Lacrimal hypofunction as a new mechanism of dry eye in visual display terminal users

Abstract

Background: Dry eye has shown a marked increase due to visual display terminal (VDT) use. It remains unclear whether reduced blinking while focusing can have a direct deleterious impact on the lacrimal gland function. To address this issue that potentially affects the life quality, we conducted a large-scale epidemiological study of VDT users and an animal study.

Methodology/principal findings: Cross sectional survey carried out in Japan. A total of 1025 office workers who use VDT were enrolled. The association between VDT work duration and changes in tear film status, precorneal tear stability, lipid layer status and tear secretion were analyzed. For the animal model study, the rat VDT user model, placing rats onto a balance swing in combination with exposure to an evaporative environment was used to analyze lacrimal gland function. There was no positive relationship between VDT working duration and change in tear film stability and lipid layer status. The odds ratio for decrease in Schirmer score, index of tear secretion, were significantly increased with VDT working year (P = 0.012) and time (P = 0.005). The rat VDT user model, showed chronic reduction of tear secretion and was accompanied by an impairment of the lacrimal gland function and morphology. This dysfunction was recovered when rats were moved to resting conditions without the swing.

Conclusions/significance: These data suggest that lacrimal gland hypofunction is associated with VDT use and may be a critical mechanism for VDT-associated dry eye. We believe this to be the first mechanistic link to the pathogenesis of dry eye in office workers.

Figure 1. Rat VDT user model.

(A) Image of rat VDT user model. (B) Schematic representation of daily experimental schedule for rat VDT user model. This series of treatments was repeated for up to 20 days.

Figure 2. Lacrimal function is impaired in rat VDT user model.

(A) Changes in tear secretion during 20 days of swing use. For rats repeating the daily experimental cycle, the Schirmer test was performed on days 1, 5, 10, 15 and 20. Data represent the mean ± SEM for 16 eyes. ** P<0.01, *** P<0.001 versus normal. (B) Circadian variation of the tear production in rats placed in the swing. Changes in the Schirmer score were evaluated on days 9 and 10. All data represent the mean ± SEM for 16 eyes. ** P<0.01 versus initial value. (C) Effect of stressed conditions on the Schirmer score. Changes of tear secretion were measured 10 days after treatment with or with-out swing or dry condition. Data present the mean ± SEM for 16 eyes. * P<0.05, *** P<0.001 versus the normal condition group. (D) Increase in tear secretion by systemic parasympathetic stimulation. Tear fluid secretion was stimulated by subcutaneous injection of pilocarpine hydrochloride on day 10. Increase in tear secretion was calculated by subtraction of the Schirmer value from before pilocarpine injection. Data represent the mean ± SEM for 10 to 23 eyes. * P<0.05, ** P<0.01 versus the normal. (E) Changes in protein secretion capacity in LG by parasympathetic stimulation. Changes in the protein release after stimulation by carbachol (Cch) using isolated LG on day 10. The protein secretion rate was calculated as a percentage of before Cch stimulation. Data represent the mean ± SEM for 16 LG. * P<0.05 versus the normal with Cch stimulation.

Figure 3. Rat VDT user model causes alterations in lacrimal gland morphology.

(A) Left: H & E staining. Left center: Toluidine blue staining. Right center and right: Electron microscopic analysis of acinar cells. Images showing expanded aciner cells accompanied by accumulated enlarged secretory vesicle in the cytoplasm (center), decresed endoplasmic reticulum and increase in the nuclei with dark neucleoplasm (Right) of LG on day 10. Scale bars: Left  = 200 µm; Left and right center  = 10 µm; Right  = 4 µm. (B) Changes in total cell number of LG. Changes of LG cell number were measured 10 days after treatment with or without swing or dry condition. Quantification of LG number was calculated by deoxyribonucleic acid content of the LG. Data represent the mean ± SEM for 8 to 16 eyes. * P<0.05 versus the without swing and dry condition. (C) Correlation between tear production and LG cell number. Pearsons correlation coefficient testing was r = 0.53 (P = 0.034, n = 16).

Figure 4. Recovery of tear secretion with long-term rest without swing activity.

(A) Effect of shortening the time spent on the swing. Ratios to initial value were calculated. Data represent the mean ± SEM for 16 eyes. * P<0.05, ** P<0.01 versus 0 hour riding swing group. (B) Effects of changing resting patterns in tear secretion. Ratios to initial value were calculated. Data represent the mean ± SEM for 16 eyes. (C) Effect of stimulation on the lacrimal function for the two groups, with and without pilocarpine injection. Ratios to initial values were calculated. Data represent the mean ± SEM for 16 eyes. (D) Effect of extended rest period without the swing. Data represent the mean ± SEM for 8 to 16 eyes. * P<0.05 versus day 10. (E) Effect of extended rest period without the swing on tear secretion in response to parasympathetic stimulation (Left) and recovery of protein released from isolated LG (Right). Data represent the mean ± SEM for 8 to 16 eyes. * P<0.05 versus the normal.