). There was a statistically significant change in temperature over time from baseline in test eyes (F = 46.451, p < 0.001) but not control eyes (F = 0.872, p = 0.426). Immediately after removal of the eyebag, there was a statistically significant increase in eyelid temperature at all locations from baseline in test eyes (F = 20.533, p < 0.001), with mean increases of 2.3 ± 1.2°C (4.1 ± 2.16°F) on EUL, 2.0 ± 1.0°C (3.6 ± 1.8°F) on ELL, 1.4 ± 1.0°C (2.5 ± 1.8°F) on IUL, and 1.3 ± 1.0°C (2.3 ± 1.8°F) on ILL (Fig. 3). The difference in temperature between control eyes and test eyes was statistically significant at all eyelid locations (EUL, ELL, IUL, ILL, p < 0.001).
Ten minutes after application, the increase in eyelid temperature from baseline in test eyes remained statistically significant at all locations (EUL, 1.0 ± 0.7°C, 1.8 ± 1.3°F; ELL, 0.9 ± 0.6°C, 1.6 ± 1.1°F; IUL, 0.7 ± 0.6°C, 1.3 ± 1.1°F; ILL, 0.5 ± 0.6°C, 0.9 ± 1.1°F; F = 14.247, p = 0.001) but had decreased from immediately after removal (Fig. 3). The difference in temperature between control and test eyes was also statistically significant at all eyelid locations (EUL, ELL, IUL, ILL, p < 0.001).
Noninvasive Tear Film Breakup Time
At baseline, there was no statistically significant difference in NITBUT between control and test eyes (9.2 ± 2.4 seconds vs. 9.4 ± 1.8 seconds, p = 0.468). There was a statistically significant increase in NITBUT in test eyes (F = 47.904, p < 0.001), but not control eyes (F = 0.625, p = 0.540), over time from baseline. The difference in NITBUT between control and test eyes immediately and 10 minutes after eyebag removal was also significant (p < 0.001). There was a statistically significant increase in NITBUT from baseline in test eyes (pairwise comparison, p < 0.001) with a mean increase of 4.0 ± 2.3 seconds immediately after removal. Ten minutes after removal, the difference in NITBUT from baseline in test eyes was also statistically significant (difference, 3.6 ± 2.1 seconds; pairwise comparison, p < 0.001). Although the NITBUT reduced slightly from immediate to 10 minutes after eyebag removal (Fig. 4), this was not statistically significant (pairwise comparison, p = 0.322).
Tear Film Lipid Layer Thickness
At baseline, there was no statistically significant difference in lipid layer thickness grade between control and test eyes, with only one subject showing a difference in lipid appearance between the eyes (grade, 2.4 ± 0.7 vs. 2.4 ± 0.8, Z = 0.000, p = 1.000). There was a statistically significant increase in lipid layer thickness in test eyes (χ2 = 35.313, p < 0.001), but not in control eyes (χ2 = 5.200, p = 0.074), over time from baseline. Post hoc analysis confirmed that there was a statistically significant increase in lipid layer thickness from baseline in test eyes (Z = −4.035, p < 0.001), with a mean grade increase of 2.0 ± 0.9 immediately after removal. The difference in lipid layer thickness between control and test eyes was statistically significant (Z = −4.102, p < 0.001) at this stage. Ten minutes after application, the increase in lipid layer thickness from baseline in test eyes remained statistically significant (Z = −3.835, p < 0.001), however, the mean increase was smaller than immediately after removal (1.5 ± 0.9 grades higher than baseline). After 10 minutes, the difference between control and test eyes also remained statistically significant (Z = −3.759, p < 0.001). The distribution of lipid layer thickness grade at baseline, immediately after, and 10 minutes after eyebag application is shown in Fig. 5.
No adverse events were reported during or after the study.
The temperature of test eyes covered with the heated eyebag increased significantly at all measurement locations from baseline immediately after removal. The maximum mean temperature measurement was 37.4 ± 1.6°C (99.3 ± 2.9°F), which occurred on the external upper eyelid immediately after eyebag removal. This was not unexpected given the relatively closer and larger surface area in contact with the eyebag compared with that of the lower external eyelid when the eyes were closed. The temperatures presented here are lower, and the treatment duration was shorter than those reported to cause thermal injury to the eyelid tissue.22 In addition, heat (infrared) radiation is also known to cause cataract23; in an animal model, it has been shown that lens protein changes start to develop at 40°C (104°F) after 2 minutes of direct infrared exposure.24 Eyebag therapy applies heat through the closed lids so the crystalline lens will be protected to some extent. It was not possible in our study to measure the maximum eyelid temperature because the eyebag obscured the view of the eye; it is likely that the eyelids were initially warmer during earlier stages of heating as the eyebag naturally decreased in temperature during the 5-minute treatment protocol (Fig. 1). Thus, the maximum eyelid temperature achieved during heating is unknown, and it is unclear if thermal injury is possible. However, no adverse events were reported during or after the study. Eyelid-warming therapies have also been associated with transient visual blur caused by pressure exerted on the eyelids. Solomon et al.25 found that the corneal polygonal reflex of Fischer-Schweitzer in experimental eyes receiving warm compress therapy combined with gentle eyelid pressure was significantly positively correlated to visual blur and visual acuity decrease. More recently, increases in corneal temperature after eyelid rubbing or massage has been associated with corneal deformation and subsequent visual degradation.26 It is purported that eyelid-warming therapy where heat is applied directly to the eyelids combined with massage may exacerbate this effect.26 Therefore, we did not apply any additional pressure to the eyelids other than to stabilize the device, and our treatment protocol did not require massage. Until the safety of eyelid-warming therapy with devices making contact with the eyelids such as the eyebag is fully established, patients should be instructed carefully as to their use and advised not to rub the eyes after treatment.26 The internal lower eyelid was less warm immediately after eyebag removal, and the internal upper eyelid temperature was also less warm than the external upper eyelid, providing further support for noncomplete heat transfer between external and internal eyelid surfaces.15 This is likely to be caused by blood flow within the eyelid vasculature, which carries heat away via convection as well as the different insulating properties of the ocular tissues.22,27 Although not statistically significant, the small increase in temperature in control eyes immediately after removal is somewhat surprising given that the nonheated eyebag was much cooler than eyelid temperature. It is likely to have been caused by the insulating properties of the nonheated eyebag and/or the creation of a layer of air between the eyelid and eyebag surfaces minimizing heat loss.
Ten minutes after application, the temperature of test eyes remained significantly warmer at all test locations from baseline (F = 20.533, p < 0.001) but lower than immediately after application (F = 24.889, p < 0.001). The heat source had been removed 10 minutes earlier and heat energy dissipated to neighboring tissues and the surrounding air, causing the surfaces to cool. The rate of cooling seemed to be slightly faster for the external eyelid surfaces compared with that for the inner eyelid surfaces in test eyes. This finding could be explained by the larger temperature gradient between the external eyelid surface and air compared with that of the inner eyelid surface and globe. In addition, the inner eyelid surface is insulated by the eyelid tissue from exposure to the surrounding air.
A potential source of error in our measurements could have resulted from everting the upper and lower eyelids to obtain internal eyelid surface temperature; although brief (∼5 seconds), the time taken from everting the eyelids to measurement may have allowed heat loss via convection after exposure to the air. In addition, the time taken to measure one eye may have impacted on the contralateral eye measurement in a similar fashion, but the order of which eye was measured first was randomized, negating any bias. Manipulating the eyelids during eversion may have also caused changes in vascular blood flow and subsequent changes in heat loss.15 In addition, although cotton buds were used to evert the upper and lower eyelids, the heat radiated from the eyes after treatment may have been detected by the examiner so they were no longer masked to the treatment eye. Therefore, bias may have been introduced in our measurements. However, given the nature of eyelid-warming therapies, it is not possible to completely eliminate this limiting factor where measurements are taken by human examiners. Tear film breakup time was measured noninvasively rather than using fluorescein sodium because the latter may disrupt the tear film by artificially increasing the volume of the tear film and the application technique inducing reflex tearing, all of which may affect the measurement.28 In addition, the TearScope uses a cold light source that prevents/minimizes any additional evaporative loss.21
Tear film stability has been demonstrated to correlate significantly with the thickness of the tear film lipid layer, derived from secretions of the meibomian glands.1,29,30 Increasing the tear breakup time and lipid layer thickness may therefore help prevent dry eye symptoms associated with MGD.4,9 In our study, immediately after removal of the heated eyebag, a significant increase in NITBUT from baseline was detected (p < 0.001), which was sustained at 10 minutes after application (p = 0.965). Previous studies on eyelid-warming therapies have also demonstrated significant increases in tear film stability and improved symptoms in MGD patients and normal subjects, although the treatment methodology differed with respect to eyelid-warming device used and the frequency and duration of application.7–9,11,13
Tear film lipid layer thickness also increased significantly in test eyes, and this increase from baseline was sustained at 10 minutes after application, albeit at a slightly smaller thickness than immediately after eyebag removal. Increases in TFLLT after eyelid-warming therapy in healthy subjects and MGD patients have been demonstrated in previous studies, although the treatment methodology differed again with respect to eyelid-warming device used and the frequency and duration of application.7,9–11 Increases in TFLLT after eyelid-warming therapy have also been associated with improved signs and symptoms in patients with MGD and normal subjects.7,9,11,12
Based on the present data, the short-term increases in tear film stability and TFLLT in healthy subjects can be attributed to the heat transferred across the eyelids, consistent with previous studies and supports the use of eyelid-warming therapy for MGD treatment. The eyebag is simple to heat and handle compared with traditional moist warm compress, which are reported to require a methodical and labor-intensive protocol to optimize treatment.15 Furthermore, the eyebag is relatively inexpensive and can be reused, whereas many devices providing alternative sources of heat are commercially unavailable.7–9,11,16 More recently, the LipiFlow system, which combines eyelid heating and massage while minimizing the effect of heat damage to the cornea, has demonstrated short- and long-term improvements in meibomian gland secretion and tear breakup time after a single in-office treatment, including in severe cases of MGD.13 In comparison, the eyebag can be used for home therapy in milder cases or as adjunctive therapy after in-office procedures in severe cases where more intensive treatment is required.26 Although the eyebag requires the eyes to be closed during application, which minimizes exposure to direct heat, further research is required to determine the maximum eyelid temperature induced to evaluate the safety of this eyelid-warming device, with particular focus on the cornea. After 5 minutes of application, the heat transferred from MGDRx EyeBag produces increases in tear film stability and lipid layer thickness that are maintained up to 10 minutes after application. These effects on the tear film have improved the signs and symptoms of MGD in previous studies and support the use of eyelid-warming therapy in MGD. However, these short-term effects were observed in healthy subjects without MGD, who are likely to have lower meibomian secretion melting points and few, if any, obstructed glands. In addition, long-term and frequent treatment may be necessary to produce a clinically significant improvement in signs and symptoms of MGD because of its chronic nature, although short-term effects are encouraging. Future longer treatment duration clinical studies are therefore required to determine the efficacy and safety of this eyelid-warming device in patients with MGD.
Shehzad A. Naroo
Ophthalmic Research Group
School of Life and Health Sciences
Birmingham, B4 7ET
The study received no funding from external sources. The authors acknowledge the EyeBag Company Ltd. (UK) for providing the MGDRx EyeBags at no cost for this study. The authors have no financial or commercial competing conflicts of interest relating to this study.
Received November 27, 2012; accepted September 17, 2013.
1. Bron AJ, Tiffany JM. The contribution of meibomian disease to dry eye. Ocul Surf 2004; 2: 149–65.
2. Foulks GN, Bron AJ. Meibomian gland dysfunction
: a clinical scheme for description, diagnosis, classification, and grading. Ocul Surf 2003; 1: 107–26.
3. Nelson JD, Shimazaki J, Benitez-del-Castillo JM, Craig JP, McCulley JP, Den S, Foulks GN. The international workshop on meibomian gland dysfunction
: report of the definition and classification subcommittee. Invest Ophthalmol Vis Sci 2011; 52: 1930–7.
4. Foulks GN. The correlation between the tear film lipid layer and dry eye disease. Surv Ophthalmol 2007; 52: 369–74.
5. McCulley JP, Shine WE. Meibomian secretions in chronic blepharitis. Adv Exp Med Biol 1998; 438: 319–26.
6. Terada O, Chiba K, Senoo T, Obara Y. [Ocular surface temperature of meibomia gland dysfunction patients and the melting point of meibomian gland secretions]. Nihon Ganka Gakkai Zasshi 2004; 108: 690–3.
7. Mitra M, Menon GJ, Casini A, Hamada S, Adams D, Ricketts C, Fuller ET, Fuller JR. Tear film lipid layer thickness
and ocular comfort after meibomian therapy via latent heat with a novel device in normal subjects. Eye 2005; 19: 657–60.
8. Goto E, Monden Y, Takano Y, Mori A, Shimmura S, Shimazaki J, Tsubota K. Treatment of non-inflamed obstructive meibomian gland dysfunction
by an infrared warm compression device. Br J Ophthalmol 2002; 86: 1403–7.
9. Mori A, Shimazaki J, Shimmura S, Fujishima H, Oguchi Y, Tsubota K. Disposable eyelid-warming device for the treatment of meibomian gland dysfunction
. Jpn J Ophthalmol 2003; 47: 578–86.
10. Olson MC, Korb DR, Greiner JV. Increase in tear film lipid layer thickness
following treatment with warm compresses in patients with meibomian gland dysfunction
. Eye Contact Lens 2003; 29: 96–9.
11. Matsumoto Y, Dogru M, Goto E, Ishida R, Kojima T, Onguchi T, Yagi Y, Shimazaki J, Tsubota K. Efficacy of a new warm moist air device on tear functions of patients with simple meibomian gland dysfunction
. Cornea 2006; 25: 644–50.
12. Ishida R, Matsumoto Y, Onguchi T, Kaido M, Iwamuro K, Kobayashi J, Takano Y, Shimazaki J, Goto E, Dogru M, Tsubota K. Tear film with “Orgahexa EyeMasks” in patients with meibomian gland dysfunction
. Optom Vis Sci 2008; 85: 684–91.
13. Lane SS, Dubiner HB, Epstein RJ, Ernest PH, Greiner JV, Hardten DR, Holland EJ, Lemp MA, McDonald JE 2nd, Silbert DI, Blackie CA, Stevens CA, Bedi R. A new system, the LipiFlow, for the treatment of meibomian gland dysfunction
. Cornea 2012; 31: 396–404.
14. Geerling G, Tauber J, Baudouin C, Goto E, Matsumoto Y, O’Brien T, Rolando M, Tsubota K, Nichols KK. The international workshop on meibomian gland dysfunction
: report of the subcommittee on management and treatment of meibomian gland dysfunction
. Invest Ophthalmol Vis Sci 2011; 52: 2050–64.
15. Blackie CA, Solomon JD, Greiner JV, Holmes M, Korb DR. Inner eyelid surface temperature as a function of warm compress methodology. Optom Vis Sci 2008; 85: 675–83.
16. Mori A, Oguchi Y, Goto E, Nakamori K, Ohtsuki T, Egami F, Shimazaki J, Tsubota K. Efficacy and safety of infrared warming of the eyelids. Cornea 1999; 18: 188–93.
17. Tomlinson A, Bron AJ, Korb DR, Amano S, Paugh JR, Pearce EI, Yee R, Yokoi N, Arita R, Dogru M. The International Workshop on Meibomian Gland Dysfunction
: report of the diagnosis subcommittee. Invest Ophthalmol Vis Sci 2011; 52: 2006–49.
18. Purslow C, Wolffsohn JS, Santodomingo-Rubido J. The effect of contact lens wear on dynamic ocular surface temperature. Cont Lens Anterior Eye 2005; 28: 29–36.
19. Purslow C, Wolffsohn J. The relation between physical properties of the anterior eye and ocular surface temperature. Optom Vis Sci 2007; 84: 197–201.
20. Giraldez MJ, Naroo SA, Resua CG. A preliminary investigation into the relationship between ocular surface temperature and lipid layer thickness. Cont Lens Anterior Eye 2009; 32: 177–80.
21. Guillon JP. Use of the Tearscope Plus and attachments in the routine examination of the marginal dry eye contact lens patient. Adv Exp Med Biol 1998; 438: 859–67.
22. Despa F, Orgill DP, Neuwalder J, Lee RC. The relative thermal stability of tissue macromolecules and cellular structure in burn injury. Burns 2005; 31: 568–77.
23. Lydahl E, Philipson B. Infrared radiation and cataract. I. Epidemiologic investigation of iron- and steel-workers. Acta Ophthalmol (Copenh) 1984; 62: 961–75.
24. Okuno T. Thermal effect of visible light and infra-red radiation (i.r.-A, i.r.-B and i.r.-C) on the eye: a study of infra-red cataract based on a model. Ann Occup Hyg 1994; 38: 351–9.
25. Solomon JD, Case CL, Greiner JV, Blackie CA, Herman JP, Korb DR. Warm compress induced visual degradation and Fischer-Schweitzer polygonal reflex. Optom Vis Sci 2007; 84: 580–7.
26. McMonnies CW, Korb DR, Blackie CA. The role of heat in rubbing and massage-related corneal deformation. Cont Lens Anterior Eye 2012; 35: 148–54.
27. Huang HW, Shih TC, Liauh CT. Predicting effects of blood flow rate and size of vessels in a vasculature on hyperthermia treatments using computer simulation. Biomed Eng Online 2010; 9: 18.
28. Mengher LS, Bron AJ, Tonge SR, Gilbert DJ. Effect of fluorescein instillation on the pre-corneal tear film stability. Curr Eye Res 1985; 4: 9–12.
29. Nichols JJ, Nichols KK, Puent B, Saracino M, Mitchell GL. Evaluation of tear film interference patterns and measures of tear breakup time. Optom Vis Sci 2002; 79: 363–9.
30. Isreb MA, Greiner JV, Korb DR, Glonek T, Mody SS, Finnemore VM, Reddy CV. Correlation of lipid layer thickness measurements with fluorescein tear film breakup time and Schirmer’s test. Eye 2003; 17: 79–83.